nrNrlrn ilr)rff rfrnn, rfrnil
THE CEl{TER TOR A6RIGUTTURAL ECOilOUIC RESEARGH
Reprint series No. 8202
CAPITAL INTENSITY AND PRODUCT COMPOS]TION IN TT1E KIBBUTZ AND THE MOSHAV IN ISRAEL by
Yoav Kislev & Shlomir Farbstein
Rehovoth, lsrael, P.O.B. l2 12 .'r.n ,n1f1n") CAPITAL INTENSITY AND PRODUCT COMPOSITION
IN T}IE KIBBUTZ AND THE MOSHAV IN ISRAEL
Yoav Kislev and Shlomit Farbstein
Abstract
Differential capital intensity in the agricultural sectors in Isr:ae1 is realjzed in the employment of different technologies and, mostly, in different product composition. Between 1966 and 1-973 the scale of use of the different technologies declined as the family farn adopted large scale methods in several lines of production but, due to widening intensity gap, the contribution of the techological differences to differences in intensity grew. Theoretically, in analogy to the case of trade between nations, the returns to factors in sectors producing to the same market should be identical. A linear programing model was formulated and it was found that major factor (shadow) prices have been equalized in the economic environment of Israeli agriculture despite the prevalent government intervention and arbitrary administrative allocation decision making. January, 1982
CAPITAL INTENSITY AND PR.ODUCT COMPOSITION
IN THE KIBBUTZ AND THE MOSHAV IN ISMEL
J Yoav Kislev and Shlomit Farbstein
Israel's agriculture is comparatively capital intensive, perhaps even over-capi talized, but not all farm units operate at the same 1eve1 of intensiry. Significant sectoral differences exist: kibbutzim (single: kibbutz, a collective village) have higher capital-labor ratios than the moshavim (single: moshav, a cooperative village of family farms) and the traditional Arab sector operates at even lower 1evels of capital intensity.
The causes of these differences can be both external and internal; that is, associated both with supply and with demand factors: The capital market in Israel is dominated by the governrnent and long term credit is directed to ,lesired 1ines of production i-n the economy at large and in agriculture in particular. Short term credit is allocated by market mechanism. Jewish agricul ture has had belEer access to sources of long term capiCal than Arab farming. Both cooperative sub-sectors are better organized politically, than the private sector, and can better affect the government allocating process in their favor. The structure of Ehe moshav is an advantage in mobilrzLng capital as Ehe risk of defaulE of a cooperative is lower than that of its individual members singly. The kibbutzim, both large and djversified, may be even in a better position Ehan the moshavim. The kibbutzim share with the moshavim the ideology of self-labor -- not to exploit the work of ot.hers -- but adhere to it much more; private farmers -- Jewish and Arab -- have no inhilitions on this subject. -2
These differences are reflected in production in two ways: (a) the capiral intensive sectors employ different techniques in 'the production of
the same products -- kibburzim operate expensive milking parlors, dairy operators jn Ehe moshav use portable milking machines; (b) the sectors sper:ial ize jn differ:ent products -- cotton (machine picked) is grown in the kibbutzim, the moshavim produce vegetables and cut flowers, and almosI all srrawberry production (frighly labor intensive) is concentrated in Arab vi i 1ages.
Jn the study reported we attempted to estimate capital-labor ratios
for the kibbutzim and the moshavim and to break up the differences in this ratio jnto those due to different technologies and those due to different product composition. Data limitation prevented the inclusion of private farming in the analysis. If capital and labor were the only factors of production, then, in analogy to the L\ro-country model of internalional
Erade, Ehe returns to the factors in the sectors would have been equalized. However, in such a simple situation, each sector would have produced at most two products. Actual production is governed by a multitude of factors an)ong them government intervention in credit, land and water allotment and a partial production quota system. It has often been argued that exogenous resource allocation and structural rigidities caused signj ficant inefficiencies in agriculture -- differenti,al contribution of inputs (Sadan, 1963). To test whether, under the prevailing circr;mstances, the returns to the major factors were equalized in Israeli agricrrlLui-*, we constructed and ran a linear prograurning model reflecting the regional, sectoral and institutional
structure of agri crrl ture. -3
The question of capital intensity is an important part of the set of issues associ.ated with employment and "appropriate technology", discussed in developmerlt economies (Ranis, 1979). The reader will also find many poi.ncs of similarity between the discussion in this paper and the review of employment impl:ications of industriaLization by David Morawetz (1974),
Definitions and Data
Capital and labor inpuL are measured in each activity, each farm enterprise, separately and aggregated to form the overall capital-1abor ratio by sector. Since capital assets differ between enterprises, the appropri.ate measure for aggregation is the flow of services. It also happens to be the only measure available. The Ministry of Agriculture prepares occasionally a complete set of tables of costs and returns by line of production for all agricultural activities. The two most recent tables available for this study were the 1966 and the 1973 sers and they rr'r-,re the sources f or our data.
For some products, Ehe input-output coefficients in the tables of the Mini.stry of Agriculture, are the same in the kibbutz and in the moshav; for others, the tables list two different sets of coefficient, a set for each secfor. For prodrrcts of rhe first c1ass, we assumed that (within the wage-renLal spectrum in Israel) there exists only one technology that can be employed economically in Israel. The production activities for which a single technology vlas specified were termed identical-technology activities. ln the other cases, the tables of the Ministry specify Ewo sets of input- ollLplrt coefficients, reflecring different production practices in the kibhrrtzim and the moshavim. These lines \,nere classified as multi-technology a(r;\riLi,.,. Table I exemplifies the farm enterprise data of Ehe Ministry of Agri- culture and the catculation of capital-labor raEioq for dairy (a multi- technr:1ogy activity). Capital-labor ratio in the kibbutz dairy is higher by 1.59 Ol.5l5l .4) than in the moshav. Scale of operation dif fer widely: 20-60 cows in rlre family farm and 300-500 in the kibbutz. Table 1 indicates economies of scale -- in labor input and milk yield. Indeed, the industry is steadi1y shifting to the kibbutz (more on this industry can be found in
Ki s lev, Mei se I s and Amir, forthcoming) .
The strrciy clefined 39 activiries, a rhird were multi-technology activities
Appendix A lists the activities and their 1966 capital-labor raEios. Several lines of production, particularly field crops' are not carried on the sma1l family plots in the moshav. Accumulated surplus land and land located far away from the village is cultivated by the moshav cooperative on a large scale and the technology is the same as in the kibbutz. These wilt mostly be Fietd crops and, in some cases, orchards. In the analysis to fo11ow we shalr. ,;hereforc, presenL the data both for Ehe fami1y farms in the moshav anrl for the agricultural production activities as the moshav as a who1e, j,'.: lusive ofthe cooperative agricultural enterprises (service enterpr:ises i ; Lnoshav cooperat ive are not included) .
(),"rrnparaL i ve'._ Cilpi tal lntensLty
'l :rirle 2 strrnma rizes thc, clata. Overall capital intensity in the kibbutz was, in 1973, 1.64 [imes higher than in the moshav at large and 1.86 than in tlre family farui (part C of the table). The gap between the kibbrrtz and t lre rrc,shav inc reased by 6-7 percent between 1966 and 1973 . An index of ttre cost of capira.l services in agriculture is not available. Therefore, to compare jntra-secIora1 capital iniensity between the years, we inflated -5
Table L: Input-Output Coefficients in Milk Producqlon in 1973-- (annual data per cow , 0.9-.female calf , an{-the associated forage producing area')
MOSHAV KIBBUTZ
Milk Yjeld (kg) 5,280 6,500 l,abor (daYs) t7 .25 12.5 Capital Recovery in Structures and Equipment (IL) 375 s00 Capital Recovery in Forage and Irrigation Equipment (IL) 287 376 Interest on Livestock Capital (IL) 328 328 Total Capital OutlaYd (IL) 990 7 ,744 Capital Labor Ratio (IL/daY) 57 .4 91 .5
Source : Ministry of Agriculture, 1973 -6
and C ital Intensit Table 2 : Scale of Lines of Production
Moshav Kibbutz Fami ly Farma (incl . coop) 1966 I 1973 1966 | L97s 1966 | 7s73
I A. Scaleb I I ll soo sss 'l'ot al ( I ()" I r,) c 354 l+zz l++s l szo l I Identi-r cal te chnology activities( eo ). 67 lte loq lto lss los MuItj -techno logyactivities (%) 33 lzq lso lzq I +s lst
c I B. Uapital - labor ratio (IL/daY) I I ll ln identical technology activities so los lza l+o lzz lsr +a In mul ti-technologY activities 47 lzs lss l+a l:a I Overa 1 I 4s lot lsz lqz lzs lso
C. Comparative Capital Intensityd I lll I l.> 'tt In identical technology activities Ir.zo Ir.os lz.to I In nulti-technologY activities It.z+ It.so lr.z+ lr.so I Ove ra1 1 Ir.ss Ir.o+ li.zs li.ao
Ilotesj tL. Fanily farm--moshav exlusive of cooperative enterPrises;
b. Scale -- aggregate inpuE of labor and capital with the linear programming stradow prices (reporrecl below) as weights;
\)netary values -- in 197'3 prices;
d. Comparative capital intensity -- between kibbutz and moshav. 1966 monerary value by 1.90 which is the 1973 index of prices of machinery, parts, equipnrent and constrrrction (tg0e = 1.00). Capiral-labor ratios in the kibburz jncreased between the years in the table,by our estimate' by 36 percent and in the moshav by 32 arrd 29 percent, for the village and the f anri ly f arm t:espectively.
Scale of a farm activity was defined in Part A of Table 2 as the aggregate valrre of capita1 and labor input weighted by the t:orresponding sladow prices, as r:alculated in the I inear programming model tr.l he described below. ll'he strare of the multi-technology activities declined in a1l sectors
[;eLween the years. To ana11-ze fhe f ;rc:lors af f ecting the in<'reased capi ta1 ,l966 i.nf ensit.r- beLween and 1gi3, w:--c Lo o.l.1 k1 for capirai labor ratio in the two years, respectively, and ,t, ,l are the 1966 and 1973 shares of the identical technology activities in part A of Table 2. Then
o o,o k =w K. (1-wo) ko I m (1) .t 1,1 l( =w h- ( 1-r1 ) k1 I m wiic re the index 1 ni stanis for iientical and multi-technology activities
( L ire sec l- or index 1S ornitted). The symbol A indicates differences between the years, as in Ak = t1 - ko . llow,
(wo + (u: + (l-wo (to * (D kl = Aw) Ak.)lmm - Aw) lk )
Itertt't'
(3) Ak = Awkl + wo^k. - rwkl + (t-wo)rt< 1 1m
The expr:c,ssions including Aw in (3) indicate changes in weights of ac: t i vi ty type; those witlr lik chan ge s 1n intens ity. These changes in i ntensity can al so ri'sl.: 1n shares of activiti es within the groups -- icienLic::i1 rtni actrv].tles. -B
The data of Table 2 were utilized to calculate the components of equari<-rn (3) in Table 3" In the kibbutzim 87 percenr of the growth in capital labor ratio occurred in identical lechnology activities, 13 per- c:ent -- jn the other type of technology. In the moshavim, the decline of the we j glrt of Lhe rnr.rlti-technology activities, overcame the increased capital in[ensity of this type. These developments are tlre result of rhe moshav adopting the technology of the large farm in more lines of producrj.on (particularly orchards) than the kibbuEz developing and adoptirrg new large-scale methods (was done in some vegetables for processing and exports). The scope nf the different technologies was narrowed.
Hor,rever, the comparative capital intensity of the multi-technology activities increased overtime more than the intensity of the other activities. And, indeed, the share of the multi*technology activities in explaining the capital intensity gap between the kibbutz and the moshav increased. This i s detai 1c,d in Table 4.
The first entries in Table 4 should be read as fo11ows. In the first li re (z) rhe calculation is limited to the multi-technology activities. In tl ;rie activitie,s, and with actual pr<.rdr-rct composition, 16.9 percent of the c;:pira.l-1abor ratio in the kibbucz was due o in 1966, to technology differences. Jn other words, if technology in the kibbutz were in these ar:tivities the same as in the ntoshav, capi f a.l -labor raLios would have been lower by 16.9 percent. There were, of corrrse, no technology differences in identical tecltnology acfivities. Spreading the calculated difference' z, over: al1 j.on f arm ac: tiviries, we get (v) that 1+.7 percent of the capital labor rat in fhe kibbutz (compared to the moshav) were due, in 1966, to technology -9
FI1 3r( A,OP. rdi..!:r.EqJ.El ^(r,^d oF.olJ(D6 Ft lo I r,. tf loq F.. lo E I<5 lo o !D H o !O @C ts o OOOUJ o(rto O "-I O I.J o\ori o o 6 o. p P 0i F o p rt L o @cD P 10 in capital-Labor Ratlos Table 4 : The components of thg Differences of the Kihbut z and the Moshav (percent) ' 19 66 L973 (1.) Prciportion of- capital-labor ratio in the kjhbrrtz stemming from clifferent technologies Irt mttltj-technology activities (z) 16 .9 35 .0 In total farm oPeration (v) 4.7 8.9 (2) Shitr:e of dj fference in capital-1abor ratio explainecl by clifferent technologies t5. t Moshav ( j nc I rrdi-ng cooperative farm) (w) 1.3.3 FamilY Farms (in the moshav) (w') It.2 19 .0 t3) Share of difference in capital-1abor ratio expli:jned bY Product comPosition Moshav (including cooperative farm) (100-w) 86.7 76.3 lramily farm (100-w') 88. 8 fl1 .0 _z__-1_z_De fi.ni ti on s : 12 z = I -1: activitjes; z Ave::age capital*labor ratio in kibbutzim in all multi-technology 1 capital-lahor ratio in kitrbutzim if the moshav technology would have been ')- used in these activities; '1 - v2 ,l V, 0veral l caPi tal -.latror ratio in the ki bbutz.; I Over:all caPital-labor ratio in the kibl-lut.z with the moshav tecltnology L 'r'v2 \^7 = I the mo-shav) ,l ((iapi tal-1abor ratjo in the kibbutz) /(capital-1abor ratio in v (t:'ap:tal -1abot ratio in the moshav). 2 r/ - 11 'l'6e second and third groups of entries in Table 4 present the partition of the sources of the difference in capital-Iabor ratio in the kibbutz and the moshav between the technology effect and the product composition effect. By tlre r:alcrrlatjon jn the tab1e, in 1966, 13.5 percent of the j.ntensity difference were due to techltology differences; 86.7--to product composition. The share of technology in accounting for the differences almost doubled between 1966 and 1g73, despite a reduction in the scale of the multi-technology activitjes be tween tltese yeal:s . Ili.scussion The estj.mates and the calculations presented raise a host of problems. Following ar:e cornments on some of these problems. a. Approximately 75 percent of the labor force in the agricultural activities of the compared sector-s is engaged in identical-technology activities. Why did different technologies evolve in less Ehan half of the linesof production and nut in ai I of tilem? and why did the share of multi-technology activities decline berween 1966 and 1973 (fable 3) while their explanation in intensiry djfferences increase (Tab1e 4)? It seems that wo processes operate here. 0n the one trand, Israeli agricrrlture is undergoing a gradual process of adoption (mostly from overseas) and Lo a slighter extent creation of seParate technologies for the large and capifal intensive farm and for the smal.l one. Both sectors had employed identical Eechnooogies virtually in all 1i.nes of producrion up Co World War II, diversification began after it. A good exantl)le i s the dai ry enterpri.se were ttre f irst porf able milking macli jnes were jnrroduced in kihbutzim in the late 1940's. tlp to this date all milking was done by hand and strucrtures were basically identical. Today, 1'2 m()slravinr r.rsr- (.()mparatively sirnpl e milking equipnlerll, wlri 1e kibbutzim operate cost. iy arrd sopiristicated milking parlors oI carrousel s. When capi.tal-1abor jversity, rat j.o \ras 1o\^/, in both sectors, there was not mtrch roottt tor d as the ratio expands different technologies emerge' Ar:other f actor operating in the same di rection is associ;tl r',1 wi t lr the dynamics of agriculture. Kibbutzim ate compar:atively more innovatjve, tlley buy first the capital assets embodying new technology and are, flterefore, (f bei ng observed as more capital intensive at any point in time or: f rrr:tlter discussion of innovations in Israeli agriculture see Kislev and Shc"hori- tJachrach , 191 3,and literature cited there). On the other hand, increased capital intensity enabled the moshav to mechanizeits operation even at Ehe fanrily farm level -- the share of multi- technology activities declined. speciat1zai-ion is not complete; almost all products a1'e produced in both sectors. 'Iheory would predict a deviding capital*labor ratio with al i ihe activities of higher ratios in the kibbutz and those of lower r:atios actual , r; the moshav. Several explanations could be forwarded for the .cviation from the theoretical conclusion. ll.) Moshavim grow some kibbutz-technology clrops in their cooperative enterpri ses; (?) Lahor and caPi.tal are not the only factors and cottstr:aints; 1and, water protlrrction quota also affect the distribution of activjties' itttra - t5l 'l'he sectols thelt-selves are not homogeneous; sectori al averages covel' sector rljffer:ences in capital intensity' 13 (4) l,ltlr<-rt'irrprrt ls rrol lrorirogcnr:ons; some moshavim enlllloy Itir:ed labt.ll'widely, ollrcr':r rlrnrl. Itt llrr: killllrrtzim, 1oo, latlor is,not honrr:geneorrs, sonlc workers iue youllgtrl(ii's iil.td vo.l riltteers*'.rnostly visjtors from othef coUr)tries witlr r)o exl)cricrtce itt ;rgriculttrre. Ileing unskilled, the alternat ive cost o1' t.he lalror of tlrese people could be rather low even if employed on a cirl)il irl irltensive farm. It could be that without this labor force, the kihhrrlzirn would not have produced fruits and citruses, for example, and specializatjon would have been much mr:re distincr. The core of the moclern international economic theory is the assertion that, with free trade, under well-defined assumpcions, returns to capital and labor will he idenrical in the trading countries. In analogy, returns wil l be equalized in tl're lwo sectors selling to the same market, in our case. In the next section we test this proposition. Factor Remuneral ion The two sectors, the kibbutz and the moshav, produce to tlre same market and facert.herefore, identical product prices. If t.hey produce the same pr:r-rtlrrcLs, arrti rrse the sarne producEion f unction, then *- by Heckscher-0hl in llrr.rrreur (S.rnrrrr-'lsr.rrt, 1949) *- remlrnerations of f actors of productions should r',, i,l,,,rilal iu t"he Iwo secfors, Most labor in agriculture in Israel is sell eruployed and virtrral ly all capital assets are owned by the Llsers; factor rewards atre, tlrerefore, not observable. To overcome this difficulty, w€ constructed a linear progranrning model of the two sectors and compared shadow prices. Had the kibbutz and the moshav been constrained only by labor and capital, then, given that both produce at leasl some of every product (in fact each sector would therr 14 lrlve produced at most two producrs), capiEal and laborrs shadow prices in a linear programming model would have heen idencical in the sectors, as is jrrrmediately evident from consideration of the dual problem. In such a s i Lrration, a linear programming representation of the tl^/o sector economy canr)ot add insight beyond the Heckscher-Ohlin formulation; the t\4ro the()retical structures are formally identical. However' oEher factors also affect agricultural production in Israel, they will be discussed irelow, andthe question that we are asking is whether under the circumstances prevailing in Israel, equalLzatioa of the basic factor prices sti11 ho1ds. liine ar . Programning The programning model was both regional and sectoral in nature. It was composed of 5 geo-climate regions, with a moshav and a kibbutz sector within each region. Land, irrigated, dry and pesture, is given as a set of constraints to each sector in each region in accordance with the institutional alloc:;rtion of these factors in Israel. 'Ihe regional constraints are water alltollts--annual amounts, peak load amounts, and water for fish ponds. It was a,rrimed that water can move intersectorally within a region, but not between r ,.: regions. Labor and capital were defined as sectoral constraint and were pc-rmitted to shift between the regions. Total production was limited--to reflect market limitations--to not more than 110 percent of actual production in 1973 in eacft sector. A11 in all,208 constrai.nts and 546 activities were formulated in the model. In general the model's program was quite consistent with actual rli stribrrtion of activities, by sector md region. An inportant exception was da.i ry. in 1973, over 13 percent of value added j.n rnoshavim carne from 15 dairy production; the progranming model suggested elimination of this activity and transfer atr1 the production to the kibbutz sector (see the discussion of rhis enterprise in the section Definitions and Dara). on this line of production, the findings of the model conform to actual. developments -- number the of family farms with dairy enterprises declined over the last t\^/o decades from 20,000 to 1500. A syrunetric finding was that most of the cotton production will shift from the kibbutz to the moshav, to land freed from forrage growing. Cotton, that in 1973 contributed only 2.14 percent of the value added in the moshav, will contribute 20 percent, if the program prevails. Such changes requi.re transfer of land now cultivated separately by the members of moshavim to the cooperative farm; internal transfers are not unconmon, though they raise difficulties that are outside our current area of interest. Table 5 lists distribution of value added by capital-1abor ratio in the activities. The table shows the larger concentration of moshavim in the less capital intensive activities. The discrepancies between the actual and the prograrnmed distributions in the rine of s1 - 100 rL/day are due mostly to the progratnmed change in the distribution of dairy and cotton. Tlhe only factor, apart fron labor and capitar, that was effectiveJy constrained in all 5 regions was dry land. The shadow prices jn the program (in IL) were: Unit Kibbutz Moshav Dry land hectar I 050 r070 I.abor day 44 .3 41.8 Capi t atr 1L in service 0.88 0 .87 16 Table 5 : Value Added by Capi-ta1-Labor Ratio (percent,accumulated) tl ap i t a1 1 qlqq_rqtl o__]Sllllay] 0-50 lrt 15 l+s l+s 5r - 100 loa B7 lsz lta t0i - i50 lg: 96 lgo lgt 151 - 215 I roo 100 I roo I roo Tl.rere are two points to note with regard to these results: (a) The returns to capital are lower than the cost of the service. There are tlso possible explanations to this discrepancy. Partly, it reflects a few machinery items that were included among the variable costs in calculating the net returns as the coefficients of the objective function. Those coefficients were, as a result, biased dovmwards. :ihe other factor is over-capitalization due to subsidization of capital. The subsidization comes mostly through nominal rates of interest lower than che rate of inflation and, in this way, the real rate to.the farmer is negative.. In our data, the inflationary subsidy is ignored and the cost of the service is, therefore, measured in [erms of alternative costs to the national economy, not to the farmer. It is the appropriate definition of cost to measure the quantity of capital services. (b) The shadow prices are very close, the wage to rental. ratio in tlre kibbutz Ls 4.8 percent higher lhan in the moshav. 17 Concluding Remarks The similarity of the shadow prices in the last section can be inter- preted, continuing the line of the discussion throughout the PaPerr to indicate that despite the multitude of factors affecting the structure of agriculture in Israel, the administrative intervention and the non-narket allocation of important inputs and production quota, the Heckscher-Oh1in price equalization mechanism operates well with respect to the two major factors of production -- labor and capital. This conclusion does not. support the assertion that exogenous allocation of factors resulted in Israel in inefficient uEilization. On the contrary, our findings suggest that, given time, and if production is 1et to follow comparative advantage positions, factors can be optimally utilized. Thus, our conclusion is different from Sadanrs (1963) who found wide differences in factor remuneration (1and and labor) in the kibbutzim and the moshavim. Sadan employed a different analytical approach -- he estimated production functions statlstically -- and sometimes results may depend on the tools of the analysis. Also, Sadan worked with data from the 1950ts; Israel of the late 60rs and early 70ts was much more markeE oriented than the state in its first decade. Whatever the reasons for the divergencies in our conclusions, the issue is complex and deserving of further sEudy. Note, however, that even if per unit returns to factors, are equalized by specializatiot in production according to comparative advantages, income depends on capital labor ratio. Thus, if implicit wage rates and returns to capital are identical, income of an agricultural worker in the kibbutz was, in 1913, 64 percent higher than in the moshav. Since in both sectors' farming is only a part time employment (t t irl: actual over:a11 inconre dispari ty i s not necessarily identical to the djfIererrce in capital labor*raijo i.n agr:i,culture; though if equilibrium prt:vails throughout, tire two wjll be the same. Srrictly, the lleckscher-Ohlin cheorem is formulated for a world in r+1:i.ii factors cannot move between eountries -- or between sectors, in our case' An alternative interpretation of our findings will be that, since i:r"rrh l;illor and capital can move between each sector and rhe re$[ of tl)e ecorl(illly, f actors are employed in the f arm enterprises of the ser.,tors ,p to eritra I ity with economy r,riCe alternative. L{hi1e by the f irst i nterpre- fation, the comparative advantages of the sectors are mainly due to cifferential factor endowrnent, by the alternative interpretation, the comparative advantages stem from other sources -- perhaps the different economic and social institutional structures of the sectors -- and factor endor,nnents merely reflect these differences. Rearity evidently lies in h*ttseen the alternatives; factor mobility is possible but it is both s10w i * .,'Jjust ar:d af f ected by administraEional and ideological non-market -i'r-:es. Product mobility then substitutes for facfor f10ws. -19 REFERENCES Farbstein, Shlomit (1981), Capital IntensiEy and Product Composition in the Kibbutz and the Moshav, Master Thesis, Hebrew university, Rehovot, Israel (Hebrew). Kislev, Yoav and Nira Shchori-Bachrach (1973), "The Process of an Innovation Cycle," American Journal of Agricultural Economics, 55: 28-31. Kislev, Yoav, Michal Meisels and Shmuel Amir (forthcoming) "The Dairy rndustry of lsraelr" i, Barry L. Nestel, Animal Prodrrction systems, Elesevier. Morawetz, David (1974), "Employment Implications of Industrialization in Developing Countriesr" Economie Journal, 84: 491-542. Ranis, Gustav (1979), "Appropriate Technology: Obstacles and opportunities," In Samuel M. Rosenblatt (ed.) Technology and Economic Development: A Realistic Prospective, Boulder, Colorado, Westview Press. Sadan, Ezra (1963), "Cooperative Settlements in Israel: Problems of Resource Ailocationr" Journal of Farm Economics, 45: 547-557. Samuelson, Paul A. (1949), "International Factor-Price Equalization Once Again," Economic Journal 59 : 181-97, reprinted ed. J.E. Stiglitz, The collected scientific Papers of PauI A. samuelson, Vo1. 2, cambridge, MIT Press - Ch. 68. -20 :t:r!.!iigiE! d-*i:rew University, Rehovot, IsraeL. We are indebted to our colleagues f ii .i;rn:nents and suggestions. The preparation of this report was assisted i;,; a grant from the United-States Israel Agricultural Research and Develop- m*.:-rL Fund (BARI) . 21 Appendix A - Capital Labor Ratios in 1966 Kibbutz Moshav Identical Technology Activities 1. Fruit bearing orchards* 23.1 26.4 2. Yourg orchards* 13. 5 15. 1 3. Garden vegetables* 8.8 4.8 4. Pr:ocessing vegetables (excluding tomatoes)* 27 .0 8.6 5. Cut-flowers* 15.5 10.2 6. Cotton--irrigated 45 .5 7 . Imi gated sunmer grain 55.0 8. Irrigated winter grain 721.7 9. Winter grain--non-irrigated 76.0 10. Winter legumes 63 .3 1"1. Sunflowers 26,7 12. Sorghum 60. 0 13. Cotton--non-irrigated 44,0 14. Sugar beet--non-irrigated s2.0 15. Water melons--non-irrigated 5.6 5.6 16. Melons--non-irrigated 2.t 2.7 17. Fish ponds 45. 5 45. s I8. Sheep--1ocal breed L2.9 1,2.9 9. Sheep--marino breed 35.2 35.2 24. Bees 16.7 76.7 2L Poultry--eggs for hetcheries (light) 18.1 18. 1 22. Poultry-*eggs for hetcheries (heavy) 19 .4 19 .4 23. Turkey--eggs for hetcheries 12 .8 12.8 24. Broilers 22.7 22.7 25. Turkey--meat 37 .0 37 .0 26, Layers - meat 17 .5 77 .5 Average for identical technology activities 26.38 11.35 22 Kibbut Z Moshav :::l.i i "_r-,..:Bg&gI Activit ,i;Lp I e s t9.2 L7 .6 21 .7 19. s llt: ;.i. r'$ .; ;l lt*;rclre S 20.3 ';t. 27 .9 ,/l.1! rj c*t $ 33.2 25 ,t: r, i u:ns 29.5 .7 37 .8 30. 3 i) t'un s; l'\:iriit I ()E S 1-Or processing 51.9 6.2 ,.lr'*l.rlttl nuts 31.2 18.3 1) 7 . ;{ug;rr:-}:ee t- - i rri gated 28 "3 . ilir i ::y 29.0 24 .4 . iieef c:ittie 44 .4 39.0 :lheal, for mi lk t2.5 12 .8 , iroilLt]"y--eggs 18.4 L2.7 ji,verage multi -technology activities 24 .47 20.64 ijvr:.;1 j aveiage 25.73 14.79 i\l: .:-": : Aiti..rir-ies marked with an asterik (*) are aggregates and may therefore have j:fi:e r*nt ::atios even if the basic activities ale identical-technology j !-ii* ': il,il t " ,. r i',,,'t t.l*s 6 - tr4 are not found on fzurrily farms in moshavim; they are .-'ri'Iir:d i.ry the cooperative enterprises (see Table 2 in the text for average : ;; p1. t. ;,r-1. - l atror ratio in the moshav sector inclusive of cooperative enterplises) ,Ll ve r'* ge s we ::e calculated as r ; ,x l, /Lx.K- tlll whel.e a j.s the average, x. .lumber of units of activity in sector (see, for r:.raa',ptrr, cire def inition of a unit in dairy in Table t ), Li, K' labor in Aays per annum and capital services in IL per annum; t.he irtdex runs over a l 1 acrivities in grouP. JOURNAL OFCOMPARATTVE ECONOMTCS 8. 54-73 (1984) Cooperation in the Moshavl Nnvn HeRuvI aNo Yoev Krslrv Faculty of Agriculture, H ebrew University, Rehovot, I srael Received April 21, 1982; revised July 1983 Haruvi, Nava, and Kislev, Yoav-Cooperation in the Moshav The moshav is a cooperative village in which members own and operate their farms individually. The moshav supervises cooperative marketing of most of the farms' products. But its most important function is as financial intermediary since its members are handicapped when acting by themselves in the credit market' Despite occasional intemal conllicts about the cooperative marketing, in the long run the moshav enhances capital accumulation and income. This paper reports an empirical study ofcooperative marketing, credit, and capital in the moshav. A dynamic model is developed and estimates are presented. J. Comp. Econ., March 1984, 8(l), pp' 54-73. Faculty of Agriculture, Hebrew University, Rehovot, Israel. Journal of Economic Literature Classification Numbers: 052, 710. 1. INTRODUCTION AND SUMMARY The moshav is a cooperative village, typically consisting of 40-80 households that operate their own farms. Only members operate farms in the moshav village and, by being members, they participate in all aspects of collective action.2 Despite common structural traits, moshavim diffet widely in degree and nature of cooperation. Some maintain joint cash management, central planning and direction, and strong public services. Others are loosely organized ' This study is an extension of part of the doctoral dissertation of the first author (Haruvi, 1980). We are indebted to many colleagues, in particular to Claudia Dodge, Joel Guttman, Yakir Plessner, Ezra Sadan, and Dov Weintraub for helpful discussions. Comments and suggestions offered by Pinhas Zusman who read an earlier draft ofthe paper and by two anonymous referees improved the present version significantly. The remaining erron and shortcomings are oun. The study was supported by a gant from the United States-Israel (Binational) Agricultural Research and Development Fund (BARD). 2 The identity of the cooperative and the village is maintained in practice. However, an attempt to incorporate explicitly this identity into the law in Israel failed when a strong fraction in the moshavim opposed what they saw as a potential breach of members' individual freedom. Par- ticularly dangerous, in their view, was the possibility that a member expelled from the cooperative will have to leave his home and livelihood in the village. 0 r47-s967184 $3.00 Copyright O 1984 by Academic Pm, Inc. All rights of reproduction in any fom resryed. COOPERATION IN THE MOSHAV 55 farm communities in which families operate individually farm and nonfarm activities. Histonc circumstances made the moshav the most prevalent form of socio- economic organization in Israeli agriculture. In its early years Israel was flooded with a large number of immigrating refugees who were incorporated into a developing economy aided by outside capital, mostly channeled through public agencies. The moshav, with its egalitarian principles, family operation, mutual aid, and scale economies in services, was deemed the most suitable institutional form for wide agricultural settlement because it prevented ac- cumulation of national resources-land and capital-in the hands of a few individuals. Most moshavim were established with public support and all have been aided significantly by the government. This paper is a report on a study of financial cooperation-the major mode of cooperation in the moshav. The basis for financial cooperation is joint marketing of farm products. The share of the value of output marketed jointly is, therefore, taken in the paper as a measure of the degree of co- operation. The operation of the moshav, the factors determining the degree of joint marketing, and the conflict between the individual and the group are discussed in the first part of the paper. By improving the position of the moshav in the credit market, ftnancial cooperation enhances capital intensity; capital intensity raises income and the value of the member's time and, in turn, strengthens cooperation. These mutual relations are the focus of the second part of the paper. A simple dynamic model of capital and cooperation is developed and estimates are presented.3 The dynamic interdependence of capital and cooperation raises the issue of the stability of the cooperative in the moshav. This issue is of particular practical relevance since in many moshavim cooperation is not complete. The question posed is, therefore, whether partial cooperation is stable or whether a moshav faces a dichotomy of either complete cooperation or disintegration into separate family units. The conclusion reached in the empirical analysis is that partial cooperation is stable. This conclusion is qualified since it is based on observations on the operation of moshavim in a period in which agriculture was actively assisted by the government, par- ticularly through credit subsidies. It is not clear whether partial cooperation can be maintained in a less supportive environment. 2. THE MOSHAV AS A COOPERATIVE The first moshav was established in 192 1 and was based, from its inception, on a clear, well-articulated and documented doctrine: maximum self-suffi- 3 Focusing on empirical investigation, we do not discuss structural issues of cooperation in terms of groupdecision theory as, for example, in Olson (1977) or Zusman (forthcoming). 56 HARUVI AND KISLEV ciency in food for the farin family, cooperation in services, mutual aid in farm cultivation, exclusive family operation without hired labor or off-farm work, democratic government of the cooperative (Weintraub et al., 1969; Baldwin, 1972).Eighty moshavim were established before the creation of the State of Israel in 1948. More than 300 have been added since, the majority in the early 1950s. The prestate settlers were, as a rule, ideologcally motivated; the majority were immigrants from Eastern Europe. Most of the "young" (post-1948) villages were populated by newcomers from Middle Eastern countries, directed to the moshav upon arrival in the country, with no prior preparation in agriculture and lacking ideological commitment to cooperation. Being both an economic cooperative and a village, the moshav offers mu- nicipal services as well as marketing, supply facilities, and technical services such as a pool of farm equipment or grain storage. In addition to the privately run farms, most moshavim operate a joint agricultural enterprise-field crops in many villages, or orchards if distance to available land makes individual cultivation too costly. The cooperative organization of the moshav is supported by the economic environment in which it functions. Moshavim cultivate national land that is allotted to the village and subdivided among members according to detailed regulations: the plots must be of equal size, they cannot be subdivided, and only one son can continue on the father's farm. Water and production quotas (recently in milk and poultry) are also distributed nationally at the village level and realloted to individual members by the cooperative. The moshavim are organized in several "settlement movements," each associated with a political party. The movements represent the interest of the moshavim in the political arena and attempt to guide the individual moshav in cooperative life-style. Economically, the moshavim are members of regional "bu-ving organizations" that centrally purchase farm supplies and consumer goods for the village members and operate processing plants, trans- portation services, and credit facilities. The buying organizations and their subsidiaries sell to village cooperatives and collect payments for products marketed through the cooperatives; they seldom deal directly with individual farmers, thus strengthening cooperation in the moshav. 3. FINANCE The cooperative in the moshav acts as a financial intermediary that receives credit and distributes it among its members. There are many advantages to financial cooperation. National land cannot be used as credit security by the individual member and the bank cannot repossess the farm in case of failure to repay.a The village functions as a riskpool. Lenders, including banks and a Membership in a moshav can be transferred, but the cooperative has to approve the new member. A farm cannot be transferred to a commercial concern. COOPERATION IN THE MOSHAV 57 regional organizatio4s, prefer to operate at the village level. Private dealers, mostly wholesalers of farm products, may supply short-term credit directly to members in need against the coming crop. They probably can enforce repayment by methods that banks and cooperatives would not follow. How- ever, private dealers will naturally not be able to match the terms of subsidized, particularly long-term, credit and will not finance investment in structures and equipment. Sole reliance on private sources will severely restrict the volume of f,nance to the moshav farms. The operation of the cooperative as a financial intermediary greatly sim- plifies mutual aid within the moshav. Assistance to a member in distress, which was based in the past on actual work on the member's farm, now takes the form of sharing the financial burden.s Furthermore, members are allowed to draw on their credit in the cooperative in periods of lower income.6 In some cases the moshav may also decide to contribute from its own Sources in the development of a new enterprise on a member's farm. By acting as a financial intermediary, the success of the moshav depends on its credit standing, which depends. in turn. on the cooperation of the members. Moshavim often pledge their crops as collateral for credit. The onll w,a1'the cooperative can be sure of timely repayments is when marketing is actualll done through its offices; then debts are simply deducted from the member's market revenue. This explains the critical role of cooperative mar- keting in the moshav. Without joint marketing the moshav cannot operate as a financial intermediar]'. that is. it cannot function as a cooperative. On the other hand. the higher the degree of cooperation in marketing, the better the credit standing of the moshav. and the easier capital accumulation be- comes. Cooperation enhances capital intensity. The moshav can, within limits. encourage cooperation and protect itself against moral hazards. It is customary. for example. to supply feed on credit in proportion to livestock products marketed through the cooperative' But members may choose low levels of utilization of village services and only partial participation in joint marketing. Moreover. once a member has ac- cumulated debt and reduced his participation, the power of the moshirv to enforce repayment is limited and legal actions are costly. financially and 5 The bylaws of the moshav. suggested in 1951 b-v the largest movement. stated under the title of Mutual Aid: "Every member shall cultivate his farm himself with the assistance of members of his family and he shall not employ hired laborers. The cooperative shall assign another member to help in case of illness. Members shall be assigned according to a given order" (our t.anslation). No mention of mutual aid in labor appears in more recent regulations of moshavim. (We are indebted to P. Zusman for this reference.) 6 The marginal propensity to consume estimated lrom a cross section of families in moshavim was 0.293 (Arbel, 1973). This value is much lower than all the estimates for urban families in Israel and reflects maintenance of similar levels of consumption despite income variations in households in moshavim. 58 HARUVI AND KISLEV socially. As a result, marketing in moshavim is seldom completely cooperative and conflicts may arise between the cooperative and the members who have incentives to market privately. We turn now to a schematic analysis of these conflicts. 4. THE INDIVIDUAL AND THE COOPERATIVE This section outlines a schematic single-period modelT of the behavior of the cooperative and the individual member with particular attention to fi- nancial intermediation and joint marketing. Time requirement of private dealing in the product and input markets plays a major role in our analysis. Private marketing will therefore be made a function of labor devoted to this activity. We first disregard labor requirements of private market dealings. Basically, the analysis of product allocation to private and cooperative mar- keting is analogous to the analysis of effort allocation in labor cooperatives (Sen, 1966; Israelsen, 1980). The reader will recognize lines of similarity. A. The Model Output of farm i, y;, is a function of labor. L;. andcapital, K;, Yi: .f(Ki, L) (l) and, disregarding hired labor and oflfarm employment, we assume Z; : const. We also assume constant and identical variable cost Z per farm. Capital in production is the initial amount, K6;, plus investment, which is assumed for the purpose of the present analysis equal to borrowing, .B;, K,:Koi*Bi (2) The farmer markets q, percent of his product through the cooperative and si : 1 - q; privately. Some products are more easily sold privately (calves, cut flowers to local markets), and farmers may secure individually better prices. On the other hand, subsidies (recently in milk and poultry) are paid only through certified dealers and are mostly limited to cooperative marketing. Other products are marketed jointly for technical reasons: milk, for example, is collected by the tanker only in the village station. To reflect the differential advantages of private and cooperative marketing for various products we assume a constant pice p. for joint marketing, disregarding the possibility of economies of scale, and a variable price p;(s;) for private marketing, this price being a decreasing function of s;. ? We disregard consumption, saving, capital accumulation, risk in production and in financial operations. Some of these issues are discussed below, but a complete theoretical analysis of the moshav is outside the scope ofthe present discussion. COOPERATION IN THE MOSHAV 59 The value of private and joint cooperative marketing for farmer i is, re- spectively, Ii: sipi!)f(Ko, + 8,, L,), Qi: eip,-f (Koi + Bi, Li) (3) and for the moshav I": Z Ii Q,: Z Q, B,: Z Bi, @) where the summation is on all i (i : l, . . . , n). Borrowing is through the cooperative which faces an upward sloping supply curve of credit with average rate of interest, r, a function of both borrowing and the moshav's cooperative revenue: r : r(8,, Q) (5) with partial derivatives 16 ) 0, ra < 0.Increased joint revenue augments risk pooling and reduces the risk of lending to the moshav. Although efficiency consideration would dictate that cooperatives charge their members the maryinal rate of interest, most moshavim follow the pa- tronage principle and charge the average rate. There are several reasons for this seemingly suboptimal policy: it is comparatively simple to administer, it looks 'Just," and it permits indirect mutual aid-farmers in distress can expand their borrowing at a rate that may often be lower than the social cost of credit. We shall also assume here that members in debt are charged the average rate (and members holding financial reserves in the cooperative are paid the average rate). This assumption will not alter the conclusion of our analysis. B. Analysis We first assume centralized allocation (Sen, 1966). The moshav's "GNP," net of interest payments. is G,: I, + Q" - nV - r(8,, Q)8,. (6) Maximizing with respect to the two farm-level control variables, .B; and J;, we get from the first-order conditions (/5 is the partial derivative ofl(') with respect to Ki): -f*ls,p,(s,)* qip"l: rl B,(rp* rqeip"fx), Q) pi(s) + sipi(s) : p"(l - roB"). (8) Equations (7) and (8) are the conditions for maximum social income in the 60 HARUVI AND KISLEV moshav. There are n pairs of such equations, a pair for each farm. We assume that internal solutions exist. Alternatively, taking the individual member's point of view and assuming voluntary allocation, income on the farm is Gi: Ii t Q, - V - r(8,, Q,)81 (9) and maximizing with respect to ,Bi and s; we get for each farmer l^{s,p,(s) * qip,l: rl Bi(r6+ rqQip"fx) (10) and pij) + srpl(s) : p,(l - roB). (11) Equations ( 10) and ( 1 1 ) define the conditions for maximum income on farm i, disregarding externalities-the effect of the action of a farmer on other members in the moshav-or, in Sen's (1966) terminology, assuming zero sympathy. The left-hand side in Eqs. (7) and (10) is the value of the marginal product of capital on the farm, valued at the average (weighted) price. The right-hand side is the maryinal cost of credit-in Eq. (7) the marginal cost to the moshav, in Eq. (10) the cost to farmer i. Since ra> 0, and rq < 0, the maryinal cost can be either higher or lower than average cost. when the effect of capital accumulation and production expansion on lowering the cost of credit out- weighs the effect of borrowing on raising this cost, r decreases with borrowing. optimal credit under centralized allocation will then be higher than under voluntary allocation. This possibility is plausible. There are moshavim in which officers encourage members to invest and expand, explaining that ..it will be good for e'eryone." Under different circumstances, the individual member ma1' borrow "too much" from the social point of view. Equations (8) and (11) determine the optimum s;. The maryinal price of individual marketing is equated to the marginal price of cooperative marketing. The last term incorporates the lowering effect of expanded joint marketing on cost of credit. As.B; < 8., individual marketing under voluntary allocation exceeds the social optimum. The interests of member and cooperative are, at least partly. in conflict. C. Labor in hfarketing The major hypothesis of our study is that private marketing (and purchase of inputs) is time-consuming and is affected, therefore, by the member's valuation of his time. To incorporate this aspect in the analysis, we adopt the schematic assumption that the share of private marketing is equal to the percentage of labor input devoted to this activity. Normalize labor in such a way that the total amount of labor on the farm equals unity (r : I ). Let r; be the amount of labor in agricultural production COOPERATION IN THE MOSHAV 6l on farm i, and now write s; : | - Li-labor devoted to private marketing. The shares of farm product marketed individually and jointly are si : | * L and Li, respectively: Ii: sipi!i)f(Ks1 * Bi, L1), Qi: Lip,.f(Koi+ Bi, L). Total values of marketed product at the moshav level, 1,, Q, are defined as in (4) with Z, replacing qi; the moshav's and the farm's "GNP" are defined, similarly. as in (6) and (9), respectively. Maximizing G. and G, with respect to ,B,, we get from the first-order condition expressions that are identical to Eqs. (7) and (10). Maximizing with respect to 1,. we get at the moshav level (/a is the derivative of /(. ) with respect to l;): frfs,p,(s,) * Lip,f - rqp,l.f(Ki, L) + LifLlB, : .f(K,, L,)[p,(s,) - p, t srpi(s)]. (12) The expression forthe individual farmer is identical to (12), except that B. is repiaced b1, B;. The left-hand side in (12) is positive since rB < 0 and all other terms in . this expression are positive. The first term on the left. .f,_[ ], is the value of the marginal product of a unit of labor on the farm-a unit shifted from marketing to agricultural production; the second lerm, rqpcl.].B.. is the value to the moshav of the reduction in interest payments due to increased joint marketing on farm i. The additions to the product marketed cooperativel-v comes from two sources: the withdrawal of one unit of labor from individual marketing increases cooperative marketing by .f(Ki,I,) and this same unit of labor also contributes to production on the farm .ft, of which Lifl are marketed through the cooperative. The term on the nght-hand side of (12) (in the second line) is the marginal revenue of the farm-marginal with respect to a unit of labor withdrawn from private marketing. Again, assuming internal solutions, since B. > ,8,. optimal cooperative marketing from the social point of view exceeds the optimum from the individual perspective. We are interested in the effect of capital accumulation on the share of cooperative marketing. This effect can be positive or negative. Capital accumulation will shift both sides of (12). The net effect is subject to empirical investigation, which is attempted in the empirical sections to follow. We start with a description of the data utilized in the study. 5. THE SAMPLE Commonly, moshavim are divided in Israel according to period of foun- dation: established moshavim settled before 1948: young moshavim settled 62 HARUVI AND KISLEV after statehood. Another division, utilized in this study, is between developed and developing moshavim. This division is in accordance with administrative practices but it also reflects the age and level of economic development. The developing moshavim are all "young" and all the "established" villages are included among the developed group. The data, available at the moshav level, were mostly utilized in the study as averages per farm.e The most complete information was for 1976 and the sample is for this year. Included are 249 villages, I 15 developing and 134 developed moshavim, but certain information was available only for part of the sample. Only 95 developing moshavim are therefore included in the data base for the regressions reported. Excluded from the sample were moshavim for which reliable data were not available. In some cases these may have been less well-organized communities whose exclusion might have biased the sample somewhat in favor of higher levels of cooperation. We do not know the extent of this possible bias. Some of the attributes of the moshavim in the sample are reported in Table l. For easier reference, averages were converted to dollars at the 1976 exchange rate ($l : 7.94IL). Output per farm in the developed moshavim is almost twice as high as in the developing villages, despite lower levels of value of capital per farm.10 The share of farm enterprises run directly by the cooperative (not including village services) is higher in the developing mosha- vim, which are less efficient in utilizing productive resources. The discrepancy is partly due to a different natural environment: 5lVo of the developing mo- shavim are located in the hilly regions of the country, as against 9Ea of the developed moshavim. Cooperation, defined as the share of farm product marketed by or revenue passed through the offices of the moshav, was 65 and 727o of output. respectively, in the subsamples of the developing and developed moshavim. Some economic characteristics associated with cooperation are reported in the iast part of the table. A moshav is deftned as a livestock-type if more than 507o of the village output is from livestock enterprises. Of the developing moshavim, l.8Vo are populated by Western families; the rest are of Eastern origin. The efficiency of the village services was studied in a survey conducted by the extension service.lr Thirty percent of the moshavim in the sector were 8 Developing moshavim are associated with the Jewish Agency, developed with the Ministry of Agriculture. e A detailed description ofthe data and sources appears in Haruvi (1980) and can be obtained from her. '0 Capital in the developing moshavim was estimated by the extension service; capital in the developed group was estimated by us. It may be that the procedures used to construct the two measures are not exactly identical. rr Extension officers graded the services as good, bad, or ol intermediate quality. We took the "good" as being efficient. COOPERATION IN THE MOSHAV 63 TABLE I THE SAMPLE_AVERAGE VALUES Moshavim in the sample Developing Developed All Number of moshavim l15 134 249 Number of farms per moshav 60 78 ',70 Value of output per farm (lL) I 30,600 214,500 174,000 (US dollar) ( l 6,4s0) (27,015) (21,91s) Value added per farm (IL) 45,200 73,300 59,700 (US dotlar) (5,690) (9,230) (7,524\ Farm capital (lL) 292,700 267,200 278,900 (US dollar) (36,805) (33,652) (35,12s) Share ol cooperative farm enterprise' (%) In output 20 ll l4 In value added 26 l4 18 Cooperation-products marketed through the cooperative as percent of total value of output Family farms 65 '72 69 The village 72 74 '13 Structural characteristics (percent) Farm t-vpe-livestock 58 65 62 Ethnic origin-Western l8 76 49 Region-hilly 5l 9 28 Moshavim with efficient village services 30 Inequality (coelicient of variation of output in famil-v farms. in percent) 88 ' The agricultural production enterprise run b1'the cooperativei in many cases, distant orchards or field crops. See text for description of olher rariables. regarded as having efficient services. As a measure of inequality we propose to use the value of the coefficient of variation of output. This coefficient averaged 887o for the developing moshavim. (The data are not available to calculate the coefficient for the developed moshavim.) Average flows of long- and short-term credit, indicators of financial activity, are reported in Table 2 for developing and developed moshavim. Government- or Jewish Agency-subsidized credit is grouped under the heading of"directed finance." The rest is commercial credit. 6. STRUCTURAL ATTRIBUTES Moshavim vary widely by social, ideological, and economic attributes, all of which affect cooperation. In the next section we present a simultaneous- equations model of the joint determination of cooperation and capital in- tensity. Since the two variables are deterrnined simultaneously, cooperation 64 HARUVI AND KISLEV TABLE 2 FTNANCIAL Fr-ows, 1976o Long term Directed Own Short-term finance' finance' Investment finance Developing moshavim Value (IL) 18,000 9,400 27,400 34,900 (U. S. dollar) (2,267) (1,184) (3,45 r ) (4,39s) Share (percent) 66 34 100 Developed moshavim Value (lL) l 3,700 I 6,100 29,800 38,600 (U. S. dollar) (t,725) (2,028) (3,753) (4.86 1) Share (percent) 46 54 r00 o Average per farm in the moshav, b Includes credit and subsidies by govemment and the Jewish Agenc1.. 'Including savings and short-run commercial (nonsubsidized) credit. and the amount of capital are highly correlated; factors that affect one of the variables also affect the other. Moreover, the effect of most of the structural attributes was obscured in the simultaneous-equations analysis. The discussion in this section is therefore limited to ordinary least squares and the estimates are taken as measures of association. cooperation is a form of social behavior that is difficult to police. The power of the officers of the moshav to enforce cooperative action is limited. The legitimac-v of sanctions is greatly affected by the standards and interests of the groups. which vary widely across the moshavim. Individual members can often be motivated to act privately. Hence the importance of the struc- tural-social and economic-factors that enhance cooperation. we shall discuss a few of them. Income. The higher the income of the member of the moshav, the higher the alternative cost of his time. In the sample, higher income is associated with higher levels of schooling and a higher degree of modernization (Sadan and weintraub, 1980), factors which themselves affect the member's com- prehension of the advantages of cooperation. Product composition. Products differ in suitability for individual or co- operative marketing. In the long run farm structure and product composition may be affected by the willingness to cooperate; in the short run the causation goes from product composition to cooperation. Income distribution. A homogeneous moshav is stronger socially. If, on the other hand, comparatively large income differences exist, coordination and compromises are harder to achieve. COOPERATION IN THE MOSHAV 65 Ethnic origin. Moshavim with "European" settlers, who joined, in most cases, at their own initiative and were often ideologically motivated, can be expected to maintain stronger cooperation than communities populated by immigrants from Middle Eastern countries. Eficiency of services. When the cooperative is run well, members are encouraged to use its services and to trust its officers. Table 3 summarizes the association between structural attributes and the degree of cooperation. Availability of data limits the analysis to the sector of the developing moshavim. By regression 1, an increase of 1000 IL in the value of output per farm, a proxy for income, is associated with an increase of 0. I 3 57o in the share of cooperative marketing. The value of this parameter is somewhat lower in the other regressions. As hypothesized, inequality iS associated with reduced cooperation in the estimates of Table 3, while effi- ciency in village services is associated with higher levels of cooperation. The coefficient of ethnic origin is insignificant, perhaps due to insufficient variation of this factor in the subsample. Region and farm type are correlated: the hilly moshavim tend to have more orchards (deciduous fruits, mostly), quite often run jointly due to distance from the village; and typically they do not have dairy enterprises. As the findings indicate, the variable "region" is stronger than our farm-type classification in explaining cooperation; with region in- cluded in the regression, the influence of farm type is insignificant. In the long run, farm type (product composition) may be affected by wiilingness to cooperate. In the short run, the chain of causation runs from farm type to cooperation. But to the extent that farm type is influenced by geographical considerations, it is an exogenous variable, which will not be affected by cooperation, even in the long run. 7. CAPITAL AND COOPERATION Two factors affect the capital-labor ratio in the moshavim' The first is the general process of capital accumulation in the individual farms in reaction to increases in nonfarm income. Farmers either increase the amount of re- sources at their command or give up farming (Kislev and Peterson. 1982). It also often happens that exit from agriculture is incomplete. in which case the share of part-time farming increases. The second factor is the process of mutual development of capital and cooperation. Cooperation improves credit and increases capital intensity. Higher capital-labor ratios increase income and the cost of self-service in marketing and procurements. It also increases the share of capital-intensive enterprises, the products of some of which are more conveniently marketed cooperatively. However, despite past capital accumulation in agriculture, cooperation is not complete. Evidently structural attributes prevented complete cooperation. 66 HARUVI AND KISLEV TABLE 3 STNUCTUNEI ATTRIBUTES ANO COOPENETION, DEVELOPING MOSHAVIM Regression R2 0. l5 0.29 0.41 0.28 0.42 l. Intercept 53.91 64.99 ss.67 48.86 55.53 (12.33) (12.94) (ll.lt) (10.05) (9.8 l ) 2. Output 0.135 0. 13 0.073 0.121 0.076 (4.33) (4.33) (2.33) (4.28) (2.53) 3. Inequality -14.35 -6.06 -5.83 (3.42) ( L43) (1.37) 4. Village services-efficient 9.15 8.33 8.72 (2. r9) (2.r5) (2.22) 5. Ethnic origin-Western *3.46 -0.65 -0.98 (-0.73) (-0. l4) (-0.23) 6. Region-hilly 18.30 l 5.36 (4.06) (3.00) 7. Farm types Field crops 0.62 -1.46 (0. l3) (0.33) Orchards 16.37 3.53 (3.72) (0.66) Notes. obsen'alions are averages per farm in the moshav. output and cooperation, for the moshav as a rvhole, include members' larms and the cooperative enterprise. Dependent variable: share of value of product marketed cooperatively (in percent). Independent variables: output- value of average output in the moshav ('000 IL); inequality-coefficient of variation of output in family farms in the moshav. Dummy variables: village services-l = efficient, 0 : inefficient; ethnic origin-l = Western,0 = Eastern; region-l : hilly,0 = others; farm type*l = 6eld crops or orchards, 0 : livestock. The regressions are linear. , ratios are in parentheses. Also, with better roads and communication, self-service in marketing and procurement may be more efficient. The effect of economic gowth was thus both to increase and to decrease the incentive to cooperate. The crucial structural issue is the question of stability; namely, is the growth process of capital-cooperation stable or should we expect a complete breakdown of cooperation in the labor-intensive villages? In an attempt to examine this issue empirically we propose the following dynamic model. The model is restricted to conform with the available data-a single cross section. COOPERATION IN THE MOSHAV 67 Let us define q degree of cooperation (share of joint marketing, 0 < q < 1) K capital on the farm C long-term credit ' ,S short-term credit and saving D depreciation of capital. Since all variables are moshav-specifrc, the moshav index (or subscript) is omitted. Time differences are defined as Ax, : Xt+r - x, for any variable x. Cooperation is a function of capital and structural variables (in the intercept) q,: a* bK, a,b>0. ( 13) Long-term credit is a function of capital and cooperation C,:c*eq,-lfK, e,f>0. (14) Saving and short-term credit (a single variable in the data) are also functions ofqandK St= h* gq,* sK, g>0,0 D,:dK, 0 Equations ( 13)-( l7) constitute lhe dynamic model of our analysis. Rewriting (t7), L.K,: c* h*a(e +g)+ K,(b(e+ d+f + s- d) (18) from which one derives K,*r=Kr+A.Ki, =B*AK,, ( 1e) where A:t+b(e+d+f+s-d\ B=c*h+a(e+g). From (19), A.') x,: B(1.- * ,,*0. ' \t-Al : B / B \ t-A+lI(Ko- t-^) (20) In equilibrium, under the present assumptions, LK,:0 and K,*1 : K,; then, Eq. ( 19) can be rewritten as K : B l0 - l). The possible equilibrium values of capital, K, are 68 HARUVI AND KISLEV R: ,r-t for A: l B -B t*A tor0< t-A .A The last line in (2 1 ) indicates that actual capital cannot be negative; calculated equilibrium values can be. The equilibrium value of capital is further limited by the fact that cooperation is a positive magnitude. The calculated equilibrium value of cooperation equals q:a+bK (22) and since, from 0 < q < 1, capital is constrained, 0 < K< (t - a)/b. Substitute f< into 120;: K,:R+At(Ko-R), (23) and A.K,:1nr'-l'11Xo-R1. (24) Notice that At+t-Atz0 ifl= 1, (2s) which implies the following interpretation of (23) and (24).If O < A < t, equilibrium is stable. K will converge in the long run to K and cooperation to Q :, + bK. The equilibrium is unsrable if A > t.Then. ifKs < R. capital will grow and cooperation will develop to compietion. If, on the other hand, Ki < K, capital and cooperation will both continuously decline. It is also useful to note that Azl asb(e*g)+f+sad. (26) This means that ifthe rates of increase of saving and credit due to cooperation and capital intensity add up to more than the rate of depreciation, equilibrium is unstable and the solution diverges either to complete cooperation or to its breakdown. 8. ESTIMATES Past capital accumulation and degree of cooperation can be taken as ex- ogenous in determining credit availability. Therefore, Eqs. (14) and (15) of the dynamic model-the long- and short-term borrowing-are estimated by ordinary least squares. Equation ( l3) is estimated in a simultaneous-equations model to be detailed below. Data on depreciation of capital, Eq. (16), are not available; we have assurned that d : 0. 15 for the calculations in the present section. COOPERATION IN THE MOSHAV 69 The model is estimated for each subsample separately and for the sample as a whole. The OLS estimates of Eqs. (14) and (15) are reported in Table 4. Cooperation affects positively finance in all the regressions reported in Table 4. The estimates of the coefficient of capital in regressions I and 2 are insignificant. This may be a reflection of the effect of the support of public agencies in the sector of the developing villages which comes mostly in the form of subsidized finance and is biased in favor of the weaker, relatively capital-poor moshavim. In comparison, the significant and strong effect of cooperation in these regressions may be interpreted to indicate that the better organized moshavim make better use of opportunities offered by the public agencies and by commercial financial institutions. To incorporate simultaneity of size of capital and cooperation we suggest the following two-equation model (other variables will be added below): Q:ar+blKlc10, le, (27) K:az*b2q-lc20plu, (28) in w.hich the newly defined variables are 0r,0r, c,, c2, vectors ofexogenous variables: the corresponding coefficients e, il are error variables. The common exogenous variables in 0n and 0p aie farm type and geographic location TABLE 4 SHoRr- A\D Lo\G-TERM FINANCE-EQUATIoNS (14) A.\o (15) Regression Der eloping Der eloped Whole moshavim moshar im umple Parameten in the model Dependent variable CS CS CS s R2 0. 14 0.04 0.16 0.19 0. 1l 0.10 Intercept -26.t 3.99 -r.15 7.37 -12.i7 -6.02 h (-1.62) (0.38) (-0.r5) (l.0l) ( t.,17) (-r.03) Cooperation 0.64 0.22 0.20 0. 13 0.37 0. r l e (3.20) ( r.6e) /) l)r rl 5)r (3.77) (1.71) Capital 0.008 -0.032 0.022 0.026 0.021 0.023 (0.14) (-0.96) (4.23\ (5.30) (3.37) (4.03) Subsample (dummy. developed = l) -2.61 t4.46 ( 1.01) (5.60) Note-i. Dependent variables: C-long-term finance in IL per farm: S-short-tem finance and farm savings in IL per fam. Independent variables: (t) Cooperation-share of product marketed cooperatively (7o), (2) capital-net value ofcapital per fam in IL. Obseruations are for 1976, at the moshav level, for the village as a whole. including the cooperative enterprise. I ratios are in parentheses. HARUVI AND KISLEV (region) of the moshav. Specific variables are ethnic origin, affecting only cooperation, and the age of the moshav (years from establishment) affecting, by assumption, only the amount of accumulated capital. Equation (27) is the empirical formulation of the earlier Eq. (13) in which the exogenous variables were represented by the intercept a. Second-stage estimates of Eq. (27) are reported in Table 5. The coefficient of capital is significant in regressions 4 and 6. Estimates from these regressions are therefore utilized in the analysis that follows. By Eqs. (21) and (24), the long-run behavior of the system depends on the magnitudes of the parameters I and ,8. As we have seen, if A > l, the system is unstable; cooperation and capital in the moshav will tend to either grow to completion or decline to zero. The value of a that maintains A < I can be consistent with a stable long-run equilibrium of above zero but less than complete cooperation. TABLE 5 CeprrnL eNo CooprnauoN-EeuATroN (27) Regression Developing Developed Whole moshavim moshavim sample Parameters in the model R2 0.36 0.37 0.08 0.1 l 0.20 0.22 Intercept 97 .90 7 4.37 59.37 56. 19 5 r.59 47.73 (0.08) (3.16) (9.17) (10.16) (8.68) (10.84) Capital -0.t76 -0.07'1 0.009 0.02s 0.01l 0.026 (-0.30) (-0.84) (0.44) (2.08) (0.6 r ) (2.36) Ethnic ongrn r.r9 0.72 6.83 4.7 4 5.66 3.80 (0.re) (0.r3) ( 1.21) (0.e0) ( 1 .4 1) (0.99) Farm type 13.41 10.77 8.56 8.72 8.62 8.54 (0.e r ) (2.28) ( r.96) (2.04) (2.8 r) (2.82) Region 34.1t 27.63 14.49 17.29 20.52 20.28 (0.e0) (3.87) ( r.73) (2.22) (5.38) (s.37) Subsample (dummy, 7.'.72 9.09 developed : l) ( r.e4) (2.39) rVoles. Dependent variable-cooperation. Regressions are two-stage estimates of the model's Eq. (27). Estimates are for the whole sample. Qualitative variables are included as dummies: Ethnic origin-l = Western, 0 : Eastern; Farm type-l : livestock, 0 : others; Region-l : hilly, 0 : other. Settlement movements are included as dummy variables in the first stage of the estimation procedure in regressions 2, 4, and 6. COOPERATION IN THE MOSHAV 7l The calculated equilibrium values of capital, & and cooperation, 4-, are reported in Table 6 for several cases. The values are calculated utilizing the estimated parameters of Table 4 and of regressions 2, 4, and 6 of Table 5. Examples of the calculations are explained in the notes to Table 6. The calculated values of A and its standard deviation are also reported in Table 6. In all cases 0 < A < 1, indicating stable equilibrium. According to Table 6, the long-run equilibrium value for cooperation in developing moshavim (by regression 2 of Table 5), of Eastern origin, non- livestock farms and nonhilly regions, is 0.61. That is, in the average moshav of these characteristics, 6lVo of the farm product will be marketed cooper- atively. In the same line, in Table 6, a developing moshav of Western origin, with livestock-type farms and located in the hilly regions will typically market cooperatively 91% of the farm output. Cooperation will be, by our calculation, much higher in the last case, mostly because of the exogenous conditions encouraging it-region and farm type. 9. SUMMARY Cooperation has many economic advantages and the principles of the moshav fit well the prevailing ideas on the desired land-tenure system and sectoral structure of agriculture in Israel. However, a deteriorating cooperative can be extremely costiy to its members. If farmers gradually cease to market through the cooperative, yet attempt, as long as they can, to rely on its service facilities and credit, the cooperative will sink into debt. Since the cooperative is run separately from the individual farm, information is seldom complete and members rarely realize the full severity of the economic difficulty their moshav may face. If, as a result of these difficulties the cooperative will have to be dissolved, members will find themselves shouldering their shares in a debt much exceeding the size they anticipated or their ability to pay from their farm income. TABLE 6A Calcuurso VrLurs: THE PARAMETER ,4 AND Its S-r.q.NoA.RD DEVIATToN o(A) Developing moshavim Regression 2 0.760 0.054 Regression 6 0.9t2 0.01 I Developed moshavim Regression 4 0.906 0.034 Regression 6 0.9t2 0.011 72 HARUVI AND KISLEV TABLE 6B EQUILIBRIUM VALUES oF CAPITAL AND CooPERATIoN 2 Ethnic origin (Western) No Yes Yes No Yes Yes Farm type (livestock) No No Yes No No Yes Region (hilly) No No No Yes Yes Yes Developing moshavim (regression 2) R 174.32 t76.90 224.95 273.33 275.84 323.96 s 0.61 0.61 0.71 0.81 0.81 0.91 Developing moshavim (regression 6) R 59.05 80.21 127 .7'1 171.97 193.13 240.69 q 0.40 0.54 0.63 0.72 0.77 0.87 Developed moshavim (regression 4) R 263.42 280.06 3 10.63 324.12 340.76 371.83 s 0.63 0.68 0.'77 0.82 0.87 0.96 Developed moshavim (regression 6) R 243.64 264.80 3 12.36 356.58 377.73 425.28 s 0.63 0.68 0.77 0.86 0.91 1.00 Nores. Equilibrium value ofcapital is calculated asi : Al( - A), as in Eq. (21). Cooperation is calculated by (22): i : a * bk. The parameters A and B are calculated according to Eq. (19) where the parameters are the estimated values reported in Table 4 and in the indicated regressions of Table 5. Two examplgs of this calculation are given: (l) The first value of k in column I is calculated as .1 :0.760: l - 0.077(0.64 +0.22) + 0.008 - 0.032 - 0.15; B : 41.838 : -26.fi + 3.99 +',74.37(0.64 + 0.22): R: t74.32 = 41.838/(r - 0.760). (2) In the flrst value of column 2, the estimate of the dummy for the Western ethnic origin in regression 2 in Table 5 is added to the intercept, the parameter a. Accordingly, B : 42.457 - -26.11 + 3.99 + (74.37 + 0.72)(0.64 + 0.22): R = ti6.9o:42.4s710 - 0.760). Similarly for other values. The dangers of economic entanglements call for close monitoring of the operation of the moshav, both by its own omcers and members and by its creditors and public mentors. Effective monitoring is based on thorough understanding of the monitored system. The finding that partial cooperation can be stable can assist in evaluating the economic situation of a moshav. Needless to say, even ifour study contributed to better understanding ofthe COOPERATION IN THE MOSHAV 73 functioning of the moshav, we are still far from complete identification of the social and economic determinants of cooperation and its success. REFERENCES Arbet, Assaf, Patterns of Consumption and Saving in Established Family Farm. Mimeo, M.Sc. thesis, Hebrew University, Rehovot, 1973. (ln Hebrew.) Baldwin, Elaine, Differentiation and Cooperation in an Israeli Veteran Moshay. Manchester: Manchester Univ. Press, 1972. Haruvi, Nava, Cooperation in Moshavim. Mimeo, Ph.D. thesis, Hebrew University, Rehovot, 1980. (ln Hebrew.) Israelsen, Dwight L., "Collectives, Communes, and Incentives." J. Comp. Econ. 4(2):99-124, June 1980. Kislev, Yoav, and Peterson, Willis, "Prices, Technology and Farm Size."./. Pol. Econ.90:578- 595. 1982. Olson, Mancur, The Logic of Collective Action. Cambidge, Mass.: Harvard Univ. Press, 1977. Sadan, Ezra, and Weintraub, Dov, "Ethnicity, Nativity and Economic Performance of Cooperative Smallholding Farms in lsrael." Econ. Dev. Cult. Change 28:481-507, 1980. Sen, Amartya K., "Labor Allocation in a Cooperative Enterprise." Rev. Econ. Stud. 33-361- 37 t. 1966. Weintraub. Dov, Lissak, Moshe, and Azmon. Y.. Moshaya, Kibbutz and Moshav: pattern of Jewish Settlement and Development in Palestine. Ithaca. N. Y.: Comell Univ. Press, 1969. Zusman. Pinhas. "Collective Choice, Pareto Optimatitl- and the Organization of Cooperatives- The Case olAgricultural Credit Associations." -/. Econ. Behat,ior Org., forthcoming. 378.5694 C45 8902 n/N5prt i-C7 3 5 3 1pi-10 r31nn THE CENTER FOR AGRICULTURAL ECONOMIC RESEARCH WORKING PAPER NO. 8902 EXPERIENCE WITH CREDIT COOPERATIVES IN ISRAELI AGRICULTURE by Yoav Kislev, Zvi Lerman & Pinhas Zusman 7 WAITE MEMORIAL BOOK COLLECTION DEPARTMENT OF AGRICULTURE AND APPLIED ECONOMICS 232 CLASSROOM OFFICE BLDG. 1994 BUFORD AVENUE, UNIVERSITY OF MINNESOTA ST PAUL, MN 55108 Rehovoth, Israel, P.O.B. 12 12 .1.11 ,nlnini TABLE OF CONTENTS . INTRODUCTION B. THE MOSHAV COOPERATIVE ASSOCIATION AS A FINANCIAL INTERMEDIARY 6 C. REGIONAL COOPERATION 13 D. CREDIT MARKETS AND INSTITUTIONS . . . 23 E. CREDIT IN THE ECONOMY AND IN AGRICULTURE 27 F. GOVERNMENT 37 G. CRISIS 40 H. RECAPITULATION, LESSONS, AND CONCLUSIONS • 46 REFERENCES . 54 LIST OF TABLES TABLE 1. INFLATION IN ISRAEL 4 TABLE 2. BALANCE SHEET COMPOSITION OF A REPRESENTATIVE MOSHAV 8 TABLE 3. BALANCE SHEET COMPOSITION OF A SUPPLY COOPERATIVE . . 16 TABLE 4. CREDIT MONTHS, MOSHAVIM'S DEBT TO MISHORTM RELATIVE TO SALES 19 TABLE 5. MAZON AND OFF TWO REGIONAL ENTERPRISES IN MISHORIM REGIONAL COOPERATIVE . . 22 TABLE 6. OUTSTANDING BANK CREDIT: TOTAL ECONOMY, AGRICULTURE, INDUSTRY .. 28 28 TABLE 7. RATIO OF OUTSTANDING DEBT TO GROSS PRODUCT AND NET CAPITAL TABLE 8. SHARE OF AGRICULTURAL CREDIT HELD BY MOSHAVIM AND SUPPLY COOPERATIVES .. 28 29 TABLE 9. DISTRIBUTION OF OUTSTANDING BANK CREDIT BY SOURCE • • 30 TABLE 10. THE DOLLAR-LINKED COMPONENT OF OUTSTANDING BANK CREDIT TABLE 11. REAL ANNUAL COST OF CREDIT IN ISRAEL, 1979-1987 35 42 TABLE 12. CONSOLIDATED FINANCIAL POSITION OF THE MOSHAV SECTOR . EXPERIENCE WITH CREDIT COOPERATIVES IN ISRAELI AGRICULTURE' A. INTRODUCTION Eighty percent of Israel's agricultural product comes from cooperative farms. A major form of cooperation in agriculture has been financial. It has flourished for a long time; but financial cooperation found itself recently in deep trouble and will need massive public assistance to overcome its difficulties. It is now too early to predict what kind of cooperation, if at all, will emerge from the crisis. Yet, important lessons can be drawn even from this incomplete experience. Our observations can be summarized as follows: Financial cooperation was effective when appropriate conditions prevailed, supporting intensive development of the cooperatives and their members There was however danger in its strength; while credit supply was expanding with inflation, the cooperative structure, its mode of operation, and the public support it enjoyed--encouraged over- extension; but when macro economic conditions changed and credit's expansion was slowed down, the sector found itself trapped in financial impossibilities. Now cooperation, tying farmers and their institutions of the crisis. r, together, intensifies the dimensions 1. Inflation Members in cooperatives and many others in Israel view themselves as victims of government policy; for the crisis in agriculture erupted when stringent measures to curtail inflation were instituted and rigorously enforced. Indeed, inflation has been always a problem in Israel, but it reached ominous proportions in the 1970s. Readers will find it useful to familiarize themselves with the magnitudes of price increases in Table 1 and in Figure 1. Up to 1975, the exchange rates were generally fixed and corrected from time to time in step with inflation. A crawling peg policy was followed from 1975 tO 1977 when it was replaced by a free market. 'We had helpful discussions with many; particularly we acknowledge information received from A. Marvid, and I. Nun. The responsibility for data presented in the paper and for their interpretation is ours. 4 TABLE 1 INFLATION IN ISRAEL: YEARLY RATE (PERCENT) Year Inflation Year Inflation Year Inflation rate, % rate, % rate, % 1949 -12.5 1962 10.2 1975 23.5 1950 6.0 1963 5.0 1976 38.0 1951 20.2 1964 4.4 1977 42.5 1952 66.4 1965 7.1 1978 48.1 1953 19.1 1966 7.8 1979 111.4 1954 7.5 1967 0.2 1980 132.9 1955 4.8 1968 1.9 1981 101.5 1956 4.6 1969 3.9 1982 131.5 1957 5.2 1970 10.1 1983 190.7 1958 4.2 1971 13.4 1984 444.9 1959 2.1 1972 12.4 1985 185.2 1960 3.4 1973 26.4 1986 19.7 1961 9.0 1974 56.2 1987 16.1 Source: Central Bureau of Statistics, Statistical Abstract Israel, 1987. Notes: 1) Change in Consumer Price Index. 2) Yearly rate is December index over December index of previous year. 3) The annualized rate of inflation for the first 3 quarters of 1985 was 257.0 percent; the annualized rate for the last quarter of 1985 was 29 percent. The three-digit inflation was halted. abruptly in July. 1985.. The anti-inflationary. Program InCludecra price and wage.freeze, 'reduction of government deficits, and a:return to: a. regime of fixed exchange rates.. As part of the new, policy,. the Bank (:)f Israel dramatically. increased the cost . of :credit. Producers who relied on large volumes of. short term credit--as many in agriculture did--foundthemselves' facing enormous., financial- expenses, which- prevented them from rolling over their loans and 'precipitated a .painful crisis. 2. Outline In the paper we concentrate on the experience with credit in the family farm sector, the moshavim--villages run by cooperative 5 Figure 1. Inflation in Israel Annual rate, % 500 400 ••••••••••••••• SOO .•••••••••••••••...... •N•.••••••••••.••••.••••••••••••••••.••••.•.•••••••.•.•••••••...... •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 200 ...... •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• 100 ti , 't i t, II I I I I 1 I "1" 1" I 1949 1952 1955 1958 1901 1984 1967 1970 1973 1978 1979 .1982 1985 associations; the communal sector, the kibbutzim, will generally not be discussed at this time. There are 420 moshavim in Israel with, on the average, 66 farm families per moshav. Seventy seven are "veteran" moshavim established before the creation of the State of Israel in 1948 and the rest are "young"--established after 1948. The moshavim are members in second order cooperatives: 9 regional and 3 nation-wide supply cooperatives (termed "purchasing organizations" in the Israeli parlance) and they own together a large number of regional service enterprises; for short, we refer to all of them as "regionals." In addition, the moshavim are members in several "movements," each affiliated with a political party. The movements represent the moshavim in the political arena and attempt to set standards of cooperative life-style. Financial cooperation was reduced drastically in reaction to the current crisis. But some of the moshavim and a few of the regionals, those with a stronger cooperative basis, are still active in credit intermediation--perhaps in a modified mode of operation. In the paper, we survey financial cooperation as it was before the crisis, and discuss the crisis separately. Other sections are devoted to the analysis of credit supply to agriculture and to the role of public intervention in the credit market, both in strengthening cooperation 6 and in precipitating its crisis. The paper concludes with a discussion of policy issues. EL .THE MOSHAV COOPERATIVE ASSOCIATION AS A FINANCIAL INTERMEDIARY2 Raising financial resources and allocating them among members and village-level services and enterprises are pivotal activities of the cooperative association in the moshav. The association is first and foremost a credit cooperative. In fact, it may divest itself of any other economic activity and still be regarded as a "moshav," but it is no longer viewed as such once it gives up its role as a financial intermediary. 1. Background •A moshav (plural - moshavim) is a farming community in which all farms are family owned and operated, and all farmers are members of the multipurpose, democratically run, village cooperative. In principle (practice varies) the cooperative association in the moshav purchases all farm supplies for its members and markets the members' farm product. The cooperative usually owns and operates a variety of service facilities, such as water installations, farm implements, sorting and packing sheds, and storage facilities. It may also manage directly some jointly operated productive enterprises (citrus orchards, cotton fields). In addition to serving production activities in the moshav, the association encompasses all municipal and many social functions in the village. The association also serves as a financial intermediary channeling credit to the farmers. Moshavim were settled mostly by penniless individuals on national land leased by members and associations from the state; members do not own the land. Initial public allocations of capital and nationally alloted resources, such as land and water rights, livestock, and machinery, were distributed equally. However, differences in farm 'This section draws on Zusman(1988, Chapter G) 7 structure and personal preferences have evolved over the years. This heterogeneity is a major factor in creating conflicts of interest in the moshav and a complete adherence to its principles therefore requires high standards of cooperative ethic. The land and water rights and production quotas are, in principle, inalienable; but a market has developed and membership in the moshav may be transferred together with the attached rights. The buyer however has to be accepted as a member by the cooperative. It is therefore practically impossible for lenders, particularly suppliers and commercial banks, to repossess farms, and this deficiency--that land and buildings cannot be used as collateral--is the principal economic justification for the evolution of the moshav cooperative as a credit intermediary. 2. Financial structure and functioning The nature of the cooperative's activities, and particularly the pivotal role of credit intermediation, can be seen by examining its balance sheet (Table 2). A few qualifications are in order before we turn to numbers. The financial statements of a moshav are summaries of the activities, the assets, and the liabilities of the association, not consolidated reports of the whole village--the cooperative and its members. Information on the economics of individual farms is generally not available. Book-keeping follows historical accounting principles and financial statements are prepared in nominal, historical terms and are therefore bound to be highly affected by inflation. Accounting for more than 75 percent of the liabilities in Table 2, the regional supply cooperative is by far the most important source of finance for the moshav (more on regionals in the next section). The second most important source of credit in Table 2 are members' deposits with their associations, amounting to 13.5 percent of the liabilities. These deposits usually earn the same rate of interest that debtor -Members are charged, which encourages internal lending whenever market rates are high. When outside credit was available at negative real interest rates, as was often the case in the early accelerating phases of inflation, members with surpluses usually held their savings in 8 TABLE 2 BALANCE SHEET COMPOSITION OF A REPRESENTATIVE MOSHAV ASSOCIATION, SEPTEMBER 30, 1981 (in percent of total assets) ASSETS LIABILITIES AND MEMBERS' EQUITY Fixed assets 3.7 Equity 10.7 Long-term investments Long-term debt to and loans to members 3.5 Settlement Dept. 2.1 Inventories 4.0 Other long-term debt 2.1 Accounts receivable Short-term bank loans 0.6 from non-members 12.2 Short-term loans from Members,', debit balances 76.6 regional supply coop. 76.9 Suppliers' credit 3.9 Members' credit balances 13.5 Other payable 0.2 Total 100.0 100.0 Source: Zusman (1988, Table G.2). Note: The data are for an average association in a sample of 13 moshavim in the "Mountain Region" (a fictitious name). inflation-protected bank deposits or in indexed government bonds-often with the blessing and the advice of the association. Long term credit in agriculture comes mainly from the Settlement Department of the Jewish Agency and from the government. The Settlement Department supplies newly established moshavim with start- up capital until they can operate independently, generating savings and raising credit. The government supports investment in preferred agricultural development projects. By 1981, the development assistance of the Settlement Department and the .government (most of the "other 9 long term debt" in the table) dwindled to a small percentage of the liabilities of the moshavim in Table 2--partly due to the emergence of alternative sources, partly due to inflation wiping out the real value of these long term and mostly• unindexed loans. Inflation is also responsible for both creation and understatement of equity in the association's balance sheet. Equity contributed directly by members is a very small sum. Most of the equity accrued as inflationary gains when the real value of the unindexed loans the association took to finance its own investment projects eroded and these gains were not distributed to the members. On the other hand, the equity is understated because the association's assets (orchards, buildings, machinery) are reported in historical, highly undervalued terms. It was estimated that the true value of equity is approximately 15 percent of the association's liabilities (30 percent for veteran and stronger moshavim). The role of the association as a financial intermediary is clearly demonstrated on the assets side of the balance sheet by the dominant share of members' debt, which accounts for more than three quarters of the reported total. Money raised from outside and internal sources is channelled to members, who in effect have a permanent line of credit with their association. When credit supply from outside sources was limited, the association would ration the loans to the members in the moshav. When not--and especially the regional cooperative at times offered an elastic supply of credit--members were restrained only by the cost of debt, interest charged daily to their accounts. Since government and Settlement Department long term loans are mostly transferred to members back to back, members' long term debt also forms only a negligible percentage of the association's assets. Here and in other forms of credit, the cooperative transfers loans to its members together with the inflationary gains associated with their use. Although correcting inflationary distortions will reduce the proportion of external and internal sources of credit and members' debt, yet even then, items reflecting financial intermediation will dominate the moshav's balance sheet. Cooperative credit intermediation 10 has several advantages, but it also entails dangers and difficulties. 3. Advantages: :Cooperative -financial services relieves individual membersof-the: need, to handletheir finance independently. It is -reasonable to expect (as is happening these days, Section G)' that the strUcture of the farm credit', institutions' can;be•adapted to deal with individual farmers:. credit 'would be extended, as it is now, by suppliers, marketing agencies' and banks; but its amounts may be more limited and its. cost higher" if the advantages' of cooperation are not, realized. - These' advantages are 'of several kinds.- With cooperative financial intermediation, all members finances are handled by 'a single professional officer--often hired and not 'a member of the moshav. The members enjoy access to financial expertise, particulary important in an inflation-prone and high-tax economy such as Israel's. Being a comprehensive cooperative is another advantage of the moshav in its role as a credit intermediary. As members in a well run cooperative purchase all farm inputs and market their product through the association, credit, purchasing, and marketing are interlinked: factor suppliers extend credit and the marketed output serves as security both for suppliers and for bank loans. Since the moshav is a relatively small community, officers in the association are familiar with the farm operation of the members and can assess, better than any outside institution, credit risks and soundness of investment. By pooling their demand for credit, members in the moshav improve their bargaining position in the market; they can attempt to appropriate at least a part of the saving lenders realize when the association, dealing with every member separately, does the retailing phase of credit marketing. This advantage, needless to say, is not limited to the credit market. Perhaps the greatest advantage of cooperative credit in the moshav 1-laruvi and Kislev (1984) study the relations between joint marketing and capital accumulation in the moshav. 11 lies in risk pooling which is implemented in two ways. In the short run, the association can service from its own resources the debt of a farmer who happens to be in a temporary shortage. The association extends further credit to the needy member on the understanding that it will be repaid in due course. Outside lenders do not have to deal with transitory difficulties of individual borrowers. A more fundamental mode of risk pooling is mutual liability and guaranty. According to this arrangement, the association borrows funds from external lenders with all members sharing the association's liability. The association then uses these funds to extend credit to its members or to invest in its own service and production enterprises. Should a member default on his repayment obligation (or should a village enterprise fail to cover the cost of operation) the association can still honor its obligations by drawing on the resources of the other members, and ultimately, if the association fails to service its debt, lenders can collect directly from the members. The social pressure to comply with cooperative norms is stronger under mutual liability arrangements. The probability of association default is thereby reduced substantially. Banks and other lenders evidently recognize the advantage inherent in this arrangement, as credit is often conditioned on renewal of mutual liabilities. 4. Weaknesses Most crucial of the potential structural weaknesses of cooperative credit in the moshav is the possibility of moral hazard behavior. Under mutual liability arrangements, the social cost of a member's default exceeds private cost. Members may purchase inputs through the cooperative and then market the produce privately, depriving the moshav of the ability to collect the accumulated debt; or a member may simply cut all his economic ties to the cooperative, perhaps without directly repudiating his debt, but not sharing any further the village obligations. Such behavior reduces the moshav's credit worthiness; the damage to the individual member is only 1/n of the damage caused to the whole village with its n members. 12 Enforcement of the moshav's ethics; and in practice--enforcement of central marketing and channeling all the proceeds through the association is therefore critical to its continued functioning as a comprehensive cooperative. In some cases, technology or administrative consideration force cooperative marketing but in many other cases-- vegetables, fruits, eggs, poultry, beef--members can easily market privately, bypassing the cooperative and its obligations. In a moshav with high ethical standards, social pressure will ensure compliance with the cooperative moral code. But where ethics is more relaxed, when some members demonstrate that they can "get away" with a breach of the cooperative contract, others are tempted to follow suit. . Decision makers are often reluctant to take punitive actions against defaulting members: the penalty may have to be imposed on a neighbor, a close relative, or a potential voter whose support may be needed later. Cost Allocation As we have seen, credit transferred by the moshav to its members can be divided into two parts: Credit specified by the lender for a particular project or a particular member (Settlement Department and government development credit is often so specified), supplier credit attached to a factor, or credit from marketing agents for specified products--is transferred by the association back to back at the same terms it was received. Other sources of credit raised by the association are pooled and transferred to members in the form of drawing rights. The association faces an upward sloping supply curve of credit for its pooled resources. Typical steps in this credit supply function (in ascending order) may. be Medium term bank loans Current bank drawing rights Members' credit balances Bank overdraft Regional supply cooperative Other open market credit. 13 These steps raise the issue of the cost allocation rule: what interest rate or rates should the association charge its members? (In a moshav this will be a single, uniform rate regardless of risk differentials, if such exist.) Economically, the optimal rule is to charge members the marginal rate: the association will tap credit to sources according to members' demand and the association's readiness borrow, and charge the highest rate it has to pay. The same rate will also be paid on members' credit balances (thus determining the ranking of this step in the supply function). This rule may cause some difficulties: (1) How to distribute surpluses? (what side payments?) (2) A technical difficulty may arise if the marginal rate varies over time and the association has to keep track of members' loans and deposits and charge their accounts differently on a day by day basis. (3) The moshav is governed democratically, and it may be hard to convince the members to adopt marginal rate charges. The marginal rate is usually viewed as "too high" even if coupled with redistribution of surpluses. As a consequence, the practice in the moshavim has been to charge the average rate of interest. Since most of the non-specified, pooled credit came from the regional supply cooperative, which was often also the most expensive source for the moshav, average charging was not far from optimum. C. REGIONAL COOPERATION4 Regional cooperation started on a modest scale and developed into a set of "empires" with a large number of enterprises around the country serving kibbutzim, moshavim, and private farmers. The regionals have the played a major role in the development and growth of agriculture in last 20 years, but they were also among the main causes of the current crisis. 'PThis section draws on Zusman (1988, Chapter K) and on Cohen (1984), Kadishay (1985), and Kislev, Marvid (1988). 14 1. Secondary Cooperation In terms of .a modern economy, the moshav is a small unit. To increase scale of operation in farm services, the moshavim established secondary regional cooperative ventures. These include regional supply ("purchasing") cooperatives and a variety of service establishments such as fruit and vegetable sorting and packing operations, storage facilities, poultry processing and packing plants, machinery maintenance centers, and feedstuff elevators. Some ventures with little or no connection to agriculture were also established to increase employment and income in the regions. The regionals, both the supply cooperatives and the service enterprises, are associations of associations owned and operated by participating moshavim; individual farmers are not directly members in these second order cooperatives. The regional supply cooperatives are almost always the supplier of inputs to the service enterprises, often they are also partners in the ownership of the regional ventures. Like the moshav associations, the supply cooperatives are also active in financial intermediation--channeling credit to the moshavim and to the service enterprises in their regions. The mode of operation of the supply cooperative is similar to that of the moshav association: it extends credit to the moshavim, mostly in conjunction with the supply of inputs; it relies on mutual liabilities; it requires all its members -tcYchannel- the marketing proceeds through its offices to facilitate debt monitoring and repayment. But size, complexity, and organizational remoteness create some fundamental control problems which, while negligible in the context of the moshav association, attain significant proportions in higher order cooperatives. 2. Roles of Regional Cooperatives At the time of their establishment, the goal of the regional supply cooperatives was to pool together the purchasing power of several moshav associations in order to strengthen their bargaining position in the markets and to ensure reliable deliveries of farm inputs and 15 household goods. Relying on mutual liabilities of their members, the supply cooperatives became major instruments for raising credit--initially it was chiefly factor suppliers' credit, then the government began channeling directed short term loans through the regionals, and as the apparent stability of the regionals was established, commercial banks also became willing and eager to offer credit. Financial intermediation thus expanded over the years. Serving as a focal point for regional activity with an improved access to credit, the supply cooperatives assumed a leading role in initiating and operating the regional enterprises and other cooperative ventures. In addition, the supply cooperatives became centers for political, particularly lobbying, activity and as a result regional public service became the accepted route to national politics. The economic and the political roles of the supply cooperatives reinforce each other; economics provides a firm basis for political operation, and political lobbying promotes the economic interests of the moshavim and their regional activity. 3. Financial Structure Like the moshav balance sheet, that of the supply cooperative also reflects its role as a financial intermediary. In Table 3, 60.9 percent of the assets of the regional organization is members' debt and an additional 18.3 percent is debt incurred by regional service and production enterprises. The sources of short term credit are commercial banks, suppliers, and gray market lenders (Section D). Up to the end of the 1970s, the supply cooperatives had relied mostly on short term credit. But then, as part of policy reforms, the• government curtailed its supply of directed short term loans, and banks were also constrained in the expansion of credit. At the same time, the banks were permitted to raise capital on the stock exchange and on the *world markets. They began offering medium and longer term loans linked to the exchange rate. This was the major source of the long term debt (19.5 percent in Table 3) which in 1981 was a relatively recent phenomenon. Unlike other sources of long term credit, mostly development assistance, these loans were not tied to specific users or 16 TABLE 3 BALANCE SHEET COMPOSITION OF A SUPPLY COOPERATIVE, SEPTEMBER 30, 1961 (in percent of total assets) ASSETS LIABILITIES AND MEMBERS' EQUITY Fixed assets 3.5 Equity 3.0 Long-term investments Long-term debt 19.5 and loans 13.7 22.5 17.2 Short-term loans 34.5 Receivables: members 60.9 Suppliers' credit 21.8 regional enterprises 18.3 marketing agents 3.6 Notes payable 21.2 Total current Total current assets 82.8 liabilities 77.5 Total 100.0 Total 100.0 Source: Kadishay (1985). Note: The data are for the regional supply cooperative in the "Mountain Region." projects and were therefore not transferred back to back to the individual moshavim. The shift to long term debt, which may have been more expensive than unindexed short term credit, reduced the need to secure continuously large volumes of new credit, volumes that grew at least in their nominal value as inflation accelerated. Again as in the moshavim, due to inflation and the historical accounting practices, the financial statements underestimate fixed assets in the supply cooperatives. But even if adjusted, they will not occupy an important place in the balance sheet. For the supply cooperative acts as an intermediary, a traffic control in factors, consumer goods, products and credit; and apart from some office space, equipment, and a few cars, it does not hold property. Similarly, • equity is understated but it will not become a substantial source of 17 capital even if adjusted for the effect of inflation. Not all the liabilities of the supply cooperative are reported in its balance sheet. In a study of Mishorim, another supply organization (Kislev, Marvid 1988) it was found that, in addition to extending outright credit, the cooperative guaranteed loans and co-signed notes for the moshavim and the regional enterprises. The sum of these off- balance-sheet liabilities amounted, also in 1981, to 64 percent of the cooperative's reported liabilities. Credit to its moshavim was the principal item among the supply cooperative's assets (60.9 percent in Table 3). The regional cooperative also occupies a prominent place in the economy of its member moshavim: 76.9 percent of the moshav liabilities in Table 2 is debt to the Mountain Region cooperative. This ratio, however, varies over time and with the nature of the cooperative and the participating moshavim. Figure 2 depicts the ratio of debt to the supply cooperative to total current debt over several years for two groups of moshavim: Mountain Region and Valley moshavim. The first group increased its reliance on the regional cooperative between 1977 and 1981; the second, if anything, reduced it. Valley moshavim are veteran and stronger economically than the moshavim in the Mountain Region, their cooperatives--both at the moshav and at the regional level--are also healthier. Evidently they directed their comparative advantage to strengthen their cooperatives independently--not to increase inter- reliance. The financial dependency of the Mountain Region moshavim on their supply cooperative did not increase the regional's control over its members. On the contrary, it reduced control: because the moshavim in that region had only limited access to alternative sources of credit, the supply cooperative was committed to their continuous funding and the managers of the regional organization could hardly afford the dire financial, social, and political consequences of members' bankruptcies. This is borne out by the following observation. A reasonable rough measure of the financial exposure of a member moshav is the ratio of the outstanding debt to its monthly sales through the regional office ("credit months"). Kadishay (1985) found 18 Figure 2. Reliance of Moshaviin on Supply Cooperatives urrent moahav debt to supply cooperative that, during the period 1977-1981 13 out of 24 observed moshavim in the Mountain Region exceeded 12 credit months and for relatively long durations. By contrast, only 3 out of 20 moshavim observed by him in the Valley Region exceeded 12 credit months and that too only for short durations. Thus, while the "strength" of the supply cooperative's power over the individual moshav association is considerable, the "cost" of its application is also very high. The financial power of the supply cooperative over its members, the moshav associations, is thereby blunted. As a result, heavy financial inter-reliance of first and second order cooperatives led to moral hazard behavior in many of the borrowing associations and was one of the major factors that exacerbated the severity of the current crisis. The alarm signals were building up for a long time and are worth examining further. Evidence comes from the third regional organization we are studying. Data for Mishorim, a supply cooperative with young moshavim similar to those of the Mountain Region, show that financial exposure is affected strongly by economic performance of the members of the For a relevant theory of social power, see Harsanyi (1962) 19 TABLE CREDIT MONTHS, MOSHAVIM'S DEBT TO MISHORIM RELATIVE TO SALES 1970 12.1 1974 5.9 1978 6.0 , 1971 9.9 1975 5.3 1979 9.0 , 1972 7.5 1976 4.6 1980 10.2 1973 5.9 1977 5.7 1981 10.7 Source: Kislev, Marvid (1988). Note: _Credit months = (average annual debt of moshavim to Mishorim)/(annual moshavim's sales through Mishorim)x12. regional organization. Mishorim's credit months ratio (Table 4) improved significantly in the mid 1970s with the expansion of export crops, in which the region specializes, but evidently the moshavim found it convenient to enhance their borrowing in the regional organization in 1979-1981 when agricultural terms of trade deteriorated with the weakening of the European currencies against the dollar and the region's production was reduced. Differences are also observed between moshavim in the same region. In a sample of 13 moshavim, members of Mishorim, one moshav owed [I nothing to the supply cooperative--it did all its purchasing for cash. Another held the record for 1981 with 55 credit months. The next in line had 42 credit months. There is no way that moshavim with such heavy burden of debt compared to their production capacity will ever be able to repay it. Heavy debt burden does not happen overnight, it evolves gradually as the moshav purchases factors and consumer goods in the regional cooperative and either fails to reach reasonable levels of agricultural production, or diverts its produce to private channels, bypassing the regional offices. Moshavim with 55 or 42 credit months testify to the weakness of the regional organization, a weakness that breeds permissiveness and lax financial discipline. 20 4. Regional Enterprises Regional service enterprises were ordinarily organized as limited liability cooperative associations. Their membership consisted of moshav associations, mostly potential patrons of the service offered. The establishment of the regional enterprises was financed mainly by development grants ,and soft loans from the government and the Settlement Department; the rest was financed by small amounts of equity capital and loans from a variety of sources, including supply cooperatives. In addition, the supply cooperative became naturally the agent providing the regional enterprises with factors and credit; in many cases, it also invested directly in equity capital. The two kinds of regionals, the supply cooperatives and the regional enterprises, have thus maintained strong economic connections; a relationship that proved detrimental when the recent financial crisis developed. Zealous adherence to rural development by public agencies, easy access to credit through the supply cooperatives, and strong political regional lobbies--resulted in over-expansion of service enterprises. This occurred particulary in the 1970s when credit was in ample supply and economic optimism ran high. Consequently, in the early 1980s many service enterprises operated under capacity. Estimated capacity utilization for 1982-86 in slaughter houses and feed mills run by regional enterprises was SO - 60 percent (Marvid, Arian 1988). Moreover, this was a conservative estimate, based on the assumption of one shift a day in the slaughter houses and two shifts in the feed mills. True capacity utilization was much lower. Not unlike firms in a cartel, regionals scrambled to get their share in the service enterprises and the expected benefits they were to yield. The final outcome, however, was that many of the enterprises did not cover their operating costs. The supply cooperatives, in their role as the financiers of last resort, found themselves financing not only operating losses but also debt service of the regional enterprises. To see how this happened, consider the feed mill (Mazon) and the slaughter house (Off) in the Mishorim region (Table 5). Both are owned by the moshavim in the region and initially Mishorim, the supply cooperative, had no share in 21 their ownership. Plans for the establishment of both enterprises were initiated in the early 1970s; the aim was to provide farmers in the region, particularly poultry growers, with convenient, reliable service. In terms of cost, the establishments were of similar size, investment was $9.7 and $10.9 million in Mazon and Off, respectively. Construction of Mazon began in 1976 and of Off in 1977; by 1981 both enterprises were in operation. Differences in the fate of the enterprises emerged right from the beginning. Financing in the early stages of investment was with concessionary unlinked credit, but in 1978 the government modified its credit policy and new development loans were price indexed or dollar linked. Late financing, and particularly budget overruns, were expensive. Off, being constructed one phase behind Mazon, had to rely comparatively more on linked loans. Mazon also enjoyed a government grant of one million dollars, which was not available to Off. By 1981, Mazon accumulated inflationary capital gains of $4.9 million on its unlinked debt and a negligible operating surplus; Off accumulated only $2.4 millions in inflationary capital gains, and had $1.5 million in operating losses in its first year of production. As a result, in September 1981 the enterprises emerged with equity (including retained earnings) of $8.6 and $2.9 million for Mazon and Off, respectively. In addition, in 1981 both enterprises serviced their past debt and accepted new short term obligations, most of which was suppliers' credit passed-over to their own customers. Outstanding debt at the end of 1981 was $3.7 and $9.4 million, respectively. By 1982, Off also succeeded in balancing its costs and revenues, but. with no surpluses. Both enterprises had to raise additional loans to service their debt. Most of the loans were short term, borrowed from the regional supply cooperative. Mazon borrowed $200,000, and Off $3.3 million. No figures were available beyond 1982, and costs, revenues, and financial flows were estimated using information on volume of operations, equipment renewal, going rate of interest, and repayment schedules of past debt. By these estimates, at the end of 1985, the debt of Mazon was $4.2 million of which $2.7 million was short term, 22 TABLE 5 MAZON AND OFF • TWO REGIONAL ENTERPRISES IN MISHORIM REGIONAL COOPERATIVE (in millions of US $, September 1981)4A MAZON OFF Investment and finance Equity (reported) 2.7 2.0 Grants 1.0 Long-term debt Indexed 1.4 5.2 Not indexed 4.6 3.7 Total investment 9.7 10.9 Changes during construction period Inflationary capital gains 4.9 2.4 Operating surplus 0.03 -1.5 Total 4.9 0.9 Equity, adjusted to include grants and inflationary capital gains, Sept. 1981 8.6 2.9 Debt, Sept. 1981 Long-term Indexed 1.9 5.6 Not indexed 0.5 0.3 Short-term 1.3 3.5 Total debt 3.7 9.4 Financial flows, 1982 Uses of funds Operating expenses 9.0 10.2 Debt service (interest+repayment of principal) 0.5 3.4 Total uses 9.5 13.6 Sources of funds Sales revenue 9.3 10.3 Increase of short-term loans 0.2 3.3 Total sources 9.5 13.6 Debt, September 1985 (estimated) Long-term 1.5 4.3 Short-term 2.7 19.9 Total 4.2 24.2 Source: Kislev, Marvid (1988). Notes: ) Agricultural years in Israel are from October to September. 23 mostly to Mishorim. The corresponding numbers for Off were $24.2 and $19.9 million. The differences in the fate of Mazon and Off are typical of many investment projects in recent years. The key word is timing. Producers (and households) that were lucky enough to invest in the early stages of accelerating inflation saw their debt evaporate. Latecomers who, relying on recent experience accepted projects whose feasibility depended on negative cost of capital, ended up with enormous debt burden. However, earlier construction did not rescue Mazon either. With over-capacity in feed mills in the country, Mazon could not cover both the operating costs and its comparatively modest debt service. Thus two enterprises constructed with the initiative of the economic regional elite and with the assistance of public agencies striving to promote economic development and well being became a drain on the economy of the region. So long as Mishorim could finance Mazon and Off, even if at relatively high rates of interest, farmers in the region's moshavim did not feel the burden (particularly that financial reporting was always late and, being prepared in historical values, was considered irrelevant). But once further finance was not forthcoming, Mishorim's members found that they were being asked to foot the bill-- they, through their moshavim, signed mutual liabilities guaranteeing loans to Mazon and Off. D. CREDIT MARKETS AND INSTITUTIONS Israel has fairly developed money and capital markets. The markets are strongly dominated by government intervention, but most of the financial transactions are handled by commercial banks, supplemented by a well-developed securities exchange. 24 1. The Banking Systeme' Commercial banking in Israel is widely accessible and at the same time highly centralized. There are 29 banks, but the industry is dominated by 5 banking groups which together account for 94 percent of all bank assets. More than 1000 bank branches offer competing service in all parts of the country. The commercial banks also act as major securities investment brokers and most of the highly developed mutual funds industry is owned and managed by the banks. Agriculture is additionally served by two specialized banks whose major activity is channelinginvestment funds. They account however for only 15 percent of total bank credit to agriculture. In the early 1950s, the number of cooperative credit societies in Israel exceeded by far the number of commercial banks. But they found it hard to compete with the banks and in the long run had to fold, many absorbed by Bank Ha'Poalim, the labor movement's bank. By now, despite Israel's deep rooted cooperative tradition, there are only two cooperative financial institutions in operation, and both are small, local, and hardly known to the public. 2. Bonds" Through most of its history the securities market in Israel was dominated by bonds, which since 1955 have all been linked to the price level or to foreign exchange. Linkage allowed the market to survive the vagaries of inflation. Also for most of its history, the bond market was dominated by the government which first mobilized resources by borrowing from the public--sometimes by fiat, but more often by open market operations--and then had to roll over the internal debt thus created. Government dominance crowded out private capital raising, and to keep the cost low, the government also limited private bond issues. Only a few financial institutions and utilities were permitted to issue d'This subsection draws on the publications of the Supervisor 'o Banks and on Heth (1966). -7For the history of Israel's securities market, see Sarnat (1966) and Levy, Smith, Sarnat (1978). 25 bonds and this also only to raise capital for government approved of projects. Consequently the government is also the biggest lender long term funds. •The situation has changed when a "general permit" was enacted in 1986 enabling all corporations to issue bonds and use the proceeds without seeking specific approval. As a result, linked bond 1984 issues of non-government corporations were in 1987 ten times their level. As part of the liberalized bond market, two major kibbutz to movements also raised $75 million in four 12-year bond issues linked and the consumer price index, with the proceeds intended to help first- second-level agricultural cooperatives roll over and convert their burdensome short term debt. 3. Stocks The market for equity issues in Israel began its accelerated development in 1959, in response to legislation encouraging investment in stocks. Since then the market has had its ups and downs, following policy measures and changes in the economic environment. When the market was bullish, new corporations went public and old public companies came out with new issues. Equity value exceeded bond value for the first time in 1977. Agriculture's representation on the stock market has always been modest. Only 4 citrus growers, 8 food processors, and the 2 the 250 agricultural investment banks have publicly traded stock out of No odd corporations listed on the Tel-Aviv Stock Exchange in 1987. not cooperative has ever raised equity on the stock market. This is - incidental; cooperatives, with their member-oriented and patronage based policies, are usually not built to have two kinds of stock and the holders, members and non-members. Some cooperatives in the UK USA have recently started raising new equity capital through stock measures. issues to the public, introducing appropriate organizational Probably, so long as credit was ample and cheap the need for such reforms was not felt in the Israeli cooperative sector. In the secondary market, farmers and farm organizations invested bonds. savings and temporary surpluses in securities, both stocks and 1980s; and Stocks were particularly popular in the late 1970s and early 26 when the market crashed in 1983, it was revealed that at least one kibbutz movement suffered a substantial loss. Up to 1982, moshavim could borrow at nominal interest rates that reflected inflation, write the cost off as a tax deductible expense, and let their members invest in stocks whose capital gains were tax exempt. Many in the moshavim held therefore bank shares that prior to the crash were considered safe and highly profitable investment. No data are available on the volume of these investments or on the associated profits and losses. 4. Other Markets and Sources of Credit Gray, non-bank credit markets have always operated in various guises in Israel. These markets channelled consumer and business savings to borrowers, bypassing liquidity, and interest regulations of the commercial banks. Part of this activity was open: insurance companies, for example, used to hold more than 10 percent of their assets in the gray market. It is not known to what extent agriculture was active in the gray market, but the involvement of the kibbutzim received some publicity recently when one of the movements suffered a considerable loss through a gray market intermediary and the case ended in court. Folklore has it that kibbutzim and moshavim were lending to each other on the gray market without being aware of the identity of the other party to the deal. Other sources of credit--as reflected in the cooperatives' balance sheets--are suppliers, marketing agents, government directed credit, and, last but not least, members' deposits. The clearing house activity of village and regional cooperatives (Tables 2, 3) has been a significant component of the agricultural segment of the credit market in Israel. Members in moshavim attempting to bypass their cooperative associations generally found it easy to get private wholesalers to finance working capital needs. Evidently, these creditors could secure their loans by methods that the moshav or the regional cooperative could not afford to follow. 27 E. CREDIT IN THE ECONOMY AND IN AGRICULTURE Inadequacy of "suitable" sources of credit has often been blamed in Israel for the financial difficulties of agriculture, particularly in cooperative farming. Contrary to this popular claim, the findings of the survey, presented in this section, on the development of bank credit in the economy and in agriculture suggest ample supply, mostly at low cost. 1. Growth of Credit From 1969 and up to 1979, outstanding bank credit to agriculture grew at the same pace as credit to other sectors in the economy (Table 6). But since 1980, agricultural credit has been growing faster than credit to industry. Despite the deceleration of inflation, credit continued to grow after 1985, and by the end of 1987 real credit to agriculture was 33 percent higher than its volume at the end of 1984; the corresponding economy-wide increase was 25 percent. Credit did not grow only in absolute terms, it also grew relative to product and capital (Table 7). In the economy at large, debt to product ratio increased from 0.30 in 1969 to 0.67 in 1986. Reported product in agriculture is not exactly comparable to product as calculated in GNP, but rates of growth are: debt to product ratio increased in agriculture between the 1969 and 1986 by a factor of 3.8, much more than the economy's average. Another measure of debt intensity is the ratio of credit to net capital. While for industry the ratio hovered around 0.7 since 1969, for agriculture it grew steadily from less than 0.20 in 1969 to nearly 0.80 in 1986. The share of the moshavim in agricultural credit grew by 40 percent from 1971 to 1986 (Table 8). Multiply this by the growth of credit to agriculture (Table 6) and one gets that real credit to the moshavim grew 7 fold between 1970 and 1986. And this calculation is only for credit received by moshavim directly from commercial banks, it does not cover growth of credit channeled through second order cooperatives. 7 28 TABLE 6 OUTSTANDING BANK CREDIT: TOTAL ECONOMY, AGRICULTURE, INDUSTRY (Selected years, end of year balances in real values) In percent of 1969 credit • In percent of total credit total credit to credit to credit to credit to Year credit agriculture industry agriculture industry 1970 138 136 124 10 32 1974 236 209 219 9 33 1979 351 335 327 10 33 1984 427 491 352 12 29 1987 535 655 402 13 27 Source: Annual Bank Statistics, various issues. TABLE 7 RATIO OF OUTSTANDING DEBT TO GROSS PRODUCT AND NET CAPITAL (Selected years, end of year data, real values) credit to total agriculture, credit to credit to Year credit, % of agricul- agriculture industry % of gnp tural product % of net capital 1969 31 48 19 52 1974 48 76 35 69 1979 57 101 48 75 1984 57 177 67 70 1986 67 182 79 64 Source: Credit balances from Annual Bank Statistics, various issues. GNP and agricultural product from Statistical Abstract, various issues. Capital stock data from Buch (1984) and privately communicated updates TABLE 8 SHARE OF AGRICULTURAL CREDIT HELD BY MOSHAVIM AND SUPPLY COOPERATIVES (Selected years, end of year data, current values) supply moshavim cooperatives Year percent percent 1971 10 14 1974 6 23 1979 10 20 1984 15 26 1986 14 28 Source: Bank Credit, various issues, and Current Bank Statistics, Bank of Israel, Supervisor of Banks, Jerusalem. 29 TABLE 9 DISTRIBUTION OF OUTSTANDING BANK CREDIT BY SOURCE (in percent end of year data) government short-term free directed directed investment earmarked YEAR credit credit credit loans loans TOTAL CREDIT 1971 38 62 18 14 30 1974 38 62 16 13 33 1979 29 71 22 7 42 1984 28 72 14 10 49 1986 35 65 . 10 9 47 AGRICULTURAL CREDIT 1971 17 83 36 33 14 1974 16 84 33 37 15 1979 23 77 28 13 36 1984 12 88 13 16 59 1986 23 77 9 12 56 Source: Annual Bank Statistics, various issues. 2. Sources Thirty to forty percent of the credit in Israel is free credit supplied by banks from their own sources and distributed to borrowers at the banks' discretion (Table 9). The rest is under government control--either originating from the government budget and the central bank or from bond issues and deposits administered by commercial banks but designated as funds for earmarked, government approved projects. Public involvement in credit supply to agriculture is even larger, with more than 80 percent government directed. The decrease in the share of investment loans and the corresponding increase of the share of earmarked credit apparent from Table 9 is a change in form more than a change in substance. The decrease in the share of short term directed credit, however, is real: it reflects gradual withdrawal of the government from the financing of working capital except for the financing of production for export and export shipments. 30 TABLE 10 THE DOLLAR-LINKED COMPONENT OF OUTSTANDING BANK CREDIT (in percent of each source, end of year data) All Sources Free Credit total total Year economy agriculture industry economy agriculture industry 1971 35 7 43 27 1 14 1974 42 15 54 54 37 46 1979 51 39 67 68 68 64 1984 51 44 67 60 39 46 1986 36 28 54 30 9 22 Short-Term Directed Earmarked Loans total total Year economy agriculture industry economy agriculture industry 1971 52 9 63 NA NA NA 1974 41 4 60 NA NA NA 1979 79 51 84 24 11 19 1.984 91 85 91 34 36 45 1986 100 100 100 27 24 39 Source: Annual Bank Statistics, various issues. 3. Dollar Linkage Foreign-currency (primarily US-dollar-denominated) loans have been used in Israel for a long time as an alternative form of linkage side by side with consumer-price indexation. Foreign currency did not exchange hands in these transactions (except for credit for imports): it was credit in domestic currency at values linked to the exchange rate. Linkage, either to foreign currency or to the domestic price level, reduces uncertainty associated with fluctuating inflation. When inflation reaches the two digit level, every day counts. Foreign currency linkage is more convenient than indexation for short term debt, because the exchange rate is known ona daily basis while the price index is measured only once a month in Israel and published with 31 a 2 week delay. Despite obvious advantages, dollar linkage of short term obligations was introduced only gradually (Table 10). Linkage of free credit was almost completely eliminated after July 1985 when the confidence in the stability of the shekel had been restored. This move, however, was costly to the borrowers: free credit carried extremely high real interest rates (see below) while linked credit was usually supplied at "civilized" rates. Restricted to export financing, directed short term credit was 100 percent linked in 1985 and 1986. 4. Term Structure It was repeatedly shown that agriculture relied on short term credit to finance production assets (Lowe 1955; Sadan, Kadishay 1967; Sang Committee 1979). It was also often claimed that the "financing gap" between the term structure of credit and the productive life of assets was the result of government policy of encouraging capital formation while supplying too little long term credit. Investment can be financed by short term credit; but then the credit may have to be rolled over for the length of the economic life of the new asset. Credit of adequate maturity reduces uncertainty as to the availability and cost of future sources of finance. Future availability was perceived to pose no problem and cost was low, even negative, for long periods of time. Farmers in the moshavim and their associations may have been tempted to finance investments with short term credit. Moreover, they were then saving the paper work and the time needed for the administrative approval of long-term development loans. Looking at the issue from another perspective, the moshav is an ongoing concern with many assets and liabilities. The association manages a portfolio of liabilities similar to that of assets and adjusts its composition as relative prices change. When short term a credit is comparatively cheap, the moshav's balance sheet will show wide financing gap. The danger is that when circumstances change and it short term credit becomes scarce and expensive, those who relied on may find that the cost of capital is• much too high for the projects they promoted. Many in agriculture enjoyed cheap credit at its time, 32 and many were caught with inappropriate term structure and paid dearly for it. In some cases, circumstances beyond the control of the moshavim or their farmers induced comparatively heavy reliance on short term credit and the responsibility, at least partly, could be the government's: Development loans were often delayed, mostly due to bureaucratic sluggishness. Investors had to turn to short term credit for intermediate financing; in periods of inflation, delayed credit created not only temporary but also permanent gaps in financing due to the erosion in the real value of the loans which were late to arrive. The government recognized the problem and even requested the supply cooperatives to bridge the gaps. Inflation and policy introduced yet another distortion. Soft loans were often offered by the government at nominal interest rates only partly adjusted for inflation. Such rates mean (a) negative real cost of debt, and (b) early repayment of the principal. As an example, consider a two year loan of 100 dollars at zero real interest, the yearly rate of inflation is 100 percent. There are two alternative payment schedules, and in both the principal is repaid at the end of the second year: (a) a 70 percent nominal rate of interest, paid at the end of the first and of the second year; (b) the loan is linked to the price index and the only payment is at the end of the second year. Debt service will then be as follows Alternative a Alternative b End of first year 70 End of second year 170 400 Present value at the beginning of the first year (discounted at 100 percent) 77.5 100 The non-linked loan carries a nominal interest lower than the rate of • inflation and involves a grant of 22.5 dollars. But it requires, in the example, repayment of 35 percent of the principal at the end of the first year. In this sense investors may be forced to short term financing of their projects--loans carrying nominal interest rates are 33 of shorter duration than their stated maturity. Whatever the origin of the financing gap, as cost or availability of credit changed, farmers turned to the government and to the Settlement Department for help, usually with the appropriate political backing. There were many cases, in recent periods almost one each year, of "conversion"--rescheduling of loans: short term credit was replaced by long term loans, mostly on concessionary terms. Many of the conversions were specifically oriented to certain groups of farms or regions, others were more general. Their recurrence was one of the major reasons for the general belief that agriculture would not be allowed to collapse. The remedy however was not always effective. Sadan and Kadishay (1967) report a case of debt rescheduling in 1956 covering 72 kibbutzim. Within 3 years all but two of the kibbutzim were back were they started in terms of the maturity structure of their debt. The State Controller (1986) reports a similar experience for a group of moshavim. Despite the lessons, today too, the prescription for the current crisis in agriculture involves large scale debt refinancing. 5. Cost of Debt Perhaps the greatest damage that inflation inflicted on the Israeli economy was the distortion of the cost of capital. Real interest rates varied markedly, cost of borrowing was at times very high while at other times and for other loans it was negative. This is not a recent phenomenon. Interest rates have always lagged behind inflation, and rates of bureaucratically administered loans lagged further and longer behind price changes. But apart from a two year burst in the early 1950s, up to 1974 inflation was lower than 20 percent per year. The distortions introduced were small compared with those reached in the second half of the 1970s (Figure 3). More detailed data than those summarized in Figure 3 are available for the last 10 years (Table 11). In this period, real interest rates were not only too low, they were also, particularly recently, too high. Real rate of interest on overdraft facilities reached the 100 percent mark in 1985. A year earlier the real rate on short term directed 34 Figure 3. Interest Rates and. Inflation 50 52 54 58 58 81 83 85 87 89 71 73 75 ,77 79 Free Credit ST Directed -4— Investment CZI Inflation credit in domestic currency was minus 59 percent. The rate of interest on investment credit was minus 25 percent in 1978 but credit from this source was then linked and its real cost became positive. Still, the cost was negative for tax paying enterprises since linkage charges were treated as tax-deductible financing expenses. For many of those, linking of investment loans reduced and did not increase the cost of credit This distortion, at the time unexpected, was eliminated in 1982 with the introduction of inflation adjusted income tax regulations which basically recognized only real interest expenses as tax-deductible. The information in Table 11 is fragmentary and collected from incomplete statistics. To double check, we consider the average cost of debt from all sources calculated from sample balance sheets. The real effective cost of outstanding debt of one of the largest industrial corporations in Israel was around 20 percent (before-tax) in the 1964-1970 period. In 1978, with accelerating inflation, the real effective cost of debt of this corporation plunged to minus 20 percent (again before-tax), reflecting the erosion of the massive .financing from non-linked government-directed sources (Levy 1979). For a sample of 13 industrial corporations, the real effective cost of all loans declined from minus 6 percent in 1971 to minus 40 in 1980 (Levy and Lerman 1987). 35 TABLE 11 REAL ANNUAL COST OF CREDIT IN ISRAEL, 1979-1987 (PERCENT PER YEAR) Year Overdraft Free bank Short-term All short- All short- Long-term facili- credit, . directed term term investment ties(1) short- credit in directed credit(4) credit term(2) domestic credit(3) currency 1978 2 NA -24 NA -11 -25 1979 -11 NA -45 NA -10 -26 1980 19 NA -39 NA 1 -9 1981 34 NA -28 NA 17 -6 1982 5 NA -38 NA 4 -2 1983 -3 NA -44 NA 8.5 0.5 1984 67 40 -S9 6.3 26 0.0 1985 100 •36 -23 0.9 20 2.5 1986 31 16 -3 -9.0 7 1.6 1987 39 27 +3 -0.6 20 0.9 Source: Bank of Israel, Annual Report, 1980--1987 issues. Notes: (1) Overdraft facilities are the most expensive source of free bank credit: this represents the marginal cost of credit after all the other sources have been exhausted. (2) Free bank credit includes the overdraft facilities as one of its components, but the average cost of the entire mix is much lower. (3) The weighted average of short-term directed credit in domestic and foreign currency (mainly US dollar-linked). (4) The weighted average cost of all sources of short-term credit. Preliminary results obtained for eight agricultural cooperatives (four moshavim and four regionals) reveal a similar picture (Figure 4).e The real effective cost of outstanding loans from all sources became progressively more negative between 1971 and 1984. In 1983- 1984, with inflation rates rising to 300-400 percent per annum, the nominal average cost of debt soared, but still not fast enough, so that the real cost of debt reached all-time lows of minus 30 and minus 50 percent in 1984, for the regionals and the individual moshavim, e)The time development of the cost of debt of the 4 cooperatives in each group was almost identioal; the figure depicts average costs. 36 • Figure 4. Real Cost of Debt: Averages for Four Irgunim and Four Moshavim 190 50 irgunim moshavim —50 —100 1 , 1 i 1 1 1 1971 1973 1975 1977 1979 1981 • .1983 1985 1987 respectively. But then inflation slowed down by government policy, and nominal rates were caught on their upward path and overshot, producing real effective cost of debt of 16 and 26 percent in 1985. They returned, for these cooperatives, to lower levels in 1986. 6. Provision for Bad Debts Here is another facet of inflationary distortions: provisions for bad debts are usually around 1 percent of total credit balances. In Israel in the galloping inflation years of 1975-83, writeoffs for questionable debts in the banking system were an insignificant amount, much lower than the normal 1 percent value. Inflation and the ample supply of credit it entailed bestowed a feeling of economic security and success even where the underlying structure was shaky. Writeoffs increased in 1981-84 in reaction to difficulties in the diamond industry, but even then they were only 0.1 - 0.3 percent of credit balances. Substantial provisions for bad debts appeared in the banks only after 1985, when many enterprises, not only in agriculture, suddenly faced sever financial difficulties. 15"The data in this subsection are from the publications of the Supervisor of Banks. 37 Bankers were familiar with over capacity in agriculture, with debts f accumulated to levels that some of the farmers and some of the moshavim could not be expected to repay, and with the lax financial discipline in the cooperatives--at both the village and the regional level. And yet they did not make provisions for bad debts in agriculture probably because they were fooled by the appearance of affluence that their own credit created and they expected to be bailed out with public money if the need ever arose. Had bankers been suspicious of the soundness of agricultural loans prior to 1985, they would have increased the allowance for this debt at that time. But, given their familiarity with cooperative financial structure, the arousal of such a suspicion would have caused them to curtail further credit to agriculture and thus precipitate an immediate crisis--as actually happened in 1985. F. GOVERNMENT Agricultural settlement in Israel was, from its inception, a public undertaking. Most of the newly settled farmers had no capital of their own and were provided with land, means of production, and some working capital from national funds. The Settlement Department had been responsible for agriculture before the establishment of the State of Israel, and after 1948 it continued with newly settled farms. Government's deep involvement was therefore a natural taking over of responsibilities. This involvement, particularly in credit, was augmented by the government being the channel for large amounts of capital transfers from abroad--Israel probably holds the world record in terms of per capita foreign assistance. Israel was governed, from 1948 through 1977, by coalitions with Labor at their center. These governments favored agriculture and 38 cooperation.10 Thus in the early 1950s, when population was doubled in 3 years and tripled in 13--mostly due to immigration--settlement in cooperative farms was seen as one of the major solutions for unemployment and for food shortage. More than 250 young moshavim were then added to the 77 veterans which had been established before the creation of the state in 1948. The settlers in the new moshavim were as a rule unfamiliar with agriculture and with cooperation; they were sometimes brought to the moshav directly from the port of entry, often without realizing where and to what kind of life they were going. New settlers were free to leave, many did, those who stayed in their moshavim despite the hardships accompanying any new agricultural venture, remained because of the relatively low alternative income that was then available in town and because of the intensive assistance they were receiving from the Settlement Department and the government-- mainly in capital and close personal guidance. Governments in all developed countries assist and protect agriculture in water supply, subsidies, planning and removal of surpluses, research, and extension. In Israel, in addition to the functions performed by other governments, the special circumstances of the early stages of development created what amounted to fosterage relations between the moshavim and the public agencies that looked after them. These special relations were particularly strong for the moshavim which were settled in agriculturally inferior regions (our Mountain Region of the earlier sections included). Over time many of the operators in the young moshavim acquired farming skills and cooperation also became well established--independence was strengthened. But the view, held not only by farmers, that it was the role of the government to maintain the welfare of the farming sector and the expectation that the government would actually shoulder this responsibility did not wane. In cooperative agriculture the government plays a special role. Partly through the Knesset (parliament) and partly through regulations 1°1-fad Labor maintained its hegemony, agriculture may have been assisted more lavishly and the current crisis postponed. It could not, however, be postponed for ever. 39 enacted by its offices, the government (in the wide sense of the term) is responsible for the laws and regulations of cooperative activity. Two instances of interest to our discussion can be mentioned: (a) Attempts to pass a "law of moshavim," strengthening the power of the association over individual members, including in matters of credit and cooperative marketing, failed because of opposition from some of the members fearing infringement of freedom and the support they received from right-wing parties. (b) A regulation was recently issued that a cooperative cannot force members to participate in covering its losses. The argument is that a cooperative is a limited liability entity and members are responsible only up to the value of their shares (the limitation does not apply to cases of mutual guaranty). A judge already applied the new regulation in one dispute and the case is now on its way to the supreme court. If upheld, it will mean a revolution in the mode of cooperation. The most profound public involvement was, however, in credit. By deciding on the allocation of subsidized credit, the government (we do not always make the distinction between the government and the Settlement Department) affected regional development, lines of production, and farmers' income (Kimhi 1986). The dependency on the government and the expectation that it will bail out farmers and moshavim in trouble created moral hazard problems, not unlike those that mutual guaranty creates in moshavim and regional cooperatives. Investors were willing to take risks that they might have avoided if the danger of bankruptcy and the entailed loss of property and income was real. Lacking the usual mechanism of collateral and effective guaranties, the government turned to close monitoring--"concentrated credit. It worked in the following way: a moshav, or a kibbutz, concentrated all its financial activity in a single bank; its operation was monitored by a steering committee which included representatives of the government, the foster bank, and the movement; credit was supplied, either for investment or for short term needs, only with the approval of the steering committee. Joining concentrated credit was voluntary, moshavim were lured by additional loans. And, indeed, the program, that started in the early 40 1960s, covered in a few years most of the moshavim in the country. However, the increasing credit supply in the 1970s, and particularly the convenience of sources offered by the regional cooperatives, eliminated the advantages of concentrated credit as seen by the moshavim and the program folded in the mid 1970s. The problem of moral hazard in the moshavim was recognized, tools to overcome it, at least partly, were created, but the will to maintain a strict policy could not withstand the flood of available credit (concentrated credit is now proposed again in reaction to the current crisis). Moreover, the government then supported the supply cooperatives and encouraged them to expand their role--they often received directed credit on favorable terms, they were assisted in the establishment of the regional service enterprises, and they were made parties to development loans by advancing short term credit to finance delays in government approved loans. Evidently, despite the experience with credit in the moshavim, the dangerous implications of unconstrained operation of financial intermediaries were not understood or, perhaps they were understood but governments are often slays of circumstances more that they are their masters. G. CRISIS The crisis developed gradually. Sensing difficulties as interest rates rose in the second half of 1985, intermediaries in the gray, market pulled their money out of many agricultural enterprises or asked for rates that even cooperatives in distress were not willing to pay. The banks were evidently not coming forth with enough funds to cover the emerging shortage and regionals were collapsing one by one. But even seasoned bankers need time to recognize a lost cause, particularly if they may be the ones to lose most. For a while banks cushioned the collapse, continuing to supply credit to the regional cooperatives. Suspicion and difficulties mounted, however, and within a few months, 9 of the 12 supply cooperatives of moshavim were insolvent, banks refused to renew credit, demanded loans to be repaid and charged the highest delinquency rates of interest on overdue moneys. With the supply 41 cooperatives in financial difficulties, they could not support the regional service enterprises and many of them also collapsed. Inflation was still running at 15 - 20 percent a year, raising the needed nominal balances. But agriculture--particularly cooperative agriculture--was cut from its sources of additional bank credit and required to honor previous obligations. This was impossible. Panic spread and the accepted view was that production in the kibbutzim and the moshavim was coming to a halt. Political pressure was mounting and at the beginning of 1986 an emergency rescue program was set up: a sum of $242 million was budgeted for the moshavim. The money was given to the supply organizations in the understanding that they will channel it to the moshavim in order to relieve the burden of short term credit. Most of the organizations kept the money, it lightened their financial burden but not directly that of the most distressed moshavim. The time gained was to be used for analysis of the situation and formulation of a comprehensive solution. Two committees were nominated in succession (Gibton Committee 1986; Ravid Committee 1987), with representatives of the major banks participating in both in an effort to reach an acceptable solution. The first committee assessed the severity of the crisis and suggested a set of basic structural changes. By its assessment (Table 12) the liabilities of the sector of the moshavim were $840 million, not including private debts of moshav members to creditors other than village cooperative associations. The estimated farm gate value of production was $870 million and off farm income $140 million. Deducting expenses, the amount left to service the debt was estimated as $120 million a year. This sum, the committee noted, seems as if it could suffice if interest rate were not too high, if most of the debt could be taken as standing, and if high repayments were not required. But the assessment in Table 12 was an aggregate estimate. Some moshavim and a few of the regionals could continue with convenient cash flows, others could not even pay the interest on their debt. The suggested solution involved rescheduling of debt with terms tailored to the repayment ability of each cooperative--the moshav and the supply cooperative. A prerequisite for assistance is the 42 TABLE 12 CONSOLIDATED FINANCIAL POSITION OF THE MOSHAV SECTOR (in millions of US dollars, September 1986) A. LIABILITIES Supply cooperatives Long-term bank loans 190 Short-term bank loans 120 Suppliers' credit 130 Other sources 40 Moshav associations Bank loans 230 Other debtors (incl. gray market and intermoshav debt) 130 Total liabilities of the moshav sector 840 B. SOURCES AND USES'IL Sources: Uses: Farm income (farm gate Farm expenses (incl. hired value) 870 labor) 490 Non-farm income" 140 Living expenses" 230 Depreciation of production assets 70 Financing expense on working capital 30 Income tax expensei= 70 Total sources 1,010 Total uses 890 Available for debt repayment and retention as equity 120 === Source: Gibton Committee (1986). Notes: a) Consolidated estimates including all moshav farmers. b) Non-farm income: it was estimated that out of 48,000 participants in the labor force in the moshavim, 26,000 hold non-farm occupations with an average monthly income of $450; living expenses estimated as 27,500 families X $700 average a month. c) Including tax liability on non-farm income. 43 implementation of structural changes believed to safeguard against recurrence of the circumstances which led to the current crisis. Among them: supply cooperatives will limit their operation to inputs supply and product marketing; no financial intermediation is to take place. Failing regional enterprises will be closed or sold. Aid will come through the association and its distribution within the moshav will be left to its members to decide. •Moshavim will deal independently and directly with banks; preferably concentrating their financial affairs in a single foster bank. Regulations will be modified to enable trade in land, water rights, and production quotas. These will then become assets for the farmers to use as collateral, reducing the reliance on mutual guaranties. By the time the second committee submitted its proposal (May 1987) the estimated debt of the moshavim was $1 billion and growing. The committee worked out the details of the proposed financial settlement: a rescheduling of approximately $600 million to long term debt (15-20 years) at 7-8 percent real interest, and writeoffs of $200 million. Most of the money was to come from the government and the Settlement Department; the banks would contribute $165 millions to rescheduling and 66 millions to writeoffs." Moshavim and farmers with positive deposits in cooperatives will also "write off" part of the debt due to them in the same proportion as the banks will do. Again, terms of assistance will depend on the individual ability to pay of each farmer and moshav. Each member in the moshav will have to provide a guaranty or collateral for his own liability plus a mutual guaranty limited to the average liability per member in the village. Debt will be written off only to farmers that will be judged unable to repay theirs even if rescheduled. They will then return their means of production and marketing quotas and will not be agricultural producers any more. An association in a moshav in which a third or more of the members will cease functioning as farmers, will be split into a municipality and a separate agricultural cooperative for the 11 These numbers were modified somewhat when the committee's recommendations were approved by the government. 44 remaining producers. It was estimated that .38 moshavim will turn into pure residential communities and 2600 farmers will have to exit agricultural production. The proposals were reached in the committees by consensus--each party accepting that the suggested solution minimized the damage it was to suffer: the government will rescue the family farm and the banking industry, the banks will get most of their debt back, and the leaders of the moshavim's movements were also happy with the proposals seeing an opportunity to rescue cooperation, even if at the cost of jettisoning some moshavim and quite a few farm families in the process. Consequently, an administration was set up to implement the new debt settlement moshav by moshav. Not asked as for their opinion were the farmers in the field, but they were expected to join gladly the new settlements, as the understanding was that they were hard pressed, lacking credit for current production and renewal of investments. The proposals were reached in the committees by consensus--each party accepting that the suggested solution minimized the damage it was to suffer: the government will rescue the family farm and the banking industry, the banks will get most of their debt back, and the leaders of the moshavim's movements were also happy with the proposals seeing an opportunity to rescue cooperation, even if at the cost of jettisoning some moshavim and quite a few farm families in the process. Consequently, an administration was set up to implement the new debt settlement moshav by moshav in accordance with the recommendations of the committees. Not asked as for their opinion were the farmers in the field, but they were expected to join gladly the new settlements, as the understanding was that they were hard pressed, lacking credit for current production and renewal of investments and bound by mutual guaranties to the debt of the collapsing regionals. Meanwhile, life continued: without declaring it, farmers in moshavim that failed financially set a moratorium on payments to the banks, thus freeing themselves of immediate financial burden. Suppliers and marketing agents continued to offer credit (broilers production, for example, can be financed completely in all stages by slaughter houses competing for the right to market the product) and 45 operators' equity was evidently added to finance gaps--if there were any. Banks that attempted to force farmers who refused to join the administered settlement to repay their debts by foreclosing on private property, found that this was not an easy task--justice is slow, the police is not exactly eager to fight sturdy farmers now cooperating fiercely, and the mere magnitude of the problem with 15,000 to 20,000 families in arrears seems to make a forceful solution hard to impose. Moreover, many banks are ready to offer credit to moshavim and farmers with appropriate collateral (saving programs and pension funds are popular). Agricultural production did not suffer, markets are bountiful. The crisis uncovered and intensified differences in the family farm sector. As of now, it seems that approximately a third of the moshavim continue with comprehensive cooperation, including cooperation in credit. In the rest, farmers are reluctant to be parties to the suggested solution. They see the problem as that of the banks, "let them worry," and they do not wish to join a settlement that will again force them into mutual guaranties, however limited. For many of the weaker farmers, this may mean exit from agriculture. The administration, on its part, promises early joining moshavim better conditions and threatens that funds will not suffice for latecomers. By the agreement between the government and the banks, moshavim that will not join the settlement will have to cover fully their share in the obligations of the regional organizations and the public agencies will help the banks force these payments. Carrot and stick not withstanding, after a year of the administration's efforts no more than 20 moshavim joined. But the administration is now negotiating with many more. As many in the moshavim see it, the crisis is alive but it subsided and the participants retreated into waiting positions. It is likely that political and not only economic factors will determine the ultimate outcome of the crisis. 46 H. RECAPITULATION, LESSONS, AND CONCLUSIONS Many businesses suffered severely when economic conditions changed with the introduction of anti-inflationary policies in 1985. But it is only in agriculture that a whole sector--the cooperative sector-- collapsed financially. The reasons for this general failure lie in the particular character and mode of operation of cooperative credit in Israel. In the last 2 - 3 decades, moshavim (and kibbutzim) developed' intensively, adopted new technologies, reacted fast to changes in market conditions, and established a network of farm services owned and operated by farmers and their representatives. Much of the credit for this development is due to cooperation. But cooperation in credit, together with a government policy of easy money, also caused over- expansion, over-investment, and enlargement of debt to crisis proportions. Examination of the accumulated experience reveals the weaknesses as well as the strength of cooperative credit and suggests possible reforms. . Experience: Cooperation Moshavim and supply cooperatives developed from transferrers of suppliers goods and credit into full fledged and powerful financial intermediaries. Cooperatives at both levels--the village and the region--face, in theory and in practice, basically the same management and control problems (Sections B, C) and, since the center of gravity of financial activity has shifted in the last 20 years from the moshavim to the regionals, we concentrate the discussion on the latter'5 experience. The supply cooperatives have done remarkably well in the last two decades. Despite a late start, they became a prominent feature of the agricultural scene in Israel. They performed a central role in the economy of the great majority of moshavim, especially the new ones, and served as the principal instrument in the development of regional economies. To some extent, their success was due to a generous and preferential public support in the form of soft loans and grants-in-aid. It is doubtful, however, that this preferential 47 treatment alone can account for the remarkable performance. The potential gains from collective actions were apparently sizeable. Yet, the great majority of the supply cooperatives failed financially. They failed when a weakness in the cooperative structure was subjected to the effects of an unanticipated shift in government policy that brought an accelerating inflation almost to a complete halt and changed interest rates form low, sometime negative, to exceedingly high values. The structural weaknesses are to be found the relationship between the moshavim and the managements of the regional supply cooperatives. Firstly, the same type of moral hazard as in the moshavim (Section B), albeit in a magnified form, is produced by mutual liability arrangements at the second order cooperatives. Secondly, goal disparities between the member-moshavim and the management of the supply organizations give rise to a costly "agency relation." In particular, the management of a supply cooperative aspires to expand its role and scale of operation beyond those that would be considered optimal by the ultimate owners and residual claimants. Thirdly, membership control over management was inadequate; in part, because no member would find the control worth the effort from his own individualistic point of view; in part, due to lack of market control mechanisms (ownership is not tradeable); and in part because of insufficient expertise. Fourthly, management, on its part, found it difficult to overcome internal political obstacles and establish financial discipline among its member-associations. Finally, it seems that the management teams in many supply cooperatives did not possess the required professional expertise to run big and complex organizations and that managers were often motivated by personal political interests rather than by the economic needs of the organizations (Gibton Committee 1986). As a result of the two-layer moral hazard behavior, the inadequate control, and the willing of the creditors to expand loans when interest rates were low and credit was in easy supply, indebtedness grew rapidly. At the beginning everyone benefited from the influx of the generous financial flows, but when in 1985 real rates of interest 48 soured sky high, and lenders were reluctant to refinance, many moshav associations and supply cooperatives found themselves in untenable financial positions. The stage was set for a full scale financial crisis. 2. Experience: Government and Policy Up to 1985, the government virtually monopolized the capital market in Israel. • When credit supply expanded in the 1970s, the government followed the trend and increased its supply too. This shift in supply was the major cause in enlarging the volume of credit in agriculture (Kimhi 1986). Further, - the danger of over-extension of credit in the moshavim was recognized already in the early 1960s, but efforts t monitor and control the moshav association and its members with "concentrated credit" failed. Despite this experience, the government encouraged uncontrolled growth of the regionals, both the supply cooperatives and the service enterprises. • It was policy together with structural weaknesses in cooperation that bred -the crisis in agriculture, but policy carries a major share of the blame. Lessons drawn from history are generally based on hindsight, but not all is hindsight in this case: overcapacity in service enterprises was identified by officials in the Ministry of Agriculture and the Settlement Department long time ago, but they were over-ruled by the political decision makers. Several factors contributed to this behavior: high national priority given to agricultural development, the need to. strengthen the weaker sectors of the family farm, and the political pressure exerted by the farmers through their regional and national organizations. Evidently, the desire to 'show progress, even if fictitious, and to satisfy local demands, even if shortsighted, overcame economic rationality. Policy and behavior created the impression, in agriculture and outside, that farmers and their organizations will be bailed-out if encountering strains. This confidence increased moral hazard behavior on part of the farmers, their organizations, and their creditors.- The resulting over-capacity was not limited to the regional level, idle capital was created on the farms too. In addition, several studies 49 suggest that during the late seventies and early eighties saving rates in moshavim declined considerably (Gal 1984, Artzy 1985, and Zusman 1988) which, along with increased investment in productive capacity, housing and other durable consumer goods led to increased borrowing and excessive farm financial leveraging (debt to capital ratios). Evidently, cooperative agriculture has followed an inefficient path: over-investment in productive capacity, especially in support services; over-investment in housing and durable consumer goods, and considerable under-saving. In addition •to the welfare losses experienced in the last decade, present conditions in agriculture are certain to entail further losses in the future. This is because the present composition of physical capital is non-optimal and because farmers' financial exposure and existing credit arrangements will impair smooth functioning of agricultural credit markets. In particular, the weaker farmers are bound to face grave difficulties in raising capital resources in the future as a result of the disintegration of first and second order cooperatives. Of course, to the extent that more efficient alternative institutions will replace those that failed, there may be a net benefit; but whether this would actually happen is not at all certain. The distributional effects are somewhat inconclusive. Welfare losses in the past were only partly borne by farmers. It is not impossible that for some time, at least, farmers even benefited. The main losers, apparently, have been banks and tax payers. The distribution of future losses depends on the way the current crisis is solved. If the debt settlement promoted now (Section G) is eventually accepted by all parties and implemented, the main losers will again be the tax payers. Otherwise, the banks and the farmers will have to share the losses in yet unpredictable proportions. 3. Evaluation It may be that we are witnessing end of cooperation in credit in the family fail') in Israel (the kibbutzim conduct all their economic activity collectively and credit will not be an exception). Clearly, for many years now financial activity in the moshavim will by and large 50 differ substantially from its mode of operation before 1986. We are already witnessing members in moshavim raising privately credit for investment in capital assets. Since a great part of the investment on the farm can be done piecemeal--particularly investment in orchards and 4' livestock--most farmers may find that they do not lack the resources to expand their operation at an optimal pace. Marketing agencies and wholesalers are also ready to finance short term needs, again mostly on a private basis. At the same time farmers are reluctant to return to any form of mutual guaranties. It may well be therefore that cooperation will continue in agriculture; but it will be limited to cooperation in production services, in purchasing of inputs, and in marketing--some specialized and partial cooperative associations, of dairy farmers for example, have already sprung up in moshavim--and there will not be a return to the earlier form of comprehensive cooperation. On the other hand, it may also turn out that some of the moshavim will attempt to maintain or revive comprehensive cooperation, including in credit. Revival of cooperative credit was the goal of Gibton and Ravid Committees (1986, 1987) and is the intention of the administration of the new settlement for the solution of the crisis. The major issues and intellectual challenges in considering revival of cooperation in credit are: (a) To identify environmental conditions conducive to successful agricultural credit cooperatives, and those under which failure is most likely; setting up credit cooperatives under the latter conditions is certainly inadvisable. (b) Where cooperation stands a good chance of succeeding, to determine the structural and procedural designs of cooperative organizations that maximize expected net benefits; that is, the expected gains from cooperative action minus expected losses due to structural weaknesses. 4. Prerequisites The principal prerequisites for successful credit cooperation are cultural. Successful cooperatives can. evolve only in a cultural environment,supportive of cooperation and collective action. Unrestrained individualism is not likely to breed successful • 51 cooperation. This applies especially to the moral basis of society.12 Credit cooperatives also require leadership and professionals, primarily accountants and financial experts. Without a proper and updated accounting system, a credit cooperative is bound to fail. In addition to intellectual and moral integrity, which are absolute requirements of the leadership, it should also possess some understanding and have familiarity with financial affairs. No internal or external control mechanism can substitute for intellectual and moral integrity; and complete leadership ignorance cannot be fully remedied by hiring professionals. Members' default on their obligations of debt repayment is the primary cause of failure of credit cooperatives. The availability of inexpensive enforcement mechanism is, therefore, a necessary condition for a successful*credit cooperative. Adequate control over the marketing of members' output is a good enforcement device. Where such control is impossible, other, easy to handle forms of collateral are required. In this respect the credit cooperative should be able to act as any other lender. 5. The Moshav Association's The recent financial crisis has brought out very clearly the failure of mutual liability; it encouraged over-borrowing when that was possible and could not be enforced when the need arose. Restricting mutual liability arrangements is, therefore, the principal structural reform necessary. In particular, the size of debt secured by mutual liability that any member is entitled to should be strictly limited. member wishing to borrow above that ceiling should be required to provide personal guaranties by way of appropriate collateral or third party guaranty notes. It is thus implied that the mutual liability obligations of every member of the moshav would also be limited. These limits should, perhaps, be established as moshav by-laws and thus be 'See, for instance, Banfield (1958). 1 Many of the conclusions presented below are based on Ravid Committee (1987). 52 integrated into the moshav constitutional setup. . Second Order Associations The underlying structural faults that made the regioanls the mainspring of the recent moshav financial difficulties are poor members' control characteristic of second order cooperation, the moral hazard behavior associated with mutual liability arrangements and the political nature of the organization. Several lessons can be derived from this experience. (a) Regional purchasing cooperatives should not function as financial intermediaries and should concentrate on requisitioning alone, passing-over suppliers credit to their patrons the moshavim. (b) Secondary cooperatives should never guarantee loans to other borrowers. Neither should they seek external guarantees for their own debt. (c) Secondary cooperatives should have sufficient equity to back their fixed payment obligations. (d) A regional supply cooperative should not own equity in other secondary cooperative enterprises, neither should it extend loans to 41 these organizations. The equity capital of all regional second order cooperatives should consist solely of contributions by member associations, who would thereby retain full .and undivided decision control over these ventures. (e) Control of the supply cooperatives should be strengthened by adding external independent directors to their boards. (f) Exit from second order cooperatives should be facilitated. To this end, a certain part of the secondary cooperatives' capital should be maintained in the form of returnable funds, which would allow a member-association to redeem at least part of its equity share upon termination of membership. Control through exit will thus be reinforced. (g) The role of auditors should be strengthened. Boards should be able to change auditors freely. 53 7. Policy Given the propensity of first and second order moshav cooperatives to over-borrow, could the government avoid the associated welfare• losses? It should first be emphasized that the government cannot and should not seek to control directly the borrowing behavior of farmers' cooperatives. The membership should see to it by creating more effective governance structure, and by setting up internal mechanisms as designed to check over-borrowing. Financial intermediaries, such banks, should be aware of their share in the risk of the ventures they support and should therefore provide the necessary external controls. Government policy seems to have erred on two counts: (i) it followed an excessively liberal credit policy; and (ii) it created the false impression that the government will always provide farmers and their cooperative organizations with a financial safety-net. We summarize our conclusions in the following recommendations: (a) The extension of agricultural credit under preferential terms should be minimized. This should be reflected in lower subsidization rates and in reduced amounts of subsidized credit. (b) Government should avoid by all means the creation of any misconception regarding its commitment to aid farmers in financial difficulties. To be sure, the government should take measures to alleviate farmers' financial distress caused by adverse environmental fluctuations or by temporarily depressed markets. However, little or no assistance should be extended when the financial difficulties are due to other reasons. Furthermore, even when financial support is Justified, it should be limited so as not to create unfounded expectations regarding future government support. (c) The government should be instrumental in creating institutional structures to control over-borrowing by farmers and farmer F) associations. The suggested revival of concentrated credit (Section is worth exploring, for it may contain the potential for productive collaboration between banks and farmers' credit cooperatives. (d) The government should legislate changes in cooperative law and the regulations that will enhance internal and external control over 54 cooperatives financial integrity. The credit cooperative governance structure should be strengthened. The registrar of cooperative societies should be authorized and required to exercise external control on credit cooperatives' financial integrity. REFERENCES Annual Bank Statistics - see Supervisor of Banks. Artzy, A. (1985). A Comparative Analysis of Moshavim Using a Village Accounts Approach, unpublished MSc dissertation [in Hebrew], The Hebrew University of Jerusalem, Faculty of Agriculture, Rehovot. Banfield, E.C. (1958). The Moral Basis of a Backward Society, The Free Press, Glencoe, Ill. Bank Credit - see Supervisor of Banks. Buch, V. (1984). Capital Stock Series in Israel, 1960-1984, Bank o Israel, Research Department, Jerusalem. Central Bureau of Statistics, Statistical Abstract of Israel, various annual issues, Jerusalem. Cohen, J. (1984). An Analysis of the Organizational Structure of Secondary Regional Cooperatives Associated with the Moshavim, unpublished MSc dissertation [in Hebrew], The Hebrew University of Jerusalem, Faculty of Agriculture, Rehovot. Examiner of Banks - see Supervisor of Banks Gal, Y. (1984). A National Accounts Approach to the Analysis of a Moshav Economy--Application to Moshav Ein Ha-Teva, unpublished MSc dissertation [in Hebrew], The Hebrew University of Jerusalem, Faculty of Agriculture, Rehovot. Gibton Committee (1986). Report of the Committee to Prepare a Plan for Strengthening the Family Farm Sector [in Hebrew], September. Harsanyi, J. C. (1962). "Measurement of Social Power, Opportunity Costs, and the Theory of Two-Person Bargaining Game," Behavioral Science, 7:67-80. Haruvi, N. and Kislev, Y. (1984). "Cooperation in the Moshav, Journal of Comparative Economics, 8, pp.54-73.- Heth, M. (1966). The Banking Institutions in Israel, 2nd edition, Falk Institute of Economic Research, Jerusalem. 55 Kadishay, A. (1985). The Financial Relationship between the Regional Purchase Cooperative and the Moshavim Associated with It, unpublished MSc dissertation [in Hebrew], The Hebrew University Jerusalem, Faculty of Agriculture, Rehovot. Kimhi, Ayal-(1986). Agricultural Credit in Israel, unpublished MSc dissertation [in Hebrew], The Hebrew University of• Jerusalem, Faculty of Agriculture, Rehovot. Kislev, Y. and Marvid, A. (1988). A Supply Cooperative MISHORIM: Plants, Economic Functioning and Finance [in Hebrew], The Maghes Press, Hebrew University, Jerusalem. Levy, H. (1979), "Capital Structure, Inflation, and Cost of Capital in Israeli Industry, 1964-1978," [in Hebrew], Falk Institute of Economic Research, Discussion Paper No. 795, Jerusalem. Levy, H. and Lerman, Z. (1987), "Estimating the Cost of Capital Under Inflation--Israeli Industry, 1971-80," Bank of Israel Economic Review, No. 59, pp. 64-87. Levy, H., Smith, M., and Sarnat, M. (1978). The Stock Exchange and Investment in Securities [in Hebrew], Schoken Publishing House, Jerusalem. Lowe, Y. (1955). Agricultural Credit in Israel in Kibbutz Settlements [in Hebrew], Ministry of Agriculture and Jewish Agency, Tel Aviv. Marvid, A. and Arian O. (1987). "Excess Production Capacity of Regional Agricultural Enterprises," The Economic Quarterly, 38:254-261, [in Hebrew]. Ravid Committee (1987). Report to the Minister of Agriculture and the Minister of the Treasury [in Hebrew]. Sadan, E. and Kadishay, Y. (1967). Financing Capital Formation in the Kibbutz Settlement, The Hebrew University of Jerusalem, Faculty of Agriculture, Department of Agricultural Economics, Rehovot. Sang Committee (1979), Inter-Ministerial Committee to Inquire into the State of Financing in Israeli Agriculture, Mimeo, Jerusalem, January. Sarnat, M. (1966). The Development of the Securities Market in Israel, Kyklos-Verlag, Basel. State Controller (1986). Annual Report. Statistical Abstract.- see Central Bureau of Statistics. 56 Supervisor of Banks, Annual Statistics of Israel's Banking System, various issues (1974-1987), Bank of Israel, Jerusalem. Supervisor of Banks, Bank Credit by Sectors of the Economy, various issues, Bank of Israel Jerusalem; as of 1986 superseded by Current • Statistics of Israel's Banking System. Supervisor of Banks, Israel's Banking System, Annual Survey, various issues, Bank of Israel, Jerusalem. Zusman, P. (1976). "The Incorporation and Measurement of Social Power in Economic Models," International Economic Review, 17, pp. 447- 462. Zusman, P. (1988), Individual Behavior and Social Choice in a Cooperative Settlement, The Magnes Press, Hebrew University, Jerusalem. ONE1k wP uniol n p n n 3 5 32 p n t) 5 T3113 n THE CENTER FOR AGRICULTURAL ECONOMIC RESEART Llialfeet0 aiiivN(C4 WORKING PAPER NO. 8601 A THEORY OF AGRICULTURAL SETTLEMENT by Eli Feinerman and Yoav Kislev Rehovoth, Israel, P.O.B. 12 12 .1./1 minirri A Theory of Agricultural Settlement* Eli Feinerman and Yoav Kislev** Abstract The paper draws on the modern theory of local public goods to develop a theory of agricultural settlement which focuses on the provision of village services. Alternative objectives of the settlement project are considered. Conditions for optimal level of services and optimum community size are derived together with rules of finance of village oeprations. The value of the right to settle is calculated for the cases considered. *The work n this study was supported by the US-Israel Agricultural Research and Development Fund (BARD). We received constructive comments and suggestions from Eitan Berglas, Meira Falkowitz, Esther Redmount and participants in seminars in Rehovot and in the World Bank. The remaining shortcomings are ours. **Hebrew University, Rehovot, Israel Yoav Kislev is currently at the World Bank, Washington, D.C.). Page 2 A Theory of Agricultural Settlement Economic development often requires planned settlement or resettlement. Farmers will be settled in villages that will supply municipal and production services. The planners have to decide on size of villages and farms, on public services and the way they will be financed. These questions are close to the questions asked with respect to urban communities in the modern theory of local public goods. Indeed, Berglas' exposition of that theory can be adopted verbatim as a theory of rural community and is the starting point of our present work. However, there are also differences between the problems discussed in urban economics and the problems of agricultural settlement. One difference is in the emphasis on production services; another is that the literature on urban communities focuses on whether, a welfare optimum exists under competition, while planned rural settlement encounters questions of the obejctive of the planning authority and the value of the settlement to the farmers. These issues will be in the center of our discussion. Focusing on services, we disregard the classical issues of geographic distribution of farm communities. Attempting to maximize settlers' income, our approach resembles in some aspects the theory of the labor managed firm (0i and Clayton; Vanek) in which,in most cases, the objective is the maximization of income per laborer. A closely related literature is on the theory of cooperation (LeVay) which should Page 3 be here particularly relevant since cooperatives share several characteristic features with local governments: non-profit entities, supply of services some of which are in the nature of public goods, decreasing costs, voluntary association through free choice of community and freedom to enter or not. The village welfare services enter directly households utility; farming services affect productivity. To illustrate that the principles of the analysis are similar whether consumption type or production type services are considered, we commence by recapitulating Berglas' analysis and move on to agricultural production. For simplicity we shall not attempt to deal with a municipality supplying both types of services simultaneously. We also do not treat explicitly in this paper private services--stores, shops--and focus solely on the public sector. A Rural Community..... A village municipality supplies two services: it operates a well supplying water to the residents and it provides a pure public good: water quality, a non-commercial .community radio station, weather forcasting service. There are n identical residents in the village and the utility function of each is (1) U U(x, y, Q) where x other, non-water, privately consumed goods Page 4 y water in gallons per year Q the public good. The cost function of the municipality is (2) C- C(ny, Q) (ny) + c (Q) In (2) fly is the total amount of water pumped in the village well. Marginal cost is increasing in both ( and c7( ). basic conclusion of the theory of local public goods is that the chargeable services (water in our case) and non-chargeable pure public goods ought to be financed jointly. The separate formulation in the second line of (2) is introduced to demonstrate that jointness in ' finance is not due to jointness in the production and the cost of the municipal services. The income constraint of each the village residents is (3) ny) c (Q)]/n I is annual dollar income, the price of the consumption good is taken be unity, and the community covers the cost of its services. We start with a community of a given size and take social welfare be the sum of the utilities of all residents. Since they are Identical, we consider a single household. Maximizing U in (1) with respect to and Q, subject to (3) yields 14 where U. is the partial derivative of U and '( ), ) are the 1 1 derivatives of the cost functions. The interpretation of (14) is straightforward at the optimum, the marginal utility per dollar (1) 1) is the same for x which is purchased on the market and for roduced by the village. Also, expenditures on Q will be carried to the point at which the sum total of the community's marginal utility per dollar is the same as in x or in y, Q is the community's pure public good. Since water, y, is a chargeable commodity it should be priced at the marginal cost i(ny); but if marginal* cost is increasing,, user charges can be applied to finance, at least partly, the provision of the public good, Q. A ,lump sum tax or subsidy, o" - t dollars per resident, will complete the financing of the municipal operation ny +c,(Q) -c ny)y To determine optimum village size, maximize (1) subject t (3) with 1 respect to n. The first order condition is that the right hand side in (5) is zero. By the same condition, t > 0 for suboptimal community size, Page 6 and ,t < 0 for oversized communities. Hence, at optimum village size all municipal costs are covered by user charges. Moreover, as we shall see . presently, at this optimum per capita municipal costs are minimal. Economics of the Municipality Consider a village of n residents supplying optimal amounts of Q and n*y gallons of water. Municipal costs are drawn in Figure 1 as a function of the village water supply. Since Q is given, constant 2(0 - is the "fixed cost" element in the figure and cost averaged per gallon is a U-shaped function (AMC I). As drawn, n* is smaller than optimal and a positive lump sum tax is levied. In a larger community, optimal Q and total village water supply will also be larger. The average municipal cost curve will be higher. In Figure 1, .n is optimal community size at which with marginal cost pricing, in e . (5), t 0 and user charges exactly cover all village costs. In a community larger than optimal, the municipality accumulates financial surplusses which it may distribute to residents as lump sum subsidies. The municipality functions as a firm with a U-shaped cost curve. But this is a zero profit firm the patrons of which have to cover its cost: selling too little, the firm has to tax its patrons; selling more than at minimum AMC, the firm can distribute refunds. The difference between the regular firm and the municipality is the shift of AMC with Page 7 n--the number of "patrons". Cooperatives Zero profits and patronage refunds are typical of cooperatives. And indeed the village can be run either as a regular municipality or as a cooperative. Several issues can be incicated when considering cooperative supply of village services. Strict patronage refunds are inefficient, as with them marginal becomes average cost pricing. We avoided this distortion by resorting to lump sum taxes or subsidies. Efficient management of cooperatives requires marginal cost pricing and then surpluses and deficits are distributed as side payments. The democratic choice .of methods for allocating side payments or subsidies raise difficult and important practical and theoretical issues--particularly in cooperatives with non-homogeneous membership--and is the subject of a major branch of the modern theory of cooperation (see, for example, Zusman 1982). We shall not consider in this paper questions of group choice, but note a conclusion stemming from the previous analysis: in comprehensive cooperative communities the issue of side payments is mitigated or even eliminated. By a comprehensive cooperatives we mean a cooperative that provides its members with many of their needs. An example can be the moshav in Israel in which members operate private farms and cooperate in finance, marketing, municipal services, some production activities, and Page 8 several other aspects of rural and communal life. In such a cooperative efficiency will dictate, as we have seen, using surplusses in joint financing of the services and at the optimal size no side payments or taxes will be used. This avenue of escaping the need for often arbitrary •• • allocation rules for side payments is not open to single line cooperative--for example to a cooperative supplying only water, y, and 4 no Q service. In agricultural villages taxes will sometimes be imposed even in 5 communities of optimal size. We turn now to the analysis of these cases. Objectives and Distribution A farmer joining a newly established village foregoes alternative earnings and, in most cases, will also have to invest in captial assets and land. His real, alternative cost will therefore be labor earning foregone and interest on capital. Farmers will join a settlement project if their alternative costs can be covered. They will be willing to pay for the right to join if the income generated will more than cover the cost Considering earlier in the paper a consumption type public good, we adopted the traditional view that the goal of the authorities was to maximize the utility of the individuals in the community. In settlement planning we, take up several alternative objectives: maximize total 6 income in the village, in the region, and per farm income. These objectives are interrelated with each other and with the issue of the Page 9 appropriation of rent; that is, of the marginal product of the land. Settlers joining a new project will .either buy their plots of land or get them free. There are several ways in which the rent can be appropriated in the second case (Figure 2). The national government can *collect the rent as a tax (point 1 in Figure 2). In Israel, for example, farmers cultivating national land pay rent (although probably lower than the marginal product in many cases) the proceeds of which are to be used for land reclamation--generally not in the same village or region. Alternatively, the rent can be left in the region and the village government permitted to collect it as taxes (point 4). Another possibility is that villages may be refrained from taxing the rent and it will be, all or part, appropriated by the farmers (points 2, 3). In the last case the rent will be capitalized into the value of the farm. A settler, if receiving land free of charge, receives in this case grant, capital gain. We shall see below the relation of the value of joining a settlement project to the rent of the land.. A Village aft41.0 We are turning now to the planning of settlement, we start at the village level and then move to a region. We shall consider two alternative objectives: maximizing income per farm, and maximizing income per village or region. The findings are summarized in Table 1. The well behaved farm level production function is (6) q- F(y,A/n,L,Q) Page 10 where y water, supplied from the village well A/n land; A is total area cultivated by the village L Labor Q public service provided by the village: a road, communication center, weather information, pest control, water quality. We think of L as self employment, with alternative cost, w, the urban wage rate. To accomodate a degree of indivisibility and for simplicity, we assume that farmers accepted to the project have to work LL on their farms. An example may be that farmers are required to spend all their time on the project and hired labor is not permitted. Farmers are charged for water and taxed to complete local finance according to the linear cost allocation rule ci(ny) + c (Q) (7) -- ..... am...ft .11.1.11 oty Rent can be charged by the national or the local governments. Income per farm We start by maximizing income per farm. The maximization is done at two levels. On the farm the settler decides on water and labor input given costs, technology and land allotted. At the planning level, the control variables are a, t, Q, and n. The Lagrangian is Ony)+c2(Q) (8) H - PF(y, A/n, L,Q) wL - s(A/n) + CL-L) 4.. Page 11 where P is the price of the product; s is constant per-acre rent • collected by the national authority,, not the village; and e lagrangian multiplier. For a given number of farmers, the first order conditions are (8a) PF c (ny) (8b) nPF c 2(Q) (8c) PF w + On the farm PF -a, hence by • (8a c(ny) and optimal water utilization will be with marginal cost pricing. Equation (8b) characterizes Q as a public good. With the present assumptions, equation (8c) has no operative contents, it specifies the value of the shadow price of the labor constraint. Consider now A A (9) •••••• - c (ny)y n n 1 Solving (8) we maximize income per unit, 3H/3n summarizes the change in income on an existing farm caused by a new entry. The negative expressions are due to reductions in production as land is shared and to increased cost of water; the positive expressions--sharing of cost and taxes. Maximizing (8) with respect to n (i. . setting 3H/3n - 0), implies Page 12 • (ny)+c (Q) 1 2 (8d (ny)y (A/n)(PF -s) 1 A/n It is useful to examine (8d) in relation to. the financing of the village services. Here-o: eq.(5) applies: ny c 1 2(Q) (ny Combining with (8d wo alternative rental situations will be analyzed. Alternative 1:5 - 0 Assume that in (8) s 0, no outside rent taxation and the village can tax the farms to finance the provision of the public services. Then, (10) t PF A/n) A n with optimum number of settlers, the village exactly covers its cost collecting user charges and a head tax equal to the rent of the land (see the first column of Table 1). Examining the sign of BH/On with - 0, the tax is higher than the rent if n is lower than optimal and lower than the rent if n is higher than optimal. When the village is of optimum size, the total "tax (for water and other services) is equal to average municipal cost. By the construction of (8), if income is positive, a settler covers the alternative cost of labor. . But the project may yield further returns. We name them settelr rent SR. As F( ) is well behaved witp constant returns to scale Page 13 ,A , (11) PF P(F y + F A n n and, utilizing (8a)-(8d), deducting actual and alternative costs,. ny) + c (Q) (12) SR .PF( ) A -PF( wL c ny y A n n -6L + PF Q When the number of settlers is optimal the value of the right to join the settlement project is 6L+PF Q. That is, if farm size and production technology are such that in the solution of (8) L L, labor requirement is not effectively constraining and 6-0, then SR-PF Q--the value of the right to settle is the productive value of the public good provided by the village (Table 1). Alternative Now the head tax is (A/n)(PFA/n - s) particular, if s PF and for optimum village size A n' fly) (8d') C ny Page 14 find that whether the village is constrained by regulation finance solely through user charges (t-0 is imposed exogenously) or if the rent is appropriated by an outside agency, community size and level of services will be the same. In Figure 2, settlement plans are the same for points 1 and 2. See Table 1 for a summary and settler's rent. Income in Village We assume now that s-0 and t?40, the planner's objective is to maximize total income in village, then the Lagrangian will be 7 (13) H nPF( - wnL - c(ny) c(Q) + On(L L) Optimizing community size, skipping the other conditions which are identical to (8a)-(8c), (13d) P ) w -c(ny) + A/n) PF.An Rewriting Euler's equation, aby-product of this result is Table 1) (14) 0L+PF Q-0 Q which means that when total village income is the objective, settlers will be called in until the cost of over settling (from the point of view of the single farmer) exhausts the value of the public services. In this case the village will be financed according to (5) and t Page 15 will in general differ from F,(A/n). Settler's rent can in this case be expressed as (15) SR- A/n FA n Comparative Statics The comparative statics of the model developed in this paper is complex and only several of its aspects are presented. We start by tracing the changes in the village services as the number of farms (not necessarily optimal) is modified, that is n is taken to be an exogenous parameter not a control variable. We do this for the case in which the planning objective is to maximize income per farm. For convenience, define in equation (8) --total amount of water pumped at the village well. We also let s 0. The relevant first and second derivatives'and their.signs are 1 1 - n ,z n n 1 = ç c2(Q) H L. <0 zz 2 z/n 1 P _ (Q) < 0 HQQ Page 16 1 PF >0 n, 1 A aram...• z/n n 2 z n,A/n -TT n'? PF - P + c(Q) Q,z n n FQ • n The determination of the signs of these derivatives rely on the assumption that the production function F( ) is linear homogeneous with Fi > 0, Fii < 0, Fij > 0, and that the first order conditions are maintained. The two last cross derivatives„ H and demonstrate zn HQn' the difference between consumption type public good and productive service. Consumption type services enter directly the utility of the household and in general the more households in the community, the larger the marginal contribution of the public good. Production service affect income in combination with other factors. In the expressions for H and H as n increases, zn Qn the amount of land for farm decreases because land is shared with new entrants and this reduction in areas subtracts from the contribution of the services. For the reduction HQn of water per farm adds to the negative effect of area contraction and 8 the sign of this cross derivative cannot b determined a-rpiori. That the sign of H can be determined is due to zn F( ) being linear homogeneous; with a different technology, could be negative or more zn likely also indeterminate. By comparative static analysis, the effects of a change in village size on the supply of the public services are Page 17 dz (16a) Sign a-) - sign(- H H H ) zn QQ Qn z (16b) - sign H ) Sign dn Z. Z •zn Qz and these signs depend on the sign of H n in the following way: (17) if H sign n( Qn 'sig cin") otherwise Sign sign indeterminate. dn dn Hence public services, both chargeable and the pure public good will increase with the number of farms only if a growth in number does not reduce the marginal contribution of the public good on the individual farm (recall that the analysis is done for the case in which the maxim and is income per farm). Similary, and without presenting the detials, one can show that, for given n, optimal supplies of both public services z and Q will increase with the price of the product. Incorporating the possibility that n is a control variable, one canshow that (18) dri. as 0 As the village ability to finance its services is constrained, optimum village .size rises." The intuitive explanation is that, for any level Page 18 of the services,, the village has to collect more water . charges as its rent revenue is curtailed. With a larger number of farmers, total labor input is higher and consequently the marginal contribution of water and the use of water on the farm increase.. However, in general it cannot be asserted that optimal water supply will increase with s. Analysis shows that dz/ds > 0 and dQ/ds > 0 for H 0 and the sign of the derivatives is Qn indeterminate if HQn < 0. The intuitive explanation: n goes with s, whether z and Q will increase with s depends on the effect of n as expressed in equation (17). Further analysis, again not detailed here, reveals that the effect of a change in P, the price of the product, on optimal n, z, and Q is indeterminate. Intuitively, this could perhaps be expected since in general an increase in the price of the product does not increase equilibrium level of every single unit. Similarly, for a given n, output per farm--product supply--will increase with P. This brings us to the boundary condition: a break--even P will be the price at which settlers will cover alternative cost, of labor and of capital if they are required to purchase the right to settle. At a price lower than the break even level, settlers will not join the project. In the same way, given P, a break even can be calculated, now with urban wage rate higher than break-even level settlers will not join. The Region ,YY The planning authority decides now also on the number of villages Page 19 in the region. A region is defined by a given amount of land A which can be divided between M villages with n farms in each. We disregard village services and consider two alternative planning Lbjectives. a. Maximum income per farm This is the trivial case; the solution is a single village in the region. Needless to say, more than important in its own right, this case highlights the anlytical danger of overstretching "desired" assumptions such as no transportation cost or the absence of regional services, economies of scale and external effects. b. Maximum income in region. The Lagrangian is now (19) •H MnPF -wMnL-Mc ny (Q)+Mne L-L +n A--MA) For optimum n (19d) PF 1(nY)Y A n A/n) equation 19e approximates the condition for optimum M 11 .(ny)+c2- (Q)• (19e) PF( PF A/n) A n . Page 20 Combining the last two equations 1 (20 ny)y a planned region, disregarding discontinuities, villages will operate • at minimum AMC with user charges exactly covering costs f municipal services. When a region expands, additional villages will be added, each of identical area and number of settlers. This is in analogy to the expansion of a competitive industry, as demand rises, with entry of identical firms, each operating at minimum AC. Settler's rent is again the marginal product of the land. Note that this value of margainal product is in general the net income of a farmer who owns a plot of land, purchases on the market all other inputs and operates with a well behaved production function. In our analysis, this general result is obtained only when the settlement plan is optimized at the regional level (see Table 1). Monopoly A region may sometimes be planned with a monopoly position in the product it produces. Maximizing per farm or regional income is a straightforward exercise and most of the results of the last column of Table 1 carry over to the monopoly case with marginal revenue, MR, replacing price, ID, One exception is settler's rent which is now (ny)+6 ( ) 1 2 (21) SR - PF( ) - Page 21 MRF A n A/n +(P-MR The magnitudes in eq. (21) reflect optimization of number of settlers. In interpreting the equation it will be convenient to define as the unit of the product the amount produced in equilibrium by one rarm (Figure 3). Settler's rent is the difference between the price of the product (per unit as defined) and cost of outside factors: labor and village services. Given the monopoly position, the regional optimization is done in the following way: all land will always be cultivated, the amount of the product to be produced by the region will be determined by the amount of labor (settlers), water and services; in equilibrium, marginal revenue, MR, equals marginal cost in the production of the product, MC. Marginal cost has a horizontal components, similar to the cost function of a cartel of identical firms. The other components of the cost is the rent, and the shadow price of land (Figure 3). To see why, consider adding a settler: the addition the cost of production in the region is L + C/n and also--to make room for the new settler the established farms will have to contract, reducing production the value of which is graph MC i drawn Figure 3 for the equilibrium value of MRFA/n * The in MR and its positive slope is due to the increasing value of FAin as production is expanded by cultivating the regional land endowment more intensively. Page 22 Special Purpose 'Cooperatives A problem often discussed in Israel is the possibility of special purpose cooperatives replacing the comprehensive moshav. It is argued that functioning of the cooperatives will be much improved if they are . comprised of members engaging in identical lines of production--a dairy cooperative, a poultry cooperative, and others. The issue is similar to the issue of special-purpose authorities in local governments. Consider a village with a well and a public service with two groups of producers, one produces product 1, the other--product 2. Maximizing village income, we write (22) H wn 1 Cy 1 ' ,Q) W n L2 c y +n 2y 1 2 2(Q) L ) n -L 01n1(L--- 1 2 2 -- 2) A-n a -n a ) The interesting results are •(22c) n P + p F 1 2 2Q 2(Q) and Pi 1 (22d) -- a l P F. )-wL -c n y 2 2 2 1 1 1 .______...... 1. OM ...... 11.11...... = vmp a 2 a Page 23 By (22c) the public service will be prbv..t.de4 to 'satipfy both groups and • by (22d) land will be allocated to equate Marginal .product (VMPa) in farms of both groups. If the village is subdivided into two, separate communities smaller amounts of the public good will be provided and land may be allocated in suboptimal manner. It may be socially and politically easier to run special purpose cooperatives, but the comparison of the advantages of special purpose authorities is beyond tbe scope of this paper. Concluding comments We attempted to produce a beginning of a theory of settlement. We hope that it can provide the framework for the analysis of many real world problems of agricultural organization. Examples that come to our mind: (a) the analysis of the economics of a marketing board with monopoly power, were processing plants function similarly to the operation of the villages of the present paper; (b) an analysis of the value of farms in existing villages and the degree to which this value reflects the quality of public services in the village; (c) settlement and market forces; (d) democratic choice of level of public services and individual choice of community--the Tiebout component of the theory of local government which we have yet to incorporate in the analysis; (e) the possibility that farmers with different amounts of initial capital will join the project (the present analysis will not be altered if settlers bring identical capital endowments or if capital can obtained on the market at a fixed cost--an assumption often made in the literature on the labor managed firm Page 25 Footnotes 1. Maximizing individual utility with respect to n, the social objective is no more maximum sum of the community's utility levels. Now the social welfare function includes a single argument: the utility of the single individual member of the community; a community in which all members are identical. 2. Average Municipal Cost (AMC) is total cost of services per gallon of water; marginal cost c (ny) water 1 is in suprly. 3. For similar cost structures consider a supermarket with parking space and air conditioning as functions of the number of patrons and other costs associated with the volume of trade. 4. Zusman (1985) reports that moshavim tend to finance municipal services from a general local tax and often do not collect user charges for individual services even for those that can be taken as private goods. The practice is explained on egalitarian and social grounds. User charges are however common in production services. 5. Such taxes are also imposed in towns with location rents site values), see Berglas. considering these objectives of the planning agency we are deparating from the literature on urban economics that concentrates on Page the feasibility of competitive solutions to community services. The village 7. cost allocation rule, ,still applies. 8. For many cases HQn > can be expected. For example, using the first order condition, write z n - Q,z n Q,A/n The function is homogeneous of degree zero. z A + F Q,z n + r sQ,A/n n • QL For F( Cobb-Douglas with a the exponent of Q, 1 > 1 Hence for the Cobb-Douglas case, HQn > 0. 9. Divide Hzn y z/n to get + n•A \ 2 z/n z z n,A/n fl n which is equal to P. Q) .z z n, n,Q > 0 Page 27 See Allen, p. 317 and P. 324, excercise 25. 10. A similar conclusion was reached by Oi and Clayton. 11. The regional objective function is not continuous in M and methods of calculus are strictly appropriate for its maximization. Define 0(M) the regional income with M villages M maximizes O(M) if 0(M-1) O(M ) 0(M+1). Consider a region with M village; adding one village will add to the regional income nPF( ) nwL Ci(ny) C (Q). The addition of a village will have a negative effect on income in the existing villages the area of which will have to contract. The negative effect is PFAin (A/n). For optimum, the net effect will be positive for , the change from M-1 to and negative thereafter. Eq. 19e) approximates this criterion. Page 28 REFERENCES • Allen, 1938, Mathematical Analysis for Economists, Macmillan. Berglas, Eitan (1982) "User Charges, Local Public Services, and Taxation of Land Rents," Public Finance, 37 178-88. * LaVa , Clare (1983) "Agricultural Cooperative Theory: A Review," 12=1 of Agricultural Economics, 34:1-4. 01, Walter Y. and Elizabeth M. Clayton (1968) "A Peasant's View of soviet Collective Farm," American Economic neview 58:37-59. Vanek, Jaroslav (1970) The General Theory of Labor-Managed Market Economies, Cornell, Ithaca and London. Tiebout, C.M. 1956) "A Pure Theory of Local Expenditures", Journal of M.= Political Economy, 64: 416-424. Zusman, Pinhas (1982) "Group Choice in, an Agrciultual Marketing Cooperative," Canadian Journal of Economics, 15: 220-234. Zusman, Pinhas 1985) "The Moshav as a Municipality," '(mimeo). 6 . Table : Summary Findings for Optimum Community Size Planning level VILLAGE .REGION Objective Farm Village Outside taxes s = 0 s= PF A/n =0 Optimization rules farmers in village, n F (Ain) = c PF( ) wL = c F A n A (A/n)n villages in region, = PF(. ) 7 W PF A/n) derived rule of finance, PF (Ain) by-product OL + F Q = 0 OL + Settlpr's A rent, SR e + PFQ + PFQQ PF(—) - t A n n Ain) c (ny) + c 1 Notes: ' = f(ny)y ••• dollars per gallon : Municipal average and Agricultural Economics, 6 (1991) 1-13 1 Elsevier Science Publishers B. V., Amsterdam Agricultural settlement with joint production servtces Eli Feinerman and Yoav Kislev Department ofAgricultural Economics, The Hebrew University, P.O. Box 12, Rehovot 76100, Israel (Accepted 13 July 1990) \ABSTRACT Feinerman, E. and Kislev, Y., 1991. Agricultural settlement with joint production services. Agric. Econ., 6: 1-13. A theory of settlement planning focussing on village-level production services is pre sented, and settlement plans are compared. The analysis draws on the modern theory of local public goods. Conditions for optimal level of services and optimum settlement size are derived, together with rules of finance of collective operations. The value of the right to settle is calculated for the cases considered. INTRODUCTION Economic development in rural areas sometimes requires settlement or resettlement. Examples are opening of new land, developing of irrigation schemes, or moving farmers when dams flood arable land. Settlements are centers of economic activity; they are often villages, but scattered home steads also exist. Whatever the spatial structure of the settlements, and whether they are planned or develop spontaneously, their size, distribution, structure and functioning are affected by economic forces. In the past, a major consideration was geographic distribution, and settlements were planned to minimize distance to the cultivated plots. Today, distance is no longer an over-riding consideration: more important are the services pro vided. We concentrate on production services. To exploit scale economies, the services are often supplied publicly. But these economies may be exhausted, and the average cost curves of the services are generally U-shaped. This is the rationale for the existence of separate local commu nities - villages in our case. 0169-5150j91j$03.50 © 1991 - Elsevier Science Publishers B.V. All rights reserved 2 E. FEINERMAN AND Y. KISLEV The historic example of village-level public services is the management of grazing 'commons'. Among the recently described examples relevant to our discussion are the FELDA land-settlement program in Malaysia, the Lilongwe land-development program in Malawi (Kirsch et al., 1980), and the specialized agricultural cooperatives in Egypt (Rochin and Grossman, 1987). The example we are more familiar with is the 'moshav' in Israel (Zusman, 1988), a cooperative village, typically with 60 to 120 farms. The first moshav was established in 1920, but the majority were settled in the 1950s. They offer their members a variety of production services, such as provision of farm inputs, financial intermediation, bookkeeping, water distribution, drainage, machinery and implements, produce collection and trucking, extension, and village-wide pest control. Moshavim are coopera tives, but our analysis is not limited to cooperatives; we deal with joint public supply of production services in agriculture, whatever the organiza tional structure of the group of farmers using these services. In a typical settlement project, a region is divided into villages whose land is further subdivided into farms. The planner's problem is to solve simultaneously for the number of villages and farms in a village (that is, land area per farm), level of services, and mode of cost allocation. We lay the basis for the analysis of these problems by considering first a single village with a given amount of land. The village is planned once when the objective function is to maximize income per farm, and once when it is the maximization of total income in the village. We turn to regional p)anning after the village-level analysis. The paper continues with a discussion of land taxation and an analysis ofcomparative statics. We draw on the analysis of the modern theory of local public goods (Arnott and Stiglitz, 1979; Berglas, 1982, 1984; Rubinfeld, 1987; Mieszkowski and Zodrow, 1989). When attempting to maximize settlers' income, our approach resembles in some aspects the theory of the labor managed firm (Oi and Clayton, 1968; Vanek, 1970) in which, in most cases, the objective is the maximization of income per laborer. There is a closely related literature on the theory of agricultural cooperatives (LeVay, 1983) since, even if settlements are not organized as such, often they are, like cooperatives, associations of a voluntary nature operating collective ven tures. STRUCTURAL ASSUMPTIONS The participation of settlers in the project, in the analysis below, is in the form of labor input in production. When this variable is continuous, the farmer in the model can work part-time, hire labor, or even be an absentee SETILEMENT PLANNING 3 settler. In most real-world cases these alternatives are not actually open to settlers and are not included in the menus of activities planners have in mind. To avoid these possibilities, we assume in the analysis that settlers devote all their time to the farms; part-time farming and hired labor are not permitted. A settlement project is not viable unless farmers recover the costs of labor and capital. Urban wage is therefore included as a cost in the analysis. Capital is not treated explicitly, though the settler's income (termed in the paper settler's rent) can be seen as covering the cost of this factor - if settlers bring their own equity to the project. Settlers joining a new project either buy their land or, as in many development projects, rent it from the development authority or from another agency of the national government. Sometimes they may get it free. As shown below, the magnitude of the land rent, and whether it is actually collected, affects the value of the farm and settlement plans and services. BASIC LAYOUT For simplicity, the analysis is limited to a village with two public services [the ensuing principles are the same for a multi-service settlement (Berglas, 1984)]. The village operates a well supplying water to the farms (household consumption of water is negligible) and it provides a pure public good: inner roads, joint pest-control program, weather forecasting, extension service, price information, or others. There are n farms in the village, all of identical size. The farm-level production function is 'well behaved' (linear homoge neous, twice differentiable, etc.): q =F(y, Ajn, L, Q) (1) where q is farm output; y water, supplied from the village well; Ajn land, A total area cultivated by the village and n the number of farms; L labor, constrained - according to our earlier assumptions - to full employment L = L 0 , and with urban wage as the alternative cost w; and Q public service provided by the village. The cost function of the village services is: C = C(ny, Q) = c1(ny) + c2 (Q) (2) In (2), ny is the total amount of water pumped at the village well. Marginal cost is assumed to increase both in c1( ) and in c2( ). This assumption is introduced here for simplicity, it is not necessary. For an internal solution, 4 E. FEINERMAN ANDY. KISLEV marginal costs may be decreasing so long as they are not decreasing faster than marginal products (see equations Sa-Sc below). If, however, marginal costs are decreasing everywhere faster than marginal products (second order conditions are not met) the agricultural sector will not be divided into villages and all the services will be provided at the national level. A basic conclusion of the theory of local public goods is that the marketable and chargeable services (water in our case) and the non chargeable public goods ought to be financed jointly. The separate formu lation in the second line of (2) is introduced to demonstrate that joint financing is not due to jointness in production or costs. The village is a non-profit entity. Farmers are charged for water (a dollars per unit) and, in addition, taxed to complete local finance according to the linear cost-allocation rule: (3) The lump sum tax, t, covers the gap between per-farm cost of services and user charges. The tax t is local; we consider below land taxation by the national government. Income per farm We start with Objective F, maximizing income, net product, per farm. The maximization is done at two levels. On the farm, profits are maximized by equating marginal product of water to unit price, a. At the village level, the planner's problem is to set the policy variables a, t, Q, and n consistently with the objective while taking into account farmers' behavior. To calculate these values, start with the Lagrangian: H=P F(y, Ajn, L, Q) -wL- [c 1(ny) +cz(Q)]/n + O(L 0 -L) (4) where P is the price of the product and 0 is a Lagrangian multiplier. Equation (4) is maximized with respect to the village policy variable. For a given number of farmers, the first-order conditions are (a total derivative is marked by an apostrophe, a partial by a subscript): PFY=c~(ny) (Sa) nPFQ = c~(Q) (5b) PFL =w+O (Sc) Equations (Sa)-(Sc) are the factor allocation rules. On the farm PFY =a, hence by (Sa) the village sets a = c~(ny) and optimal water utilization will be with marginal cost pricing. Equation (Sb) characterizes Q as a public SEITLEMENT PLANNING 5 good, and equation (Sc) specifies the value of the shadow price of the labor constraint. Maximizing (4) with respect to n, we set: aHjan = [ -PFA;n(Ajn) + Cjn- c~(ny)y ]In= 0 (6) Equation (6) summarizes the effect a new entrant has on income of an insider - a farmer already in the project. The negative elements in (6) stand for the reduction in farm income due to the sharing of land and water with the newcomer, and the positive element - for the sharing of village cost. Equilibrium is attained when marginal costs and benefits are matched. Rewriting (6) and combining with (3), the lump-sum tax is: t = PFA;n( Ajn) (7) With optimum number of settlers, the settlement exactly covers its cost collecting user charges for water and a lump-sum tax equal to the marginal product of land. Examining now the sign of the derivative in (6), we observe that the tax is higher than land's marginal product when n is lower than optimal, and lower than the rent otherwise. In Fig. 1, a village of n* farms is supplied with optimum amount (for its size) of the public service, Q, and n* y ·gallons of water. The cost curves of the services are drawn as functions of the village water supply. Since Q is given, ciQ) (=constant) is the 'fixed z 0 _J _J <( <:> 0: w a. Ill 0: <( _J _J 0 0 ny Fig. 1. Cost function of village services. Average service cost is: ASC = [c 1(ny )+ Cz(Q)]jny. 6 E. FEINERMAN ANDY. KISLEV cost' element in the figure and average cost per gallon is U-shaped (Asc I). As drawn, n* is smaller than the optimum, and the positive lump-sum tax levied in a village of this size is higher than land's marginal product (which increases with labor and water). Optimal village size will be n 0 and, at this size, the lump-sum tax (CD) is equal to the marginal product of land (AB), both divided in the diagram by y. The village functions as a firm with a U-shaped cost curve. But this is a zero-profit firm, the patrons of which exactly cover its cost and it does not necessarily operate at minimum AC (it will, under different circumstances, as shown below). Another difference between a regular firm and the village service sector is the shift of the average-cost curve Asc with n - with the number of 'patrons.' Settler's rent and limited entry We turn now to the value of joining the project. By the construction of (4), if income is positive, a settler covers the alternative cost of labor. But the project may yield for its participants further returns. We name them settler's rent SR and, in a viable project, SR z 0. Settler's rent is revenue minus actual and alternative costs: SR =P F( ) -wL- [c 1(ny) +c2(Q)]/n (8) · By Euler's theorem, as the production function is well behaved: PF( ) =P[ Fyy +FA;n(Ajn) +FLL +FQQ] (9) and making use of (5a)-(5c) and (8), the value of the right to join the project is: SR = ()L + PFQQ (10) By Objective F, the village is planned to maximize income on the settler's farms. So long as sR is positive, outsiders may wish to join, but since additional entrants may reduce income of the insiders, entry is limited to the number n consistent with the equality in (6). Moreover, if () =I= 0, farm income can be increased by either hiring labor at the wage rate w (for () > 0) or farming part-time and seeking off-farm work at the same wage (if () < 0). As already indicated, these possibilities are not permitted in the present model. We return to this point below. Income in village We assume now that the planner maximizes total income m village (Objective V). The Lagrangian is: H = nP F ( ) - wnL - c 1( ny) - c 2 ( Q) + () n ( L 0 - L) ( 11) SETILEMENT PLANNING 7 The first three necessary conditions are identical to the factor allocation rules (5a)-(5c) and are not repeated. The conditions for optimal number of farms is now: aH;an = P F( ) - PFA;n(Ajn)- wL- c~(ny )y = 0 (12) Consistent with the current Objective V, the entrance of an additional settler is considered from the point of view of the planner. Two elements are added in (12) when compared with (6): the contribution of the marginal settler to production, and the newcomer's alternative cost of labor. These elements are taken into account by the planner but disregarded by insiders. The following properties can also be shown to hold (detailed derivations in Feinerman and Kislev, 1987): · (a) A shift of the settlement project from Objective F to objective V will change its plan only if under Objective F SR =I= 0. And then, if the settler's rent was positive, the shift will entail an increase in n. Increasing n will reduce income on the individual farms, since farm income was at its maximum under Objective F. (b) By (13), under Objective V, the cost of over-settling exhausts the value of the public services: fJL +PFQQ = 0 (13) (c) But settler's rent is not eliminated altogether, as: SR = PFA;n(Ajn)- t (14) (d) Under Objective V, sR > 0 (provided it was positive under Objective F). To prove the inequality, substitute t( = C jn - c~(ny) y) into (6). Under Objective V, n is larger than under F; hence under V, aH;an < 0 and t < PFA;n(Ajn). THE REGION Beside determining village size and services, the planner decides now on the number of villages in the region. A region is defined by a given amount of land, A 0 , which can be divided among M villages with n farms in each. For simplicity we disregard regional services, which are often provided in larger-scale projects; similarly, we disregard negative externalities such as that wells of several villages draw water from the same limited source. We consider several alternative planning objectives to be maximized. Maximum income in farm A natural objective is to maximize income per farm (Objective F) but this leads to unacceptable plans. To see it, add to the Lagrangian in (4) the 8 E. FEINERMAN ANDY. KISLEV constraint A 0 = MA. The condition for optimal plot size is, with this constraint, PFA/n = 0: if the production function is well behaved and marginal products are positive at any range, .there will be a single village in the region and one or a few fortunate settlers will receive maximum income. If not, and marginal product vanishes at finite area per farm, village size will be determined by the condition PFA;n = 0; still an unac ceptable condition - if land is in any sense a scarce resource in the economy. Predetermined area of farm land Often, a regional program will allot a certain amount of land per farm. The planning problem is then to set optimal village size, and number of villages in the region (the variables n and M). We are still working under Objective F and maximizing income per farm. Substitute now a ( = constant) for Ajn in F( ) in the Lagrangian in (4). Maximization leads again to the factor allocation rules (5a)-(5c) and to the condition for optimal village size (n): Cjn = c~(ny )y {15) which implies t = 0 - all service costs are covered by user charges. With a given plot size, determining n sets area per village and the number of villages in the region (the setting is only approximate as A 0 j(an) = M will most often not be an integer). Equation (15) also implies that in a planned region, with predetermined land allotment per farm, villages operate at minimum Asc (Fig. 1). When the region expands, additional villages are added, each of identical area and number of settlers. This is in analogy to the expansion of a competitive industry, as demand rises, with entry of identical firms, each operating at minimum AC. Settler's rent is, with the current objective and constraints: SR =PFaa + ()L +PFQQ {16) where a is again the predetermined area of land per farm. Parity income A typical objective in regional planning is to set farm income on par with alternative income, say urban wage rate. Farms and villages are then planned to yield net income at the parity level for settlers fully employed on the land allotted to them by the development authority. Translating this procedure to the terminology of the present discussion, the region is SETTLEMENT PLANNING 9 planned for maximum number of farmers (call it Objective N) provided that the alternative cost of labor, w, is exactly covered. The immediate implication is zero settler's rent, SR = 0. With this objective, farm area a becomes an endogenous variable and planning can be done by varying plot size, maintaining the first-order conditions of the previous problem, until the condition SR = 0 is met. Now, by equation (16): (17) Combining with (Sc), PFL < w: farmers settled in the project region will find that they can increase their income by working part-time off their land - provided alternative employment is really available. Thus, for example, a settlement project aimed at reducing the pressure on the local labor market in a rural area may disappoint its planner when the settlers will be found competing with the non-farm laborers in the region. This eventuality can be avoided by letting the settlers reach income exceeding parity levels (at the cost of improving the lot of the settlers compared with those who have to stay with urban wages and increasing the pressure of outsiders to join the project). Income (product) in region (Objective R) Relaxing the previous assumptions of given plot size or parity income, we attempt now to maximize total income in the region (this can be the objective of a government trying to maximize the region's contribution to national product). The Lagrangian is: H =MnP F( ) - wMnL -M c 1(ny) -M c2(Q) +Mn 8(L0 - L) +~t(A 0 -MA) (18) The factor allocation rules (5a)-(5c) still prevail. For optimum n: PF- wL = c~(ny)y +PFA;n(Ajn) (19) Equation (20) is the (approximate) condition for optimum M: P F( ) -wL = [c 1(ny) +c2 (Q)]/n +PFA;n(Ajn) (20) Combining the last two equations: (21) When the region is planned for maximum income, user charges cover cost of services and t = 0. 10 E. FEINERMAN AND Y. KISLEV Settler's rent is the marginal product of the land, PFA;n(Ajn). This is the only case among those considered in the paper in which the value of the right to settle is equal to the value of the land allotted to the settlers. LAND TAXES COLLECTED BY THE NATIONAL GOVERNMENT We introduce now the possibility that the development authority charges rent in the form of land tax. The revenue due to this tax is not used to finance village services, but is taken out of the region. Let the land tax be s dollars per unit of land. We examine the effect of the tax once for the plan of a village and once for the case of regional planning. To consider village planning under Objective F, subtract sAjn from the Lagrangian in (4). The factor allocation rules (5a)-(5c) are not affected, but equation (7) is now: t = (Ajn)(PFA;n -s) (7a) In particular, if the land tax is equal to the marginal product of land, the lump sum tax will vanish (t = 0) and all village costs will be covered by user charges (for water in our example). How will levying a land tax affect village size, the number n, and farm income, sR? To examine size first, consider a village with optimal n, in the absence of a land tax, with aH;an = 0 in (6). Imposing land tax adds s(Ajn) > 0 to the right-hand side of (6); the derivative aH;an is then positive, implying an increase in planned n. This means that the imposition of the tax increases village size. Intuitively, the imposition of the tax increases the cost of land services; consequently, farm-level demand for land decrease and the number of farms rises. To check farm income when s > 0, we examine settler's rent. The tax sAjn has to be deducted from the right-hand side of (8). Therefore, SR in (10) is reduced by the amount s(Ajn): the tax imposed by the development authority or the national government reduces income of settlers in the project. Similarly, the imposition of a land tax in a regional project with parity income as the objective reduces income and settler's rent, and modifies the expression in (17) which is now: ()L = - (PFaa + PFQQ) + s(Ajn) = - [ a(PFa- s) + PFQQ] (17a) Comparing (17) with (17a), one observes that levying an exogenous (to the region) tax reduces the attractiveness of off-farm work. The explanation for this effect is that when the planner aims at maintaining parity, the area a becomes an endogenous variable and the plan calls for larger plots the SETILEMENT PLANNING 11 higher the tax rate s. With larger plots, labor's marginal product increases, and with it the associated shadow price e. On the other hand, with a fixed area per farm or when the objective is maximum income for the region, s is in fact a lump-sum tax; it reduces settler's rent by s(Ajn), but it does not affect allocation considerations [a similar point is made by Mieszkowski and Zodrow (1989) for urban property tax with strict zoning]. COMPARATIVE STATICS In this section we turn to examine the effect of changes in planning variables. Our aim is only to exemplify the approach in order to gain a better grasp of the operation of the model. Assume that the question posed is: in a village plan under Objective F with village land given as A and with s = 0, authorities desire to change the number of settlers per settlement; how will the change affect the plan? Now n becomes an exogenous variable, and the tool of the examination is the analysis of comparative statics. For convenience, define z = ny - total amount of water pumped at the village well. The signs of the relevant second derivatives of the Lagrangian, H, in (4) are [detailed derivations are omitted here (see Feinerman and Kislev, 1987)]: Hzz < 0 HQQ < 0 HzQ > 0 Hzn > 0 HQn indeterminate. The last two cross-derivatives, Hzn and HQn' demonstrate the difference between consumption-type public goods, which we do not consider here, and production services. Consumption services directly enter the utility of the household and, in general, the more households in the community, the larger the marginal contribution of the public good. Production services affect income in combination with other factors. In the expressions for Hzn and HQn' as n increases, land per farm decreases because land is shared with new entrants and this reduction in area subtracts from the contribu tion of the services. For HQn' the reduction of water per farm adds to the negative effect of area contraction and the sign of this cross-derivative cannot be determined a-priori, though in many cases HQn > 0 can be expected. That the sign of Hzn can be determined is due to F( ) being linearly homogeneous; with a different technology, Hzn could be negative or, more likely, also indeterminate. By the analysis, the effects of a change in settlement size on the supply of the public services are: SIGN(dzjdn) =SIGN( -HznHQQ +HQnHzQ) (22) SIGN(dQjdn) =SIGN( -HzzHQn +HznHQn) (23) 12 E. FEINERMAN AND Y. K.ISLEV and these signs depend on the sign of H 0 n in the following way: If HQn ~ 0 (dzjdn), (dQjdn) > 0 (24) Otherwise, SIGN(dzjdn), SIGN(dQjdn) are indeterminate. Hence public services, both chargeable and the pure public good, will unquestionably increase with the number of farms only if a growth in numbers does not reduce the marginal contribution of the public good on the individual farm (recall that the analysis is done for Objective F). A similar analysis shows that, for a given n, optimal supplies of both public services z and Q will increase with the price of the product, P. Returning to the case where n is an endogenous variable, further analysis, again not detailed here, reveals that the effect of a change in P on optimal n, z, and Q is indeterminate. Intuitively, this could perhaps be expected since in general an increase in the price of the product does not affect farm size when land supply is completely inelastic (Kislev and Peterson, 1982). Similarly, for a given n, output per farm - product supply - increases with P. This brings us to the boundary condition: a break-even P will be the price at which settlers cover alternative costs of labor and of capital - if they are required to purchase the right to settle. At a price lower than break-even, settlers will not join the project. In the same way, given P, a break-even w can be calculated, now with urban wages higher then the break-even level settlers will not join. CONCLUDING REMARKS The paper presents a theory of agricultural settlement and examines several cases. We find that village structure - size, services, and finance - vary with the planning objective. In particular, we show that a regional settlement project cannot realistically be planned to maximize farm in come, as this objective leads to the trivial solution of allotting all the land to a single farm. Planning for parity income is possible but it encourages off-farm work away from the project. Levying land rent increases optimal number of settlers and reduces income per farm. The model and the analysis draw on the modern theory of local public goods, but unlike the literature expounding that theory we do not focus on household utility, but rather on production. We are dealing with a simple version of the theory, in which all units are identical. Relaxing this assumption introduces the possibility of heterogenous villages supplying different services to groups of settlers: milk producers and orchard grow ers, for example. Alternatively, settlers may be drawn to villages 'catering' to their specific needs; this is the essence of the Tiebout model of local public goods. The model we presented can also be applied to the analysis SETILEMENT PLANNING 13 of cooperatives; an example can be cooperation in credit, where the public good is the goodwill created by joint efforts and responsibility. But these, as well as empirical applications of the theory, are left to other occasions. ACKNOWLEDGEMENT The work on this study was supported by the US-Israel Agricultural Research and Development Fund (BARD). We received constructive com ments and suggestions from Eitan Berglas, Meira Falkowitz, Esther Red mount, participants in seminars in Rehovot and in the World Bank, the Journal's referee, and the Editor. Remaining shortcomings are ours. REFERENCES Arnott, R.J. and Stiglitz, J., 1979. Aggregate land rents, expenditure on public goods, and optimal city size. Q. J. Econ., 43: 471-500. Berglas, E., 1982. User charges, local public services, and taxation of land rents. Public Finance, 37: 178-88. Berglas, E., 1984. Quantities, qualities and multiple public services in the Tiebout model. J. Public Econ., 25: 299-321. Feinerman, E. and Kislev, Y., 1987. Agricultural settlement with alternative objectives and constraints. Work. Pap. DRD 236, World Bank. Washington, DC, 28 pp. Kirsch, O.C., Benjacov, A. and Schujmann, L., 1980. The Role of Self-Help Groups in Rural Development Projects. Breitenbach, Saarbriicken, 232 pp. Kislev, Y. and Peterson, W., 1982. Prices, technology, and farm size. J. Polit. Econ., 90: 578-95. LeVay, C., 1983. Agricultural cooperative theory: a review. J. Agric. Econ., 34: 1-44. Mieszkowski, P. and Zodrow, G.R., 1989. Taxation and the Tiebout model: the differential effects of head taxes, taxes on land, rents and property taxes. J. Econ. Lit., 27: 1098-1146. Oi, W.Y. and Clayton, E.M., 1986. A peasant's view of a Soviet collective farm. Am. Econ. Rev., 58: 37-59. Rochin, R.I. and Grossman, J.C., 1987. Peasant cooperatives and government control in Egypt. J. Rural Dev., 25: 27-49. Rubinfeld, D., 1987. The economics of the local Public Goods. In: A.J. Auerbach and M. Feldstein (Editors), Handbook of Public Economics. North-Holland, Amsterdam, pp. 571-645. Vanek, J., 1970. The general theory of labor-managed market economies. Cornell, Ithaca, NY, 409 pp. Zusman, P., 1988. Individual Behavior and Social Choice in a Cooperative Settlement. Magnes Press, Jerusalem, 324 pp. Recent Experience with Cooperative Farm Credit in Israel* Yoav Kislev, Zvi Lerman, and Pinhas Zusman Hebrew University, Rehovot Governments and aid agencies in many developing countries channel credit to agriculture. The goal is to promote development, help the poorer peasants, encourage innovations, and eliminate the reliance on the local allegedly monopolistic moneylenders. Credit is deemed appropriate for achieving these goals since it is easy to administer, its effect is immediate, and it can be directed to the desired groups in the population. However, experience has shown that credit programs tend to fail.' Subsidized credit introduces inefficiencies, most of the credit goes to the richer farmers, and it encourages capital intensification and replacement of labor by machinery-hardly the goals governments aim for.2 Cooperative credit was introduced in several places to improve distribution of subsidized loans and to increase repayment rates, but most of these efforts also failed.3 These failures of outside intervention can be explained within the perspective of modern theory, which has improved our understanding of (and increased our sympathy to) the working of rural credit markets and their institutions in poorer coun- tries.4 Israel is not a poor country; its agriculture is well endowed with capital and is technically advanced, and while informal (gray) credit markets exist, most of the financial services are provided by commer- cial banks. Still, there are many similarities between the basic structure and problems of the credit market in Israel and those of developing countries. Central among them is the problem of control, which lies in the basis of the theoretical analysis and is of major practical impor- tance both in rural credit markets for developing countries and in the cooperative-intensive Israeli farm sector. In this article we summarize lessons drawn from many years of experience with cooperative agricul- tural credit in Israel. The discussion is limited to the sector of the moshavim, cooperative villages of family farms; the experience of the large-scale communal kibbutzim is not considered at this stage. ? 1991 by The University of Chicago. All rights reserved. 0013-0079/91/3904-0033$01.00 This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 774 Economic Development and Cultural Change Background and Summary Eighty percent of Israel's agricultural product comes from cooperative farms. A major form of cooperation in agriculture has been financial. It flourished for a long time, but financial cooperation has recently been in deep trouble and will need massive public assistance to overcome its difficulties. It is now too early to predict what kind of cooperation, if any at all, will emerge from the crisis. Yet, important lessons can be learned even from this incomplete experience. In a nutshell, inflation in Israel accelerated steadily from a yearly rate of 12% in the early 1970s to more than 500% (annualized) for the first half of 1985. It was then halted by an abrupt change of direction in government policy, and since then inflation has run at approximately 20% per year. The rising prices were fueled by expanding supply of credit, much of it imported, market interest rates lagged behind infla- tion, and real rates were negative for most of the decade ending in 1985. These conditions encouraged overinvestment and discouraged saving. But interest rates also lagged when inflation decelerated in 1985, and, as a result, real rates skyrocketed. Agricultural coopera- tives, as well as many other business enterprises, collapsed. A financial crisis breaks out when lenders realize that the real economic value of the debtor's assets is too low to service the out- standing debt. In general, three groups of factors are involved in the creation of a crisis: (a) The basic fault: included here are the factors that reduce the income streams and hence the value of the assets-a decrease in the price of the product, increase in input cost, increase of the long-run rate of interest. In Israeli agriculture (as we shall show), the basic fault was due to undersaving, overinvestment, and overca- pacity on the farms and in regional cooperative enterprises. (b) The trigger: this is the change in the economic environment that exposes the basic fault and triggers the crisis. In Israel, this was the abrupt rise in the rate of interest.5 (c) The manifestation of the crisis: it always takes the form of a financial failure. Banks and other lenders that share the risk with the borrowers will precipitate a crisis in an effort to salvage at least part of their capital. A high proportion of equity capital cushions the crisis, and firms with a high ratio of equity to assets are more likely to weather difficul- ties, although possibly at some loss to their owners. Extensive reliance on loans, on the other hand, with fixed interest and tight repayment schedules, increases the variability of the net residual incomes, finan- cial stress, and risk. Agriculture with its small units, and particularly cooperative agriculture, is especially vulnerable: it cannot raise equity by stock issues in the capital market and expansion relies entirely on the sources of the families or the members in the cooperatives and on debt financing. Saving and accumulation of equity are thus particularly important in agriculture. This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 775 Indeed, agriculture in Israel enjoyed many years of low, even negative interest rates, in which the real value of its debt was eroded. If the capital gains due to the erosion of the debt had been retained, equity capital would have increased. But most farmers and coopera- tives did not exploit this opportunity; in agriculture, low interest rates resulted in overinvestment and overreliance on debt financing. Partly this was an intended policy-removal of equity to enjoy inflation- related tax benefits; partly these were investment and consumption based on ample supply of credit. At any rate, excessive expansion of investment and credit was a myopic and dangerous policy, gambling with the property of the farmers. And despite the opportunities for accumulation of equity, debt to assets ratios increased and the stage was set for the coming crisis. Agriculture, and particularly cooperative agriculture, was thus the victim of inflation and the measures implemented to halt it. It was lured into overinvestment and excessive leverage by availability of cheap credit under inflation and found it could not service its debt when the real interest rates shot up. There was, however, a deeper factor involved. Basically the crisis is the result of control failures in cooperative agriculture-failure of the cooperatives to control bor- rowing by their members, failure of the members to control the associ- ations' officers, failure of the banks to control the cooperative borrow- ers, and above all failure of the government to control the affairs of a sector in the financing of which it was intensively involved. These failures reflected, on the one hand, the weaknesses inherent in the cooperative structure with its group decisions, internal politics, and the public good nature of its operation and, on the other hand, the inflationary environment and government policies that encouraged in- vestment and intensive reliance on debt financing. The Moshav and Second-Order Cooperation A moshav (plural, moshavim) is a farming community in which all farms are family owned and operated, and all farmers are members of the multipurpose, democratically run, village cooperative. In principle (practice varies) the cooperative association in the moshav purchases all farm supplies for its members and markets their farm products. It may also own and operate a variety of service facilities and manage directly some jointly operated productive enterprises. In addition, the association encompasses all municipal and many social functions in the village. Moshavim are members in two types of second-order coopera- tives: supply cooperatives (requisite societies, purchase organization) set up to purchase farm requisites for their member moshavim, and regional service enterprises (feed mills, slaughterhouses, transporta- tion services, and others). Both types operate on a regional basis, This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 776 Economic Development and Cultural Change though some nationwide cooperatives also exist. The supply coopera- tives act also as spokesmen of their regions in government offices; they engage in intensive lobbying and have acquired a strong political standing. The basic structure and mode of operation of the moshavim and the supply co-ops are similar, though differences in size, scope, and complexity create significant differences in the organizational prob- lems the cooperatives face at the primary and the secondary levels. But we shall mostly overlook the differences in this short survey and focus the discussion on the major common characteristics. The re- gional service enterprises, whose nature is more that of industrial com- plexes than of regular farm cooperatives, will be discussed separately. Moshavim were settled on national land that cannot be used as collateral; cooperative credit was therefore a natural alternative to a commercial credit market that could perhaps function efficiently if farmers had full property rights to the land they cultivated. Indeed, starting with the transfer of suppliers' credit to their members, both the moshav and the supply cooperative expanded into full-scale financial intermediation. In the last 2 decades and up to the eruption of the current crisis, the associations in the moshav and in the supply cooper- ative were first and foremost credit associations. Financial Intermediation The pivotal role of credit intermediation in the activities of the moshav and the supply cooperative is demonstrated in their balance sheets, as shown in table 1.6 Members' debit balances are by far the largest assets the associations hold-76.6% of the total in the moshav and 60.9% in the supply cooperative. The moshav and the regional co-op raise capi- tal and transfer it to their members. The associations also function as clearing houses, accepting deposits from members with financial surpluses (members' credit balances as in table 1) for use by others. The supply co-op and its moshavim are strongly linked through credit, as shown in table 1, and through joint ventures in the regional service enterprises. The government was the major source of long-term credit and the major lender in the early stages of the development of moshavim.7 Government credit was generally supplied on concessionary terms, and its minuscule share (see table 1) is a reflection of both inflationary erosion and the growing availability of alternative sources of finance. Financial reports in the moshavim were prepared in historical val- ues, and equity capital is under-represented. It was estimated that, if adjusted for inflation, equity will reach between 15% and 30% of the value of the associations' liabilities. But even then, items reflecting financial intermediation will dominate the cooperatives' balance sheets. This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms TABLE 1 BALANCE SHEET COMPOSITION OF A MOSHAV ASSOCIATION AND A SUPPLY CO-OP, SEPTEMBER 30, 1981 (in % of Total Assets) Supply Supply Assets Moshav Co-op Liabilities Moshav Co-op Fixed assets 3.7 3.5 Equity .7 3.0 Long-term investments and loans to Long-term debt 4.2 19.5 members 3.5 13.7 Short-term loans .6 34.5 Inventories 4.0 ... Short-term loans from supply co-op 76.9 \1 \1 Accounts receivable: Suppliers' credit 4.1 21.8 \1 Nonmembers 12.2 3.6 Members' credit balances 13.5 21.2 Regional enterprises . .. 18.3 100.0 100.0 Members' debit balances 76.6 60.9 Total 100.0 100.0 SouRcE.-Pinhas Zusman, Individual Behavior and Social Choice in a Cooperative Settlement (Jerusalem: Magnes Press, Hebrew University, 1988). NOTE.-The data for the supply co-op are for the regional cooperative in the "Mountain Region" (a fictitious name). The data for the moshav are for an average association in a sample of 13 moshavim in the same region. Balance sheets are prepared in historical values, not adjusted for inflation. The financial reports of the cooperative in the moshav are for the association, not for the whole village. Information on individual farms is not included and is generally not available. This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 778 Economic Development and Cultural Change The supply cooperatives thus succeeded to establish for their members the moshavim-and the moshavim, for their members, the farm operators-financial services with steady lines of credit and con- venient saving facilities. Because of proximity and familiarity, asym- metric information, which is generally of major importance in credit markets, is not a significant problem in cooperative agricultural credit in Israel. Still, interlinkage of credit and marketing is practiced, farm- ers are expected to channel the proceeds of their marketed products through the moshav association, and it receives the market revenue through the supply co-op. Interlinkage forms part of the institutional set-up that replaces collateral for loans. Proximity, central purchasing of inputs, product marketing, and financial interdependency should, in principle, allow close monitoring and control of the member bor- rowers. Advantages and Weaknesses of Financial Cooperation The "classical" discussion of the theory of cooperatives struggled with the definition of the goals of the cooperative firm and its behavior.8 Difficulties created by the cooperative's egalitarian democracy were recognized but not formulated explicitly and not examined analyti- cally. In the modern approach the cooperative is seen as a collection of individuals, each guided by personal tastes and interests, but all agreeing to perform jointly certain economic functions.9 This contrac- tual perception enables theoretical analysis which enhances our under- standing of two central facets of the cooperative model of action: (a) Laws and regulations governing cooperative life are often compro- mises and are not necessarily first best, Pareto efficient. This subject is treated at length by P. Zusman and will not be elaborated further here. (b) Members in cooperatives-in our case farmers in a moshav or moshavim in a supply co-op-are not subordinates in a centralized hierarchy; they are free to act within wide limits. By treating explicitly individual behavior and group decisions, the modern, contractual per- ception of cooperation throws new light on the advantages of coopera- tive credit intermediation and particularly on its weaknesses. Advantages. With cooperation, members in the moshavim, and moshavim in the supply co-ops, enjoy economies of scale in loan pro- cessing, professional financial management (particularly important in a high-inflation high-tax economy such as Israel), and a stronger bar- gaining position in the credit market (as well as in other markets). Perhaps the greatest advantage of cooperative credit, both in the moshav and in the supply co-op, lies in risk pooling implemented in two ways. In the short run, the association can use its own resources to smooth over the credit needs of its members. Outside lenders do not have to deal with transitory difficulties of individual borrowers. A more fundamental mode of risk pooling is mutual liability and This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 779 guaranty. Members in the moshav sign mutual guaranty agreements for the moshav association, and representatives of moshavim pledge similarly for loans raised by the supply co-op. The social pressure to comply with cooperative norms is strengthened under mutual liability arrangements. The probability of association default is reduced. Banks evidently recognize the advantage inherent in this arrangement, as credit is often conditioned on renewal of mutual liabilities. Weaknesses. Several kinds of structural difficulties afflict the moshav and the supply co-op. (a) Moral hazard-members may tend to invest on their farms and in the moshav's enterprises in risky proj- ects knowing that with a mutual liability arrangement they will be bailed out should the returns on the investment be disappointing. (b) Free riders-a member in the moshav, or a moshav in a supply co-op, may choose to market a farm product privately, thus weakening the association's standing in the credit market. (c) Agency cost-banks and other lenders view the cooperative associations as their agents and expect them to protect their interest (by limiting credit to failing members, e.g.), but the associations are guided by their own interests which are not always identical to those of the lenders. Similarly, offi- cers in the associations may be tempted to expand operations and to assume risks which prudent members would avoid. (d) Horizon problems-members may favor short-term gains in expectation that in the long run they may exit, leaving those who stay to carry the re- maining debt. Enforcement of the moshav's norms and rules-in practice, mainly enforcement of the interlinkage arrangements of collective mar- keting through the moshav and through the supply co-op-is critical to its continued functioning as a credit cooperative. However, compli- ance with the moshav's code requires high standards of cooperative ethics and will to enforce. Conceivably, interdependence of the degree revealed in table 1, and close monitoring due to interlinking of credit with product marketing, would allow effective control. However, this was not the case. Particularly where interdependence was strong, the moshavim and their members had only limited access to alternative sources of credit and, consequently, the cooperatives were committed to continue funding their members. Their elected officers could hardly afford the dire financial, social, and political consequences of mem- bers' bankruptcies. These difficulties are reflected in the behavior of members of coop- eratives. A reasonable rough measure of the financial exposure of a member is the ratio of outstanding debt to monthly sales through the cooperative ("credit months"). During the period 1977-81, 13 out of 24 moshavim in the "Mountain Region" of table 1 exceeded 12 credit months and for relatively long durations.'0 Moshavim with 55 and 42 credit months were observed in another supply co-op." Similarly, This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 780 Economic Development and Cultural Change more than a few farm operators owe their moshav cooperative associa- tions several times their yearly production capacity. Moshavim or indi- vidual farmers with such heavy burdens of debt compared to their production capacity will never be able to repay their loans or service them adequately. Heavy debt burden does not happen overnight; it evolves gradually. Moshavim or individuals with 55 or 42 credit months testify to the weakness of their cooperatives, a weakness that breeds permissiveness and lax financial discipline. Evidently the cooperatives did not function efficiently as financial intermediaries; they could not enforce credit rationing, they allowed cost of debt to increase when conditions changed, and they encouraged negative selection of investment projects, thus increasing the risk in their portfolios-a risk that all their members shared.'2 This behavior is not incidental; it stems from the nature of cooperation, and it is the result of the distorted incentive system created by the mutual guaranty arrangements. Moral-hazard behavior and weakness of cooperative control increase the risk to the lender and may even outweigh the advantages of cooperative credit. This indeed has happened as a result of the last crisis-lenders have recently been reluctant to extend credit to moshavim and second-order cooperatives even for loans backed by mutual guaranties. Regional Enterprises Regional service enterprises were ordinarily organized as limited liabil- ity cooperative associations, and their establishment was financed mainly by development grants and soft-term loans. Their membership consisted of moshav associations, mostly potential patrons of the ser- vice offered. Often the regional supply co-op was also a member, and in all cases it provided the enterprises in the region with short-term finance and purchasing services. Strong economic connections devel- oped between the two kinds of regionals-the supply co-op and the service enterprises-a relationship that proved detrimental when the recent crisis erupted. Zealous adherence to rural development by public agencies, easy access to credit through the supply cooperatives, and strong political regional lobbies resulted in overexpansion of service enterprises. This occurred particularly in the 1970s when credit was in ample supply and economic optimism ran high. Consequently, in the early 1980s many service enterprises operated under capacity-50%-60% by the estimate of the state controller. Not unlike firms in a cartel, regionals scrambled to get their share in the service enterprises and the expected benefits they were to yield. The final outcome, however, was that many of the enterprises did not cover their operating costs. The supply cooperatives, assuming the role of the financiers of last resort, found themselves financing not only operating losses but This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 781 also debt service of the regional enterprises. As early as 1981, the share of credit to the regional enterprises in the assets of the co-op, shown in table 1, was 18.3%; it grew substantially thereafter. In an- other co-op we found that a regional slaughterhouse, which started operation in 1981 with equity equaling 25% of the value of its capital,'3 accumulated losses right from its start, and by 1985 its debt reached 2.5 times the value of its assets, most of that short term and to the supply co-op. This was an extreme case but not atypical. When the enterprises went bankrupt in 1985, they took the supply co-ops down with them. Cost of Debt Perhaps the greatest damage that inflation inflicted on the Israeli econ- omy was the distortion of the cost of capital. Real interest rates varied markedly; cost of borrowing was at times extremely high while at other times, and for other loans, it was very low. For more than a decade, since the early 1970s, real rates of interest for most sources of credit were negative. This was particularly true for government-supported development loans until they were linked to the price index and tax- deductible interest payments were adjusted to inflation (these mea- sures were introduced in 1978 and 1982, respectively). In a preliminary survey of eight cooperatives, both village and regional (fig. 1), we found that average effective real cost of outstanding debt was zero in 1971, and it declined gradually thereafter as interest lagged behind the accelerating inflation; it was minus 40%-50% per year in 1984. When inflation was halted in 1985, interest again lagged, and this time average real rates jumped to plus 15%-20%. Current cost of credit fluctuated even more and varied greatly between sources; in 1984 the real rate of interest on the subsidized government-directed short-term credit was minus 59%. In 1985 the real cost of overdraft facilities, one of the most expensive sources, was plus 100% per year. Credit Supply to Agriculture It has often been claimed in Israel that agriculture suffers from short- age of credit. However, examination of the available information re- veals that credit has been in ample supply. Between 1970 and 1987 credit to agriculture increased more than it did for the economy at large and more than for industry (table 2). While the share of agriculture in net domestic product of the business sector has been 6%-7%, over the last 2 decades its share in the volume of credit was higher than 10%. With inflation, financial leverage increased. In 1986 the ratio of outstanding economy-wide credit to GNP was twice its 1969 value; in agriculture the corresponding ratio increased by a factor of 3.8. The ratio of credit to net capital increased in agriculture between 1969 and 1986 by a factor of four (fig. 2), while in industry it rose over the same This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 782 Economic Development and Cultural Change 100 I I 50 z u 0 Regionals a- Moshav im - 5- -100III I I 1972 1975 1978 1981 1984 1987 FIG. 1.-Real rates of interest: four Moshavim and four regionals period by only 20%. As we have seen, credit was underpriced, and the low and negative real rates of interest, encouraging demand, evidently contributed to the feeling of shortage. Another claim often made in Israel relates to inadequate term structure of credit. Not enough long-term loans were available, and investment projects had to be financed by short-term credit, creating a financing gap between the expected life of the assets and the duration of the loans. Again, with negative real interest rates and easy access to short-term credit, many farmers and cooperatives knowingly financed investment with short-term loans and knowingly created financing TABLE 2 OUTSTANDING BANK CREDIT (End of Year) INDEX OF REAL VOLUME OF CREDIT (1969 = 100) CREDIT TO PRODUCT RATIO Economy Agriculture Industry Economy Agriculture 1970 138 136 124 31 48 1974 236 209 219 48 76 1979 351 335 327 57 101 1984 427 491 352 57 177 1987 535 655 402 67 182 SoURCEs.-Bank of Israel, Annual Bank Statistics, various years, and the Research Department of the bank; Central Bureau of Statistics, Statistical Abstracts of Israel, various years. NOTE.-The economy-wide credit to product ratio is the ratio to GNP (the public sector is included both in the numerator and the denominator of the ratio); in agriculture, the ratio is to the sector's product. The credit to product ratios in the first line of the table are for 1969. This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 783 100I I 80 z 60- w To Net Capital 20 To Gross Capital o 1970 I1973 II1976 1979I 1982I 1985, FIG. 2.-Financial leverage in agriculture: ratio of debt to capital gaps. In part, however, the government was also responsible for the financing gaps. Government-approved development loans were often supplied with delays, forcing reliance on short-term finance. In periods of inflation, delays create not only temporary but also permanent gaps in financing due to the inflationary erosion of the real value of the loans, which were late to arrive. In addition, loans at nominal interest rates that reflected inflation even partly were of shorter than their specified duration. In real terms they were retired earlier than speci- fied, and investors had to turn to short-term credit to finance the gaps thus created.'4 Whatever the origin of the financing gaps, farmers and their orga- nizations turned to the government for help, usually with the appro- priate political backing. There were many cases, in recent periods almost one every year, of "conversions"-rescheduling of loans. Short-term credit was replaced by long-term loans, mostly on conces- sionary terms. The recurrence of the conversion episodes, sometimes general and sometimes specific to certain farms or regions, was one of the major reasons for the widespread belief that agriculture would not be allowed to fold. However, the remedy was not always effective. In many cases farmers and cooperatives returned to the preconversion maturity structure just several years after rescheduling. With negative interest rates and ample supply of credit, invest- ment expanded to the range of negative returns. Returns on capital in agriculture, which were 6%-7% in the mid-1970s, decreased gradually and were minus 4% in 1987. Detailed information on investment in consumption assets is not available but, as visitors to moshavim (and kibbutzim) could easily observe, houses grew and the number of cars increased. It seems that at least part of the capital gains due to negative cost of capital were channeled to consumption. This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 784 Economic Development and Cultural Change Government The particular circumstances of the early stages of development in Israel, in which penniless immigrants were settled with public assis- tance on national land (many of them in the early 1950s), created spe- cial fosterage relations between the moshavim and the public agencies. Over time many of the newly settled operators acquired farming skills, and cooperation also became well established; but the view, held not only by farmers, that it was the role of the government to maintain the welfare of the farming sector and the expectation that the government would shoulder this responsibility did not wane. Cooperation when agriculture expanded, was, and still is, sup- ported by the government; for example, new immigrants were settled in moshavim as a matter of policy, and land and water were allotted to the moshav to be distributed equally among the members. Today production quotas are allocated on a village basis, and the moshav decides on internal distribution; government agencies usually consult with the cooperative association in the moshav on the allocation of long-term loans to farm operators. The most profound public involvement has been in regulating credit. By deciding on the allocation of subsidized credit, the govern- ment affected regional development, product mix, and farmers' in- come. The dependency on the government and the expectation that it will bail out farmers and moshavim in trouble created moral hazard problems, not unlike those that mutual guaranty created in moshavim and regional cooperatives. Lacking the usual mechanism of collateral, the government turned to close monitoring-"concentrated credit." In a program started in the early 1960s, a moshav or a kibbutz concen- trated all its financial activity in a single bank, and credit was supplied, either for investment or for short-term need, only with the approval of a steering committee consisting of representatives of the bank, the Ministry of Agriculture, and other public agencies. Joining "concentrated credit" was voluntary and moshavim were lured by additional loans. And indeed, within a few years most of the moshavim in the country participated in the program. However, the increasing credit supply in the 1970s, and particularly the convenient sources offered by the regional cooperatives, eliminated the advan- tages of concentrated credit for the moshavim, and the program folded in the mid-1970s. Thus the problem of moral hazard in the moshavim was in fact recognized, and tools were created to overcome it, at least partly, but the will to maintain a strict policy could not withstand the flood of available credit. (Concentrated credit is now proposed again in reaction to the current crisis.) Moreover, the government supported the supply cooperatives and encouraged them to expand their role- they often received directed credit on favorable terms, were assisted in the establishment of the regional service enterprises, and were made This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 785 parties to development loans by advancing short-term credit to bridge over the delays in government-approved loans. Evidently, even before the recent acute crisis, and despite the experience with recurrent fi- nancial difficulties and loan rescheduling in the moshavim, the danger- ous implications of unconstrained operation of financial intermediaries were not understood or perhaps they were understood but not acted on by the government. Crisis On July 1, 1985, the government adopted a package of policy measures designed to curb inflation, which was then running at 500% per year. Wage and price rises were halted by decree, government outlays were curtailed in order to balance the budget, credit supply was squeezed- the banking system's reserve ratio rose by more than 75% between 1985 and 1986-and the exchange rate was pegged. These measures created an unprecedented increase in the real rate of interest; it reached 100% (annualized) on some kinds of credit. The rise in interest increased the cost of capital to holders of nonlinked credit, but the real burden of index-linked loans with fixed real interest rates also increased. Loans are mostly linked to the "index known," which means in Israel that linking is with 1-month delay; when inflation is stopped, indexation calculation goes on for 1 additional month (the indexation on a 2-year loan taken on July 1, 1984, was calculated for 13 high inflationary months and 11 low inflationary ones, while the loan was kept for equal periods of high and low inflation). This factor alone increased the real value of index-linked debts by 10%-20%. Once creditors realized that agriculture, particularly cooperative agriculture, could not continue to service its debt with the exceedingly high, postreform real rates of interest on short-term loans, and that the government could not bail the sector out any more, the crisis erupted. Private lenders and commercial banks were not willing to extend addi- tional credit and insisted that first loans be repaid. This was impossible, and most regional cooperatives and many of the associations in the moshavim collapsed. Farm production has continued, often with pri- vate credit arrangements (wholesalers, e.g., pay in advance for farm products) and the farmers' own resources. But this cannot be a com- plete solution to the crisis. In most cases, the available sources will be insufficient for investment in equipment and machinery, and farm- ers will find it hard to renew their production assets; and, in addition, banks and other creditors are still demanding repayment of their loans. For most members in the cooperatives the heavy burden is not their own debt but their share in the mutual liabilities-their share in cov- ering the debt of several heavy borrowers in the moshav and the debt of the regional enterprises. Agriculture could not repay or service its debt in full; the question This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 786 Economic Development and Cultural Change was how to distribute the losses. Once this was realized, the govern- ment stepped in offering support in an effort to reach a debt settlement between the lenders, particularly the banks, and the moshavim. An agreement was formulated in 1988, but its implementation has been slow, as farmers still hope that they can gather political support for a more favorable settlement. Recapitulation Inflation created a special opportunity for agriculture in Israel and particularly for cooperative agriculture. With negative real interest rates and erosion of loans, agriculture could increase its equity capital and emerge from the inflationary period economically stronger. This did not happen; as we have seen, financial leverage increased in agri- culture, returns on capital, and probably also savings, were negative,15 and farmers sank deeper in debt, partly to finance investment in pro- duction assets (often with overcapacity), partly to finance housing and consumer durables, and partly to increase current consumption and standards of living. Considerations of short-run inflationary gains dom- inated long-run economic health. Myopia is common; but it afflicts cooperatives more strongly than individuals and private enterprises because of the cooperatives' politi- cal structure and the ensuing characteristics, which lead to moral haz- ard behavior, free riding, agency costs, and horizon problems. But the cooperatives were not the only ones at fault. Credit was also distrib- uted by the commercial banks; it was their money that was lent, and it was their responsibility to secure the loans and to control their use. Evidently they neglected this responsibility. The main blame for the breakdown in agriculture, however, lies with the government. By its policy, which was a result of its effort to accelerate development and its yielding to political pressure, it created the false impression that it would bail agriculture out of any difficulty. Moreover, the government carries the major blame for overcapacity in agriculture. Farmers and regional officers naturally tend to increase their share in aggregate capacity. Since the funding of most of the development projects was with government approval and assistance, it was the duty of the government to examine the aggregate picture and to balance the desire to invest against the needs. This was not done; the decisions of the policymakers and even the recommenda- tions of the Planning Authority of the Ministry of Agriculture encour- aged overinvestment. Governments are often too weak and they yield to the pressure of short-run interests. As a result, the crisis in coopera- tive agriculture is largely the outcome of the favoritism it enjoyed for a long time. Attempts were made to divide quantitatively the blame for the crisis, and the responsibility for its magnitude, among the parties-the This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 787 government, the banks, and the farm sector. A typical approach was to simulate financial development with alternative rates of interest and to attribute to macroeconomic policies excess debt over the level that the debt would have reached with "acceptable" rates of interest. Cal- culations of this sort overlook the role of credit cooperation in the creation of the debt, in reliance on short-term and risky sources of finance, and in the erosion of equity capital in agriculture. But then again, the cooperatives and the banks were encouraged-explicitly and implicitly-to follow risky and evidently reckless strategy by gov- ernment action and advice. We are left, therefore, with the qualitative and somewhat subjective judgment that, though each of the sectors was responsible for its part in the creation of the conditions that led to the crisis, the major blame lies with the government. Cooperation in general, and cooperation in credit in particular, has many advantages but, as we have seen, it also suffers from in- herent weaknesses. It is not clear whether intensive cooperation in agriculture-if not heavily assisted by public funds-can succeed or even survive in the long run the economic test of competitive markets. But the test of the current crisis is much harsher. Even if it is basically viable, cooperation may now be destroyed because of the particular crisis conditions. Lessons and Recommendations The failure of cooperative agriculture was a failure of control. If coop- eration in agriculture and particularly cooperation in credit is to sur- vive and succeed, control has to be tightened. Control is expensive, however, and often inconvenient; both incentives and appropriate structures are needed to assure optimal control. First and foremost, the government cannot and should not take explicit or implicit responsibility for agriculture or for cooperatives. Once the government sheds its responsibility, both farmers and lenders will know that they are the sole residual claimants for success and for failure. It will be in their direct interest to tighten control and to follow prudent economic policies. Mutual guaranties should be severely lim- ited to reduce moral hazard behavior at the farm and in the coopera- tive; and external market control of cooperatives should be established wherever possible. A necessary condition for effective control is the availability of accurate information. Financial reports, including balance sheets and income accounts, should be prepared and published regularly. The reports will have to be adjusted for inflation as prices are still rising in Israel at 15%-20% per year. Supply co-ops will have to be limited to commercial activity; they should no longer act as financial intermediaries. The regional service enterprises ought to be incorporated into limited liability companies This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms 788 Economic Development and Cultural Change with members of the owner-moshavim receiving marketable shares. Moshavim and their members should be free to patronize service enter- prises of their choice, whether in their region or elsewhere. Members in the moshavim should also be free to leave their coop- eratives and to operate privately or to form alternative organizations, revealing and realizing in this way their preferences for stronger or weaker cooperation. Exit is expensive; it raises the average cost of services to the remaining members, but it is often the only way for patrons to enforce efficiency and for minorities to voice their opposi- tion.'6 Lack of control may be more expensive. The structural changes that we are proposing-and in many cases we adopt proposals that have already been made in Israel-are not easy to implement. Exposing the regional enterprises to market compe- tition may seem extremely painful in the short run; and indeed the Debt Settlement Administration is attempting to cure the enterprises by erasing their debt and assuring capacity operation by tying mosha- vim to their services. In the long run this is a recipe for inefficiency. Similarly, in moshavim, exit is not simple. The exiting farmer may forfeit his allocation of land and water and, more probable, his produc- tion quotas and development loans. The implementation of the changes we are proposing will require modifications of both law and attitude in Israel. Notes * This article draws on our report to the World Bank, "Experience with Credit Cooperatives in Israeli Agriculture," World Bank Working Paper WPS 156 (1989). The preparation of the article was partly supported by the United States-Israel Agricultural Research and Development Fund, BARD. We are indebted to Avishay Braverman for raising the issues discussed in the article, for assistance, and encouragement. Responsibility for the analysis and the opinions expressed is ours. 1. J. D. Von Pischke, Dale W Adams, and Gordon Donald, eds., Rural Financial Markets in Developing Countries--Their Use and Abuse (Baltimore: Johns Hopkins University Press, 1983); Avishay Braverman and J. Luis Guasch, "Rural Credit Markets and Institutions in Developing Countries: Les- sons for Policy Analysis from Practice and Modern Theory," World Develop- ment 14 (1986): 1253-67; Dale W Adams, "The Conundrum of Successful Credit Projects in Floundering Rural Financial Markets," Economic Develop- ment and Cultural Change 36 (1987): 355-86. 2. For a successful experiment with rural credit (so far, at least), see Mahabub Hossain, Credit for Alleviation of Rural Poverty: The Grameen Bank in Bangladesh (Washington, D.C.: International Food Research Policy Insti- tute, 1988). 3. Avishay Braverman and J. Luis Guasch, "Institutional Analysis of Credit Cooperatives," World Bank Discussion Paper (World Bank, Washing- ton, D.C., 1988). 4. Clive Bell, "Credit Markets and Interlinked Transactions," in Hand- book of Development Economics, ed. Hollis Chenery and T. N. Srinivasan (Amsterdam: North-Holland, 1988), 1:763-830; Joseph E. Stiglitz, "Economic This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms Y. Kislev, Z. Lerman, P. Zusman 789 Organization, Information, and Development," in Chenery and Srinivasan, eds., pp. 93-160. 5. Deteriorating terms of trade were also blamed in Israel for inducing the crisis. The terms deteriorated steadily over the decades of the 1960s and the 1970s (at an average yearly rate of 2.4%), but this trend was reversed in 1981 and the terms of trade of agriculture have improved since then. They could not have caused the crisis in 1985. 6. For additional details, see Zvi Lerman, "Capital Structure of Agricul- tural Co-operatives in Israel," Yearbook of Co-operative Enterprise 1989 (Ox- ford: Plunkett Foundation for Co-operative Studies, 1989). 7. We lump together government and other public agencies. 8. Clare LeVay, "Agricultural Cooperative Theory: A Review," Journal of Agricultural Economics 34 (1983): 1-44. 9. P. Vitaliano, "Cooperative Enterprise: An Alternative Conceptual Ba- sis for Analyzing a Complex Institution," American Journal of Agricultural Economics 65 (1983): 1079-83; Pinhas Zusman, Individual Behavior and Social Choice in a Cooperative Settlement (Jerusalem: Magnes Press, Hebrew Uni- versity, 1988). 10. Zusman. 11. Yoav Kislev and Arie Marvid, A Supply Cooperative Mishorim (in Hebrew) (Jerusalem: Magnes Press, Hebrew University, 1988). 12. On the theoretical basis for these arguments, see Joseph E. Stiglitz and Andrew Weiss, "Credit Rationing in Markets with Imperfect Informa- tion," American Economic Review 71 (1981): 393-410. 13. Kislev and Marvid. A large part of the equity accumulated was due to inflationary erosion of unlinked loans. 14. As an example, consider a 2-year loan of $100 at zero real interest; the yearly rate of inflation is 100%. There are two alternative payment sched- ules, and in both the principal is repaid at the end of the second year: (a) a 70% nominal rate of interest, paid at the end of the first and of the second year; (b) the loan is linked to the price index and the only payment is at the end of the second year. Debt service will then be as follows: Alternative a Alternative b End of first year 70 End of second year 170 400 Present value at the beginning of the first year (discounted at 100%) 77.5 100 The nonlinked loan carries a nominal interest lower than the rate of inflation and involves a grant of $22.5. But it requires, in the example, repayment of 35% of the principal at the end of the first year. This is the sense in which investors may be forced to short-term financing of their projects-loans car- rying nominal interest rates are of shorter duration than their stated maturity. 15. Negative returns and consumption of capital gains may reflect optimal behavior when interest rates are negative, but erosion of equity capital can hardly be optimal. 16. A. O. Hirschman, Exit, Voice, and Loyalty (Cambridge, Mass.: Har- vard University Press, 1970). This content downloaded from 109.67.249.69 on Fri, 15 Sep 2017 06:58:19 UTC All use subject to http://about.jstor.org/terms SUSTAINABLE AGRICULTURAL DEVELOPMENT: THE ROLE OF INTERNATIONAL COOPERATION PROCEEDINGS OF THE TWENTY-FIRST INTERNATIONAL CONFERENCE OF AGRICULTURAL ECONOMISTS Held at Tokyo, Japan 22-29 August 1991 Edited by G.H. Peters, Agricultural Economics Unit, Queen Elizabeth House, University of Oxford, England and B.F. Stanton, Cornell University, USA Assisted by G.J. Tyler University of Oxford INTERNATIONAL ASSOCIATION OF AGRICULTURAL ECONOMISTS QUEEN ELIZABETH HOUSE UNIVERSITY OF OXFORD 1992 Dartmouth YOAV KISLEV* Family Farms, Cooperatives and Co/lectives1 THE FAMILY FARM What is more natural than to expect modem agriculture to be organized in large-scale food factories? Such expectations notwithstanding, a large part of agricultural production - though not all of it - is still done on family farms. Evidently, economies of scale in production, to the extent that they exist, are outweighed by countervailing forces. Transaction costs and control Family organization of production m1mm1zes transaction costs through 'specializaton by identity' and the harmony and trust that comes with it (Ben Porath 1980). 2 In agriculture, this advantage is augmented by the specific nature of control as discussed already by Brewster (1950) who observed that (a) the biological production process in agriculture is time-dependent: food and fibres cannot be produced in an establishment in which different stages of the product are manufactured simultaneously; and (b) the farm product cannot be moved along a production line: rather the worker has to go to the field to perform the necessary tasks. Both these characteristics make control of labour in agriculture difficult relative to manufacturing and increase the comparative advantage of the family farm versus large units relying mostly on hired hands. Labour according to Brewster, is a fixed factor on the family farm. The farmer strives, therefore, for a balanced product mix with stable labour re quirements throughout the year. As a result, the advantage of the family farm is visible in areas suitable for diversified farming, while large farms dominate the regions of monocultural agriculture. Brewster's analysis can be augmented with modem theoretical insight and historical experience. Where land owner ship is highly unequal, the problem of control can be solved by sharecropping (Otsuka and Hayami, 1988). That this is only a partial solution can be seen from the fact that even livestock production is still mostly a family enterprise. The staged production nature of the biological processes, with waiting periods in between seasonal tasks which prevents factory-like organization of produc tion in agriculture, enables farm operators and other family members to seek part-time, off-farm employment. Monoculture is no more an obstacle; it is an *Hebrew University, Rehovot, Israel. 520 Family Farms, Cooperatives and Collectives 521 advantage. Professional career, farm size, product mix and employment are now determined simultaneously - affected by considerations of income and risk. The above arguments are analytical and hypothetical, but would not size associated efficiency gains on large farms compensate for the loss of the aforementioned advantages? To answer this question, we have to examine the empirical evidence. Economies of scale andfarm growth In an often quoted paper, Griliches (1963) found the sum of the coefficients in a Cobb-Douglas production function for the American farm sector to be 1.28; others report a similar value. Consequently, growth of farm size has been attributed to economies of scale. Hence we now face two questions: should these high-scale estimates be taken at face value and, if not, how can farm growth be explained? I wish to argue that the evidence does not support unequivocally the exist ence of economies of scale in agriculture. Because of data shortcomings, every measurement and method of estimation can and should be questioned. When this is done, the case for economies of scale is weakened significantly. The argument is detailed in Kislev and Peterson (forthcoming). Only the main points are repeated here. Conventional production function estimates, Griliches' included, do not allow for differences in ability and local conditions on farms. Indeed, virtually all reported estimates of covariance analysis - taking care of the unobserved specific factors (Mundlak, 1968) - fail to find economies of scale in agriculture. The alternative method of synthetic firm analysis, which has also produced large estimates of economies of scale, assumes away the crucial issue of control. For further examination, consider expected consequences. If scale econo mies exist in a competitive industry, they must be a disequilibrium phenomenon and growth should bring farms down the average cost curve towards its minimum. Yet re-estimates of the Cobb-Douglas production function for American agriculture with data spanning the pedod 1949-87 repeatedly came up with sums of coefficients of 1.3. No convergence to constant cost is discerned in the data with this method of analysis. Between 1929 and 1987, output per farm in constant dollars grew in the USA by some 6.4 times. If economies of scale are not important, what can explain such fast growth? Changing prices appear to be the answer (Kislev and Peterson, 1982): between 1929 and 1969, machine rental declined relative to alternative labour cost in agriculture by 3 per cent per year. Farmers reacted by increasing the machine-labour ratio. An operator with more machines cultivated a larger area, and farms grew in size. The trend changed when the cost of operating machines increased during the 1970s and, indeed, farm growth was then halted. Farm size seems to have resumed its upward trend in the 1980s, when increases in real wages again overtook growth in machine rentals. 522 Yoav Kis/ev Hence structural characteristics of production and control make the family farm the dominant form of organization (though not the only one, as should be emphasized) and price ratios affect optimal size of operation. The amounts of land, capital and labour on the farm are determined simultaneously with lines of production and non-farm employment. 3 Decision making The economic decisions of the farm household in a market economy are guided by prices and optimization can often be done recursively. In the first stage, income is maximized by deciding on allocation in the production 'de partment' of the family farm; in the second stage, income is distributed between consumption items and saving (Singh, Squire and Strauss, 1986). However, the recursive nature of decisions is not always maintained. Exam ples to the contrary arise when household members draw different utility from working on or off the farm (Lopez, 1986) and when uncertainty affects the price of a product that is both produced and consumed on the farm (Finkelshtain and Chalfant, 1991). Though recursiveness and relying on market signals simplify decision making, with experience and necessity even traditionally bounded peasants allocate optimally the resources at their command (Schultz, 1964). COOPERATIVES AND COLLECTIVES The advantage of the family organization is usually presented in comparison with commercial farms relying on hired labour. Cooperation facilitates the realization of scale economies in services and trade, overcoming local mo nopolies (Sexton, 1990), risk sharing and credit enhancement. The problem of control can be much reduced in cooperatives and collectives (voluntary, not forced cooperation of course), but it cannot be eliminated completely, as evidenced by the Israeli experience which culminated in a severe financial crisis in the mid-1980s. I draw on that experience and base my discussion to a large extent on Zusman (1988). For an analysis of the economy of the kibbutz, see Barkai (1977). Four types The major types of cooperatives in Israel are moshavim, kibbutzim, regional cooperative enterprises, and supply cooperatives. They are now undergoing revolutionary changes and their description is somewhat outdated. The recent changes will be discussed towards the end of the paper. A moshav (moshavim is plural) is a farming community in which all farms are family-operated, and all farmers are members in the multi-purpose, demo cratically run, village cooperative. In principle (practice varies) the coopera- Family Farms, Cooperatives and Collectives 523 tive association in the moshav purchases all farm supplies for its members and markets their farm products. A kibbutz is a commune. Members work together and receive from the kibbutz all their needs. Again in principle, a member in the kibbutz owns his or her personal belongings but no other property. Moshavim and kibbutzim are members in two types of second-order coopera tives: regional service enterprises (such as feed mills, slaughter houses and transport services) and supply cooperatives set up to purchase farm requisites for their members the moshavim and the kibbutzim. Owing to space limitations, the supply cooperatives will be considered only to the extent necessary to explain developments in the first-order cooperatives- moshavim and kibbut zim - and the regional enterprises will not be dealt with at all. Starting with the transfer of suppliers' credit to their members, the village associations of the moshavim and the supply cooperatives expanded into full scale financial intermediation and the domineering position that financial activities occupied among their functions greatly affected both their well being and their structure. It is useful to commence the discussion with a review of balance sheets. Structure and accounting Differences in their organizational forms are reflected in the accounting prac tices and balance sheet composition of the cooperatives (Jensen, 1983). Three randomly chosen cases are presented in Table 1. TABLE 1 Balance sheet composition (percentages, September 1984) Moshav Kibbutz Supply coop Fixed and financial assets 2.8 62.4 8.3 Loans to members 75.7 0.1 77.8 Other current assets 21.5 37.5 13.9 Total assets 100.0 100.0 100.0 Equity 2.9 33.7 0.4 Outside debt (including suppliers) 51.7 54.7 84.1 Loans from members 5.4 0.1 15.5 Loans from supply coops 40.0 ll.5 Total liabilities 100.0 100.0 100.0 Source: Lerman (1989) The accounting framework of the moshav and its balance sheet are for the cooperative association of the village; the economic activities of the family farms are not covered. In this way, the privacy of the members is respected, but the practice also limits the moshav's monitoring ability, representing a 524 Yoav Kislev weakness which contributed to the financial crisis. The importance of credit intermediation can be seen from the share of loans to members in its balance sheet, estimated at 75.7 per cent of the assets. The moshav raises credit from outside sources, almost a half of it from the supply cooperative, and distrib utes it to its members. To a lesser extent, the moshav also functions as a clearing house, receiving deposits from members with surpluses for others in need (5.4 per cent of the liabilities). Like the association in the moshav, the supply cooperative is also mainly a provider of services, including financial intermediation. The functional resemblance is reflected in the similarity of the balance sheet composition. The kibbutz conducts its accounting like a family farm, combining its business and household books. The assets in the balance sheet are machines, buildings, orchards, livestock, plus the members' dwellings, pension funds and other savings. Consistently consumption, not labour, is considered as part of the cost of running the economic enterprise. Such mixed accounting prac tices mask the distinction between business and household, and between ownership and management, and are obstacles in the control of the economic affairs of the kibbutz. Kibbutzim are profit-maximizing entities, engaged mostly in production and aiming to accumulate equity capital-a third of the liabilities in the kibbutz in Table I. As zero-profit cooperatives, concentrating on financial intermedia tion, the moshav and the supply cooperative naturally accumulated smaller share of equity. STRUCTURAL CONSTRAINTS AND ORGANIZATIONAL BEHAVIOUR The moshav, the kibbutz and the regional cooperative are contractual institu tions. Members contract, explicitly or implicitly, with the cooperative asso ciation and with each other to perform together certain economic and social activities. In principle, members are bound by the rules and regulations of the cooperative or the collective they joined. Practice is dictated by the democratic governance of the kibbutz and the moshav and by the cooperative ethics of their members. Conflicts of interest - the moshav By tradition, cost (of marketing, for example) is allocated in cooperatives according to 'patronage'. This results in average pricing, which may differ from optimal pricing if, in the quantity of services provided, marginal costs differ from their average. In an attempt to improve upon this rule, the moshav may choose a two-part cost allocation rule: each member is charged a given amount a (perhaps to cover part of the fixed cost) and an additional sum f3 per unit of product marketed through the cooperative. Now there is room for conflicts of interest. Members with a large volume of production will try to allocate most of the cost to the fixed element a, small producers will favour Family Farms, Cooperatives and Collectives 525 charging mostly on a per unit basis. If this issue comes to voting and the distribution of members by size of production is not symmetrical, the value of ~chosen will reflect the interest of the majority. The minority may find itself shouldering a larger than proportional share in the cost of the marketing service. Consider now the construction of a feed-processing plant by the moshav for the service of its members. When the investment is financed by the general fund of the moshav, the risk of the venture will be shared by all members. If only livestock farmers participate in the investment, others will avoid the risk, but the moshav will not make use of one of the great advantages of coopera tion. The possibility of any degree of risk sharing raises new issues of moral hazard which are nowhere more pronounced than in credit and will be discussed below. Conflicts of interest- the kibbutz A major source of structural conflict in the kibbutz is its set of operational constraints: equality, own labour (no exploitation of hired employees) and standard of living in parity with the standard of reference groups outside the kibbutz. These constraints are not always consistent. Own labour implies that unskilled work in agriculture and manufacturing is also done by members. These tasks seldom yield the income needed to support the expected standard of living. During the 1970s and early 1980s, ample supply of credit assisted in maintaining the desired private and social consumption levels, but also cre ated the background for the financial crisis to follow. By sharing income equally among its members, the kibbutz avoids much of the kind of conflict of interest that plagues the moshav. The outcome of an economic undertaking affects similarly all members and it is generally not in the immediate interest of any one group to tilt decisions in its direction. And if conflicting interests arise, since personal consumption is not affected, the intensity of the conflicts and the social antagonism they may generate are more often than not much weaker than in similar situations in the moshav. The identity of the society and the community with the economic activities is the source of another kind of conflict. The majority of the members cannot comprehend fully the economic situation of the kibbutz - particularly with accounting practices that do not separate business from the community - but everyone understands social problems. Consequently, unlike the rational tra ditional peasants whom Schultz (1964) praised, kibbutzim are often subject to the logic of collective action (Olsen, 1965). They tend to have bloated serv ices (particularly in children's care), readily purchase new and convenient machines, continue with failing activities to avoid painful labour re-allocation, and invest in dwellings and community services even if the necessary capital is too costly. They also tend to permit their young members long leave periods to experience outside life. The result is an unequal age distribution of the labour force, which is manifested particularly as the kibbutzim get older. 526 Yoav Kislev Recapitulation As we have just seen, the structure of the kibbutz and the moshav breeds conflicts which may hamper optimal operation. Other difficulties can be re garded as free-riding and moral hazards. They are manifested strongly in the financial activities of the moshavim and the kibbutzim to which we now tum (for details, see Kislev, Lerman and Zusman, forthcoming). THE COOPERATIVE AS A FINANCIAL INTERMEDIARY Moshavim and kibbutzim cultivate national land leased to the farmers on a long-term basis. Farms cannot be used as collateral against loans. The alterna tive is cooperation in credit. Both the associations in the moshavim and the supply coops of the moshavim and the kibbutzim function as credit intermedi aries. To augment monitoring and to facilitate convenient collection, members are required to market farm product through the cooperative. Advantages With financial cooperation, members in the moshavim, and moshavim and kibbutzim in the supply cooperatives, enjoy economies of scale in loan processing, professional financial management, and stronger bargaining posi tion in the credit market. However, the greatest advantage of cooperative credit, both in the moshav and in the supply cooperative, lies in risk pooling, the implementation of which is founded on mutual liability and guarantee. Members in the moshav sign mutual guarantee agreements for the moshav association and representatives of moshavim and kibbutzim pledge similarly for loans raised by the supply cooperative. The social pressure to comply with cooperative norms is strengthened under mutual liability arrangements. The probability of default is reduced. Banks evidently recognize the advantage inherent in this arrangement, as credit is often made conditional on renewal of mutual liabilities. Weaknesses Several kinds of structural difficulties afflict the moshav and the supply coop erative, particularly (a) moral hazard - members tend to invest on their farms in risky projects knowing that with mutual liability they will be bailed out should the returns on the investment be disappointing; (b) free riders - a member in the moshav, or a moshav and a kibbutz in a supply cooperative, may choose to market farm products privately, thus weakening the associa tion's standing in the credit market; and (c) agency cost- banks and other lenders view the cooperative associations as their agents and expect them to protect their interest (for example, by limiting credit to failing members) but the associations are guided by interests which are not always those of the Family Farms, Cooperatives and Collectives 527 lenders. Similarly, officers in the associatiOns may be tempted to expand operations and to assume risks which prudent members would avoid. Enforcement of the cooperative's norms and rules - in practice, mainly enforcement of the inter-linkage arrangements of product marketing through the moshav and through the supply cooperative - is critical to its continued functioning as a credit cooperative. However, compliance with the behav ioural code requires high standards of cooperative ethics and will to enforce. But enforcement is difficult in the internal political environment of the coop eratives. PUBLIC POLICY Cooperation in agriculture has always been supported by the government in many ways, but the most profound public involvement was in credit. The government raised capital on the markets in Israel for its budgetary needs, thus crowding out private sources of investment capital. To remedy the short age of its own creation, the government distributed credit and subsidized it. Farm cooperatives were among the beneficiaries of this policy. The depend ency on the government and the expectation that it would bail out moshavim and kibbutzim in trouble created moral hazard problems. Cooperatives at all levels were willing to rely on large amounts of debt, and banks were willing to lend, all trusting the government to save them in case of difficulty. This problem of moral hazard was recognized by the government, but the will to maintain a strict policy could not withstand the flood of credit in the late 1970s, when Israel participated in the global credit expansion. The situation was aggravated as inflation accelerated (to an annual rate of 440 per cent in 1984) while interest rates lagged and real rates were negative on many kinds of loans for most of the 1970s and early 1980s. The combination of ample supply of credit with the weaknesses of cooperative financial intermediation resulted in over-expansion and excessive reliance on debt in moshavim, kib butzim and regional cooperative enterprises. In July 1985, inflation was abruptly halted by severe measures, including tight monetary policy. Real interest rates rocketed. Crisis The financial crisis in agriculture erupted at the end of 1985, when creditors realized that agriculture, particularly cooperative agriculture, could not con tinue to service its debt in view of the exceedingly high real rates of interest and that the government- bound by a stringent fiscal regime - could not bail the sector out any more. Most regional cooperatives and many of the associa tions in the moshavim collapsed. Farm production has continued, often with private credit arrangements and the farmers' own resources. But this cannot be a complete solution to the crisis, and banks and other creditors are still demanding repayment of their loans. For most members in the cooperatives, the heavy burden is not their own debt but their share in the mutual liabilities 528 Yoav Kislev - their share in covering the debt of several heavy borrowers in the moshav and the debt of the regional enterprises. Agriculture could not repay or service its debt in full. Once this was realized, the government moved in, offering support in an effort to reach a debt settlement between the banks, on the one hand, and the moshavim and the kibbutzim on the other. Agreements have been formulated but their imple mentation has been slow, as many in the sector still hope that they can gather political support for a more favourable settlement. A major victim of the crisis has been cooperation. Many of the village associations in the moshavim ceased to function as cooperatives and most of the supply cooperatives had to give up financial intermediation. It is practi cally impossible to get credit guarantees, and banks became suspicious of borrowers. These changes, coming in the wake of the crisis, are affecting different sectors of agriculture in various ways. Wealthier farmers in moshavim can offer collateral in the form of private property and saving. Poorer mem bers have only their farms to offer, but these are not acceptable and such operators are often driven to expand off-farm work. The crisis also accentuated the differences between moshavim and kibbut zim. A moshav can function as a village even if the farmers desert the coop erative and each fends for himself. Kibbutzim are made up of their membership and, should the members leave, the kibbutz will disappear. Some young members are already leaving (not all for economic reasons, to be sure). The crisis is therefore much more dangerous for the kibbutz. And indeed, many of the kibbutzim reacted by adopting revolutionary structural changes including the division of the kibbutz into several semi-independent economic units, each with its own board of directors and reporting obligations. The kibbutzim are showing here - not for the first time - both their commitment to the collective idea and their practical flexibility. It seems that, economically, most of them may survive the crisis and emerge from it strengthened, provided the younger generations stay and continue in their parents' tradition. CONCLUSIONS Cooperation was in the forefront of agricultural development in Israel. Much of the sector's institutional building and technological advancement was achieved through cooperative efforts. For many years, members in moshavim and kibbutzim reached satisfying income levels and maintained stable coop eratives. The late 1970s were particularly favourable for cooperative agricul ture: with its access to credit, it succeeded in accumulating large amounts of equity capital, much of it due to inflationary gains resulting from negative real interest rates. With prudent housekeeping, moshavim and kibbutzim could have emerged from the inflationary experience stronger than ever. Instead, driven by weaknesses of cooperative action, combined with irresponsible government policy, they have sunk deeper into debt and prepared the ground for the devastating crisis. It may well be that, with time and with changing government attitudes and public atmosphere in Israel, cooperation would have lost in the evolutionary Family Farms, Cooperatives and Collectives 529 struggle to private modes of organization in agriculture. However, in the intensely unstable economic environment of the last 20 years, the institutional competition has been unfair and cooperation may have retreated too much. Time will tell whether cooperation will return to its pivotal position in Israeli agriculture. NOTES 1Writing was supported by the Maurice Falk Institute for Economic Research in Israel. The paper reflects ideas I received from Zvi Lerman, Willis Peterson, Gadi Rosenthal and Pinhas Zusman. The responsibility is solely mine. 2Schmitt (1990) also attributes particular importance to transaction costs in explaining the survival of the family farm. 3Here I differ with Schmitt (1990) who assumes, perhaps implicitly, that land and capital are given exogenously and farm size is determined (actually, defined) solely by labour allocation. REFERENCES Barkai, Haim, 1977, Growth Patterns of the Kittbutz Economy, North-Holland, Amsterdam. Ben Porath, Yoram, 1980, 'The F-connection: Families, Friends, and Firms, and the Organization of Exchange', Population and Development Review, 6, pp. 1-30. Brewster, John M., 1950, 'The Machine Process in Agriculture and Industry', Journal of Farm Economics, 32, pp. 69-81. Finkelshtain, Israel and Chalfant, James A., 1991, 'Market Surplus Under Risk: Do Peasants Agree with Sandmo?', American Journal of Agricultural Economics, 73, pp. 557-67. Griliches, Zvi, 1963, 'The Source of Measured Productivity Growth: United State Agriculture, 1940-60', Journal of Political Economy, 71, pp. 331-46. Jensen, Michael C., 1983, 'Organization Theory and Methodology', Accounting Review, 58, pp. 319-39. Kislev, Yoav and Peterson, Willis, 1982, 'Prices, Technology and Farm Size', Journal of Politi cal Economy, 90, pp. 578-95. Kislev, Yoav and Peterson, Willis, 1991, 'Economies of Scale in Agriculture: A Reexamination of the Evidence', in John M. Antle and Daniel A. Sumner (eds) Essays on Agricultural Policy in Honour of D. Gale Johnson. Kislev, Yoav, Lerman, Zvi and Zusman, Pinhas, forthcoming, 'Recent Experience with Coop erative Farm Credit in Israel', Economic Development and Cultural Change, 39, pp. 773-89. Lerman, Zvi, 1989, 'Capital Structure of Agricultural Co-operatives in Israel', Yearbook of Co operative Enterprise, Plunkett Foundation, Oxford, UK. Lopez, Ramon E., 1986, 'Structural Models of the Farm Household That Allow for Interde pendent Utility and Profit-Maximizing Decisions', in Singh, Squire and Strauss (I 986). Mundlak, Yair, 1968, 'Empirical Production Functions Free of Management Bias', Journal of Farm Economics, 43, pp. 44-56. Olsen, Mancur, 1965, The Logic of Collective Action, Harvard University Press, Cambridge, MA. Otsuka, Keijiro and Hayami, Yujiro, 1988, 'Theories of Share Tenancy: A Critical Survey', Economic Development and Cultural Change, 36, pp. 30-68. Schmitt, Giinther, 1990, 'Why Collectivization of Agriculture in Socialistic Countries Really Has Failed: A Transaction Cost Approach', Discussion Paper9002, G Zusman, Pinhas, (1988) Individual Behaviour and Social Choice in a Cooperative Settlement, The Theory and Practice of the Israeli Moshav, Magnes Press, Jerusalem. DISCUSSION OPENING- LAURENT MARTENS* It is a common occurrence to accept the role of discussion opener on the basis of nothing but a title. As long as the discussant does not have any other information, he may have rather conflicting expectations concerning the con tent of the paper on which he will eventually have to comment. On the one hand, he can hope to find a paper with which he substantially agrees, since we all like our biases to be reinforced by others. In such a case the discussant can explain how much he enjoyed reading an excellent paper and he can comple ment the paper from his personal experience. On the other hand, he may look forward to a paper leaving plenty of opportunities for disagreement with the author, enabling the discussion opener to point out major theoretical or em pirical shortcomings and leaving all participants with the impression that the topic still holds scope for further research. In this particular case it appears that there is ample scope for both ap proaches. The title alone suggests that the theme is an evergreen one and an old classic as well. It is evergreen because the issues covered remain excep tionally topical in the present debate on structural transformations in agricul tural production world-wide, not just in what used to be centrally planned economies in Central Europe, but also in less developed economies and in the European Community. It is an old classic because throughout the 70-year history of IAAE one can hardly find a single conference during which the institutional organization of agricultural production was not on the agenda. Yet the overall scope of the paper has been scaled down considerably from the initial title to the final content. The initial title sounded like 'Competitive Institutional Arrangements in Farming: Theory and Evidence', from which one could expect a paper developing a universal theory, based upon empirical evidence, which would cover all aspects of alternative institutional structures. Participants who have chosen to attend this particular session on the basis of such high expectations might feel somewhat misguided or even disappointed by the much narrower scope of the paper which has been presented. Indeed the concept of family farms has been scaled down to that of developed market economies, and even mainly to United States experience, while agricultural cooperatives and collectives are even further reduced to the very specific institutional arrangements of moshavim and kibbutzim in Israel. The section on family farming focuses almost exclusively on economies of scale and is introduced by the question 'What is more natural than to expect modem agriculture to be organized in large-scale food factories?' I would rather have expected this question to be phrased in the opposite sense: 'Is there anything natural about expecting large-scale food factories?' Indeed, it is soon made clear that the evidence does not support unequivocally the existence of economies of scale in agriculture. Kislev is certainly in good *University of Ghent, Belgium. Family Farms, Cooperatives and Collectives 531 company in suggesting that fast growth in farm size and the maintenance of the dominant position of the family farm can be reconciled. In his stimulating paper presented at the 19th IAAE conference, Boussard (1985) concluded that farm size heterogeneity is a consequence of absence of economies of scale, and that farm structure heterogeneity is a result of the interactions between a dynamic process of adjustment towards optimal price-dependent structures and of market constraints which perturb this adjustment. At the same confer ence, Newby (1985) stated that the tendency towards increased concentration of production has not been accompanied by the disappearance of the family farm or peasantry, and that the emergence of a dual farming economy can be witnessed in many countries. At the 5th Congress of the European Association of Agricultural Econo mists, Nielsen (1985) concluded that its ability to adjust to changing external conditions is the main reason for the family farm having been such a persistent institution. Schmitt (1989, and at this Conference), focuses more specifically on the argument that transaction costs related to farming are smaller in family farms, so that economies of scale are insufficient to compensate higher trans action costs in commercial farms. All this points us towards household pro duction theory as an integration of the neo-classical theory of producer and consumer behaviour. However, here too there is hardly anything new. Almost four decades ago Heady (1953) wrote: 'Motivational forces behind the farm producing unit are consumption-inspired as well as profit-inspired . . . The optimum use of resources in production or the optimum allocation of income in consumption cannot be defined unless the two basic sets of economic relations are related'. This approach also clarifies the similarity between the kibbutz and the family farm. Indeed, Kislev points out that in both cases business and house hold accounts are combined and that the labour of the extended kibbutz family is not recorded as an expense. Some similarities could also appear with farming systems based on extended families throughout the world. Here, too, the unity of the firm and the household can be a source of conflict, either because it can result in the exploitation of family labour or because it leaves ample scope for free-riding. The institutional framework of the moshavim is clearly that of a coopera tive characterized as a voluntary association of people, in which capital sub scription does not form the basis of voting power and in which the reward is primarily seen as the patronage rebate or discount based on the value of business done with the society. Kislev points out some conflicts of interest in the management of the moshavim, such as cost allocation according to pa tronage. Cobia (1989) formulates this problem in more general terms, in relation to the heterogeneity in size and structure of farms mentioned earlier, stating: 'The size disparity among farmers challenges existing cooperatives to serve patrons with very different needs'. Olson (1965) is even more specific: 'Unless the number of individuals in a group is quite small, or unless there is coercion or some other special device to make individuals act in their common interest, rational self-interested individuals will not act to achieve their common or group interests'. This phenomenon of conflicting individual and group interests is also experienced in some cases of group farming in Western 532 Yoav Kislev Europe (Martens, 1973) and certainly helps us to understand the financial crisis of the kibbutzim to which Kislev is referring. Somewhat to my surprise, the author states that 'institutional competition has been unfair to cooperatives'. I would like to know how the author arrives at such a conclusion, especially since in the same paper we also read that 'cooperation in agriculture has always been supported by the government in many ways', and that the greatest involvement was in credit, with negative real interest rates. Moreover, kibbutzim cultivate national land leased on a long-term basis. My personal, very subjective, impression is that in Israel institutional competition has been unfair to family farms and to non-family corporate farms. If fair competition prevailed, and considering the unbalanced age distribution of the labour force as well as the fact that some of the initial objectives behind kibbutzim and moshavim are somewhat outdated, I would rather expect that family farms or non-family corporate farms would take over. Of course, the answer could be a matter of definition and it is my impression that many arguments concerning institutional arrangements in farming have to do with a lack of clearly defined concepts. REFERENCES Boussard, J.M., 1985, 'Changing environment and structural heterogeneity in agriculture', in Proceedings, 19th IAAE Conference. Cobia, D., 1989, Cooperatives in Agriculture, Prentice-Hall, New York. Heady, E.O., 1953, Economics ofAgricultural Production and Resource Use, Prentice-Hall, New York. Martens, L., 1973, 'Nouvelles formes de collaboration dans le domaine de la production agricole'; in Commission Communautes, European, Information internes sur l' Agriculture, 110. Newby, H., 1985, 'The changing structure of agriculture and the future of rural society', in Proceedings, 19th IAAE Conference. Nielsen, A.H., 1985, 'The family farm in a changing technological and economic environment', European Review of Agricultural Economics, 14, (1). Olson, M., 1965, The Logic of Collective Action, Harvard University Press, Boston. Schmitt, G., 1989, 'Farms, farm households and productivity of resource use in agriculture', European Review of Agricultural Economics, 18, (3/4). 10 CooperatiqreCredtt in Agria,tlnne- The IsraeliExperience YoaqtKuleo,, Zvi Lemnrl, and PinhasZu.mwn Cooperation in credit offers a.d"vantagesin risk pooling throughmutual liability and guarantJ, but it also poses serious control problems. The debt crisjs in tlrc family farm sectorof Israel, triggered in 1985 by onti-inflationary policies, revealed weak- nessesinherent in the cooperative structure. Mutrnl ltabtlity encouraged ouerbor- rowing when possibleand could not be enforced when needed. Cooperative credit could not surq,)iuein a ilgh|y unstable macroeconomic enuironment. PeNwu-rssIMMIGRANTS wERE sETTLED on national land with public assistance in the early stagesof agricultural development in Israel, thus creating what amounted to fosterage relations between the farm sector and the public agencies that looked after it. Modern Jewish agriculture had received sub- stantial public support since its inception in the nineteenth century. Ideol- ogy and expedience made cooperation the preferred form of organization within the agricultural sector. Cooperative agriculture developed gradually in the 1930sand 1940s, and its growth accelerated dramatically in the early 1950s, when the newly established state directed arriving immigrants to agricultural cooperatives and furnished them with housing and farm tools. Today 80 percent of Israel's agricultural product comes from the cooperative sector, both family farms and the collective kibbutzim. A major form of cooperation in agriculture has been financial. lt flour- ished for a long time; but financial cooperation has found itself recently in deep trouble and will need massive public assistance to overcome its diffi- culties. It is now too early to predict what kind of cooperation, if any, will 214 emerge from the crisis. Yet important lesso", ."";.;: ;1t "t:::- incomplete experience. This chapter concentrates on Israel'sexperience with credit in farming villages run by cooperative associations, the so-called moshavim. In each moshav, all farms are family-owned and -operated and all farmers belong to the multipurpose, democratically run village cooperative. The communal sector-kibbutzim-will not be discussedin this chapter. We summarize our observations in the following. Financial cooperation supported intensive development of the family farm sector when stable financial conditions prevailed. When credit supply expanded with inflation, and when it was augmented bv government support, overextension was encouraged, particularly in the cooperatives. \ilhen credit expansion was slowed down, the cooperative sector found itself trapped in financial impos- sibilities. Now cooperation ties farmers and their organizations together and intensifiesthe crisis. Inflation in Israel accelerated steadily from a yearly rate of i 2 percent in the early 1970sto more than 500 percent (annualized) in the first half of 1985.It was then halted by an abrupt change of direction in policy, and since then inflation has been at approximately 20 percent per year. The rising prices were fueled by the expanding supply of credit. Interest rates lagged behind inflation, and real rates were negative for most of the decade ending in 1985. These conditions encouraged overinvestment and overborrowing and discouraged saving. But interest rates also lagged when inflation decele- rated in 1985 and, as a result, real rates skyrocketed. Unable to service their debt, agricultural cooperatives collapsed. Agriculture, and particularly cooperative agriculture, was thus the victim of inflation and the measures implemented to halt it. But the crisis reveals weaknessesinherent in the cooperative structure itself, as well as weaknesses in government lending policy to agriculture. Many businesses suffered severely when economic conditions changed with the introduction of the anti'inflationary policy in 1985. But it is only in agriculture that a whole sector-the cooperative sector-collapsed financially. Cooperative financial intermediation was founded on mutual liability arrangements and the crisis brought out clearly the failure of these arrangements. Mutual liabilities encouraged overborrowing when that was possible and could not be enforced when the need arose. The Moshav and Second,OrderCooperation In principle (practice varies) the cooperative association of each moshav purchases all farm supplies for its members and markets their farm products. It may also own and operate a variety of service facilities and manage directly some productive enterprises. In addition, the association encomtriassesall 216 coopERATrvEcREDrr IN AcRTcuLTURE-THErsRAELr EXpERIENcE municipal'a.,d many social functions in the village. Besidesall these, the associatidn also serves as a financial intermediary through which credit is channeled to the farmers. The moshav is therefore a supply, production, service,municipal, and credit cooperative. Moshavim are members in two types of second-order (mostly regional) cooperatives: supply cooperatiues(requisite societies, purchase organizations) set up to purchase farm inputs for their member-moshavim, and regional seruice enterprises(feed mills, slaughterhouses, transportation services, and others). The supply cooperativesact also as the spokesmenof their regions in the government offices. They engage in intensive lobbying and most have acquired a strong political standing. Starting with back-to-back transfer of suppliers' credit to their members, both the moshav and the supply cooperative expanded into full-scalefinan- cial intermediation. In the two decadespreceding the eruption of the crisisin 1985, the associationsin the moshavim and the supply cooperatives were first and foremost credit associations. Moshavim were settled on national land; it was practically impossible for lenders, particularly suppliers and commercial banks, to repossessfarms. This deficiency-that land and build- ings could not be used as collateral-was the principal economic justification for the evolution of the moshav cooperative as a credit intermediary and for its mode of operation. Financial Intermediation The pivotal role of credit intermediation in the activities of the moshav association and the regional supply cooperative is demonstrated in their balance sheetsin table l0-1 (for additional details, seeLerman 1989).Mem- bers' debit balances-accounts receivable from members-are by far the largest assetthe associationshold: 76.6 percent of the total in the moshav and 60.9 percent in the regional supply cooperative. The moshav and the regional coop raise debt capital from outside sources and lend to their members. The associations also function as clearinghouses, accepting depositsfrom members with financial surpluses(members'credit balancesin table l0-1) and lending to members with credit shortages.The supply coop and its moshavim are closely linked: through credit, as can be seen in table 10-1 where 76.9 percent of the moshav's liabilities is short-term credit from the supply coop, and through joint ventures in the regional service enterprises. The government was the major source of long-term credit and the major lender in the early stages of the development of the moshavim.l Govern- ment credit was generally supplied on concessionary terms and the YOAV KISLEV AND OTHERS 217 Table 10-i . BalanceSheers of a Mosluu Associationand a RegionalSupply Cooperatiue,September 30, 1981 (percentage of total assets) Assets Liabilities Suppll Supply Mosh.at, cooperative Moshar., cooperatiue Fixed assets 3.1 ),) Equity v.t J.U Long-term investments and loans to members 3.5 Long-term debt 4 ) lq 5 Inventories 4.0 n.a. Short-termloans 0.6 34.5 Accounts Short-termloans receivable from supply Nonmembers t2.2 3.6 cooperative 76.9 n.a. Regional enterprises 18.3 Suppliers'credit 4.1 2r.8 Members'debit Members'credit balances 76.6 60.9 balances lJ.) zt.z Total 1000 100^0 100.0 100"0 n.a. Not applicable. Note: The data for the moshav are for an average association in a sample of thirteen moshavim, all members of the regional supply cooperative. Balance sheets are prepared in historical values, not adjusted for inflation. The financial reports in the moshav are for the cooperative association,not for the whole village. Information on individual farms is not included and is generally not available. Source: Zusman ( 1988). minuscule share of long-term (mostly government) credit in table 10-1 is a reflection of both inflationary erosion of unindexed debt and the growing availability of alternative sourcesof finance. The balance sheet is prepared in historical values, and as a result equity capital in table l0-1 is understated. [t was estimated that if adjusted to current values,equity would reach between 15 and 30 percent of the associa- rions' liabilities. But even with fully revalued equity, items reflecting finan- cial intermediation still dominate the cooperatives'balance sheets. The regional supply cooperatives thus established for their members the moshavim-and they in turn for their members the farm operators- financial serviceswith steady lines of credit and convenient saving facilities. Because of proximity and familiarity, asymmetric information was not a significant problem in cooperative agricultural credit in Israel. Still, inter- linkage of credit and marketing was practiced (Bell i98B): farmers were expected to channel the proceeds of their marketed products through the moshav association and it, in turn, through the supply coop. Interlinkage formed part of the institutional setup that replaced collateral for loaqs. 218 coopERATrvEcREDrr rN AGRTcuLTURE-THErsRAELr ExpERTENcE Financial Cooperation The "classical" discussion of the theory of cooperatives (LeVay 1983) strug- gled with the definition of the goals of the cooperative firm and its behavior. Difficulties created by the cooperative's egalitarian democracy were recog- nized but not formulated explicitly and not examined analytically. In the modern approach (Vitaliano 1983; Royer 1987; Zusman 1988), the coopera- tive is viewed as a collection of individuals, each guided by personal prefer- ences but committed to joint performance of certain economic functions. The modern approach enhances our understanding of two central facets of the cooperative mode of action. First, laws and regulations governing coop- erative life are often compromises and are not necessarily first-best, Pareto- efficient. For example, the optimal cost allocation rule is generally marginal cost pricing; but unable to agree on the distribution of side payments, the moshav most often settles for average cost pricing. This subject is treated at length in Zusman (1988). Second, members in cooperatives-in our case, farmers in a moshav or moshavim in a supply coop-are not subordinates in a centrally managed hierarchy; they are free to act within wide limits. By treating explicitly individual behavior and group decisions, the modern, contractual perception of cooperation throws new light on the advantages of cooperative credit intermediation and particularly on its weaknesses. Advantages With cooperation, members in the moshavim and moshavim in the supply coops enjoy economies of scale in loan processing and professional financial management-particularly important in a high inflation, high tax economy such as Israel-and a stronger bargaining position in the credit market (as well as in other markets). Perhaps the greatest advantage of cooperative credit, both in the moshav and in the supply coop, lies in risk pooling, implemented in two ways. In the short run, the association can use its own resources to smooth over the credit needs of its members. Outside lenders do not have to deal with transitory difficulties of individual borrowers. A more fundamental mode of risk pooling is mutual liability and guar- anty. Members in the moshav sign mutual guaranty agreements for the moshav association and representatives of moshavim pledge similarly for the loans raised by the supply coop. This creates explicit and implicit peer monitoring. The social pressure to comply with cooperative norms is strengthened under mutual liability arrangements, and in general the proba- bility of the association's default is reduced. Banks evidently recognized the advantage inherent in this arrangement, as credit to cooperative associations was often conditioned on renewal of mutual liabilities. YOAV KISLEV AND OTHERS 219 Weoknesses Several kinds of structural difficulties afflict the moshav. First is moral lwzard-members may tend to invest in excessively risky projects on their farms, safe in the knowledge that mutual liability will bail them out should the investment fall. Free riders pose another difficulty-a member in the moshav may market the product of his farm privately, thus weakening the association's standing in the credit market. Agenq cosrs are another problem-banks and other lenders view the cooperative associations as their agents and expect them to protect their interest (by limiting credit to failing farms, for example), but the associations are guided by their own interests, which are not always identical to those of the lenders. Similarly, officers in the associations may be tempted to expand operations and to assume risks that prudent members on their own would avoid. Finally, there may be horizon problems-members may support policies favoring short-term gains in expectation that in the long run they may exit, leaving those who stay to carry the cooperative'sliabilities (Vitaliano 1983). Enforcement of the moshav's norms and rules (which in practice implies enforcement of the collective marketing interlinkage arrangements) is critical to its continued functioning as a credit cooperative. However, compliance with the moshav's code requires high standards of cooperative ethics ("sym- pathy" in the usage of Sen 1966) and willingness to enforce the rules. Inter- dependence of the degree revealed in table 10-l and close monitoring due to interlinking of credit with product marketing could be expected to allow effective control. However, this was not the case.Particularly where interde- pendence was strong, the moshavim and the farm operators who belonged to them had only limited accessto alternative sources of credit and, conse- quently, the supply cooperatives were committed to continue funding their member moshavim. however weak and close to failure. Their elected officers could hardly afford the dire financial, social, and political consequences of members' bankruptcies. These enforcement difficulties are reflected in the behavior of members of cooperatives. A rough measure of the financial exposure of a member is the ratio of outstanding debt to monthly salesthrough the cooperative ("credit months"). During the period 1977-8I, thirteen of the twenty-four moshavim in the regional supply cooperative of table 10-1 exceeded twelve credit months (Zusman 1988). Moshavim with fifty-five and forty-two credit months were observed in another supply coop (Kislev and Marvid igBB). More than a few farm operators owe their moshav cooperative associations several times their yearly production capacity. Moshavim or individual farmers with such heavy burdens of debt compared to their production capacity will never be able to repay their loans or service them adequately. Heavy debt burden does not happen overnight; it evolves g5adually. 220 coopERATrvEcREDrr tN AcRIcuLTURE-THElsRAELt ExpERTENcE Moshavim or individuals with dozens of credit months testify to the weak- ness of their cooperatives, a weakness that breeds permissiveness and lax financial discipline. Moral hazard behavior and other weaknesses increase the risk to the lender and may even outweigh the advantages of cooperative credit. This indeed is observed in the aftermath of the recent crisis-lenders that were not previously involved are reluctant to extend credit to moshavim and to regional and national cooperatives. Regional Enterprises Regional service enterprises were ordinarily organized as limited liability cooperative associations and their establishment was financed mainly by government investment grants and subsidized loans. Their membership con- sisted of moshav associations, mostly potential patrons of the service offered. Often the regional supply coop was also a member and in all cases it pro- vided the enterprisesin the region with short-term financing and purchasing services. Strong economic relations developed between the two kinds of regionals-the supply coop and the service enterprise-a relationship that proved detrimental when the financial crisiserupted at the end of 1985. Zealous pursuit of rural development by public agencies, easy accessto credit through the supply cooperatives, and strong political regional lobbies resulted in overexpansion of the service enterprises. These trends were particularly prevalent in the 1970s,when credit was in ample supply and economic optimism ran high. Not unlike firms in a cartel, regionals scrambled to grab their share in the service enterprises, with all the expected ensuing benefits. Consequently, in the early 1980s many service enterprises operated under capacity: 50-60 percent by the estimate of the state comp- troller. The final outcome was that many of the enterprisescould not even cover their operating costs. The supply cooperatives, assuming the role of the financiers of last resort, found themselves financing not only operating losses but also the debt serviceof the regional enterprises.In 1981,the share of credit to the service enterprisesin the assetsof the supply coop in table 10-1 was already 18.3 percent; it grew substantially thereafter. In one case (Kislev and Marvid 19BB)we found that a regional slaughterhousethat started operation in 1981 with equity representing 25 percent of its inflation-adjusted capitalz began accumulating losses,and by 1985 its debt reachedT.5 times the value of its assets, most of that short-term loans from the supply coop. This was an extreme casebut not atypical; when the service enterprisesbegan failing in 1985,they took many of the supply coops down with them. YOAV KISLEV AND OTHERS 221 Cost of Debt Perhaps the greatest damage that inflation inflicted on the Israeli economy was the distortion of the cost of capital. Real interest rates varied markedly due to both market and administrative lags in adjusting nominal interest to the rate of price changes.Cost of borrowing in some channels was at times very high, while at other times and for other loans it was negative. For more than a decade,since the early 1970s,real rates of interest on most sourcesof credit were negative, primarily due to government intervention in cost of debt for preferred uses. Particularly well subsidized were government- supported development loans until they were linked to the price index in 1979. Moreover, as both interest expenses and indexation linkage of the principal were tax-deductible, taxpaying farmers and cooperatives enjoyed a negative effective cost of capital even for index-linked loans or for loans with interest rates fully adjustable to the rate of inflation. It was only in 1982that tax regulations were introduced requiring inflation-adjusted accounting and thus eliminating loopholes that inflation created. In a preliminary survey of eight cooperatives,both village and regional, we found that the auerageeffective real cost of outstanding debt was zero in 1971 and it declined gradually thereafter; it was minus40-50 percent per year in 1984. \ilhen inflation was halted in 1985, interesr again lagged and real rates jumped to plus 15-20 percent. Current cost of credit varied even more: the real rate of inreresr on directed short-term credit in 1984.with inflation at its peak, was minus 59 percent. In 1985,the real cost of overdraft facilities was plus 100percent per year. Credit Supply to Agriculture It has often been claimed in Israel that agriculture suffered from credit shortage. E,xamination of the available information reveals, however, that credit has been in ample supply. The share of agriculture in the net domestic product of the businesssector has been 6-7 percent; but over the past two decades,its share in the volume of credit was higher than 10 percent. With inflation, financial leverageincreased, particularly in agriculture. ln 1986the ratio of outstanding economy-wide credit to gross national product was twice its 1969value; in agriculture the same ratio increasedby a factor of 3.8. The ratio of credit to net capital stock increasedin agriculture between i969 and 1986 by a factor of four, while in industry it rose over the same period by only 20 percent. As we have seen,credit was under-priced and the low, even negative real rates of interest evidently contributed to the feeling of *rortage. 222 coopERATrvEcREDIT IN AGRTcuLTURE-THErsRAELr ExpERTENcE Another claim often made in Israel was that the maturity structure of loans did not match capital needs. Not enough long-term loans were avail- able and investment projects had to be financed with short-term credit, creating a financing gap between the expected life of the assets and the duration of the loans. Again, with negative real interest rates and easy access to short-term credit, many farmers and cooperatives knowingly financed investments with short-term loans and knowingly created financing gaps.J Whatever the origin of the financing Baps, farmers always turned to the government when financial stress became a cause for concern, usually with forceful lobbying and politicalbacking. There were many cases,almost one a year until i985, of "conversions" (rescheduling of loans): short-term credit was replaced by long-term loans, mostly on concessionary terms. The recur- rence of the conversion episodes, sometimes general and sometimes specific to certain farms or regions, was one of the major reasons for the widespread belief that agriculture would not be allowed to fold. The remedy was, how- ever, not always effective. In many casesfarmers and cooperatives returned to the preconversion maturity structure just several years after rescheduling. Government Cooperation in agriculture was encouraged by the government as a matter of policy: new immigrants were settled in moshavim; land and water were allotted to the moshav and distributed equally among the members; produc- tion quotas in milk, eggs,broilers, fruits, and other products were allocated on a village basis and the moshav decided on internal distribution; govern- ment agencies usually consulted with the cooperative association in the moshav on the allocation of long-term loans to farm operators. Over time many of the newly settled operators acquired farming skills and cooperation became well established. Yet the view-held not only by farmers-that it was the government's role to maintain the welfare of the farming sector persisted, and the expectation that the government would actually shoulder this responsibility did not wane. Government (in the wide senseof the term) is responsible for the laws and regulations of cooperative activity. Two instances of interest to our discus- sion can be mentioned. First, attempts to pass a "law of moshavim," strengthening the power of the association over individual members and improving their ability to control financial activities, failed because the law was deemed to infringe on the freedom of the members. Second, a regulation was recently issued that a cooperative cannot force members to participate in covering its losses.The argument is that a cooperative is a limited liability entity and members are responsibleonly up to the value of their shares(in principle, the limitation does not apply to casesof mutual guaranty). A judge already applied the new regulation in one.,rri :::;.:r. ., * ":: way to the Supreme Court. If upheld, it will mean a revolution in the mode of farm cooperation in Israel. The most profound public involvement in agriculture was in credit.4 By deciding on the allocation of subsidized credit, the government influenced regional development, lines of production, and farmers' income. The depen- dency on the government and the expectation that it would bail out farmers and moshavim in trouble created moral hazard problems, not unlike those that mutual guaranty created in moshavim and regional cooperatives. Lack- ing the usual mechanism of collateral, the government turned to close mon- itoring in the form of "concentrated credit": under this system, a moshav or a kibbutz concentrated all of its financial activity in a single bank; credit for both investment and short-term needs was granted only with the approval of a steering committee consisting, among others, of representatives of the bank and the Ministry of Agriculture. Participation in the concentrated credit scheme was voluntary and moshavim were attracted by the additional loans they could get. lndeed, the program, which started in the early 1960s,covered in a few years most of the moshavim in the country. However, the increased credit supply in the i970s, and particularly the convenient alternative sources offered by the supply cooperatives, eliminated the advantages of concentrated credit from the point of view of the moshavim and the program folded in mid-1970s. Thus the problem of moral hazard in the moshavim was recognized and tools to mitigate it were devised, but the will to maintain a strict policy could not withstand the flood of available credit. Concentrated credit is now proposed again in reaction to the current crisis. Crisis The crisis erupted at the end of 1985 once creditors realized that agriculture, particularly cooperative agriculture, could not continue to service its debt in view of the exceedingly high, post-reform real interest rates and the inability of the government to continue to bail the sector out. Private lenders and commercial banks refused to extend additional credit and insisted that loans be repaid. This was impossible and most regional cooperatives and many of the associationsin the moshavim collapsed. Farm production has continued, often with private credit arrangements (wholesalers, for example, pay in advance for farm products) and the farmers' personal resources. But this .annot be a complete solution to the crisis: (a) in most cases,the available iources will be insufficient for investment in equipment and machinery and :armers will find it hard to renew their production assets;and (b) b-anksand ',ther creditors are still demanding repayment of the outstandinqloans. For 224 coopERATIvE cREDIT IN AGRTCULTURE-THE ISRAELI EXPERIENcE most farqrers, the heavy burden is not their own debt but their share of the mutual liabilities-their share in covering the debt of several heavy bor- rowers in the moshav and the debt of the regional service enterprises. Agriculture cannot repay or service its debt in ful[; the question now is how to distribute the losses.Once this was realized, the government stepped in offering support in an effort to reach a debt settlement between the banks and the moshavim. An agreement was formulated in 1988, but its implemen- tation has been slow becausefarmers still hope that they can gather political support for a more favorable settlement. Recapitulation Inflation created a special opportunity for agriculture in Israel, particularly for cooperative agriculture. \Uith negative real interest rates and erosion of debt, agriculture could have increased its equity capital and emerged from the inflationary period economically stronger. This did not happen; as we have seen, financial leverage increased in agriculture: farmers sank deeper into debt, partly to finance investment in production assets (often with overcapacity), partly to finance housing and consumer durables, and partly to increase current consumption and standards of living. Considerations of short-run inflationary gains dominated those of long-run economic health. Myopia is common, but it afflicts cooperatives more strongly than individ- uals and private enterprises becauseof the cooperatives' internal politics and becauseof its incentive structure, which leads to moral hazard behavior, free riding, agency costs, and horizon problems. Still, the cooperatives were not the only ones at fault. Credit was distributed by the commercial banks; it was their money that was lent and it was their responsibility to secure the loans and to controltheir use. Evidently they neglected this responsibility. However, cooperation is not the sole cause of the crisis. Government, by its policies to accelerate development and by ultimately yielding to political pressures,created the impression (which has since proven false) that it would bail agriculture out in case of difficulty. Government also carries the major blame for overcapacity in agriculture. Farmers and regional officers naturally tend to increase their share in aggregatecapacity. Becausemost of the devel- opment projects were funded with government approval, it was the duty of the government to consider the aggregatepicture and to balance the desire to invest against the needs. This was not done; policymakers and even the Ministry of Agriculture Planning Authority encouraged over investment. The crisis in cooperative agriculture is to a large extent the outcome of the favoritism it enjoyed for a long time. Structural weaknessesin the moshavim and irresponsible behavior on the part of the government and the commercial banks reinforced each other in precipitatingcrisis. It is impossible," the "oo"lffir*Jl1'li* r,::: and it is probably not important. The significant question is what inference can be drawn from the analysis about the future of agricultural cooperation. Although cooperation in general, and financial cooperation in particular, has many advantages, it also suffers from inherent weaknesses.It is not clear if cooperation-unless heavily assisted by public funds-can succeed or, in the long run, even survive the economic test of competitive markets. But the test of the crisis is much harsher. Even if cooperation is basically viable, it may now be destroyed becauseof the particular crisis conditions. Much will depend on the willingness of the members to maintain cooperation in agri- culture and on their ability to make the required structural modifications that will increase the stability and reduce the probability of failure of cooperatives. Lessonsand Recommendations The failure of cooperative agriculture was a failure of control. If cooperation in agriculture and particularly cooperation in credit is to survive and suc- ceed, control has to be tightened. But control is expensive and often incon- venient. Both incentives and appropriate structures are needed to assure optimal control.5 First and foremost, the government cannot and should not take explicit or implicit responsibility for agriculture or for cooperatives. Then both farmers and lenders will know that they are the sole residual claimants of profits or losses.It will be in their direct interest to tighten control and to follow prudent economic policies. Mutual guaranties should be severely limited to reduce moral hazard behavior at the farm and in the village and regional cooperative association, and external market control of coopera- tives should be establishedwherever possible. A necessarycondition for efficient control is availability of accurate and rimely information. Financial reports, including balance sheetsand income accounts, need to be prepared and published regularly. The reports should be adjusted for inflation; prices are still rising in Israel at 15-20 percent per Vear. Supply coops should be limited to commercial activity; they should not irct as financial intermediaries. The regional service enterprises should be :ncorporated as limited liability companies and the members of the owner- :roshavim should receive marketable shares. Moshavim and their members .hould be free to patronize service enterprises of their choice, whether in :heir region or elsewhere. \{embers in the moshavim should be free to leave their cooperatives and rerate privately or to form alternative organizations. Exit is expbnsive-it 226 coopERATrvEcREDrr rN AcRICuLTURE-THErsRAELr ExpERIENcE I raisesthe average cost of servicesto the remaining members, and the exiting farmer may forfeit his allocation of land and water and, in addition, his production quota and development loans. But exit is often the only way for patrons to enforce efficiency and for minorities to voice their opposition (Hirschman 1970).Lack of control may be more expensive. The structural changes that we are proposing-and in many cases we adopt proposals that have already been made in Israel-are not easy to implement. Exposing the regional service enterprises to market competition may seem extremely painful in the short run; and indeed the Debt Settle- ment Administration is attempting to cure the enterprises by erasing their debt and assuring capacity operation through tying moshavim to their ser- vices. In the long run this is a recipe for inefficiency. The implementation of the changes we propose will require modifications of both law and attitude. Notes The authors are indebted to Avishay Braverman for raising the issues discussed in this chapter and for his assistanceand encouragement. Karla Hoff, Michael Lipton, and the referees offered constructive comments and suggestions. The responsibility for the analysis and the opinions expressed is ours. l. \Ue lump together government and other public institutions. 2. A major share of the accumulated equity was due to the inflationary erosion of unlinked loans during the construction period, from 1976to 1981. 3. ln part, however, the government was also responsible for the financing gaps. For example, government-approved development loans were often dispensed with delays, forcing reliance on short-term bridging finance. In periods of inflation, delays create not only temporary but also permanenr gaps in financing because of the inflationary erosion of the real value of the loans that were late to arrive. 4. Thirty to forty percent of the credit in lsrael is supplied by banks from their own sources and allocated to borrowers at the banks'discretion. The rest is under government control-either originating from the government budget and the central bank or from bond issues and deposits administered by commercial banks but designated as funds for earmarked, government-approved projects. Public involve- ment in credit supply to agriculture is even larger, with more than 80 percent government-directed. 5' \ile focus on the Israeli experience with its particular characteristics; chapter 3 discussesthe problem in the context of developing countries. References Bell, Clive. 1988."Credit Markets and Interlinked Transactions."ln Hollis Chenery and T. N. Srinivasan, eds., Handbookof DeuelopmentEconomics. Vol. 1. Amster- dam: North-Holland. YOAV KISLEV AND OTHERS 227 Hirschman,A. O. I970. Exit,Voice,andLoyahy. Cambridge, Mass.: Harvard Univer- sity Press. Kislev,Yoav, and Arie Marvid. 1988.A SzpplyCooperatiue MrsHoRrM. Hebrew Uni- versity,Jerusalem: Magnes Press. Lerman, Zvi. 1989."Capital Structureof Agricultural Co-operativesin Israel."Yecr- book of Co-operotiueEnterprise 1989. Oxford, U.K.: Plunkett Foundation for Co- operativeStudies. LeVay,Clare. 1983."Agricultural CooperativeTheory: A Review."Journal of Agri- cuhural Economics34: I-44. Royer, JeffreyS., ed. 1987.Cooperdtiue Tlvory: New Approaches.acs ServiceReport 18.Vashington, D.C.: U.S. Departmentof Agriculture, Agricultural Cooperative Service. Sen, Amartya K. 1966."Labour Allocation in a CooperativeEnterprise." Review of EconomicSrudies 33 : 361-7l. Vitaliano, Peter. 1983."Cooperative Enterprise:An Alternative Conceptual Basis for Analyzing a Complex Institution." AmericanJournal of Agriathural Economics 65: 1079-83. Zusman, Pinhas. 1988. Indiuidual Behnuiorand Social Chaicein a CooperativeSecrle- menr.Jerusalem: Magnes Press, Hebrew University. t3 Experience with Collective Action and Cooperation in Agriculture in Israel Yoav Kislev* Cooperation in agriculture in Israel was maintained successfully for many years, enabling growth and development of the farm sector. But agriculture in Israel, particularlyits cooperativesector, has recentlysunk into a deep financial crisis. The immediate causes of the crisis were inflation and macroeconomic policies, but the crisis also revealed basic faults in the organization of the agricultural sector. The paper presents the thesis that these faults were mostly due to excessive doses of government intervention, to narrow rationality on the part of farmers and members in cooperatives, and to the tendency of public agencies to yield to short-sighted political pressures. I wish to put in this paper the problems and experience of our cooperative agriculture in a wider perspective. My main thesis is that Israel's agriculture was subjected to excessive doses of cooperation and collective action wittr rnost of the excess due to government intervention. The government made membership in cooperatives the only option available to new farmers when cooperation was deemed necessary 40 years ago and failed to create the conditions fbr individual action when circumstances changed. The government encouraged cooperatives to over- expand and forced collective action in marketing boards, export monopo- lies, and other forms. Israel's experience also demonstrates that "govern- ment" is not a single-minded, rational, decision-making body with national welfare its only goal; but rather government's actions reflect the outcome of political struggles - often between forces of narrow interest. *Department of Agricultural Econornics, The Hebrew university of Jerusalem, Rehovot, Israel. 270 Y. Kislev The record of cooperation in agriculture in Israel is not entirely negative. On the contrary, cooperation and active government policies contributed significantly to the impressive achievements of the farrn sector: the creation of a sophisticated and technically advanced agriculture producing abundant quantities of food and fibers for home and export markels. At the same time cooperation and government intervention propeiled agriculture into substantial difficulties which have come to the ior" in the last several years. The argument in the paper is not balanced. Focusing on the more recent experience, it is a critique, somewhat dramatiied by emphasizing the crisis proportions that several of the problems have reached. Israel: Growth and Inflation Israel is a small country; its population is 4.5 million. Half of the country is desert, with a cultivated area of 430,000 hectares, of which approximate\ 5A% is under irrigation. Israel is a medium-income counfy with per capita GNP of $9,500. Agriculture contributes 3% of the net national product and 6% of the country's export. Its share in the labor force is 5Va (1990 figures)' Struggle for survival, reconstruction, and the absorption of large numbeis of immigrants, who arrived almost simultaneously with the flight of 600,000 Arab refugees during the war of Independence in 1948, were the major economic efforts of the newly established strte. By the mid- 1950s Israel embarked on a path of economic growth which was to last at record rates for 20 years, but growth has slowed down significantly since 1974. The country experienced two periods of severe inflation. The first was in the early li50s when a fledgling government strove to finance war and reconstruction with a small tax base and a poor administration. Prices increased 56% in 1952. The second wave of inflation started in the early 1970s and accelerated thereafter. It was halted in 1985 after reaching an annual rate of close to 800%. Since then inflation in Israel has been approximately 2A% per year. The rising prices were fueled in the inflationary periods by exianding supply oi ciedit, much of it imported,l market interest rates lagged behind inflation and real rates were negative I The increased availability of credit on convenient terms to Israeli banks in the 1970s reflected, at least in part, global shifts in supply due to recycling of oil money. Collective Action in Israel 271 for most ofthe decade ending in 1985. These conditions encouraged over- investrnent and discouraged saving. But interest rates also lagged when inflation decelerated in 1985 and, consequently, real rates reached extremely high levels. Agriculture was greatly affected by these macro- economic and monetary developments. The sector enjoyed growth and rising incomes when credit was in ample supply, but it found itself in a deep crisis when inflation was halted. Agriculture: Four Decades The 1950s began with food shortage and rationing. Agricultural expan- sion, especially an extensive settlement drive and increased utilization of factors of production, was made possible when the foreign currency constraint of the newly established state was gradually alleviated. The number of moshavim and kibbutzim and the cultivated area more than doubled, irrigated area quadrupled, and the product grew since then ten fold (Table 1). Terms of trade and real output prices improved at the beginning of the decade but they have followed a downward trend thereafter. The 1960s was mostly a period of consolidation. The political response to increased supply and reduced profitability was to establish marketing boards, to implement planning, to limit production, to withdraw surpluses from the markets, and to support farm prices and their stability. Increased sophistication in production and expansion ofexports in botlr quantity and diversity were the principal characteristics of agricultural development in the 1970s. Investment in the rural sector increased substantially toward the end of the decade. [n the kibbutzim a major part of the invesfrnent was in manufacturing enterprises. This upsurge paved the way for the coming crisis. The beginning of the 1980s was not much different from the preceding decade, but in 1982 income in agriculture fell dramatically and never recovered and, gradually, the sector sunk into a financial crisis: productiv- ity did not rise with investment, and agticulture - particularly within its cooperative sector - accumulated a debt burden that eventually would prove unbearable. The crisis erupted when, as part of an anti-inflationary policy in July 1985, credit was severely squeezed and rates of interest sky rocketed. I review below the crisis and the changes it forced. The future is clouded with uncertainty for the farm sector still burdened with heavy debts and facing the need to adapt to a new economic environment. Agriculture may emerge in a few years reformed, stronger, and healthier; but the pill may also be too hard to swallow. )7) Y- Kislev TABLE I Agriculture - Key Data Irrigated Employ- Gross Net Terms of Real Areau mentb Capital Domestic Trade"'d Output Stoclf Producf Pice"'' t955 890 102 35 19 1965 1510 114 59 51 119 99 1975 1800 80 95 95 100 LOz 1985 2327 89 llz 177 90 85 1990 2057 7t 112 234 88 75 uThousand hectares dRatio of output to input price index. bThousands. eOutput price deflated by the Consumer Price Index. clndex, 1976:100. Source: Statistical Abstract oJ Israel, 1991. Government Intervention in Agriculture Israel is a free market econorny mixed with government intervention, which is especially intensive in agriculture and in the capital market^s. Government's goals in agriculture are to secure farm income, to improve food supply, and to maintain rural population. Over the years, whilst Israel's agriculture was being built up mostly by immigrants without resources, ignorant offarrning practices, public support was only natural. The government intervenes in planning, the supply of public services, price support, and trade. The government is involved in almost all aspec8 of farm life, particularly in cooperative agriculture. . The majcr irxtrurnent of policy implementation on the production side is the marketing boards. These are semi-democratic bodies, rvith nominated members representing growers, traders, and consumers. The boards are responsible for production control, marketing, exports, and the distribution of subsidies to product prices (investment capital and water are aiso subsidized, but not through the marketing boards). Planning and subsidization go hand in hand. However, only livestock products have been subsidized on a regular basis, and livestock production is controlled quite effectively. Efforts to control the production of non- subsidized vegetables and fruits has largely failed. The rate of support determines the effectiveness of planning in agriculture, but subsidies have not been stable: product price supports in agricuiture in 1986 were only one-tenth of ttreir 1984 level. It may well be that such wide fluctuations Collective Auion in Israel 273 will not be seen in the coming years, especially if inflation is kept at its current levels. But agriculture cannot expect to get in the future the same kind of support it enjoyed in the past: government budgets are now tighter, the sector lost much of its glory and political attraction, and the government is burdened with its share in the resolution of the financial crisis. Routine planning, particularly in the livestock industry, has been quite successful despite recurring failures and difficulties. Most of the time, the goals of ample supply at stable prices with reasonable income to the growers have been achieved. Planning largely failed in the major policy areas. As I attempt to show below, an optima! water policy was not implemented; not only did the government not prevent cooperative agriculture from sinking into debt, but also it contributed directly to the accumulation of excess capacity and thus to *re creation of the crisis conditions. The government was paralyzed in dealing wi& marketing boards and agricultural expors and adamantly rejected all proposed innovations (citrus marketing is an exception to be discussed below); the government was similarly ineffective in dealing with structural problems in cooperative agriculture and prevented long overdue changes from occurring. These "government failures" are not incidental; they reflect yielding to narrow interests (Zusman and Rausser, l99l), arrogant rejection of professional advice, basic mistrust of the market process, favoritism (often motivated by good intentions), and the inability to implement necessary but painful changes. It seems that the government modifies its ways only if dragged into a deep crisis or if it clearly and demonstratively loses the ability to enforce its policy. Water management, the Citrus Marketing Board, and the financial crisis in cooperative agriculture are relevant examples. WaIer Water is a common resource. It belongs by law to the state and is controlled by the government. There is no private ownership of water in Israel. The national system of reservoirs and conduits connects all the importrnt sources and users into a single network, stores water from winter to summer and between years, and moves water from the rainy north to the dry south. The government allocates pumping quotas and user's righs. Water charges are set by the government in consultation with a parliamentary 274 Y. Kislev committee in a process open to political pressures, which the agricultural lobby applies skillfully. Irrigation water is subsidized at two levels: operating costs of the national water company and of regional suppliers aie s,rpported, and the national system is not charged for the capital cost of the main conduits which were constructed with public funds. "Water carries ttre subsidy to the end of the pipeline," is the argument often made in support of water subsidy as a means to promote farming in remote regions. But water subsidizations do more than that; they have changecl the pattern of agricultural production in Israel" The production of .otton, the country's most important field crop, would have been much smaller if water had been charged at cost. Similarly, a great part of the citrus production would have been eliminated. Subsidization increases the demand for water ancl the political pressures to allocate more of the available quantities and to invest more in the development of water supply' The Water Authority, the agency in charge, operates mostly wittr the farmers' interest in mind, and it often yields to short-run pressures. The consequences have been over-utilization, hydrological deficits, intrusion of sea water to the coastal aquifers, contamination of other reservoirs' and a reductron of the carry-over capacity of the system. These detrimental effects have been accumulating gradually and are hard to detect and comprehend by non-professionals, but in the last two summers the combination of three dry years and poor reserves forced drastic curtail- ment of supply with harmful effects on farms, particularly orchards. The winter of l99ll92 was most unusually rainy and some of the reservoirs were replenished. But long-term damage cannot be amended in a single season. The Citrus Industry Citrus, particularly oranges, was the economy's most important export at the beginning of the 1950s, and at that time the area planted with citrus trees occupied half the irrigated land in the country. In the 1950s the citrus area was expanded as new orchards were planted in reaciion to increased demand in Europe On the one hand, and expanded irrigation, skills, and availability of capital in Israel on the other. In more recent years, however, the industry has declined (Table 2). Orchards have been uprooted and over the last 15 years exports of fresh fruits have decreased to half their volume. Part of the decline in tonnage reflec* a shift from the traditional bulky varieties to newer, more expensive types, but most of the reduction in TableZ is the result of cuts in production and diversion of fresh fruit to processing. Collective Action in Israel 275 TABLE 2 Recent Developments in the Citrus Industry Average 198i 1988 1976-1979 Area of orchards (000 hectares) 42 37 36 Export of fresh fruit (000 tons) 925 700 452 Share in production (%) 61 45 40 Value of export (million 1986 dollars) fresh fruit 428 248 189 processed 2t0 226 344 Terms of trade in export (ratio of FOB 100 87 89 price to input price index) In cutting production farmers reacted to changes in the terms of trade which deteriorated at the farm gate even further than indicated in Table 2. For the growers are the residual claimants in the chain of revenue from the consumer downward, and as prices in Europe declined with increasing supply of fruits by Israel and by is competitors, the processors, packers, and shippers between the farm and the harbor (where FOB prices are gauged) continued to cover their costs unaffected by the market fate of the fruit. Moreover, fruit handling cosB were high. The Citrus Marketing Board was composed mostly of packers and, though some of them are growers in their own right, these were the packers' interests that dominated the operation of the Board. Consequently, the packing and shipping industry was nm as a cartel with over-capacity and inefficiencies whose cost the citrus producers were made to cover. By law all marketing of citrus, both domestic and export, was handled by the Board. The Board accepted the fruit from the packers and shipped it to Europe; lower grades were delivered to industrial processors. Relatively small quantities went to the local consumer markets where monopoly prices were charged. The growers received pooled prices, modified to reflect quality and harvesting time. The Board was also responsible for pest control and planning ofareas, locations, and varieties. It functioned as the industry's political lobby and participated in the financing of research and its direction. The citrus industry in Israel missed, under the leadership of the Board, two major developments that occurred in the world markets in the last decades: Israel lagged in developing and adopting new varieties, particu- larly easy to peel types and sweet grapefruits, and Israel ignored the shift 276 Y. Kislev in consumption from fresh fruits to reconstituted concentrated juice. Consequently, by the time the Israeli growers came to the markets with their new varieties, prices were already down to competitive levels, and Israel did not even try to develop processing-dedicated orchards of the kind grown by the major producers of concentrates in Florida and Brazil. The deterioration in terms of trade and the disillusionment with cooperative action fueled a "growetrs' mutiny"; farmers grouped together to struggle for free marketing and competition in exports. Perhaps naively, they were willing to give up the advantages of a monopoly position in the local market, the bargaining power of the Board in the export markets, the economies of scale in shipping and handling, and much of the expertise accumulated through the years in the Marketing Board. Eventually they had their way. A sympathetic Minister of Agriculture stripped the Board of its monopoly powers and opened the market to competition. The winter of l99ll92 was the first season in which marketing was in private hands. Six export companies are now operating in the market and there are no restrictions on domestic transactions. The Board was retained with a small staff to monitor developments, to supply public services, io influence research, and to collect levies to finance these functions. It is too early to assess the new experience, but two aspects of this recent development are already encouraging: operators at all levels pay more attention to quality now that ptroling has been narrowed substantial- ly; and smuggling fruin to the local markets - an ugly aspect of the domestic monopoly of the Board - disappeared. The Financial Crisis in Cooperative Agriculture Eighty percent of Israel's agricultural product comes from cooperative farms. A major form of cooperation in agriculture has been financial. It flourished for a long time; but financial cooperation has found itself recently in a deep crisis the roots of which lie in the structural weaknesses of cooperation, in government action and inaction in agriculture, and in macroeconomic factors, particularly inflation and the policies adcrpted to fight it. Debt settlement agreements were reached, but it was doubtful right from the beginning whether the sector would be able to honor ils obligations. And, indeed, a new debt settlement law to replace the original agreement in the family-farm sector was recently passed. Collective Action in Israel 277 Moshavim and Kibbutzim A moshav (plural moshavim) is a farming community in which all farms are family owned and operated, and all farmers are members of &e multipurpose, democratically run, village cooperative.2 In principle (practice varies), the cooperative association in the moshav purchases all farm supplies for its members and markets ttreir farm products. It may also own and operate a variety of service facilities and manage directly some jointly operated farm enterprises. A kibbutz is a commune. Members work together and receive from the kibbutz food, shelter, health care, education, clothing, and a relatively small monetary allowance which they may allocate at their discretion. Again in principle, a member in the kibbutz has no property other than personal belongings. Differences in the degree of cooperation induced many other operational differences between the kibbutzim and the moshavim. Two examples are noteworthy: 1. As the labor force exited from agriculture, members in the mosha- vim shifted mostly to part-time farming and found additional employ- ment in other sectors. The kibbutzim, on the other hand, invested at their own risk in ttre creation of employment opportunities for ttreir members in manufacturing and services.3 2. Since consumption in the kibbutzim is communal, the management of a kibbutz has much larger control over the consumption level of its members than the cooperative association in the moshav. With control comes responsibility. The modern kibbutz cannot lag too far behind the country's standard of living - its members, particularly young mernbers, will leave. Some are already leaving (not all for economic reasons, to be sure). These considerations dictated and still dictate much of the behavior of the kibbutzim in economic and financial affairs. Moshavim and kibbutzim are members in two types of second-order cooperatives: supply cooperatives (sometimes named requisite societies or 2Thi. ir a transition period in the moshavim, and many are undergoing wide- ranging strucfural changes. I shall discuss these changes below. 3There is an interesting similarity in the numbers. Only a third of the operators in the moshavim are full-time farmers, and tarming contributes on the average a third of the total income in the kibbutzim. 278 Y. Kislev purchase organizations) set up to purchase farm requisites for their members, the moshavirn and the kibbutzim; and regional service enterpris- es (feed mills, slaughterhouses, transportation services, and others). Both types operate on a regional basis, though some nation-wide cooperatives also exist. Starting with the transfer of suppliers' credit to their members, both the associations of the moshavim and the supply cooperative expanded into full-scale financial intermediation. This tendency was reinforced by the fact that most farrn land in Israel is nationally owned, and moshavim and kibbutzim cannot use it as collateral. Fin m cial I nte rm e diatin na The pivotal role credit intermediation occupied in the activities of the moshav and the supply cooperative is demonstrated in their balance sheets in Table 3: members' debit balances were by far the largest assets the associations held - 763% of the total in the moshav and 60.9% in the supply cooperative. The moshav and the regional coop raised debt and transferred it to their members. The associations also functioned as clearing houses, accepting deposis from members with financial surpluses (rnembers' credit balances in Table 3) for use by others. The supply coop and its moshavim were linked strongly together: through credit, as can be seen in Table 3, and through joint ventures in regional service enterprises. The relations between the kibbutzim and their supply coops were similar to those depicted in Table 3 but there are in general no financial transac- tions between the kibbutzim and their rnembers. The supply cooperatives provided their members - the moshavim and the kibbutzim - with financial services, including steady lines of credit and convenient saving facilities. The moshavim provided their members with the same kind of services. Interlinkage between marketing through the cooperatives and credit operations provided the institutional set-up that replaced collateral for loans in cooperative agriculture. In addition, virtually all members - individuals, kibbutzim, and moshavim - were parties to mutual guarantees, and all were mutually responsible for loans raised by their cooperatives. Proximity, central purchasing of inputs, 4This section depicts intermediation as it was practiced before 19E5 . One of the consequences of the crisis has been a great reduction in the financial interconnections between cooperatives, especially in the sector of the moshavim. 279 $ o\n\ 6N O E dd+ t -i-l o ir.1 I NN o g \t B u;o a4 xx ao aao= @ '5 o\ 3s \] .:oq) r*N\o( F\a o o ,;d-e cf'l q o$o\o rf(i o k F-O )a {Yg d) 0) r-* E ;; ,e 6) o'.= U) --B-- 04e g o 5 E E .EH i o o o olH'q? tr ,a o () o o H:== F ts *.* i b x h556hh6xFF460r\ J E€ R* aoi i' I cB ;: U) .5-r.IIEa€! d e6ttE B,gs SFE 3f € lil)u)q) q)z '5sIc5!go (! xE =Ei6or.i: !!, h- d H.;€',;b I E \.lf- \Oc..}O\ O o ..i d t c.i o.i cj d $ , _\O O l3eEF, $ E.9.;E.3 S 6HEo! V2 a & F'E € € H q) tr 6 > tr.d A 4 uoo>5e Ht'6€ 6 \ * 9'F ;6 f-\n O N \O () .oi:!itr - o o o'- d."i + oi rf\9t.o o o >'= o-: .=q (j---xtrLd-A9 .< 6) () ^ :i '- a o.d68 OdH9EE'IEsi -:qr r) O 4groE 'v=a 6) O^ ao 3d L3= o- d..qa a E s 9l a (46 9- g 9+ o# d tl.! @ 6)o XL & 6 a v-a O\ -di4;LtrOcdXv 9b Eg0'd.aaH.i E E: EA > tr o H ru.o .li o: ii'- tr dtr 6{.)0(l)=o m'i2 ff'; E B sE q: !€T E E E (rl d9,aEH593E9'iEedf f f E:E ,t H -CiJd.C$ J E A'E E E E SE E i;F€cqF F I m Ir.J ,.:r< Z Fo F Soeoo€ : 280 Y. Kislev TABLE 4 Outstanding Bank Credit (percent) Voluntc of Real Credit Rarto of Debt to Net Capital Agiculture Industry Agriculture Industry 1969 r00 100 t9 52 t974 209 219 35 69 1979 335 327 48 75 1984 491 352 67 7A t987 655 402 79 64 product marketing, and financial interdependency should have, in principle, allowed close monitoring and control of the economic affairs of the member-borrowers. For a long time they did (as witnessed by the increased capital intensity in agriculture compared to industry in Table 4), but they failed the test of extreme economic circumstances in the inflationary period and their weaknesses were brought to light in the recent crisis. Regional Enterprkes These are second-order cooperatives, the members of which are moshav associations and kibbutzim, mostly potential patrons of the service offered" Often the regional supply coop is also a member; and in all cases the two kinds of regionals - ttre supply coop and the service enterprises - are strongly interconnected financially, a relationship that proved detrimental when the recent crisis eruPted. Zealous adherence to rural development by public agencies, easy access to credit through the supply cooperatives, and strong political regional iobbies resulted in over-expansion of many of the service enterprises. This occurred particularly in the 1970s when credit was in ample supply and economic optimism ran high. Consequently, in the early 1980s many service enterprises operated under capacity and could not cover their operating costs. Inflation in the late 1970s and early 1980s eroded most of the debt of the regionals, but circumstances gradually changed with the adjustment of the financial markets to ttre inflationary environment. The supply cooperatives, assuming the role of the financiers of last resort, found themselves financing not only operating losses but also debt services of the regional enterprises. A few of the enterprises collapsed an{ went bankrupt Collective Action in Israel 281 in the crisis of 1985, and ttren they took the supply coops down with them. Government Cooperation in agriculture has always been supported by the govern- ment: new immigrants settled in the cooperative moshavim as a matter of policy, land and water allotted to the moshav and distributed equally between the memberso production quotas allocated on a village basis, and the moshav decides on internal distribution; government agencies usually consult with *re cooperative association in the moshav on the allocation of long-term loans to farm operators. The deepest public involvement was in credit' The government raised debt in capital markets to finance its budget, 'Jrus crowding out private sources of iong-term investment credit. To remedy the shortage of its own creation, the government distributed credit to approved borrowers and subsidized it. Moreover, it was also often ready to supply additional credit to enterprises - farm cooperatives in particular - which ran into difficul- ties. The dependency on the government and the expectation that it would bail out troubled moshavim and kibbutzim created moral hazard problems. Cooperatives at all levels were willing to rely on large amoun8 of debt and banks were willing to lend, all trusting the government to save them in case of misfortune. This problem of moral hazard was recognized' by the government, and tools to controi its consequences were created in the 1960s and implemented vigorously, but the will to maintain a strict policy could not resist the flood of credit in the late 1970s. Moreover, the government itself encouraged uncontr Aggravatfug Polity Factors The roots of the financial crisis in agriculnrre are in excessive expansion - of investrnent and debt - of the agricultural sector and particularly of cooperative agriculture in moshavim, kibbutzim, and their regionals' The deep crisis and the difficulties agriculture is facing were aggravated by several policy factors. One of the anti-inflationary policy measures adopted in July 1985 that was particularly hard on the production sector was a severe credit squeeze which caused an unprecedented increase in the rates of interest (l0O% pet year on overdraft facilities, for example) and a reduction in credit 282 Y. Kislev availability. These developments hastened the coming of the crisis and intensified its consequences. Another such measure was an exchange rate pegging policy adopted to stabilize the local price level (creating a "monetary anchor"). Despite several devaluations, the real exchange rate has lagged significantly behind domestic price levels in recent years and prices farmers received for exports declined markedly. Simultaneously with the deterioration of the terms of trade, the quantity index of fresh agricultural exports (not including processed food) decreased by fi% between the second half of the 1970s and the 1980s. Citrus exports suffered particularly (Table 2). Since the domestic demand for agricultural products expands only slightly, the reduction of export is severely harmful to agriculture. And indeed, the 1980s saw a substantial reduction in the sector's income (see Table 5, below). Crtsis The crisis erupted at the end of 1985 once creditors realized that agriculture, particularly cooperative agriculture, could not or would not continue to service its debt in view of the exceedingly high, post-reform, real rates of interest on short-term loans and that the government could not bail out the sector any more. Most regional cooperatives and many of the associations in the moshavim collapsed. Farm production has continued, often with private credit arrangements and the farmers' own resources. But this could not be a complete solution to the crisis, and banks and other creditors continued to demand repayment of their loans. For most members in ttre cooperatives the heavy burden is not their own debt but their share in the mutual liabilities - their share in covering the debt of several heavy borowers in the moshav and the debt of the regional enterprises. Agriculture cannot repay or service is debt in full; the question therefore is how to distribute the losses. This was alreadv realized in 1986 and, consequently, the government stepped in offering its support to reach a debt settlement between the banks, on the one hand, and ttre tarming sector on the other. Agreements were formulated in 1988 with the moshavim and in 1989 with the kibbutzim,s but their implementation was 5The principa! component of the debt settlement is a rescheduling of loans according to accepted measures of ability to pay. The agreements are sector level framework contracts to be implemented with every kibbutz and every moshav (sometimes with every farm operator) separately. Collective Action in Israel 283 slow as many in the sector kept hoping that they could gather political support for a more favorable settlement. Indeed, they succeeded. But before we turn to the most recent events, let me consider the sector's ability to rnaintain the settlements originally reached. Can Agriculture Honor the Debt Sefrlements? By the available estimates at the end of 1988 (the reference point for debt settlements) agricultural debt was NIS 6.5 billion, and the value of is net capital was then NIS 6 billion (the exchange rate was then US $1.8 to the New Israeli Sheqel). Agriculture as a whole has by these figures no equity of its own; all its capital is financed by debt. With the 1988-1989 debt settlement agreements, close to a third of the debt is erased and the remaining obligations are rescheduled for a period of 15-20 years; the new loans are linked to the price index and they carry low interest rates. Assume accordingly that agriculture is left with a debt of NIS 4.3 billion to repay over 20 years at 4.5%; the annuity is then NIS 331 million. If agriculture's debt is actually repaid at this rate, the sector will redeem its equity capiEl at the 1988 level over the next two decades. This is an impossible undertaking. For even if the calculated annuity is an over-estimate, if part of the short-term debt is rolled over, not subject to repayments, and even if debt forgiveness is somewhat larger - to be able to maintain its part in the settlement agreements - agriculture will have to returnto the levels of profiS itenjoyed in the 1970s (Table 5). In that decade the operating profits were upward of NIS 300 million (at 1987 prices). At such levels, with replacement of only necessary capital assets, agriculture will be able to repay is rescheduled debs. But profitabiliry has been deteriorating in recent years. Terms of trade that were improved several years ago are worsening again, technological improvements can be expected to slow down with the reduction of investment, competition in foreign markes is toughening, and domestic demand is expanding only slightly.6 The parties to the debt settlements were aware of these difficulties, and they based a Ereat part of their optimism on structural changes to come in 6Estimates of operating profits such as in Table 5 are not available for more recent years, but other indicators show no sign of improvement in agricultural profitabiiity. 2E4 Y. Kislev TABLE 5 Operating Profits in Agriculture (million MS in 1987 prices) Output Purchased Deprecia- Operating lnputs tion Profits (t) (2) Own (3) Hired (4) (s) (6) 1965-1967 2116 829 854 292 188 -47 1975-1977 3927 1792 1094 374 302 365 1985-1987 4648 2334 1367 503 457 -13 Noles'. Column (3) is imputed according to the sum of per laborer consumption and saving levels in the economY. Column (6) is Column (l) - (2) - (3) - (4) - (5). The exchange rate in 1987 was NIS 1.6 to the US dollar' Source: Lerman and Kosto (1990). the wake of the crisis and the obligatiors implied by the agreements. Among them: the supply cooperatives ceased to function as financial intermediators; the regional enterprises were constrained to direct services to agriculture; some enterprises were closed down to reduce excess capacity; the kibbutzim adopted a policy ofredirecting labor from services to income generating activities; weak farms in the moshavim, judged unable to repay their debts were to be closed and their factors of production distributed among the remaining members; investment was to be limited to necessary replacements and only to carefully analyzed expansions. Increased efficiency can improve the ability of a firm in difflrculties to repay its debt. This, however, is not necessarily true fcrr an industry' In agriculture it can reasonably be expected ttrat if the structural changes are realized and efficiency is improved, terms of trade will worsen and profitability will not increase. Moreover, the recent crisis resulted in the collapse of part of the agricultural support systems and a reduction in the political willingness to budget subsidies for agriculture. It is impossible to see how agficulture can tax its consumers (for this is what it amounts to) and gather the necessary profits needed to service its old debts. If agriculture cannot cover its debt from profits, farmers may still do it by reducing consumption. At recent levels (last line in Table 5) f'armers will have to divert 74To of their income - returns to own labor - to debt service. It is, however, hard to expect this to occur, particularly because ttre debt is not evenly distributed and many will have to divert much larger Collective Action in Israel 285 shares of their farm income to the repayments stipulated by the agree- ments. The situation is particularly diffrcult for a large number of the kibbutzim, as the required reduction of ttre standard of living may be more than their members will tolerate. Still another alternative is for farmers to cover their debt from non-farm sources. This is possible in the moshavim where most farmers are part- timers, but it is not clear that they can be forced to do so. The kibbutzim have to cover debt incurred by manufacturing and service activities as well as by their farming sector. They have no outside sources of significant magnitude that can be diverted to the redemption of their capital. A,ftershocks Two recent developments markedly changed the economic environment of agriculture and in particular of cooperative farming. The first was a ruling of the Supreme Court (actr'rally two similar rulings) that a cooperative association cannot distribute its debt to its members against their will. The basis for the ruling is that a cooperative is, like a corporation, a limited liability association; members are responsible only for their share of equity capital. Lenders should know that they cannot collect from the private shareholders to cover the associa- tion's debt. This was a revolutionary ruling. Before it the assumption was that the majority of the members could decide democratically on the distribution of the association's losses to its members. The ruling made this practice unlawful. In the past, with easy finance, losses could always be covered with new loans, there was no need to make members actually pay cooperative debts, and distribution of losses was therefore never challenged. The crisis and the reorganization implemented in is wake forced clarification of the legal basis of the functioning of the coopera- tives. The decision of the Supreme Court did not apply to debt covered by mutual guaranties to which members were signatories. But this "shortcom- ing" was also, at least partly, soon to be amended. Lenders accept in debt settlements certain costs in an attempt to rescue the borrowers (and the capital lent) from immediate default. This recontracting does not eliminate all risks; certain aspects of risk are even intensified. Lenders now have to balance their readiness to assist borrowers in difficulty against the danger of throwing good money after bad. Borrowers may be tempted to moral hazard behavior in an effort to improve the conditions of the settlement agreement and to induce further 286 Y. Kislev debt forgiveness. These tendencies were particularly apparent in the farming sector in Israel: the agreements were widely conceived as "a plan to rescue the banks" and not the farmers; many could not or would not accept the reduced role of the government in bailing agriculture out of diffrculties; the fact that the government was party to the original agreements made it subject to political pressure to improve their condi- tions. The backlash was particularly strong in the moshavim. With the widespread belief that the settlements may be improved' many members in &e moshavim opted for a policy of wait and see and delayed the implementation of the agreements. Politicians could then point to the small volume of implementation as an indication that the agreements were unrealistic. Since members in the moshavim are diversified politically, it was not too difficult to rally a wall-to-wall coalition in their support, particularly in an election year. In March 1992 the Knesset (parliament) passed, against the expressed will of the government, a law defining a new debt settlement fof the moshavim (several private farmers and a few small and newly established kibbutzim were also included). The law grants more generous forgiveness and many improved conditions compared to the original agreements. In particular, the law nullified retroactively mutual guarantees that members signed for the debt of their cooperative associa- tions. Following the enactrnent of the new law, several of the banks in Israel are now writing-off significant parts of their equity capital as lost debt. Fearing that the new law may now apply to funrre loans as well, the banks are also withdrawing credit offered to members in moshavim' For several reasons, the implementation of the settlement agreements in the kibbutzim has faced fewer diffrculties than in the moshavim. The kibbutzim are well organized, and the strong ones support the weaker unis; being larger enterprises and lacking out"side sources, the kibbutzim cannot afford to sever their relations with the banks. The majority of the members in the kibbutzim vote for Labor and left-wing parties, and they cannot hope for the hnd of wide-based political support the moshavim got. As a result, the implementation of the settlement agreement, though slow, is proceeding gradually in the kibbueim. However, while individual kibbutzim are joining the agreement, it is becoming clearer that they many will not be able to fulfill their obliga- tions. The improved efficiency parameters incorporated into the settle- ments were too optimistic, and ttre outlook is now gloomier than it was at the time the sector-level agreements were signed. Evidently, the crisis is not over yet. Collective Action in Israel 287 Implicaions The crisis and its aftermath had devastating effects on cooperation in the family-farm sector. Most cooperative associations in the moshavim and at the regional level ceased functioning, and many farm operators now prefer to work individually and to avoid cooperation. Yet at the same time many recognize the advantages of collective action, and new cooperatives are springing-up in moshavim. The new cooperatives are generally narrower in scope, ttreir members join to perform specified functions such as milk marketing or the provision of feed, and they operate on cash basis or with limited sums of outside debt. Consequently, the interest of the membership is less diversified than in the comprehensive moshav cooperative, free riding is easily spotted and stopped, and the operation is smoother. It will be interesting to see how widespread and stable this new form of cooperation can be. The situation in the kibbueim is different. Financial cooperation in the regional and national level was eliminated, but other forms of collective action - such as in regional enterprises - continue and in many cases successfully. The kibbutz is still a commune, though some changes aimed at increasing members' awareness of outside economic signals were introduced. Despite the apparent stability and functional continuity, particularly compared with ttre land-slide effect of the crisis on the moshavim, the kibbutzim face real dangers to their existence. A moshav will still be a moshav even if its members become private farmers. The kibbutz will cease to exist if its members leave. Concluding Remarks Israel's farmers are skilled and highly motivated. College education is the norm in the kibbutzim, and many in the moshavim are also profession- als with thorough understanding of ttreir work. Farmers are also used to act together; cooperation and involvement in public affairs come naturally to them. They react speedily to economic and technological changes, adopt new varieties and methods, and reach record yields. But they are also fast to seize opportunities which may turn out to be misguided: when the rate of interest was negative and credit seemed to be in unlimited supply, farmers invested excessively. When water prices are low, farmers make the best use they can of this resource and they also find the way to cooperate politically for the expansion of the low cost supply. On the other hand, when growers received pooled prices for citruses and quality Y. Kislev was only partly rewarded, farmers curtailed production and did not reveal their technical ability in this line. As I read the record of agriculture, its potential is very high. Past achievemens testifu to this assessment, but the risk that it will be misguided is also high. This is particularly true for the cooperative sector where the strength of interdependence carries with it the dangers stemming from moral hazard behavior. A major responsibility rests therefore with the government. It must have the wisdom and the power to limit its involvement in agriculture, to let farmers be responsible - for better and for worse - for their acts, and to follow only sustainable policies. Given the cliversified political forces that shape its actions, it may be an illusion to think that the government can be expected more often than not to follow responsible policies. Farmers will act rationally and responsibly as individuals, but they will easily follow myopic, even irrational behavior, collectively. This difference between the individual and the collective perspective stems from a natural limited ability to internalize external effects and just refusal to let everybody else be the free rider. The examples we considered demonstrate these lessons, of particular relevance is the financial crisis. Inflation created a special opportunity for agriculture in Israel. Wittt negative real interest rates and erosion ofloans, agriculture could increase its equity capital and emerge from the inflationary period economically stronger. This did not happen; financial leverage increased in agriculture, returns to capital and probably also saving were negative, farmers sank deeper in debt, partly to finance invesfinent in production assets (often with overcapacity), partly to finance housing and consumer durables, and partly to increase current consumption and the standard of living. Considerations of short-run inflationary gains dominated long-run economic health. The crisis is a clear example of the consequences of cooperative myopia. But the cooperatives were not ttre only ones at fault. Credit was provided by the commercial banks; it was their money that was lent and it was their responsibility to secure ttre loans and to control their use. Evidently they neglected this responsibility. lrarning from experience, they relied on the government and did not realize that the scale of the problem was in the 1980s beyond its ability to solve. It was the government's responsibility to see the impeding crisis, just as it is the government's responsibility to guard water, Israel's most precious natural resource. And it was the government which failed. By its policy, which was a result of its effort to accelerate development and its Collective Action in Israel 289 yieiding to political pressure, it created the false impression that it would bail agricultrre out of any difficulty. Moreover, the government carries the major blame tbr over-capaciry in agriculture. Farmers and regional government ofticers naturally tend to increase their share in aggregate capacity. Since the funding of most of the development projects was rvith government approval and assistance, it was the duty of the government to examine the aggregate picture and to balance the desire to invest against the needs. This was not done; the decisions of the policy makers and evert the recommendations of the Planning Authority of the Ministry of Agriculture encouraged over-invesfinent. The crisis in cooperative agriculture is therefore largely the outcome of favoritism it enjoyed for a long time. Cooperation has many advantages and significant weaknesses. There are two preconditions for cooperation to survive the market test. The first is that members must have high levels of cooperative ethics, and they must be ready to give up short-run gains for the long-run benefits of coopera- tion (Zusman, 1993). The second is that in the final analysis, members must be responsible for their actions, individually or collectively. This means that there must be a maximum degree of internalization of externalities. In Israel mutual liability must be constrained to levels that members in cooperatives - farmers in the moshav and moshavim and kibbutzim in the regionals - can actually cover. Private ownership of land should now be established so that farmers may have more to lose if they fail. They may have then more to gain if they succeed. It seems that the organization ofagricultural cooperation will now be rationalized in Israel. The crisis made the need for reforms clear for everyone to see. The question still remains whether ttrere will be much cooperation left after such a traumatic experience. Acknowledgements This paper draws heavily on my work with Lerman and Zusman (1989), my survey of agricultural developments in Israel (1990a), and a papet prepared for the World Bank Conference in Budapest (1990b). Zvi Lerman suggested many modifications and corrections to earlier versions of the article. Detriled commen$ by Maarten D.C. Immink are also greatly appreciated. 290 Y. Kislev References Kislev, Y. 1990a. "Studies in the Development of Agriculture in Israel,' in Y. Kislev, Technology, Cooperation, Growth, and Policy - Studies in Agicultural Economics. (In Hebrew). W. 13-77. Jerusalem: Magnes. 1990b. "Structure and Reforms in Agriculture in Israel," a paper delivered to the Worid Bank Conference on Agricultural Reforms in Eastern Europe and the USSR, Budapest, Hungary, August, 1990. _, Z. I*rman, and P. Zusman. 1991. "Recent Experience with Cooperative Farm Credit in Israel." Economic Developnwnt and Cultural Aange 39(4): 773-789. Lerman, 2., and N. Kosto. 1990. "Profits and the Ability of Agriculrure to Repay its Debt," in Y. Kislev, Technology, Cooperation, Growth, and Policy, Studies in Agricultural Economics. (In Hebrew). Pp. 113-199. Jerusalem: Magnes. Zusman, P. 1988. Individual Behavior and Social Aoice in a Cooperative Settlement. Jerusalem: Magnes, Hebrew University, Jerusalem, _, and G.C. Rausser. 1991. "Organizational Equilibrium and the Optimality of Collective Action," Working Paper 9105, The Center for Agricultural Economic Research, Rehovot, Israel. 1993. This volume. Working Paper No. 9915 Farmer Co-Operative snd the Cofree Auction in Tanzania by Anecleti I( Keshuliza And Yoav Kislev TIIE CENTER FOR AGRICT'LTI'RAL ECONOMIC RESEARCH P.O. Bor 12, Rehovot FARMER CO-OPERATIVES AND THE COFFEE AUCTION IN TANZANIA- Anacleti K. Kashuli za andYoav Kislev** Abstract The role and functioning of farmer co-operatives in the coffee industry in Tanzania is examined with data from the coffee auction. It was found that private dealers colluded in the auction to lower the price of the coffee they purchased from farmers and repurchased in the auction. The resulting difference in prices is however not large. Co-operatives are too weak to perform effectively their role as a competitive yardstick. No evidence has been found to support the assertion that small-scale farmers, detached from clear market signals, produce lower quality coffee than estate growers. Introduction Farmer co-operatives have had a long history in Tanzania (Van Cranenburgh, 1990). The first co-operatives were established in the 1920s under the British colonial rule. They were set-up in a three layer structure: village level, primary societies; regional co-operative unions the members of which were the primary societies; and apex, national organisations comprising of regional unions. The co-operatives were often granted monopoly power as the sole providers of marketing services for cash crops in their villages or regions. After independence (1961), the official one-party policy was to strenglhen co-operation, even to force it. However, as the one-party policy was fuither pursued, independent co-operatives were dissolved (in 1976) and their functions were turned over to crop authorities. Since the early 1980s Tanzania has followed a liberalisation policy: co-operatives were re-established in 1982, still with monopoly power and public support, but they lost these privileges in 1991 and are expected now to be "viable" financially and compete in free markets. One of these is the coffee market. ' The research reported in this paper was supported by the Netherlands-Israel Development Research Program (NIRP). .. Sokoine University of Agriculture, Morogoro, Tanzania. The Hebrew University, Rehovot, Israel. 2 Coffee is the single most important export product of Tanzania. It is grown in several regions and sold for export in an auction held twice a month by the Tanzama Coffee Board (TCB) in Moshi, northem Tanzania. This paper summarises statistical analysis of information gathered from six auctions of Mild Arabica coffee.l The analysis is part of a larger study aimed at finding how liberalisation affected farm co-operatives and their functioning. Coffee and its Auction Coffee is grown on smallholder farms and larger estates. The beans are collected from the farmers either by the primary co-operative society in the village or (since liberalisation) by private traders. Most primary societies transfer the coffee they collect to co-operative unions for fuither handling. The co-operatives accept the coffee on consignment, they pay the farmers an advance when the coffee is brought in and a second and third payment usually follow after the coffee has been auctioned and the co-operative coffee account realised. Private traders buy the coffee and, in general, the price they pay is final. The harvested coffee beans are treated and dried on the farm and moved to curing plants where they are further treated, sorted into homogeneous lots and graded according to internationally accepted standards. Samples may then be examined by "liquering" and testing the brewed coffee for smell and taste. The participants in the auction are owners (sellers) of the coffee lots and the buyers of coffee. They are private traders, co-operatives and estates. Most coffee is bought for processing or export by the private traders, some export is done by Tanzania Export Company (TEC), the export arm of TCB, and by Kilimanjaro Native Coffee Union (KNCU!-a secondary co-operative union the member of which are the village primary co-operative societies. To determine a price and collect a TCB levy of I The first to collect detailed information on coffee auctions was Mr. L. N. Donge from Moshi Co-operative College. Mr. Donge let us have his data for the research reported here. The project he started was continued by Mr. Joseph Tesha and Ms Renalda Salum. The Excel data base and background calculations for this paper were prepared by Mr. Asangye Bangu. We are also indebted to many other people in the coffee sector, particularly in TCB, for detailed information and assistance. Auction data were also analysed by Temu (1999) in her study of the coffee market. J1 ZYo of its value, all the coffee sold as green beans (ready for processing or shipment overseas) moves through the auction. Some of the private traders (and also KNCU) may function in the auction both as owners and as buyers. In many cases they repossess their coffee; that is, they "purchase" the coffee they themselves offered in the auction. Before the auction the buyers receive catalogues specifuing, by lot, the source of the coffee, size and grade. Relying on Reuters most recent world price information, TCB's sales committee set reservation prices. These prices are given to the auctioneer but are not revealed to the buyers. Major Ouestions Coffee is grown on a large number of farms, most of them of small size, and it is traded by a relatively small number of dealers. Hence conditions of perfect competition cannot be expected to prevail in the coffee economy. Private dealers may collude----explicitly or implicitly--to influence the market and lower the price of coffee (compared to perfectly competitive prices) in two of the stages of the marketing process: (a) when purchasing coffee beans from the farmers, (b) when repossessing their own coffee in the auction. The damage to the farmers is done in the first stage; repossessing at low prices "only" reduces the levies TCB collects with no direct cost to the growers. The behaviour of the private dealers in the auction is examined in this paper with the anticipation that it will be possible to deduce from this behaviour on their performance at the village level. The question is, then, do the data indicate collusion of private dealers in the auction? The market role of the co-operatives-being the non-profit entities in he coffee sector-is to provide a competitive yardstick (Sexton and Iskow, 1993). In the village the co-operatives may take the coffee for its full worth and similarly at the auction, the co-operatives unions, if they also export coffee, may offer to purchase lots brought by private dealers at their full value. Do we learn from the auction that the co-operatives succeeded in their function as competitive yardsticks? 4 The structure of the coffee economy poses a difficulty that would not have been resolved even if the market had been perfectly competitive. As many of the farmers produce small quantities, their coffee has to be collected into pools of relatively large size, rvhether by the co-operatives or by private traders. In this way the coffee is taken form the grower before it was sorted, graded and priced. Some attention is given to quality at the village level, but it is impossible to achieve an accurate transmission of the quality and value of the coffee as sold in the auction. Consequently, farmers, whose coffee is pooled, are not rewarded accurately for higher quality coffee and are not penalised for lower quality. Given the distorted incentives, do smallholder farmers produce and supply comparatively lower quality coffee? Summary Statistics Table 1 summarises the salient feature of the six auctions analysed in this report; they were (column 2) the first two auctions on which we have data (summer 1996), two auctions in October 1997, ar:d the most recent auctions in our data set-from December 1998 and January 1999. The f,rrst two auctions were relatively small, less than 500 tons of coffee in each, the other four were larger with more than 1,000 tons per auction. All in all, 942lots with upward of 6,000 tons of coffee were sold in the six auctions covered in this report. The owners (sellers) of coffee in the auction were divided into four groups. By far, the largest quantities of coffee (90% of the total) were brought by private traders, the other owners were independent large growers (estates), KNCU and other co-operatives-both unions and primary societies. Close to three-quarters of the coffee in the auctions (74o/o, column 5) was repossessed, "purchased" by the owners. The catalogues identified six regions in which the coffee originated, they will be included in the statistical analysis to follow but omitted from Table 1. The coffee lots were assigned 14 grades. Most of the coffee was in the best three grades AA, A, and B. Column 6 reports average prices, in dollars per kg, for the categories in the table. We report separately the average prices of the highest three grades. The best, TABLE 1: Summary Statistics Lots Weight Share Average Price (ke) (%) Actual Normal Auctions 11.07.96 84 405,556 6 1.86 2.26 22.08.96 68 382,779 6 2.s2 3.07 09.1,0.97 194 1,736,944 25 3.47 2.97 23.t0.97 158 1.450,675 21 3. 1 5 3.04 17.t2.98 228 1,285,392 19 2.9r 3.04 07 .01.99 2t0 1,557,467 23 2.65 2.96 Owner Estates 42 234,572 3 3.79 3.08 KNCU 39 226.342 3 3.70 3.09 Other Co-ops 59 207,450 J 3.54 2.89 Traders 802 6,135,786 90 2.88 2.95 Repossessed 629 5,018,102 74 2.89 3.01 Grade AA t77 r,663,402 24 3.26 A r76 1,700,611 25 3.i6 B 164 1,455,819 2T 3.09 Other 425 1992,98r 30 2.46 Total or average 942 6,812,813 100 2.96 2.96 Source: Auction data. Notes: Column 5 is share by weight. Columns 6,7 , in US dollars per kg. grade, AA, was sold for $3.26 per kg. Other grades fetched lower prices. The overall sample average for the six auctions was 52.96 per kg. The entries in column 6 indicate differences in average prices between groups in categories. Thus average auction price ranged from S1.86 in the first auction to $3.47 in the third. Among the owners, private traders received the lowest price ($2.88 per kg) and---since most of the private coffee was repossessed.--essentially the same average price was received for lots "purchased" by their owners. Column 7 reports an attempt to gauge the quality of the coffee in the categories. The entries in the column were calculated by first assigning to each lot the average price for its grade and then taking the average on all the lots in the group. The result is termed "normalised price" and it is actually a quality index. Quality is here defined by average grade composition and measuring it in units of dollars per kg facilitates easy comparison with actual price. Thus the first auction, with quality 6 measure of $2.26 per kg, had on average a lower quality coffee-as indicated by grade composition-{han the other five auctions. The actual average price of the coffee in the first auction was even lower, $1.86 per kg, probably reflecting comparatively lower world price at the time of that auction. Also, by our quality measure, estates and KNCU brought to the auction higher-grade coffee than the other co-operatives and private traders. We turn now to the statistical analysis. Statistical Analysis The tool of the statistical analysis is a regression explaining prices and quality by the attributes of the coffee. All the attributes entered the regressions as dummy variables and the procedure is therefore an analysis of variance. The observations were lots sold, and the dependent variables were the logarithm (natural) of the actual or the normalised price. The dependent variable in regression 1 in Table 2 is the price by lot. The coefficients indicate relative differences from the omitted variables. Thus prices in the first auction, held in11.07.96, were higher by 26.9% than those in the omitted auctions (17.12.98 and 07.01.99). Among the grades, the price of the omitted grade, A, was l.4Yolower than the price of the best grade, AA. Grade B's price wasby 4.3o/o (0.014+0.029) lower than AA's. The other grades were of minor importance and lower prices. Clearly, grades do not tell the whole story, coffee prices differ significantly by region. Tanzania is particularly proud of coffee grown on the slopes of the Kilimanjaro mountain; and indeed Northem Coffees fetched he best price in the auctions, 7o/ohigher than coffee from Mbeya in the south. It is however surprising that coffee from Moshi, also in the Kilimanjaro region, fared badly (some of it may have been assembled from outside the region). For the purpose of the present analysis, the most interesting coefficient in Regression 1 the last in the column. Repossessed coffee was "sold" for a price that was on average 2.4o/o lower than alienated coffee. A lower price for repossessed coffee indicates possible collusion of buyers, but the effect of the collusion is not large. 7 We turn now to quality. By Regression2, in which the dependent variable was the normalised price, the fourth auction, held on 23.10.97, had the best grades,79.4yo better than the omitted auctions. Moshi region had the lowest quality coffee, as well as the lowest price (Regressionl); but for the other regions, grade composition and prices did not go hand in hand. Tuming to owner (seller) groups, estate coffee was better by 4.1% and KNCU's coffee was 74.7Yo better than coffee purchased by private traders (the omitted group). Discussion Before we turn to the major questions posed earlier, note the differences in prices between the auctions. These differences probably reflect changes in world prices (not included explicitty in our analysis). Historical data seems to indicate a ten-year cycle of coffee prices on the world market. Most recently prices peaked at the beginning of 1997 and have been falling for the last couple of years (a good source is Best Investment Coffee Newsletter, http://binews.com). It is only since liberalisation that domestic prices of coffee reflect world prices; in earlier periods local prices were isoiated from international changes (Lofchie, 1994). Needless to say, auction prices are not necessary the prices paid to producers. The transmission of world signals to the farm gate has still to be analysed. Collusion of private dealers in the auction will show up in lower prices for repossessed coffee relative to alienated lots. As we have seen repossessed coffee fetched in the auction prices that were on average lower by 2.4% than coffee bought from others. This is an indication of collusion, but not a strong indication. It shows, by the way, that TCB is acting effectively in setting reservation prices that prevent substantial under-pricing of repossessed coffee. Auction data do not explicitly show collusion at the procurement stage but, as indicated earlier, if the traders collude in the auction, they may also collude in other segments of the market. As for the yardstick function, KNCU purchased in the six auctions 5 lots (including 3 it repossessed), less than 30 tons of coffee, against 39 lots and 226 tons that it brought to the auctions. KNCU is clearly a small "player" in the coffee economy, particularly in export. It can hardly function as an effective competitive yardstick. TABLE 2: Statistical Analysis Variable Regression 1 Regression 2 Coeff. t-Stat. Coeff. t-Stat. Intercept t.177 62.846 0.898 29.575 Auction 11.07.96 0.269 2.795 0.37t r.726 22.08.96 0.396 22.568 -0.010 -0.267 09.10.97 0.238 14.047 0.021 0.542 23.10.97 0.344 3.468 0.794 3.662 Grade F -0.449 -22.559 PB -0.066 -3.292 E -0.448 -7.3r3 CB -2.523 -30.774 UG -0.558 -16.450 HP -0.73t -4.609 C -0.258 -1 1.1 18 TT -0.324 -11.603 AF -0.3 l3 -1 1.498 TEX -3.475 -52.283 B -0.029 -1.719 AASP -0.186 -4.786 AA 0.014 0.812 Region Moshi -0.532 -5.448 -0.574 -2_694 Mbozi -0.212 -8.595 0.120 2.150 Makambako -0.227 -t2.383 0.141 3.443 Northern 0.070 3.960 0.1 89 4.837 Coffee Dar es -0.485 -6.223 -0.581 -3.9 1 3 Salaam Coffee Reposs'd -0.024 -1.940 Owners Estates 0.041 0,704 KNCU 0.t47 2.409 Co-ops -0.028 -0.538 Adjusted R2 0.117 Notes: Dependent variables: Regression 1: Natural logarithm of price; Regression 2: Natural logarithm of normalised prices. Number of observations 942. Omitted variables: Auctions 17 .12.98,07.01 .99; Grade A; Coffee alienated; Region Mbeya; Owner group Private traders. As for the price transmission and its behavioural implications, one would have expected estates, bringing their coffee to the auction and being fully aware of its grades and prices, to have the best coffee and co-operatives to suffer from free rider problem and deliver relatively lower grade coffee. The finding of our analysis do not support fully these hypotheses; KNCU's coffee had the best grade composition and 9 estate coffee was not significantly better than coffee brought by private dealers and mostly bought from small producers in the villages. In conclusion: the regional union KNCU collected 17,000 tons of coffee in 1993, in 1997 it got only a fifth of this quantity. It is losing its market share, mostly to private traders. Our analysis seems to indicate that co-operatives may have a vital service to provide when parties to the market are not equal but it seems that liberalisation created an impossible environment for the established, formerly supported, co-operatives. A detailed analysis of this assertion, and assessment of possible remedies, are beyond the scope of this paper. References Lofchie, Michael F., 1994, The Politics of Agricultural Policy. In: Joel D. Barkan (editor), Beyond Capitalism vs. Socialism in Kenya and Tanzania, East Africa Educational Publishers and Lynne Rienner, London, UK. Sexton, Richard J. and Iskow, Julia, 1993, The Competitive Role of Cooperatives in Market-Oriented Economics: A Policy Analysis. In: Csaba Cshaki and Yoav Kislev (editors), Agricultural Cooperatives in Transition, Westview, Boulder, co. Temu, Anna, A.,1999, Empirical Evidence of Changes in the Coffee Market After Liberalisation: The Case of Northern Tanzxia, A doctoral dissertation, the University of Illinios, Urbana, IL. Van Cranenburgh, Oda, 1990, The Widening Gyre-The Tanzanian One-Party State and Policy towards Rural Cooperation, Ebum, Delf, The Netherlands. C:\Tanzania\Six-A-03; October 13, 1999 Recent Experience with Agricultural Co-operatives in Israel* Yoav Kislev Although agricultural output in Israel has grown steadily over the last fifty years and per-capita production kept increasing, the sector contributes today less than three percent of GNP. Compared with other parts of the economy of Israel, agriculture is small. Partly, at least, this tiny share is due to relatively low prices. Producer prices have been gradually falling, particularly since the early 1980s, in line with expanded supply and with trends in world markets. These industry-wide changes have affected the fate of the farm sector and of co-operative agriculture. Israel has been, since its establishment in 1948 and even before that, a testing ground for institutional settings in agriculture. Approximately eighty percent of agricultural output in Israel is produced on co-operative and communal farms; the rest is produced by privately owned enterprises. Time and shifting circumstances have forced significant structural changes on co-operative agriculture. In preparation for the discussion of these changes, the survey is opened in the next section with a basic, telegraphic description of the co-operatives. To simplify the exposition, the description is written in the present tense although, as will become clear below, some of the early institutional characteristics have disappeared and others are being modified with ongoing reforms. * The preparation of this paper was supported by NIRP, Netherlands-Israel Development Research Programme in the project THE TRANSFORMATION OF FARMER COOPERATIVES IN TANZANIA: A STUDY OF THE ORGANIZATION AND MANAGEMENT PROBLEMS OF EMERGING COOPERATIVES IN SELECTED REGIONS (95-12.1). 2 Co-operatives There are four major forms of agricultural co-operatives in Israel: kibbutzim, moshavim, communal moshavim, and secondary service co-operatives. Many other small co-operatives also exist, but they will not be surveyed in this report. A kibbutz (pl. kibbutzim) is a village of 200-2000 people. Today there are 269 kibbutzim in Israel. Basically the kibbutz is a commune in which members work on the collective farm and in the non-farm enterprises—manufacturing and services. Consumption (food, lodging, clothing, education, health care, etc.) is provided on egalitarian principles and “according to needs”. Strictly speaking and by conventional definitions, the kibbutzim are not co-operatives, but they are an integral part of the institutional setting of agriculture in Israel and their story complements the examination of the development of the farm co-operatives. It is therefore included in the survey. A moshav (pl. moshavim) is a co-operative village made of 60-120 member families. There are 411 moshavim in Israel. The farms in the moshav are private, but all farmers are members of the village co-operative and provided by it with local services: marketing, provision of inputs, finance and municipal services. A communal moshav is a village where the farm or non-farm enterprises are run collectively, kibbutz style, while families own their dwellings. Members receive equal pay for working in the village enterprises, but run their own households privately. There are 45 communal moshavim in Israel. Most of the secondary co-operatives are regional organisations, the members of which are either kibbutzim or moshavim (seldom both in the same co-op), the services they provide are produce collecting, sorting, storage, transportation, cotton ginning and financial services. One secondary co-operative, Tnuva, is the largest manufacturer of dairy products in the country. We will consider its case separately below. 3 Early History Throughout their 2000 years in the Diaspora, Jews viewed Israel (then Palestine) as their holy homeland but only a small number actually lived in the country. For centuries, Palestine was part of the Turkish Empire, inhabited mostly by Arabs. The first waves of significant Jewish immigration and the re-emergence of Jewish agriculture are dated to the last third of the 19th century. A great part of the immigrants who then established agriculture in Israel were comparatively poor, had no capital of their own and were assisted, in their first steps, by national funds. Even before 1948, more than half the land area purchased for settlement was public. The legacy of national land was kept alive when many Arab farmers fled the country in the wake of the 1948 war and their land was taken by the state. Consequently, with limited exceptions, agricultural land in Israel is publicly owned and leased to cultivators at nominal prices on long-term basis. Naturally, national land and capital were more easily allocated to groups committed to agriculture, kibbutzim and moshavim, than to private parties who might have used them, or so it was feared, to further individual enrichment. The first kibbutz was established in 1912 by a small group of young farm workers who chose communal life as a way to mitigate hardships and poverty. Ideology followed and many other kibbutzim were later established by people educated and trained for that purpose. The first moshav was established in 1921, aimed, at least partly, to provide a suitable framework for families wishing to settle the land without accepting the rigor of communal life. The communal moshav was a halfway solution. Tnuva, the marketing co-operative, was founded in 1926. For years it had marketed all the farm products of co-operative agriculture in the country, but then it has gradually lost its market share in fruits and vegetables and has concentrated mainly in the collection of milk and the processing of dairy products. Regional secondary co-operatives were first established in the 1940s and grew in number and size as agriculture developed. 4 Functions As indicated earlier, co-operatives, both in the moshav and the regionals, provided their members with marketing services, they also purchased wholesale farm requisites and operated sorting, packaging, transportation, and storage facilities. The most important service they provided was however financial intermediation. The co-operatives cultivated national land and, as the farm units were small (even in the relatively large kibbutz, but particularly in the moshav), they could not use their assets as collateral and lacked direct access to the capital market. The instrument developed to replace the missing collateral was a set of mutual guaranties implemented at several levels: all members of a moshav mutually guarantied credit the moshav co-operative raisedfor joint ventures or to be distributed to individual farms. Similarly, moshavim and kibbutzim were guarantors to credit their regional co-operative associations raised. The purpose of mutual guaranties was to reduce the risk banks incurred in lending to co-operative farms and their secondary co-operatives. Experience reduced the subjective risk (the risk as seen by the creditors) even further: again and again, particularly in the 1950s and 1960s, the government bailed out kibbutzim and moshavim in financial difficulties. The assistance was usually provided in the form of government-backed, long-term loans replacing short- and medium-term credit. With inflation running high, a great part of the real value of this long-term debt was eroded before it was repaid. Co-operative agriculture flourished and banks were therefore not only willing to service agriculture, they were often eager to have co-operatives among their clients. Credit raising became particularly easy during the 1970s when accelerating inflation was accompanied by ample supply of loanable funds in Israel—probably a reflection of easy access to world glut of petrodollars. The regional secondary co-operatives raised capital, relying on mutual guaranties, and channelled large quantities of credit to their members. They became financial intermediaries. Banks required mutual guaranties but, evidently, mainly relied on the government “that will never let co-operative agriculture fall”. 5 Expanded credit supply was not the only manifestation of inflation; another, not less important, was the erosion of the information value of financial reports. Though methods to adapt accounting practices to inflation were developed, their adoption was slow and both members of many co-operatives and their creditors did not really have accurate pictures of their economic performance. It seems however that lack of economic information did not deter the banks from extending increasing volume of credit to the seemingly safe co-operatives. Crisis By mid 1985 the accelerating inflation was reaching more 500 percent on an annual basis and the government felt obliged to bring it to an end. The change of direction came on 1 July 1985 with severe monetary and fiscal measures: price increases were halted and interest on short-term credit was raised to rates of up to 100% annually. A great part of the credit channelled through the secondary co-operatives to kibbutzim and moshavim was short-term and it had to be renewed at the higher rates. No business could stand such sky rocketing rates for a significant period of time and most of the kibbutzim, co-operative associations in the moshavim and the regionals became insolvent almost overnight. By mid 1986 it was clear that co-operative agriculture was in a deep financial crisis. The crisis was triggered by the measures taken to halt inflation and was realised as financial insolvency but it had deeper roots. Four inter-related problems surfaced: a. Lack of control. Secondary co-operatives and associations in moshavim transferred credit to their members disregarding their ability to repay loans on the terms received. Members in co-operatives, who mutually guaranteed loans taken by their associations, did not exercise the appropriate control over the actions of the officers running their financial affairs. Banks continued to extend credit to co-operatives that could not demonstrate stable economic and financial standing. And above all, despite some earlier attempts to limit the sector’s debt, the government lost control of co-operative 6 finance. But banks and other agents continued to rely on the government’s implicit safety net and neglected sound financial practices. b. Diminished ideological commitment. Originally, members in moshavim, and particularly in kibbutzim, were highly motivated ideologically; they viewed themselves as pioneers in a national movement of the return of Jews to the land. National ideology and co-operative commitments reinforced each other. Members could easily be called to follow communal and co-operative norms of the kibbutz or the moshav: “Together we are building a nation”. Once the state was established and its economy stabilised, the national argument for co-operation lost its force, particularly with the second and third generations who took the moshav and the kibbutz for granted and did not have the same zealous attachment to the original ideology their parents had. In the wake of reduced ideological commitment came a reduction in the adherence to old norms of action. c. Moral hazard behaviour and free riding. Ample credit supply, mutual guaranties and reliance on the government to rescue co-operatives in trouble—encouraged moral hazard behaviour: farmers, co-operatives and kibbutzim, willingly went into debt to finance investment in production and consumption assets (particularly housing) even if repayment was far from assured. Moral hazard behaviour was not demonstrated only in the financial area: in the kibbutzim, loss of strict ideological commitment resulted in deteriorated work ethic, members were seeking easy tasks and short days. Increasingly, kibbutzim have been facing difficulties in getting talented and experienced people to undertake responsible tasks. Able young members either leave or go to work outside. d. Poor economic performance. Easy credit and faulty control lead to poor economic performance in kibbutzim and co-operatives. Political and social considerations took precedence over efficiency and income. Survival was deemed secured with the government safety net. Consequently, when inflation was halted and rates of interest rose, many of the kibbutzim and the co-operatives discovered that they were operating 7 in the red. More than a few of their economic activities were making losses and the debts they accumulated were enormous and rising as interest charges continued piling. Debt Settlement As the crisis erupted, a great majority of the farm operators in the moshavim and many of the kibbutzim found that they were either themselves in heavy debt or they were guarantors of debt incurred by otherstheir peers and, particularly, the secondary co-operatives to which they belonged. Mutual guaranties were however meaningless under widespread breakdown; no one had the resources to pay anybody’s debt. Moreover, the supreme court, where the problem eventually landed, freed the guarantors of their obligations; mostly on grounds that co-operatives were limited liability associations and as such members were responsible for the co-operative’s debt only up to the value of their shares, which was very small. The government found itself in a dilemma. On the one hand, it could not simply bail out the co-operatives as it had done previouslythe magnitude of the crisis was beyond the state budget and the public would not have tolerated pouring large sums of money on the small farming sector. On the other hand, if unattended, the crisis could destroy co-operative agriculture and two or three of the largest banks in the country might have gone under with their debtors. The government had to step in. The question was how to allocate the losses and at the same time to secure continued functioning of agriculture. Two separate major settlements were reached after long political debates. The settlement for the moshav sector took the form of a law enacted by parliamentparties both on the right and on the left had constituencies in this sector. The kibbutzim, on the other hand, were and still are identified with the left. Their debt settlement could not have been reached in the political arena; it was concluded as an agreement between debtors, creditors and the government. 8 The core of the settlements, both for the moshavim and the kibbutzim, is made of two-parts: (a) assessment of income potential of farms in moshavim and kibbutzim and rescheduling the part of the debt deemed repayable; (b) forgiveness of the debt that cannot be repaid. In the kibbutz sector, where the debt is larger, the government shoulders approximately a third of the loss; two thirds is taken up by banks and other creditors. The settlement agreements were enacted and reached some ten years ago. Their implementation for the secondary co-operatives was relatively simple. In the moshav sector, these co-operatives closed and ceased operating; their assets sold to private parties. The secondary co-operatives of the kibbutzim have continued to function but in many cases in modest forms and none as financial intermediaries. Although the farm by farm implementation of the settlement agreements is not complete yet, the acute crisis atmosphere has calmed down. Still, the immediate consequence of the crisis was a significant change in the financial environment facing co-operative agriculture. Kibbutzim and farmers in moshavim have to deal directly with banks; they cannot rely anymore on familiar financial intermediaries. Operating on national land, the farms cannot serve as collateral and credit is provided only to units demonstrating sound economic basis. The financial crisis was accompanied by a sharp turn in the terms of trade in agriculture; they have been worsening since the early 1980s. Today’s real prices of farm products are only fifty percent of their level twenty years ago. The combined effects of the financial crisis and worsening terms of trade created a tight selection environment, particularly in the moshavim. Many left farming, some to seek off-farm employment others simply retired; production has concentrated in the hands of a relatively small number of operators. Structural Changes Moshavim, although originally co-operatives, may and do survive economically as villages of independent family farms. Reliance on collective provision of services and 9 finance was convenient, particularly in early stages, but present day’s farmers will not accept forced co-operation. The role of the moshav association has been limited to mere necessities (collection of milk or municipal services); some associations were dissolved completely. But, the changes at the village level and in the co-operative association did not much affect the mode of operation on the individual farms. They were family units in the past and continue operating as such today. Structural transformation in the kibbutzim seems to be more profound, painful and slow. A kibbutz is a commune; it is a community of people, an economic set of enterprises and a social entity. In the past, the community was not separated from the economic enterprises. As on a big family farm, members worked together in the economic enterprises—agricultural, manufacturing and services—shared consumption and housing and cared for their elders and children. The toughening economic environment, deterioration of the ideological commitment, spread of free riding and particularly, exit of the youngsters, changed all this. Some kibbutzimespecially those that through hard work, wisdom or luck attained economic wellbeingstill cling to the communal ideology and practice, but most are changing. The changes are gradual, many of the members find then hard to accept, and the road to reform is obstructed with economic, psychological and social obstacles. A visitor returning these days to a kibbutz after 10 or 20 years of absence may not recognise any change: members go about their work, children to school, and people dine together—just as in the past. But most kibbutzim are undergoing drastic economic and social changes and many are searching intensely for a way to follow. If one may extrapolate current trends, the kibbutzim are moving to become communities of individual households. If the reformers have their way, the emerging communities will be located in rural areas and the residents will choose either to be members in the kibbutz that will form the core of the community or to keep private household. Those who elect to be members will earn their living individually in the kibbutz enterprises or elsewhere and contribute community taxes at relatively high rates to supply social and welfare services, thus striving to maintain the mutual 10 responsibility principles on which the kibbutzim were founded. Starting in this direction, several kibbutzim are already introducing differential wages, many more are charging for services, which were for years supplied freely—members will not anymore receive identical basic budgets or consumption goods. Members are also encouraged to seek outside employment to expand sources of income. On another front, the kibbutzim are making preparations to register family dwellings as private property. Next they will probably transform the economic enterprises into limited liability companies with shares distributed to members. One of the most difficult questions is the issue of retirement funds. When the kibbutzim were regarded as stable and successful economic units, it was thought that the kibbutz was strong enough to care for its older members from its own resources. As a result, kibbutzim did not contribute to retirement funds and are now finding that they lack the necessary basis for support of the older generation. Several kibbutzim solved the problem by accepting outside shareholders to their manufacturing enterprises and using the money received to establish retirement funds. But this route is open only to the few very successful manufacturing units. Kibbutzim in tight financial conditions will probably never solve the retirement problem. The fears of the older generations hinder significantly the efforts to reform the economic structure of many of the kibbutzim. It is interesting to note that many kibbutzim turned to external professional help in the reform process. People that for two or three generations relied on internal, democratic collective decisions making are now leaning on outsiders to guide them through the new unfamiliar maze. The hypothesis I offer as an explanation starts with the observation that a significant, though perhaps not always recognised, aspect of the reform is a drastic change in the effect of collective decisions on the distribution of cost and benefits between the members. In the old kibbutz, consumption, dwelling, education, health care and other amenities were distributed on egalitarian principles. Most economic decisions—such as where to invest or what enterprise to developaffected equally all members. With privatisation, most decisions have far reaching personal consequences: salary assigned to one group of workers or another, 11 pension benefits, registration of houses—all may affect favourably or unfavourably future individual wealth and income. Privatisation is therefore a painful process not only in drastically changing old ways of life but also in testing the delicate social fabric of the community of people living in the kibbutz. Professional guidance is now accepted where it would not have been dreamed of a few years ago. The reform of the communal moshavim is relatively simple. The major change has been the introduction of differential wages in the collective enterprises and encouragement of members to seek outside employment. Families continue to maintain individual household as they have been doing all along. Tnuva Tnuva, recall, is a secondary marketing co-operative that grew into the largest dairy in the country. It collects close to ninety percent of the milk, distributes a third of it to smaller dairies and offers the largest selection of milk products in supermarkets and grocery stores. Although officially belonging to moshavim and kibbutzim, Tnuva was run independently for many years. Its management was appointed on political grounds and the leadership did not feel obliged to report to the membership. Its success owed more to sheer size and monopoly position than to economic efficiency. But successful it was. To maintain its leading position and fight off increased foreign and local competition, Tnuva had to become more flexible and efficient. In today’s globalised markets, companies of Tnuva’s size share knowledge and ownership across international borders. To operate effectively under these conditions, Tnuva has to become a joint stock company. I will not go into details here apart from indicating that the reform is already under way. Concluding Remarks A great part of the survey was devoted to the financial crisis and its aftermath. That emphasis was natural; the crisis was the most dramatic and traumatic economic event 12 agricultural co-operatives have experienced. The crisis hit hard, it imposed costs—in terms of interest charges and financial starvation—that even strong economic entities could hardly sustain, but it also revealed fundamental weaknesses in the structure and mode of operation of agricultural co-operatives in Israel. The agricultural co-operatives were never petrified; they were living entities, always changing to adapt to new economic, social and political environments. But the historical changes were gradual, evolutionary, while the reforms implemented in the wake of the crisis have been revolutionary. Many co-operatives, in the moshavim and their regionals, closed down and most kibbutzim can be expected to cease functioning as communes in the near or perhaps the distant future. Are these drastic changes due only to the crisis? Would we have still observed moshavim and kibbutzim functioning as co-operatives and communes, the way they had done for generations, had the anti-inflationary policies not been implemented? In my judgement, the changes we have been witnessing would have come, perhaps on a different day, even if anti-inflationary policies had not been implemented and interest rates had not been raised to impossible levels. Moreover, even under more calm economic policies, the revolutionary reforms would have followed a financial crisis. Let me try to explain. Basic to my explanation is the recognition that active business units, co-operatives included, invest money they receive, often through the banking system, from saving households. The savers and their agents in the banks earn interest but put their capital at risk. Once they suspect danger, they call the credit back and a financial crisis erupts. Having made this point, I return to our story. The fundamental weaknesses referred to in the last section and described above—lax monitoring, reduced ideological commitments, free riding, moral hazard behaviour, youngsters leaving kibbutzim—had similar effects, they were all the roots and causes of economic inefficiencies. So long as credit flew essentially free, new money kept covering old losses. But this could not have lasted forever, one day the government’s safety net would have been tested and found lacking (worsening terms of trade could 13 have been the trigger) and “suddenly” the realisation would have downed that the co-operatives could not cover their debts. Such a change in perception would have then ignited a financial crisis: banks would have stopped lending and call outstanding loans back, co-operatives would have fallen into arrears and bankruptcies would have followed. The revolutionary changes in agricultural co-operatives must have been implemented; basic shortcomings must have been corrected. But so long as credit was easy to come by, difficult reforms could be postponed. It takes a crisis to implement even necessary changes. Whenever drastic reforms would have materialised, it would have been in the wake of a financial crisis. Should one conclude from the recent experience and changes that agricultural co-operation in Israel was a failure? I cannot be party to this conclusion. For years, co-operation and collective action were the instruments of development of agriculture, from traditional to modernthis was not a failure. To my mind, the lesson to be drawn from the experience surveyed in this paper is that institutions have to adapt on time to changing circumstances. Inefficient secondary co-operatives have to close down and successful large organisation have to turn into business-like limited liability companies, owned by the public and subject to the discipline imposed by the stock market. It should also be remembered that, even after the reforms, there is still room for a great deal of co-operation in agriculture where the typical economic units are small relative to the suppliers of services in the product and input markets. The future will tell whether the spirit of co-operation is still alive and the institutions of mutual help can be reconstructed on a sound economic and social basis to the benefit of members and communities. April, 2000 AGRICULTURAL COOPERATIVES IN ISRAEL: PAST AND PRESENT1 YOAV KISLEV Close to eighty percent of the agricultural output of Israel is produced on coope- rative farms. In this chapter, I review the principal features of the past and more recent, still evolving, history of the two main forms – though not the only forms – of farm cooperatives: the moshav, a cooperative village, typically of 80-100 fami- lies, and the kibbutz, a commune with 100-800 members. An in-between type is the collective moshav, where the land is farmed collectively but households are owned privately. Associated with these units are second-order cooperatives – organizations whose members are themselves cooperatives, not individuals – regional cooperative associations serving moshavim and kibbutzim (the plural forms), and several nationwide bodies. Evidently due to its idiosyncrasies, the kibbutz has been the subject of intensive scrutiny and research; a recent count found more than five thousands publications. Much less has been written about the moshav although, as I trust the reader will realize, it also offers interesting economic, social, and institutional lessons. 1 HISTORIC MILESTONES The second half of the 19th century saw pogroms in Jewish communities in Eastern Europe, driving many to migrate westward; a small trickle went to Palesti- ne, then part of the Turkish Ottoman Empire. They formed the first wave of the Zionist movement – to repopulate the land of the Bible with the people of the 1 I am in debt for information and advice received from many knowledgeable people. Particular thanks go to Shlomit Arbel and Gadi Rosenthal, who shared with me written documents and spent hours discussing issues raised in the review. Dvoira Auerbach edited the manuscript. The responsibility for errors, shortcomings, and opinions is mine. 282 Yoav Kislev Bible. Most of the newcomers did not join the small, existing Jewish communi- ties in Palestine; they established separate towns and villages. Some brought capital from abroad but most were assisted in their efforts – land purchasing and investments in agriculture – by money from the Jewish Diaspora, both private philanthropy and national funds. This mode of operation continued even after the establishment of the State of Israel, with new settlements set up by the Jewish Agency (the major Zionist organization that operated in Palestine and still operates in Israel) and delivered to the auspices of the government only after "maturation". The Ottoman Empire collapsed in World War I, English and Australian forces took over Palestine, and by a decision of the League of Nations the land became a British mandate. The Zionist project met with resistance, Jews and Arabs clashed violently and recurrently, and when the British government limited immigration Jewish under- ground groups turned to illegal shipping of refugees and to terrorism. The country was not easy to rule and after World War II "the Palestine question" was handed over to the United Nation. In 1947, the UN General Assembly resolved on a partition of the land into two states, Arab and Jewish. The Arabs did not accept the verdict. The State of Israel was established in May 1948 and it was immediately invaded by armies from several Arab countries. The war ended in a ceasefire with Israel stretching over an area considerably larger than envisaged in the UN resolution. Another consequence of the war was that 600,000 Palestinian Arabs left their homes and their fields and moved as refugees to areas outside Israel. The young country opened its gates to Jewish immigrants and 700,000 came within the first four years, doubling the number of Jews in Israel. Others followed thereafter. After a period of shortages and difficulties, Israel’s economy enjoyed growth and development for close to two decades, but by the early 1970s prices started rising and inflation escalated. It was halted in 1985 with drastic policy measures that hurt a great number of economic entities. Most agricultural cooperatives experienced a traumatic financial crisis. For many of them, the period since the crisis has been a period of reconstruction. 2 BEGINNING AND LONG RUN TRENDS The history of the kibbutz started with the economic failure of a farm administra- ted by a public organization on the shores of the Sea of Galilee. Part of the land of the failed farm was given over, at their request, to the farm’s workers. They were replaced a year later, in 1910, by a group of twelve youngsters who esta- blished on that spot the first kibbutz (Degania). Agricultural cooperatives in Israel: Past and present 283 Although the twelve saw themselves as socialists and called their group a com- mune, the kibbutz was born spontaneously: the opportunity arose and they grasped it. Theory and ideology evolved later; but the idea of communes of pioneers settling the Land of Palestine caught the imagination of many through- out the Jewish diaspora, particularly in Eastern Europe, and by the end of World War I there were already several hundred youngsters preparing for farm life in Palestine. In due course, they came over, joined existing kibbutzim, or established communes of their own. Reflecting political affiliations, kibbutzim were divided into federations ("move- ments" in Hebrew). The three major federations spanned the spectrum from Labor (center) to the left, a smaller federation was orthodox religious, and one single kibbutz was associated with the Communist Party. For decades the fede- rations were headed by charismatic leaders articulating ideology and political guidance. The first moshav (Nahalal) was established in 1921 and it was preconceived, modeled after the American family farm supplemented with cooperative ideas. The farms in the moshav were to be worked by family members, with mutual assistance if needed. Marketing and purchase of farm inputs was to be done cooperatively. The second-order marketing cooperative, Tnuva, was founded in 1926. It served all agricultural cooperatives, started by marketing general farm products but later specialized in dairy, poultry and beef. Today it is by far the largest dairy in the country. Gradually, the kibbutzim and the moshavim grew in numbers and spread over the country. In 1947, just before the establishment of the State of Israel, there were 127 kibbutzim and 87 moshavim (Table 1). Today their numbers have reached 266 and 443 respectively, including 35 collective moshavim, with a population of 145,000 in the kibbutzim (nearly half of them adult members) and 290,000 in the moshavim. Table 1: Cooperative communities in agriculture 1947 2011 Kibbutzim 127 266 Moshavim 87 443 Most of the growth came after 1948 when many more moshavim than kibbutzim were set up. This was a period of "absorption" of large numbers of immigrants, the country had land, water projects were being developed, but the kibbutzim had lost in the holocaust their main source of pioneers – young people from 284 Yoav Kislev the Jewish diaspora in Europe. The immigrants that did come were not ready for or willing to pursue kibbutz life. Family farming was deemed more appropriate. Agriculture in the kibbutzim is diversified – field crops, horticulture, and live- stock. Most family units in the moshavim also started as diversified farms; but with time, many specialized in single lines: dairy, orchards, or flowers. Both kib- butzim and moshavim started off very poor and were assisted by national funds, but the kibbutzim had gained better access to the capital markets and accor- dingly they tended to rely on mechanical activities and left labor-intensive lines such as vegetables and flowers, particularly in greenhouses, to moshavim and private farming. Agriculture in Israel, as in many other countries, experienced significant techno- logical improvement and increased capital intensification. As the State economy grew, the opportunity cost of farm operators in agriculture increased; the cost of hired labor was reduced, especially when, after 1992, laborers from Thailand came to work on Israeli farms. Consequently total labor input in agriculture was reduced over time but the share of hired labor increased markedly. Many mem- bers in moshavim ceased farming, or their children did not continue in agricultu- re, while the remaining operators increased the scale of their farms. In kibbutzim the changes were manifested in a reduction of the number of members working in agriculture and expansion of manufacturing and services. Kibbutzim and moshavim were also organized in second-order cooperatives, mostly regional associations that, among other activities, took care of water pro- vision, transportation, or cultivation of remote fields. The "purchasing organiza- tions" occupied a special place. The function of these associations was to act for moshavim or kibbutzim in the markets, particularly buying farm inputs on wholesale terms for distribution to their members. As agriculture developed the purchasing organizations established regional enterprises providing auxiliary services such as feed mills, slaughter houses, fruit packaging facilities, and cold storage. Being intermediaries in products and commodities, the purchasing organizations were naturally drawn into credit intermediation; later, the growth of this activity placed them at the heart of the financial crisis of the 1980s. Other second-order cooperatives operated on the national scale; among them, Tnuva, the independent marketing enterprise, several credit funds, and purcha- sing cooperatives that extended across regional boundaries. These cross-regional associations were usually run by a kibbutz or moshav federation to serve the needs of its members. Agricultural cooperatives in Israel: Past and present 285 3 KIBBUTZ, PRE-CRISIS: SOCIETY The kibbutz was a commune.2 In principle, members had no private property; they received their perquisites in kind. The dwellings were modest – a small single room per couple, food was served in a collective dining room, clothes were washed in the kibbutz laundry and, in ideologically strict kibbutzim, mem- bers did not have private clothes: they wore whatever they received each week from the general pool. Again, in principle, members were assigned to work wherever the need arose, some in the kibbutz and others outside – if remunerative employment could be found. With time, as the kibbutz economy developed and diversified, members began to specialize in lines of production or services. Kibbutzim were generally ready to send their members to be trained, formally and informally, and to ad- vance their technical qualifications. The structure of the kibbutz was here of great help, a member leaving temporarily could in most cases be replaced on his job. A family farm operator, in the moshav or elsewhere, could not so easily go away for a significant length of time. Education occupied a special place in the life of the kibbutz. Children lived in communal homes and spent but two or three hours a day with their parents (only in a handful of kibbutzim did children stay with their parents overnight). Investment in schooling was intensive and the kibbutzim built teaching pro- grams aimed at preparing their youngsters for life in the community. This regime was the subject of numerous academic studies and popular descriptions. Most often it was judged favorably; a book by an eminent Chicago psychologist was titled The Children of the Dream and an author who grew up on a kibbutz named her memoir We Were the Future. Although a twelve year curriculum was the common standard, for many years schools in kibbutzim did not prepare for academic studies and higher education was allowed to a limited number of members and restricted to subject areas the kibbutz deemed necessary for its development. This attitude changed as the economic standing improved in most kibbutzim in the 1960s and the 1970s, and college or university training of their own choice became the right of all kibbutz youngsters. Other standards were also relaxed: dwellings expanded and mem- bers received budgetary allowances to spend at their choice on clothes, furniture, and other needs. Economic conditions improved in most kibbutzim but, due to natural and social circumstances – or sheer luck – some kibbutzim did significantly better than 2 The discussion of early periods is conducted in past tense; tenses will shift when more recent events are reached. 286 Yoav Kislev others. These differences did not much affect standards of living; guided by instructions coming down annually from the federations, the kibbutzim main- tainned similar consumption and welfare outlays. As a result, kibbutzim in better shape accumulated equity and those that stayed behind gathered debts. There were no direct monetary transfers between kibbutzim. A landmark change was the shift to family lodging: children moved to live with their parents. The change was gradual, in one kibbutz after the other, mostly in the late 1970s and early 1980s. It was a renunciation of a fundamental principle in the established educational doctrine and was adopted only after long and sometimes heated deliberations. In more than a few kibbutzim, these were the young mothers, women who themselves had grown up on the kibbutz, who did not want their children to relive their own experience of heavy peer pressure, strict discipline, distant parents, and, particularly, lack of privacy and the insecu- rity of being left alone at night. The shift to family lodging marked and was one expression of a profound though informal structural change: the rise of the traditional family as a fundamental social unit in the kibbutz. The first-comers saw their groups as comprising collec- tive families of brothers and sisters; but when the second generation, and then the third, appeared on the scene, families naturally clustered into "clans" that often acted in unison on internal kibbutz issues. These changes highlighted the gender question. It was said that the educational regime originated with the first mother handing over her newborn child to another woman to care for while she returned to milk the cows. This precedence notwithstanding and despite recurring expressions of the desire of women in kibbutzim to share with men responsibilities and tasks, most were assigned as a matter of course to education and service functions. The rise of the family and the return of the children only intensified differentiation. The implementation of the shift to family dwelling required heavy investment, but the common feeling was that money was not a problem as unrestricted credit was easily available. In fact, however, for many kibbutzim this project was one of the last major capital outlays before the outbreak of the severe financial crisis of mid-1980s. 4 KIBBUTZ, PRE-CRISIS: MISSION AND ASSESSMENTS Viewed from the present perspective, the kibbutz mission, summarized in slogan style below, seems obvious and natural. Historically it emerged from the expe- rience of the founders and not least from soul searching and deliberations, often long into the night, of young people struggling with the weight of tradition, sudden freedom, separation from families, backbreaking work, and intoxicating philosophies: Agricultural cooperatives in Israel: Past and present 287 Nation building; Revival of Jewish agriculture; Back to manual labor; Creation of a just society; Promotion of socialism. This was a difficult undertaking. By some estimates 80 percent of those joining kibbutzim left after a short or longer stay. 4.1 Nation building When Palestine was under the British mandate, kibbutzim were settled wherever land could be purchased and in this way they often defined the borders of the state to be. Indeed, the partition map of the 1947 UN Resolution included most of the kibbutzim within the area of the future Jewish state, and some that had been left out by the resolution were later incorporated within the borders of the State after the 1948 war. The kibbutzim were also instrumental in the development of a military capabi- lity, should an armed conflict erupt. Groups of young people stayed on kibbutzim combining work and (clandestine) training; they were ready to go the moment the war broke out. The creation of the State inaugurated a new era although at the time many in the kibbutzim did not recognize (or refused to recognize) how profound a change this was. As indicated above, quite a few kibbutzim were established right after 1948 – many on the borders where their presence was vital. Still, their mission as nation builders lost its urgency and pressure. The roots of the long- run process of erosion of the kibbutzim’s position at the forefront of the Israeli society were planted when it was realized that the new immigrants would not join them in masses. 4.2 Socialism Being communes, the kibbutzim accomplished the mission of socialism to its fullest; but only internally. Outside it was mainly slogans and rhetoric; the call "To Zionism, to Socialism, to the Brotherhood of Nations" was printed daily on the front page of the newspaper of a kibbutz party. But, except for the very early days, members of kibbutzim did not belong to the proletariat; they may have marched on May Day, but they did not share the experience of the country’s working class. The gap was manifested with the issue of hired labor. In the early days of the State, when food was in short supply and employment was hard to find, the kibbutzim were asked to expand production, particularly of vegetables, with the help of hired hands. Most refused; by ideology they could not exploit the labor of others. One group of kibbutzim went part of the way: 288 Yoav Kislev officially farm production was done by an outside company created specifically for this function. But the façade did not fool anybody; the work was on kibbutz land and under the supervision of its members who returned at the end of the day to what could be seen from the outside as an oasis of greenery and social care. The practice of hired labor was later tolerated and spread to all kibbutzim, but the rift between the kibbutzim and the new immigrants – who constituted most of the working class of the country – did not abate and the gap, opened generations ago, is still wide. Whatever the practice, ideological spirits were often high and controversies heated, sometimes reaching crisis proportions. In one case, seventy members from several kibbutzim left in 1927 for the Soviet Union, the land of true social- lism, and set up a kibbutz on the Crimean Peninsula. Only three or four survived the Soviet regime, World War II, and the murderous German occupation. A deeper crisis that affected the largest kibbutz federation had been develop- ping gradually for almost two decades; the line of demarcation was between left and central-leaning members. The differences reached a crisis proportion in the early 1950s, and once again the attitude to the Soviet Union was the immediate cause. The Eastern Block supported Israel in the UN and in its early steps; as the Cold War intensified, members in kibbutzim were split between Soviet sympathi- zers and the majority who agreed with the Labor Party that Israel should belong to the Western Camp. The breakup occurred mainly on the issue of education, as parents of one side refused to let their children sit in classes taught by teachers holding a different opinion. In several cases kibbutzim split up physically, in others members moved to another kibbutz close to their political affiliation. People who for decades worked and suffered and celebrated together could not continue to live on the same piece of land. It took however only a short time for the Soviet block to change its policy and for sympathy and admiration to turn into disappointment and criticism; other differences also mellowed. With a little patience the crisis could have been avoided. 4.3 Politics Members of kibbutzim were active politically. Again, the collective structure conveniently supported fulltime activists. They were also admired by the society of the young country for their pioneering endeavors. In the 1950s, when they were less than 4 percent of the population, kibbutz members formed a fifth of the representatives in the early Knessets (parliaments) and occupied up to six seats in the cabinet. The power base thus created helped to support policies favoring kibbutzim (and often also moshavim) in agriculture and in other econo- mic spheres, finance in particular. However, as time passed, the glamor and the power faded. Only a single kibbutz member was voted to the Knesset in the Agricultural cooperatives in Israel: Past and present 289 most recent elections of January 2013 and even he does not belong to a party that favors kibbutzim or cooperation. 4.4 Fundamental dilemma The effort to adhere to the principle of self-labor in agriculture, manufacturing, and services – that is, all tasks are performed by members and hired workers are not employed – raises a fundamental dilemma, even if the issue of new immigrants is disregarded. In a competitive market, the return to labor in the kibbutz will be the same as the wages of unskilled, and relatively poor, workers in the economy. But by their background and aspirations, the members of the kibbutzim compared themselves with the urban middle class. So long as the country was poor and the economy egalitarian, adherence to the principle did not pose much of a problem; but development brought differentiation, and the kibbutzim were threatened with the possibility of being left behind their non-farm reference group. The moshavim faced the same threat. The urgency of the dilemma – whether to stick to the principle and risk losing members, or to compromise ideologically – was mitigated in several ways: production quotas were imposed by the government in the late 1950s, raising prices and increasing returns to farmers; state budget subsidies were later added, in particular for live- stock products; and, as we shall see below, credit was also made easy, which sup- ported capital intensification and consumption-oriented outlays. As we shall also see below, ideology was not always strictly adhered to and, when quotas were partially relaxed, subsidies eliminated, and credit restricted – ideology had to give way to economic reality 5 CLASSIC MOSHAV The founders of the early moshavim asserted repeatedly that they were true socialists and would have surely agreed with the stated missions of the kibbutzim. But a seemingly minor qualification in the list of moshav principles – "with family units" – created significant differences between the two forms of organization. Both the kibbutz and the moshav functioned within the legal framework of the Cooperative Societies Ordinance, but the kibbutz, where members could easily join and leave, operated mainly as a normative society following conventions of behavior and social habits – whether they were written or only generally ac- cepted. The moshav, with families more heavily attached to their farms and pro- perties, had to be founded on a stronger formal basis; it was a contractual society, and explicitly so. The original fundamental principles of the moshav were, Farming as the main source of income; Family units; Self-labor; 290 Yoav Kislev Mutual aid; Mandatory cooperation in services; National land. To implement the principles, each moshav adopted by-laws and appropriate institutions with authority to manage the cooperative association and its affairs. The mode of activity of the moshav, internal and external, reflected its nature and was aimed at augmenting its particular structure. Unlike the conventional way in which a cooperative is formed by farmers volun- tarily pooling together some of their tasks – in the moshav, the cooperative preceded the individual units. The association received land from the Jewish Agency and it, in turn, allocated the land to its members. Similarly, the moshav received bulk quantities of water, distributed it to its members for irrigation or household use, and paid the provider. When food supply was judged to have reached surplus levels and production quotas were imposed, they were assign- ned to the moshav for allocation to its members. The association acted as the village municipality, building schools, paving roads, and providing other public services. It also employed auxiliary workers, teachers, accountants, nurses, and other "professionals" who often lived in the moshav but were not active farmers. The costs of these activities were covered by the members, who paid taxes that the moshav imposed on products marketed or inputs purchased through the association. Relying on cooperative spirit and practice, the moshavim created lively com- munities and advanced economically despite natural and political obstacles. However, they also encountered internal difficulties. The constraints imposed by the structure and rules of the moshav were occasionally biting. A farm could not be subdivided; only one son or daughter could continue on the parents’ farm, others had to leave the moshav and, unless the farm was sold, the parents lacked resources to help non-succeeding children. Farm products had to be mar- keted only through the cooperative association even if private dealers offered higher prices. Farmers were prohibited from using hired labor even if their harvest rotted in the fields. Members in need of help often felt that they were at the mercy of the association functionaries. Operators who consistently lagged behind were shown the way out. Such hardships, when encountered by indivi- dual families, could be no less onerous than the collective but more diffused social controls of the kibbutzim. Despite the obvious economic and social advantages of cooperation, members sometimes could not resist the desire or need to act separately: seek outside employment, market privately, allow a neighbor to cultivate their land. Some associations treated violators harshly; others cared less or were too weak to impose discipline. Carried by the logic that minor breaches open the door to Agricultural cooperatives in Israel: Past and present 291 complete abandonment of norms, the moshav federation (there was one general federation and several minor ones) tried since its early days to strengthen adherence to the rules and even attempted to write the ideology of the moshav into state law. The proposed law would augment the power of the federation over individual moshavim and the power of the moshav over its members. The bill was vehemently opposed by many of the young, second-generation mem- bers of moshavim, who agreed with the norms but objected to state enforce- ment. The, Labor-centered government supported the bill but failed in its attempt to get it passed by the Knesset. By that time (1969) the strict normative frame- work of the moshav was already coming unstitched – in particular, hired workers started appearing in large numbers – and no law could arrest the rising econo- mic forces. 6 IMMIGRANTS’ MOSHAVIM In its early years, Israel experienced unemployment, food shortage, and lack of financial resources. Immigrants, arriving in large waves, were housed in tempo- rary camps, only a few worked, and they were fed and cared for by public agen- cies; to alleviate the situation, more than fifty thousand of them were directed to agriculture to set up new moshavim. It seemed so simple: each family received a plot of land, a little house, basic equipment and livestock – with proper advice and instruction they could soon start farming and rely on the institutions of the moshav for support and services. The reality was different. The newcomers had no experience in agriculture or in cooperation, many left and went back to the camps or sought better opportunities in town; those who stayed had to struggle with primitive living conditions and lack of basic necessities. The government (actually the Jewish Agency) helped, but its means were limited. The established veteran moshavim were ambivalent about the new ones. They were proud that the moshav, and not the rival and more popular kibbutz, was winning the day, and scores of their members went to live temporarily with the new immigrants and assist them in their first steps in the unfamiliar environment. But distance was also kept, the newcomers were made to form Immigrants’ Moshavim, they were not invited to join the old, established villages. Following several years of teething pains, the new moshavim grew into farming communities and contributed to the expansion of Israel’s agricultural produc- tion, which quickly reached surplus proportions. Leadership naturally grew in the new moshavim, but although cooperative associations were formally instituted, in most places the ideology of the moshav was not strictly adhered to. Realizing that the veteran regional associations would not accept them for fear of jeopar- dizing financial stability, the immigrants’ moshavim set up, with the help of the government, second-order regional cooperatives of their own. This had been done by the late 1950s and early 1960s. Up to that point, the moshav as a whole 292 Yoav Kislev was responsible for its members: water provision could be cut off when some of the farmers failed to pay. To avoid this eventuality, the new regionals, unlike the established ones, dealt directly with individual farmers. Working at the regional level opened the road to the political arena. With large numbers of potential voters behind them, new immigrants became part of the leadership of the federation of the moshavim and the first active moshav farmer to become, in 1974, a member of the cabinet as the Minister of Agricul- ture was from an immigrants’ moshav (Aharon Uzan). Starting in the late 1960s, agriculture – particularly that in the moshavim – enjoyed a flourishing period that lasted for approximately 15 years: subsidies were expanding, the country’s real rate of exchange more than doubled, local prices of export products were rising, and credit was easily available. Veterans and new farmers responded by expanding into the production of export crops. Cotton, mechanically picked, was grown mostly in kibbutzim, and horticulture – fruits, vegetables and flowers – was the domain of the moshavim. At that time, exports were managed by three specialized agencies and this concentration increased the role of the cooperative associations at both the region and the village level. At the same time, the cultivation of labor-intensive crops had to rely on large numbers of hired hands, compromising the principle of self-labor. 7 COOPERATIVE FINANCE Cooperation in agriculture was originally built on necessity and ideology; however, with time and economic development, financial issues came to the fore. As indicated above, the cooperative in the moshav and the second-order associations of moshavim and kibbutzim provided purchasing, marketing, and other services. They were also involved in finance. The financial activity was encouraged, or even necessitated, by the ownership structure in cooperative agriculture. Land in moshavim and in kibbutzim was, and still is, owned by the state; there is no private landownership. Therefore farms cannot use land as collateral to obtain credit. The lacuna was replaced by so-called mutual guaran- ties: members in a moshav guaranteed the debt incurred by their associations, and moshavim and kibbutzim signed mutual, i.e., reciprocal, documents of guaranty. The cooperatives functioned as the hubs of the system of mutual gua- ranties. In some instances, credit was extended to individual famers or kibbutzim, in which case the village association or the regional cooperative were the instru- ments facilitating the comprehensive guaranties. In other cases, the village and the second-order cooperatives themselves raised credit and distributed it to their members, again under mutual guaranty. Consequently, cooperation in finance was strongly intertwined with the other aspects of farm activities; particularly, to back the guaranties, marketing had to be done solely through the cooperative Agricultural cooperatives in Israel: Past and present 293 channels – the association in the moshav and Tnuva for both the moshavim and the kibbutzim. The banks were freed from the need to examine every individual creditor separately. The guaranties were signed routinely on the insistence of the banks but the routines were empty. I do not know of a single case were a guarantor – kibbutz, moshav, or individual member – had to cover the debt of others. Cooperative credit relied, implicitly and by experience, on government backing. The settlers in the moshavim and the kibbutzim came to Israel penniless; they had to rely on public assistance for tools and first inputs. Indeed, agricultural development was from its inception a national project creating fosterage rela- tions between the farm sector – cooperatives in particular – and the government and other supporting national agencies. Beside agricultural research, extension, and planning, the principal channel of public economic assistance was in the form of credit for investment projects and other economic activities (the govern- ment also assisted manufacturing, especially if intended for export). The expres- sed responsibility for the sector created conditions of soft budget constraints: knowing that the government would come to their rescue should they run into difficulties, cooperatives went easily into debt (and readily signed guaranties). Generally the rescue came as "conversion." When cooperatives encountered difficulties in recycling their short-term debt, it was converted to long-term credit that the debtors were supposed to be able to service conveniently. The relief however was short-lived. In one case, observed in the 1950s, debt conver- sion was implemented in 72 kibbutzim; three years later all but two had incurred larger short-term liabilities than they had previously and were again expecting government rescue. The government recognized the predicament and offered, in the early 1960s, a solution in the form of "concentrated credit": kibbutzim and moshavim that would channel their financial activities through a single bank and agree to supervision by the Ministry of Agriculture were assured stable credit lines to satisfy their (approved) needs. The program operated satisfactorily for several years, but then economic circumstances changed: the supply of commercial credit expanded (after 1974 Israel, like many other countries, experienced an inflow of recycled oil money); trusting the government to assist cooperatives in distress, banks were happily extending short-term loans to purchasing cooperati- ves and national funds who, in turn, redirected the resources to their members – kibbutzim, moshavim, and regional enterprises. Concentrated credit crumbled and its demise heralded the slide of the sector into the financial crisis. 294 Yoav Kislev 8 THE ROAD TO THE CLIFF The financial crisis took time to brew. Although most of the years were a good time for agriculture, the period following the 1973 war and up to 1985 was Israel’s Lost Decade: growth stagnated; government deficits grew; debt, internal and external, accumulated; and prices rose. Inflation accelerated from 20 % per annum at the beginning of the period to 450 % at its end. It was halted abruptly in mid-1985. Inflation was not a smooth process, it moved irregularly; likewise, policies to halt inflation or to accommodate the country to the rising prices fluctuated over a wide range of measures, introducing uncertainty and abrupt changes of economic conditions. In addition, bookkeeping and financial reporting became meaningless; farmers and cooperatives did not really know what their economic situation was. Inflation-adjusted accounting was introduced in 1982, but this was too late in the game. Another associated effect was that money became cheap; available data indicate that in 1979 the real rate of interest on commercial credit was -11 % (negative) and much lower on government approved short- and long-term debt. This was a honey trap. At the beginning of the period the value of debt eroded and the cooperatives accumulated assets. Consequently, the demand for credit expanded and farmers were pressing for additional finan- ce, which they claimed was too slow to come. The bureaucracy agreed with the farmers. An example was a committee of public officials that submitted in 1979 a detailed report on credit in agriculture. The core of the document was the calculation of the "financing gap", the difference – it was significant – between yearly investment in the sector and long-term credit. The government was called to fulfill its duty and close the gap. Whether it was sheer ignorance or favoritism, the members of the committee did not realize that agriculture was willingly expanding its capital outlays, relying on available and convenient short-term credit – trusting that the accumulating debt was the problem of the government: it would never let moshavim and kibbutzim down. Easily available credit at negative real rates encouraged overinvestment in large risky projects. This tendency was extended and exaggerated when economic conditions could not be gauged due to inflation and shortage of funds was closed by recycling short-term debt. Officials in cooperatives were judged mainly by the projects they set up and government offices approved and supported regional enterprises with little if any regard for other available institutions. The tendency was strengthened by the desire of the moshavim to imitate the manu- facturing industries in the kibbutzim. In the mid-1980s, the State Comptroller examined more than 300 regional enterprises and found that most of them did Agricultural cooperatives in Israel: Past and present 295 not fully utilize their existing capacity; they did not even cover operating costs. In one case, an enterprise was built and completed – and then left idle, never to produce anything. When credit was flowing freely, everything looked rosy: machines were new, orchards were young, houses in good state of repair, public areas green. Many operators in moshavim constructed new greenhouses; the kibbutzim increased their investments in manufacturing enterprises. Optimism ruled; even the banks were not exempted. A political upheaval sent a shock wave: the Labor-centered government was ousted in 1977 by a right-wing coalition of parties whose rhetoric was explicitly anti-kibbutz. The change sounded an alarm in the kib- butzim and the two largest federations raised significant amounts of money "to hold for rainy days", a large share of it abroad (an early action of the new govern- ment was to free the foreign currency market). The availability of these funds supported for a while the feeling of easy credit but the atmosphere changed after a short time when large sums of money were lost in the stock exchange and the gray market, foreshadowing the financial crisis to come. The gravity of the situation was realized, at least by some, but too late for amendments. When the crisis erupted, the per capita debt of the kibbutzim was several times larger than that of the moshavim. (The religious kibbutzim were an exception; they had not expanded heavily into manufacturing, had not sought new credit sources, and had invested their surplus money in solid instruments. They rode out the financial turbulence, when it came, relatively easily.) Available credit was channeled partly to consumption; it was estimated in the 1970s that the standard of living in the kibbutzim was on a par with their urban reference groups, thus augmenting social stability and attractiveness. However, easy credit also intensified the inherent inefficiencies in the kibbutzim. Although most members worked diligently, shirking could not be completely avoided and indifference to common costs and returns was also noted (it was reported that electricity bills in the residential area of a kibbutz fell by 40 % once members began to be charged individually). However, much more damaging than perso- nal slackness were inflated services, schools with small numbers of children, expensive public buildings, and lax scrutiny of investments. Personal choice was also limited: "to each according to his needs" meant that others decided, even if democratically. Reforms adopted in the wake of the financial crisis were aimed at least partly to mend these shortcomings. 9 THE FINANCIAL CRISIS AND ITS RESOLUTION In 1985, when inflation reached devastating rates, a left-right coalition govern- ment, trusting it had the support of significant parts of the public, adopted a drastic stabilization program: prices, wages, and nominal exchange rates were 296 Yoav Kislev fixed and credit was constrained. Immediately the inflation dropped from more than 400 percent to 20 percent annually and was further suppressed later; one consequence was that real annual rates of interest jumped to 30-40 percent and even higher. It was realized that the debt of agriculture was so large that it threatened the stability of the national banks; they refused to recycle short term debts and allowed the resulting overdrafts to swell with the exorbitant rates of interest. Regional and national cooperatives collapsed and all their obligations were to be attributed to their members – the kibbutzim, the village associations, and the farmers in the moshavim. Agriculture suffered doubly as the real ex- change rate fell by a third, subsidies were cut, and the sector’s terms of trade worsened. Many, in particular in the moshavim, could not continue farming. The government, the banks, and the cooperatives tried to solve the crisis. Repea- tedly they signed agreements to settle the debts and repeatedly they had to realize that the debts were not settled. Finally, a law was passed in 1992 speci- fying the way repayment ability would be assessed for each and every moshav member. An administration was set up to implement the law; soon a parallel administration started working on the debts of the kibbutzim. The work lasted twenty years and is only now (2013) coming to completion. The moshavim were assigned to repay 10 percent and the kibbutzim 30 percent of their calculated debts (no detailed information is available), the remaining sums were erased; covered by the banks, the government and some private creditors. All mutual guaranties were canceled and all purchasing organizations of the moshavim were closed down. Three quarters of Tnuva, the marketing giant, were sold several years ago to private interests and some kibbutzim and moahsvim used the money received for debt payments. In one aspect – perhaps the only aspect to count–the settlements were a complete success: no one, not a kibbutz, a moshav or an individual farmer had to leave the land in lieu of debt repayment. But far reaching changes have occurred and are occurring. 10 RESTRUCTURING The financial crisis threatened the livelihoods of a large number of kibbutzim and caused many members, particularly young ones, to leave. Kibbutzim in stress reacted by encouraging their members to find employment off the kibbutz, hiring outsiders as managers, charging members for services, and, most revolu- tionary, paying members market-rate salaries. These and other "privatization" measured raised the question of whether communities adopting such practices could still be regarded as kibbutzim. A public-government committee was ap- pointed and, after a lengthy study, a new definition was adopted by law in 2005. The old definition was a single terse sentence indicating collective ownership and leaving the details to the by-laws of the individual kibbutz. The new definition is spread over several pages of regulations. Three types are Agricultural cooperatives in Israel: Past and present 297 recognized. Collective Kibbutz is the old conventional entity. Renewed Kibbutz is a kibbutz where any of the following holds: members receive salaries; the housing is privately owned; the collective means of production are privatized by distributing shares to members – provided that voting rights remain vested in the community at large. The third type is new, Urban Kibbutz; more than a few such small communes have sprung in recent years. The bulk of the regulations deals with the renewed kibbutz and specifies in great detail the minimum requirements of assistance to the elderly or the needy and the way assets may be privatized. With these changes the kibbutz was transfor- med from a normative society to a legal-based entity; members and outsiders, among them tax authorities, know the rules of the game. Disputes have already been brought before courts. The renewed kibbutz differs from an ordinary rural community in two ways. First, it maintains an internal "safety net" of mutual responsibility for basic neces- sities and vital services to all members, financed by returns from the collective enterprises and levies imposed on the wage earners; and second, it may demo- cratically change its status. The kibbutz was never perfectly egalitarian – elites emerged and may have enjoyed preferred amenities (free use of cars was a notorious eyesore), but much greater differences can now be found in the rene- wed communities and complaints of deprivation have been heard (opinion surveys are conducted regularly but quantitative economic data are not avai- lable). Particularly controversial is the situation of the pensioners; many claim that the monetary allotment assigned to them by the new regulations is too small, especially compared to the salaries younger members receive in the kib- butz or outside. This controversy may yet tear some kibbutzim apart. Still, surveys indicate that the majority of the members are satisfied with the reforms and agreed that they improved performance and welfare. Less than a third of the kibbutzim have not changed to renewed and stayed collective, although most have gone part of the way, for example, employing hired labor in manufacturing or paying members for weekend work. In general the collective kibbutzim are well-to-do. The majority of their members naturally accept the traditional ideology, but they also enjoy certain advantages; among them, secured, comfortable standard of living; work in a familiar environment; intensive social interaction; they are not evaluated and do not have to evaluate their neighbors and friends in pecuniary terms. Kibbutzim that could afford it stayed collective despite the inherent inefficiencies. The commune that origina- ted as a means to overcome poverty, shortage, and hardships has become a luxury good. In the kibbutz, particularly in the renewed kibbutz, the crisis and the changes it caused affected drastically the way of life of the individual members, but the 298 Yoav Kislev economic activities of the community – in agriculture, manufacturing, and com- mercial services – remained collective. In the moshav the most affected level was the cooperative. All regional purchasing organizations of moshavim disappeared and the associated economic enterprises, feed mills, slaughter houses, cold sto- rage, and the like were sold or closed down. Officially, the village associations remained intact but they lost many of their functions and much of their power; in particular, they no longer provide inputs, marketing services, and – most importantly – financial intermediation. Also, they no longer extend aid to house- holds in difficulties. Farm operators in the moshavim are now served directly by private agencies and in quite a few places by small local cooperatives or partner- ships that stepped in to fill the void. As indicated, self-employed agricultural labor has decreased over time. Most moshavim are now rural communities with a relatively small number of farmers who have increased the scale of their opera- tions by expanding livestock enterprises or renting land from their neighbors (purchasing is impossible); field and barn work is usually done with the help of hired hands. 11 MANUFACTURING Manufacturing in the kibbutzim started with small workshops, a natural expan- sion in several places was into food canning as an extension of agricultural pro- duction; later it branched into other lines and spread effectively to all the kibbutzim (some established commercial services such as hotels or shopping centers). Manufacturing was deemed to have several beneficial effects: it could use labor freed from agriculture, particularly the elderly, it opened opportunities for professional development of young members, and it added to the portfolio of economic activities of the community. These expectations were, at least partly, realized but problems were also encountered. Manufacturing on a significant scale was established when credit was available; infusion of capital increased demand for labor and hired workers were called in – some to perform manual and repetitive work and others as experts. For many kibbutzim this was the first breach of the principle of self-labor. Other problems were social; with manufac- turing came a regime of professional delineation and hierarchy, alien to the nature of the kibbutz community. Manufacturing also poses risks: a kibbutz will generally operate only one or two industrial lines and it would be difficult to find in its small community people who can run and manage a complex production process in a competitive envi- ronment. Indeed, in the nature of risk, the manufacturing establishment of some kibbutzim succeeded, grew, and even achieved global reach, while many others failed and had to close down. Today, a third of the kibbutzim do not have any manufacturing activities, but for the sector as a whole, the value of output in manufacturing is much higher than the value of the product of agriculture. Agricultural cooperatives in Israel: Past and present 299 12 REGIONALS Regional, second-order cooperatives were first established by kibbutzim in the 1940s and early 1950s as purchasing organizations intended to reduce transact- tion costs and augment market power in the provision of production inputs; they were followed by similar organizations set up by veteran moshavim and later also by new immigrants’ moshavim. As indicated, the regionals were drawn into financial intermediation; this was true for both sectors, but the differences between them turned out to be crucial. In the moshavim, most of the credit came through the regionals; it was transfer- red to village associations and thence to individual members. When the financial crisis hit, everyone was responsible and everyone claimed to rely on the all- embracing web of mutual guaranties. There was no way but to acknowledge, as the law finally did, that these debts could not be collected and the regionals had to be liquidated with all their enterprises. The situation in the kibbutzim was different. Although the regionals in the kib- butz sector also engaged in financial intermediation, this was on a relatively small scale; individual kibbutzim had direct access to banks and to other major sources represented by federation-level funds. These funds were closed down after the crisis but the sector’s regionals survived, admittedly due to debt settle- ments and sufficient write-offs. A fortunate period followed. In the 1990s Israel absorbed large numbers of immigrants from the former Soviet Union and the economy expanded. Pertinent to our story, meat consumption (beef and parti- cularly poultry) doubled and the volume of operations of the regionals, the providers of feed and owners of slaughter and processing facilities, grew to un- precedented magnitudes. Growth in monetary terms was further augmented when world grain and oil seed prices rose markedly. The regionals accumula- ted surpluses which they used to assist distressed kibbutzim and to expand their own operations. Today the regionals are still organized as cooperatives, but in fact they are holding companies in partnerships with kibbutzim and private interests. One regional holds for its member-kibbutzim a 20 % share in Tnuva. 13 RESOURCE REALLOCATION AND IDEOLOGY Net income in agriculture grew significantly since the 1990s and it rose particu- larly in the first decade of the 2000s. Considering that the sector’s terms of trade were consistently deteriorating, improved income was especially remarkable; it was a realization of augmented efficiency. Since the crisis, its resolution, and reconstruction, agriculture has experienced a period of resource reallocation. Kibbutzim separated their economic activities from the affairs of the communities and hired experts to manage each segment. In renewed kibbutzim members or 300 Yoav Kislev outsiders are assigned to work only where their contribution exceeds the salary they receive, others work off the kibbutz; many kibbutzim went into partnerships with other kibbutzim in field crops and dairy livestock (partly to comply with stringent environmental regulations); the country’s broiler production has been concentrated in the kibbutzim where economies of scale are realized; several regionals manage the cultivation of orchards and marketing of fruits jointly with individual kibbutzim; and one regional created a poultry integration styled on the American model. Parallel changes occurred in the moshavim. Farm sizes are now larger than they used to be and in more than a few places two or three operators share land or livestock enterprises. Hired workers are employed wherever they may contribute. Services and inputs are purchased on the market and farmers are no longer tied to inefficient village or sector providers. Although prosperity is not shared by all, general farm income has risen. Manufacturing also witnessed significant structural changes. Several kibbutzim created partnerships and others sold parts of their activities to private interests. Such moves have brought in business expertise and capital that can be used to cover debt or to fund retirement plans. Another benefit is diversification – not to hang the future of the kibbutz too heavily on a single source of income, even if at present it is laying golden eggs. The economic landscape in the kibbutzim and the moshavim has changed mar- kedly in the last two or three decades. Shopping and service centers are seen in the countryside, many residents – members and others – commute regularly to work away from their homes, but small, local enterprises are also flourishing. Most conspicuous is rural tourism, inns in the kibbutzim and B&B in the mosha- vim; but a variety of other establishments can also be found in both sectors, hairdressing, tour-guiding, boutique bakeries, computing, and what not. These changes run counter to the wording and spirit of the original ideology of cooperation – whether in the communal kibbutz or of the family variety in the classic moshav. Naturally, ideology was set aside when the traumatic crisis hit and economic rescue became paramount, but the dilution of the ideological basis started earlier. The members of moshavim and kibbutzim were not hermits sanctifying scarcity and seclusion. They were part and parcel of the Israeli society and when the country’s living standards improved, while terms of trade of agri- culture deteriorated, the "fundamental dilemma" raised its head and the coope- ratives could not stay behind; they had to give up, gradually and even painfully, their basic principles. They accepted hired labor – first new immigrants, then in manufacturing and in greenhouses – and they agreed to family dwellings and private partnerships. For youngsters born in kibbutzim and moshavim the Agricultural cooperatives in Israel: Past and present 301 shedding of ideology was a lot easier than for their parents. The financial crisis and its aftermath sealed the departure from orthodox ideology. The reforms and restructuring do not assure the future of the kibbutzim. In parti- cular, where the collective enterprises do not generate enough income to sup- port the elderly and other common services, the kibbutz may find it necessary to impose relatively high community taxes on its wage-earning members; but this could be a vicious circle – the best and the brightest may leave. The mere existence of the locality will be further endangered where, due to legal difficul- ties, the kibbutz is prevented from accepting outsiders as residents. 15 RURAL COMMUNITIES AND THEIR LAND Kibbutzim and moshavim are small communities; their populations have aged not only with time, but also with the tendency of younger people to leave. One solution was "expansion", the development of residential areas for young families of non-members, including the younger generation who chose to live in their birthplace without becoming full-fledged members of the kibbutz or the moshav. Formally, the locality was then transformed into a municipality where some of the residents hold membership in a separate association. In addition, kibbutzim and moshavim rented out land to commercial enterprises. These real estate activities created income; again, it was used where needed to assist in debt repayment. They also sparked opposition: the land was given for farming; addi- tional income belonged to the public at large and not to a small, select minority. The opposition was led by a group of youngsters representing immigrants of the early days of the State of Israel – mostly of "eastern" origin: from the Middle East and North Africa. When the case was brought before the Supreme Court, it ruled in 2002 in favor of the opposition. The popular utterance was against the kibbutzim, but the moshavim were also severely affected. Expansion and development programs were set back and land issues, including the privatiza- tion of housing in the kibbutzim, are now in flux. In the meantime, land prices in urban areas of Israel have risen sharply and ownership of a house or an apart- ment is now unaffordable to many budding families. But this issue and its pos- sible connection to farm land policy is beyond the scope of this chapter. 16 EPILOG It is impossible to imagine today the difficulties the first-comers had to face; the shortages, sometimes starvation, diseases, hard work, and uncertain future. The new immigrants in the early days of the State did not have it easier. They were hauled by trucks without their consent to isolated places in the desert or on the hills and had to start from nothing. The achievements of these people are laudable; today the country’s agricultural output is twenty times its 1950 302 Yoav Kislev level and the lion’s share of the credit is due to cooperative farms; kibbutzim further branched into manufacturing. But cooperation, close to the way it was originally envisaged, is practiced only in a relatively small number of collective kibbutzim. In other places, cooperation is followed only where it can be justified by cold economic calculation. The founders of the kibbutzim and the moshavim hoped to spread socialism and equality to all corners of the land, but history may judge that farm coopera- tion in Israel functioned as an instrument to achieve other, associated goals more than as an end to itself. However, as an instrument it was highly successful; Jews returned to the land and the country is covered by agricultural communi- ties, most of them cooperative (some in areas occupied in 1967, but they are a different story). Ideology may have been abandoned, but the ideological candle has not been snuffed out completely – new kibbutzim are still being set up, some with ecological inclinations and many others in urban centers. Young people keep trying and will probably continue trying the experience of to- getherness and public service. SELECTED SOURCES Listed here are some of the sources I consulted when writing the review. A complete list of references in Hebrew and English is available on request. ABRAMITZKY, R. (2011): Lessons from the kibbutz on the equality-incentive trade-off, Journal of Economic Perspectives, 25(1): 185-207. APPLEBAUM, L., SOFER, M. (2012): The moshav in Israel: Agricultural communities in a process of change – A current view, Horizons in Geography, 79-80: 194-209. BARKAI, H. (1977): Growth patterns of the kibbutz economy, North-Holland. BEN RAFAEL, E. (2011): The kibbutz: survival at risk, Israel Studies, 16: 81-108. BETTELHEIM, B. (1969): The children of the dream, Macmillan. KIMHI, A. (2009): Heterogeneity, specialization and social cohesion in Israeli moshav coope- ratives, Journal of Rural Cooperation, 37(1): 124-136. LIEBLICH, A. (1982): Kibbutz Makom: Report from an Israeli kibbutz, Andre Deutsch. NEAR, H. (1992, 1997): The kibbutz movement: A history, Littman Library of Jewish Civilization. NEEMAN, Y. (2011): We were the future, Ahuzat Bait [in Hebrew]. ROSENTHAL, G. (2014): Agricultural cooperatives in Israel, Journal of Rural Cooperation, 42(1). RUSSELL, R., HANNEMAN, R., GETZ, SH. (2013): The renewal of the kibbutz: From reform to trans- formation, Rutgers. WEINTRAUB, D., LISSAK, M., ATZMON, Y. (1969): Moshava, kibbutz, and moshav, Cornell. ZUSMAN, P. (1988): Individual behavior and social choice in a cooperative settlement, Magnes.