The Limits to Land Reform: the Land Acts in Ireland, 1870-1909 Author(S): Timothy W

The Limits to Land Reform: The Land Acts in Ireland, 1870-1909 Author(s): Timothy W. Guinnane and Ronald I. Miller Source: Economic Development and Cultural Change, Vol. 45, No. 3 (Apr., 1997), pp. 591-612 Published by: The University of Chicago Press Stable URL: http://www.jstor.org/stable/1154833 Accessed: 14/10/2010 15:32

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=ucpress.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to Economic Development and Cultural Change.

http://www.jstor.org The Limits to Land Reform: The Land Acts in Ireland,1870-1909"

TimothyW. Guinnane Yale University RonaldI. Miller ColumbiaUniversity

Irelandexperienced two greatwaves of land reformduring the late nineteenth and early twentiethcenturies. Land reform was a centraldemand of Irishpolitical factions of the time andby accedingto this demandsuc- cessive Britishgovernments hoped to pacify Ireland.This articlerecon- siders the economic significanceof the several Irish land-reformmea- sures by examining them in the context of economic theory and by comparingthem to other land-reformepisodes in currentlydeveloping countriesand in Europeanhistory. The Irishreforms contained little that could betterthe allocationof resourcesand so had little impacton economic efficiency, even thoughthe end result was the creationof a class of peasantproprietors operating in a free market.The analysis focuses on the structureof the laws and their relation to Irish institutions,although we also discuss the aggregateevidence on the evolution of agriculturalefficiency following the reforms.Moreover, the Irish legislation missed an importantopportunity to improvesmallholders' access to credit.Some facets of the land legislation,interacting with some unusual institutionalfeatures of land tenure in Ireland,made credit conditions worse. Land reform in Irelandwas much more a wealth-redistribution programfinanced by Britain than a serious effort to improve the efficiency of agriculture. Landtenure and landreform occupy centralroles in both the history and historiographyof the political strugglesthat culminatedin the partition of Irelandin the 1920s. Successive waves of Irishpoliticians sought to harnessessentially nationalist ambitions to the moremundane dissatis- factionsof the Irishfarmer, convincing the latterthat economicprosper- ity requiredalterations to or complete abolition of an agrariansystem

? 1997 by The Universityof Chicago.All rightsreserved. 0013-0079/97/4503-0001$01.00 592 EconomicDevelopment and CulturalChange

wherebytenants paid rents to landlords.While Ireland'shistorians have closely analyzed the political dynamics of agrariandiscontent before, during,and afterthe LandWar of 1879-82, scholarshave paid little at- tentionto the economiccontent of the LandActs per se. The exceptions, such as B. Solow's work, have devoted their efforts to the first wave of those reforms.1By analyzingthe actualimplications of the Irishland reform more generally,in a broadertheoretical and comparativelight, we supplementthe politicalhistorians' conclusions that the Land Acts were intended less as economic policy than as efforts to compromisewith Irish political demands. Recentstudies of land tenurein developingcountries provide a useful context for the Irish case. Comparisonof nineteenth-centuryIreland to developingcountries today is neithernovel nor far-fetched.R. D. Col- lison Black noted in a neglected paperthat economic policy in Ireland duringthe period of Westminsterrule closely resembled,at least in in- tent, much developmentpolicy today in that it was an effort to secure peace and quiet throughimproving economic conditionsfor the popu- lace.2Alterations to propertyrights in land are a centralfeature of developmentpolicy today. TimothyBesley providesa useful classificationof the implicationsof propertyrights in land underthe three headingsthat follow.3 (1) Security:occupiers who lack securityof tenurein theirholdings will not expend effort in investments,the fruits of which can be seized by another.An insecure tenantmay not take any interestin the long-termproductivity of a farm if he fears being ejected withoutcom- pensationafter the currentseason. (2) Land as collateral:pledging land as collateralcan dramaticallyreduce the costs of loans, since mortgage lendersare less likely to demandexpensive risk premiumsor to engage in costly information-gatheringactivities. Yet such mortgageloans pre- suppose a lender's ability to seize the land in the case of default. (3) Commoditization:some forms of tenureimpede the buying and selling of land and so result in a suboptimaldistribution of land. The more nearlyland is like any othercommodity, the more likely it is to be allo- cated to its highest-valueuser. As will be discussedbelow, the Irishland reformsmay have had some positiveeffect on securityin the early stages of reformbut they creatednew problemsfor both collateraland commoditization. In Section I we review land tenureand the LandActs in Ireland.In Section II we analyze the economic effects of these reforms as antici- pated by simple economic theory,and in Section III we discuss the implicationsof land tenureand its reformsfor ruralcredit.

I. Land Tenure and Its Reform Priorto the Land Acts a majorityof all Irish farmerswere tenants,and most of them were eitheryearly tenants or tenantsat will, lackingformal TimothyW. Guinnaneand RonaldI. Miller 593

TABLE 1

TENANCIES AT WILL AND THE SIZE DISTRIBUTION OF HOLDINGS

THE PROVINCESOF IRELAND

VALUATIONCATEGORY Ulster Munster Leinster Connaught IRELAND Tenanciesat will as a per- centage of all holdings in a given valuation category: ?100 31.86 21.79 22.31 28.84 24.40 Holdingsof each size as a percentageof total holdingsfor all sizes: ?100 .81 2.27 4.26 .94 1.86

SOURCE.-Houseof Commons, "Returns Showing the Number of Agricultural Holdingsin Irelandand the Tenancyby Which They Are Held by the Occupier,"c. 32 (1870), tables 1-6. NOTE.-J. Mokyr (WhyIreland Starved:A Quantitativeand AnalyticalHistory of the Irish Economy,1800-1850 [London:Allen & Unwin, 1985]) shows thattenancies at will had become more common afterthe famine, so that the distributionhere may have been ratherdifferent in earlierperiods. leases. Table 1 summarizesthe results of the first national statistical study of tenancy from 1870. Tenancyat will was most common in the northernand westernprovinces of Ulsterand Connaughtand on holdings valuedat less than?15, but tenancyat will and yearlytenancy accounted for the majorityof farmerseverywhere in Ireland.4Table 1 also suggests an importantpoint to which we returnlater: a substantialmajority of the more extensive farmsin Irelandwere not tenanciesat will. A tenant at will had no legal guaranteeof continuingoccupation; his landlordcould, in theory,eject him any year. This supposedinsecu- rity lies at the heartof what J. Mokyrcalls the "LandTenure Hypothe- sis" of Irishpoverty. Without leases landlordscould always raise rents or eject a sitting tenant, giving tenants no incentive to invest in their holdings, as the landlordcould appropriatethe value of any investment the tenant might make.5Nor did tenantshave any incentive to protect theirlandlord's investments in the holding.Thus, the argumentgoes, neither partymade the long-terminvestments necessary to increaseagricul- turalproductivity. As RaymondCrotty notes, this alleged inabilityto re- cover the value of improvementsformed a central feature of the argumentfor tenantright, discussed below.6 The idea of insecuretenancy 594 Economic Development and Cultural Change informs much of the older historical research on Ireland. For example, K. H. Connell's account of demographic patterns prior to the Famine is based on the notion that peasants wanted to invest in people rather than land. Any evidence of increased prosperity, Connell claims, would be immediately soaked up by a landlord (or middleman) in the form of a rent increase. This behavior, if it actually occurred, seems unlikely to have been in the long-term interest of landlords, who would have bene- fited from a more cooperative solution. Nonetheless, according to Con- nell, "elastic rent" deterred even effort devoted to the current year's output; "If a man worked harder he was more likely to enrich his landlord than himself."7 Tenancies at will notwithstanding, Irish farmers occupied their holdings for surprisingly long times. Long occupation without a lease often reflected the landlord's recognition of tenant right, or the "Ulster Custom." This practice has vexed Ireland's historians for generations, and there is no consensus on how or why it emerged. Although most common in Ulster, the institution was widespread throughout Ireland. Three central features of the Ulster Custom, sometimes referred to as the "three F's," were "fixity of tenure" or the right to remain on a holding so long as rent was paid; the right to pay a "fair" rent; and the right to "free sale" of the tenant's interest or tenant right when a tenancy changed hands. Fixity of tenure amounted to an informal, perpetual lease. Fair rent meant to Irish peasants a rent less than the "rack" rent: the rent at which a holding would let should the landlord solicit bids on an open market. Solow and W. E. Vaughan interpret the rack rent as the Ricardian rent.8 With rent below the Ricardian rent, the sitting tenant could find incoming tenants willing to pay a lump sum for the right to rent the land at the "fair rent"; this sum, or the practice of paying it, constituted much of tenant right.9 As contemporaries noted, tenant right implied an awkward coproprietorship in land. For tenant right to have any meaning, the landlord had to concede certain rights-such as setting rents at competitive levels-normally associated with the ownership of land.

Reform of Tenancy Parliament regulated tenancy in several major steps. The Land Act of 1870 required compensation for improvements to all outgoing tenants as well as compensation for disturbance to any tenants who were ejected. In a more confusing set of clauses, the Act of 1870 also legalized the Ulster Custom where it had existed before. Legalization of the Ulster Custom had ambiguous implications, since the act never defined the Ul- ster Custom nor did the act prevent the rent increases that would reduce the value of the tenant's interest. In 1881 Parliament altered the basis of landlord-tenantrelations more fundamentally. The 1881 act extended the three F's to all of Ireland and established a system to determine and en- TimothyW. Guinnaneand RonaldI. Miller 595

TABLE 2

SUMMARY OF THE IMPACT OF THE LAND LAW ACTS ON RENT IN IRELAND

JUDICIAL TERM

MEASURE FirstTerm Second Term ThirdTerm

Numberof holdingswhere rent was fixed 382,813 144,009 5,887 Acreageof holdingswhere rent was fixed 11,385,375 4,434,640 191,105 Averagepercentage reduction in rent 20.7 19.3 9.2 SOURCE.-Houseof Commons, "Reportof the Irish Land Commissioners,"Ses- sional Papers 1920, vol. 20, app. tables 52-54. NoTE.-Table covers all judicial rents throughMarch 31, 1919. Rent reductions measuredagainst previous term; thus third-termreductions are decrementsfrom the second-termrent. force fair rents.Land courts had the authorityto fix judicialrents for 15- year terms. Rent regulation in effect guaranteedthe value of tenant right.10The 1881 Fair Rent Act gave a statutorybasis to what amounted to a coproprietorshipin Irish agriculturalland: landlords had the rightto drawan income from land, but tenantshad limitedproperty rights in the same land. Table 2 shows the impact of the rent regulationon tenancyin Ire- land up through1919. The 1911 census of Irelandestimates the area of agriculturalland in Irelandat nearly 19 million acres;by this estimate, about60% of the agriculturalland of Irelandwas let at a rate fixed by a firstjudicial tenancy.First and secondjudicial termsimplied rent reductions on the order of 20%, while the comparativelyrare thirdjudicial terms implied much smallerreductions. For a farmerpaying ?20 annu- ally in rent, a rent reductionof 20% implied an increasein tenantright of some ?130." The true impactof rent reductionon tenant-rightvalues was at least somewhatless; with price declines in the 1880s and 1890s, Ricardianrents also fell.

TenantPurchase Tenantownership of the land had been a demandof some Irish parties from the start,although only a few leaders(such as MichaelDavitt) ad- vocated outright nationalization.Several late nineteenth-centuryland acts concernedprimarily with the regulationof tenancyhad also included provisionsaimed at encouraginglandlords to sell tenantstheir holdings. These early provisions had relatively modest effects, as the terms on which land was to be sold were not terriblyattractive to landlordor to tenant. Organized somewhat differently was the Congested Districts Board, which had been establishedin 1891 with the goal of creating holdings sufficientto supportfamilies in the poorest areas of the north- west and west of Ireland,and so engagedin considerableamalgamation and restructuringof holdings.12The boardacquired the power to compel 596 Economic Development and Cultural Change landlords to sell their land in 1909 but, before that, had succeeded in acquiring and selling a large number of holdings. The Wyndham Act of 1903 produced a large increase in the number of tenant purchases, not least because it established much better terms for tenants and gave hefty bonuses to landlords who sold their estates. Under this act landlords and tenants negotiated a sale price. Once the sale was arranged, the government paid the landlord in cash and ad- vanced the tenant the cost of the land with repayment in the form of an annuity at 3.25% interest. Landlords received both a cash bonus on com- pletion of the sale and the right to sell their own demesne to the state and purchase it back under the same terms as their tenants purchased farms. This second provision permitted many landlords to exchange costly mortgages for low cost state annuities.13The immediate effect of this arrangement was salutary for most tenants: the purchase price was legally required to be such as to make the annuity payments between 10% and 30% below rents fixed in the second judicial term. The 1909 Birrell Act revised land purchase to contend with the initiative's unex- pected popularity, replacing cash payments to landlords with stock paying 3%. This made land sale less popular as landlords were unreceptive toward the securities and tenants were required to pay a slightly higher interest rate.14 Tenant purchase was well under way when the outbreak of World War I led the government to stop making the advances required for the program to work."15The importance of the 1903 and 1909 acts can be seen in the relative proportions of tenancies purchased under them, as opposed to the earlier acts. By early 1913 about 250,000 holdings com- prising 8 million acres had been sold.'6 Of these, only about one-third of the transactions had taken place under provisions prior to the 1903 act. By March 1919 the Congested Districts Board had resold some 23,000 holdings (of some 585,000 acres) in its area of operation, and together with the Land Commissioners and Estates Commissioners had sold 285,000 holdings totaling 9.3 million acres.17This amounted to about half of the agricultural land in Ireland. After the partition of Ireland, landlords were eventually compelled to sell out in both parts of Ireland; by the Second World War, landlords in Ireland were a memory.

II. Land Reform and Economic Efficiency Revisionist interpretations of Irish tenancy and land reform agree that earlier accounts exaggerated the inefficiency of prereform land-tenure practices and by implication argue that little should have been expected of the two land reforms. Land purchase itself has received so little atten- tion that historians seem to have viewed it as an afterthought. Were the economic changes implicit in the land legislation so meager? We first consider the impact of land reform on security of tenure (leaving its implications for collateral and commoditization to the next section) and TimothyW. Guinnaneand RonaldI. Miller 597 then outline the Land Acts' effects on distributionof both land and wealth.

Security If prereformtenants had been as insecurein their holdings as the older historiographyhad argued, then the fixity of tenure provisions of the 1870 and 1881 LandActs would have improvedthe incentivesto invest in and to care for the land. Connellis not alone among an older genera- tion of Irish historiansin acceptingthat pre-1870 tenurewas generally precarious.The insecurity hypothesis is, however, incompatiblewith several importantfeatures of what we now know about Irish tenancy. First,as table 1 shows, by 1870 most of the largestfarms in Ireland(and most of the land) were held on leases. The insecurityhypothesis could only explain underinvestmenton the smaller holdings, and these accounted for relatively little agriculturalland. Second, tenants without leases were not necessarilyinsecure in their holdings. Both Solow and Mokyr have emphasizedthat many formal tenantsat will remainedon theirholdings for life and that,when asked,tenants at will were not very enthusiasticabout leases. Some of these tenantsappeared to have tenant right,but some did not. Irishtenant farmers engaged in severalpractices that made sense only if the participantsthought a holding was secure. For example, tenantsoften wrote wills that bequeatheda farm to a son on the condition that he make some settlementwith siblings.18Third, Vaughanhas shown thatrent was not so "elastic" as Connelland others thought.From 1850 to the mid-1870s, a period of great prosperityin Irish agriculture,the value of outputincreased much more rapidlythan the value of rent.19Most of these Irish farmerskept the fruits of their labors. This discussion suggests that there was no economic reasonto ad- vocate tenantpurchase. Whatever insecurity might have existed was al- most entirelyextinguished by the 1870 and 1881 acts. Even Connell ac- knowledges that late nineteenth-centuryIrish tenants were, "in essentials," owners.20A land purchaseprogram that convertedinsecure tenantsto ownerscould be expectedto improveincentives to makelong- term investmentsin the land. From this viewpoint, however, Ireland's purchase programwas superfluousas Ireland's tenants were already quite securein theirholdings. In some othercontexts one of the principal gains of land reformis the eliminationof inefficientland-tenure arrange- ments. Sharecropping,for example,has been criticizedfor not giving occupiers the right incentives to work their land since the sharecropper retains only a fraction of the marginalproduct of his labor.21But as CormacO Grfidahas pointedout, Irish tenantshad always been the re- sidualclaimants of output:in termsof incentives,secure tenants have all the virtuesof owners.22 598 Economic Development and Cultural Change

The Allocation of Land Many land reforms, historical and modern, have involved considerable reorganization of the units of agricultural production. Neither phase of Ireland's reform had significant implications for the productive distribution of land, in contrast to the English enclosures of the late eighteenth century, which were, among other things, a land-reform effort that re- sulted in the amalgamation of holdings.23Neither the regulation of tenancy nor the Land Purchase Acts altered the distribution of land to any great extent. Most rents fixed under the 1881 act pertained to farms that had been in the same hands for years, while most tenant purchasers bought a holding they or their families had been operating for an equally long time. (The only exceptions were some provisions of the purchase acts designed to break up untenanted grasslands and distribute them as new farms to those lacking any holding. These redistribution provisions accounted for few of the total tenant purchases.)24For most new proprietors, tenant purchase meant neither the break up of large farms into holdings run by families for the first time, nor the amalgamation of tiny plots into larger, more viable holdings. Further, in contrast with many Continental reforms, the Irish land reform did not dispossess any smallholders. Irish land purchase was mostly a redefinition of property rights for those already occupying the land.25 In Ireland's reforms, the combination of rental agreements rather than share tenancy, prior to reform, with the lack of substantial changes in the distributionof land make it somewhatunusual. Land reform in formercolonial lands such as LatinAmerica have typicallyinvolved major changes in the distributionof land as large latifundiaare brokenup with the aim of creatinga new class of peasantproprietors. Such redistribution necessarily leads to importantchanges in agriculturalpractice whether for good or ill. Historic reforms in continentalEurope also changed the distribution of land. Revolutionary France confiscated Church and emigre lands, selling them to the highest bidders, while under the Stein-Hardenburgreorganization of Prussian tenures in the early nineteenth century, nearly 2.5 million acres of land were transferredfrom small peasant to large estate owners in East Elbia alone, eventually creating a large class of landless peasants.26Most Asian reforms have begun from land-tenure systems largely based on share tenancies as in India and Taiwan. The Japanese postwar reform is one case with both rental contracts before reform accompanied by limited changes in agricultural structure and the allocation of land and, thus, is similar to the reforms in Ireland.27As in the case of Ireland, the primary goal of reform was the pacification of rural areas, with anticommunism as the key objective in Japan.28

What Did Land Reform Accomplish? How did the measured efficiency of agriculture evolve following the Land Acts? In the absence of detailed microeconomic data from the pe- TimothyW. Guinnaneand RonaldI. Miller 599 riod before land purchasebecame general, comparingthe performance of owner-occupiedfarms with those that were still tenanted,no certain answer is possible. Nonetheless,given the generalityof the reforms,if therewere any importanteffects on agriculturalefficiency one would ex- pect a significantupward shift in the trend of aggregateproductivity. There is no evidence of such a shift. Real agriculturaloutput per acre was stagnantover the period in question:it increasedby less than 7% from the 1870s to the 1910s.29If insecurityof tenurehad been the cause of low rates of investmentbefore the reforms, then capital formation should have acceleratedafterward. Instead, the capital stock grew more slowly in the 1890s thanduring the 1850s-1870s. However,labor input fell quiterapidly over the postreformperiod so thatoutput per capitaand total factorproductivity rose sharply.In GreatBritain there was no com- parablerise in output per memberof the agriculturalworkforce.30 But this rise was alreadywell underway at the time of the land reforms,with the rate of increase in output per worker greaterin 1851-81 than in 1881-1911.31 Thereis nothingin the aggregateevidence to suggest that the Land Acts had any positive effect on agriculturalefficiency: if any- thing the data indicatea small negative effect. The empirical evidence is consistent with our argumentthat the Irish Land Acts had little effect on the distributionof productivere- sources, and so at most minoreffects on agriculturalefficiency. Yet the land-reformprogram was certainly not insignificant:it redistributed wealth among landlords,tenants, and British taxpayers.The distinction between redistributingwealth and reallocatingresources has caused some confusion in the Irish historiography.One historian,for example, claims thatby reducingrents the 1881 act loweredthe price of land relative to other inputs, therebyinducing farmers to shift from labor-inten- sive tillage to the more land-intensivegrazing.32 John P. Huttman'sanal- ysis is flawed;a reductionin the rent paid to the landlordwould be fully capitalizedin the tenantright and so have no effect whatsoeveron the opportunitycost of using land. Any new tenant would pay for the reduced rent paid to the landlordin the form of an increasedtenant-right payment to the currenttenant. Nor did the purchaseacts improve re- source allocation.The United Kingdom's taxpayerssimply subsidized the transformationof the landlord'sright to receive a rentinto a termina- ble annuity. Revisionist analysis of Ireland'sland reformshas, then, correctly arguedthat these initiativeshad little directimpact on incentivesto work and to invest. Accordingto this view, Parliamentcould have achieved the same ends by a programof nondistortionarytaxes that made landlords poorerand their tenantswealthier. Yet the revisionistshave failed to appreciatethat changingthe distributionof wealth by itself can have importanteconomic effects. There are two channels through which changes in wealth broughtabout by a land reformmight affect production. 600 Economic Development and Cultural Change

Standardeconomic theory shows that a lump-sum increase in wealth will decreaselabor supply so long as leisure is a normalgood. Tenants whose rent is reduced (or whose rents are converted to the smallerpurchase annuity payments) would be less inclined to take on additionalwork off their own farms, as had been the widespreadprac- tice. This effect may actuallyreduce agriculturaloutput, though it will necessarilyleave the farmerbetter off. The reductionin the laborsupply of formertenants will increasethe demandfor farm labor from others, and so increaseagricultural wages.33 Less likely, the increasein farmers' wealthmay improvenutritional intake and thusthe qualityof farmlabor. Models of this effect of land reformhave been presentedby ParthaDas- guptaand DebrajRay and by KarlOve Moene.34While some were certainly very poor, few Irish tenantswere likely to have been chronically malnourishedby 1881; changes in nutritioncould not have been important in Ireland's land reform.35 Changingthe distributionof wealth can also affect investmentif creditmarkets are less thanperfect. Wealthier farmers have less need of creditsources and are betterborrowers under conditions of imperfectin- formation.Efforts to improve rural credit were a staple of late nineteenth-centuryreform legislation in muchof continentalEurope. We turn to this aspect of Ireland'sland reformnext.

III. Tenancy Reform, Land Purchase, and Rural Credit The availabilityof credit for farmers,particularly smallholders, was an importantconcern of reformersin much of westernEurope during the nineteenthcentury. In Irelandthat concernled HoracePlunkett and his Irish AgriculturalOrganization Society (IAOS) to introduceGerman- style creditcooperatives in 1894. Ruralcredit became an especially importantconcern in Irelandwhen difficultieswith the IAOS and concerns about the capitalizationof tenantpurchasers led to a full-scale critique of ruralcredit facilities in 1914.36Yet, paradoxically,land reformin Ire- land did not take some ratherobvious steps to improverural credit facilities. The questionis, Why not?

Tenancy and Credit As we have seen, few farm operatorsowned their land in nineteenth- centuryIreland, and many smallholdershad no guaranteesof continued occupationbefore the LandActs of the 1870s and 1880s. Yet these tenants frequentlyborrowed money on the "security" of the land they did not own, a fact both puzzlingand irritatingto observers.Before the famine, the Devon Commissionheard many complaintsthat tenant right robbedincoming tenantsof capital.Several of the Poor Law inspectors remarkedin their 1870 reportthat debts associatedwith the acquisition of tenant right sometimes proved burdensome for the new occupier.37 One landlord,a later,bitter opponent of tenantright practice said, "The TimothyW. Guinnaneand RonaldI. Miller 601 fatal objectionto the Ulster tenantright is that it absorbs,in buying it, all or a greatpart of whatevercapital an incomingtenant has, and leaves him often withoutthe meansof farmingwell." 38Even landlordswho tol- eratedtenant right sometimestried to regulatethe amountspaid to prevent the incomingtenant from being burdenedwith large debts.39Tenant right could be extremelyexpensive. Contemporariesfigured tenant right values in "years purchase,"or the numberof years of the rent paid to the landlordthat would equal the purchaseprice; informantstestifying before the Devon Commissionstated that tenantright of 4-6-years purchase was ordinary,a substantialsum for a poor family. Between 1850 and the mid-1870s large increases in agriculturalprices were not matchedby rentincreases, leading to substantialincreases in tenant-right values. Systematicsurveys do not exist, but it was not unknownfor a farm to sell for 10-20 years of the rent. How did purchasersof tenantright acquirethese large sums?Many simply inheritedfrom parentsor otherrelatives. An Irishtenant who in- heritedhis father'sfarm was inheriting,in fact, the farm's tenantright. Other accounts mention village usurersand other informal sources of cash for tenantright. More surprisingly,tenants very clearly could and did borrowusing tenantright as collateral.Several of the Devon Com- mission's informantsstated that tenantsraised what amountedto mortgages on their tenant right, and the commission itself concluded that "debts are contractedon the securityof the tenantright."40 K. O'Neill refers to the same practicein CountyCavan.41 Tenantright also expandedthe possibility of implicit loans from a landlord.We argue elsewhere that landlordstolerated tenant right because they had firstclaim againstarrears on paymentreceived by an outgoing tenant.42Moreover, tenant right gave tenantsbetter incentives to pay their rent as they knew that any arrearswould eventually be de- ductedfrom the tenantright. Tenantright also made landlordsless con- cernedabout small arrears,and so more willing to extend small loans in the form of overdue rent.43Tenant purchaserslost this implicit credit channel.W. F. Bailey, an official of the LandCommission, undertook a brief study of the conditionsof tenantswho had purchasedholdings under the acts prior to 1903. He reportsthat "here and there a purchaser complainedthat he must pay on the moment,and had to sell cattle at an inconvenientperiod; that under the landlordsystem the tenants would have got time.""44While it is truethat the lower annuitypayments would have reducedthe need to go into arrears,it is difficult to believe that many purchasersdid not regret at one time or anotherthe passing of a landlordwho would accept the rent a few monthslate.

Mortgage Credit and Land Purchase Bailey identifiedlack of access to credit as one of the most important problemsfor tenantpurchasers: "On a large numberof estates we found 602 Economic Development and Cultural Change thatthe purchasersare greatlyhampered by the absenceof workingcapi- tal, and that they are frequentlyprevented from making desirableim- provementsfrom this cause."45 Bailey went on to worrythat if some provision were not made to meet these needs, new purchaserswould turnto informalcredit sources such as usurers.Although we have little direct evidence concerningthe uses to which loans were put, in testimonybe- fore the 1914 DepartmentalCommittee one advocateof an agricultural bank wanted to provide loans for purchaseof land, machinery,imple- ments, seed, and fertilizer, among other purposes.46 The Departmental Committeealso found that loans from banks and other formal sources were expensive for Irish smallholders.47 One way to improve credit facilities would have been to emulate the several continentalcountries that had establishedland banks to provide long-termfinancing to farmerson the security of their holdings. Most modemrnland reforms have includedexpansions of ruralcredit facilities with varyingdegrees of success. To the contrary,the LandPurchase Acts containedclauses that restrictedthe abilityof formertenants to ob- tain crediton the securityof land they now owned. The 1903 act limited mortgagedebt to 10 times the annualLand Commission payment on any holdingbought with state aid.48While the statuteis somewhatunclear as to whetherdebt to the state is to be consideredwithin the limit, it is evi- dent from the DepartmentalCommittee report of 1914 that it applies only to second mortgages.49For some tenant purchasersthis limit reduced the availabilityof credit. Before land purchasethe tenanteffec- tively owned a fairly large portionof the land's value via tenantright and was able to raise loans on this portionof the land withoutlegal re- striction.After purchase,the farmercould borrowonly 10 times the annual annuitypayment, or approximatelyone-third of the purchaseprice. But this was the purchaseprice of the remainingvalue of the land, the part not alreadyowned by the farmerthrough tenant right. This meant that so long as the value of tenantright was greaterthan one-third of the land purchaseprice, the farmercould borrowless money afterpurchas- ing his holdingthan he could before.That is, tenantright had established a coproprietorshipin land, and beforepurchase tenants could borrowon the securityof the partof the land they owned. The rules of tenantpurchase, however,forbad them to take full advantageof these older property rights. Considerthe hypotheticalexample, presentedin table 3, of a farm that would rent for ?10 in an unregulatedmarket. If the farm was under a second-termjudicial rent, and if both the first and second terms reduced the rent by 21%, then the rent paid in 1900 would be ?6.40. The value of the tenantright would be about?111, assuming3.25% interest and thatthe Ricardianrent would still be ?10. As we have seen, the tenant could borrowagainst some, perhapsall, of the ?111 in tenant-right value. If the tenantpurchased the holding from the landlordat 20 times TimothyW. Guinnaneand RonaldI. Miller 603

TABLE 3 EFFECTOF LAND PURCHASEACT MORTGAGERESTRICTIONS ON HYPOTHETICAL FARM

Amountin ?

Mortgaging tenant right: Ricardian rent 10 Judicial rent (assuming two 15% reductions) 6.4 Value of tenant right 111 Mortgage permitted for same farm purchased under land act: Purchase price (assuming price 20 times annual rent) 128 Annual payments to Land Commission 4.7 Maximum total mortgage on the holding 47

NOTE.-Assumes tenant right capitalized at 3.25%, the interest rate charged by the Land Commission. the annualrent, paying ?128, fully financedby an advancefrom the Land Commission,his annuitypayments would be about?4.7 so thatthe maximum he could borrowwould be ?47. The largerthe proportionof the farm's total value that the occupierowned before purchase-that is, the largerhis tenantright relative to the rent he paid the landlord-the more severe this restrictionon mortgaging.The unusualinstitutional feature of tenant right, and the ability to borrow against it, meant that mortgage restrictionthat mighthave been neutralunder a more standardtenure ar- rangementcould actuallyreduce the availabilityof ruralcredit. The mortgagingrestriction was not absolute. A tenant purchaser could apply to the Land Commissionfor permissionto exceed the debt limit. Few did so, however, and of those who did between 1904 and 1913 about 20% were refused and many that were grantedpermission were allowed to mortgageless than the amountrequested.50 While the low numberof applicationsmight seem to indicatethat the restrictionon mortgagingwas not binding,the DepartmentalCommittee states that the constrainton borrowingwas bindingand that indeed it shouldbe, other- wise there would be "overmortgaging.''"51More likely purchasers,when faced with the choice of an applicationto the LandCommission or raising funds in some otherfashion, chose to deal with a bankor some other nonmortgagelender. A later legal decision also provideda loophole by decidingthat so-calledjudgment mortgages were outsidethe purviewof the 1903 act. In the case of a judgmentmortgage, if an unsatisfiedcredi- tor could show that his debtor owned an interestin land, the creditor could petition a court to give him, effectively, ownershipof that land subjectto satisfactionof the debt.52 The questionis, Why did the governmentadopt a policy intended to restrictaccess to capital?In part, it was the action of a paternalistic governmentthat simply did not trustpeasant proprietors not to overbur- den their holdings with debt, leading either to their own ruin or to the 604 EconomicDevelopment and CulturalChange

ruin of theirheirs. Two objectionsto this argumentreceived little atten- tion. First, the reasonableconclusion to draw from the evidence pre- sented to the DepartmentalCommittee or the Royal Commission on Congestionin Irelandwould be that those seeking credit would find it one way or another.The real choice, from the government'sviewpoint, was betweena peasantrywith debtssecured on land and a peasantrywith debts securedin some other way. Since nonmortgagedebts were, by all accounts, likely to be more expensive, one would think the mortgage debts would be preferable.A second objectionis more obvious: Why should the governmentcare more about excess indebtednessamong peasantsthan among any other social class? Another motive for the mortgage restrictionswas perhaps more transparent:the governmentwanted to maintain the seniority of its claims on purchasers.Bailey admitsas much in rationalizingcontinued state supervisionof land purchasers:"For its own security so long as the purchasersare liable for the repaymentof their State annuities,the Governmentis boundto see that nothingis done that will endangerthat security. . . . If land purchase is to be largely extended it is advisable that furtherconditions be providedthat will limit the power of charging and mortgagingholdings in a mannerthat is injuriousto the occupier and to the generalcommunity."53 In other words, if mortgagingendan- gers the purchaser'sability to repaythe state, then the state shouldpre- vent the practice.Nowhere does Bailey considerthe possible economic benefits of mortgaging.54In the event, very few tenant purchasersde- faultedprior to the partitionof Ireland.In its reportfor 1920, the Land Commissionnoted thatonly 0.23% of purchasers(owing 0.23% of payments) were in arrearsthat year.55 Problemswith Title to Land The DepartmentalCommittee argued that anotherobstacle to securing debts on Irishland was a poor systemof title registration.The committee claimed that Continentalsmallholders were better-servedby mortgage credit because of more effective title-registrationsystems, althoughno such system was in place in England.56This commenthighlights a second missed opportunity;rural credit facilities would have been greatly aided by a methodof registeringboth titles and mortgagesthat enabled potentialborrowers to establish clearly and with little cost what prior claims existed on a given holding.Ireland had no system of compulsory title registration,although purchasers receiving governmentassistance were requiredto registertheir new property.Even registryof title in Ire- land was problematic,since it requiredthe discharge of "equities," claims by emigrantrelatives and others with a potentialinterest in the land.57"'An individualwho contemplatedmaking a mortgageloan had first to make sure that the borrowerhad clear title to the land and, then, had to establishwhat priorliens had been placed on the holding. The latter TimothyW. Guinnaneand RonaldI. Miller 605 was virtuallyimpossible if the borrowersought to conceal prior loans; registrationof mortgageobligations was possible but not required(ex- cept for mortgageson land purchasedwith state aid underthe 1903 act), and the method of registrationrequired costly searches throughmany volumes. Much more efficientmodels were alreadyin place in continental Europe.Bavaria presents a typical case; there, each locale kept one set of recordsrecording property rights in land (the Grundbuchor Katas- ter), and anotherrecording mortgages and other encumbrancessuch as family charges (the Hypothekenbiicher).Both were up-to-datepublic documents,based on contractsdeposited in public archives, and each was organizedby holding. To establishprior claims on a holding, a po- tentiallender had only to inspectthese documents.Complaints about the availabilityof rural credit were common in Bavaria at the end of the nineteenthcentury, but it is also clear that individualswere able to raise mortgageson very small properties-in some cases consisting of little more than a gardenplot.58

Irish Joint-StockBanks and MortgageLoans Parliamentcan be faultedfor the creditlimitations in the tenantpurchase rules or for failing to establishcompulsory registration of title and mortgages. Anotherlimitation on mortgagingcame from a differentsource: ruralsocial normsthat preventedbanks from playing a role in mortgage credit. The second half of the nineteenth century witnessed a great expansionof joint-stockbanks in Ireland.By 1900 most small towns had a branchbank open for at least part of each week, yet these banks did not providemortgage loans to Irishfarmers.59 The banksthemselves told the DepartmentalCommittee that it was not theirpolicy to lend on mortgage security.The inabilityto use this well-developedfinancial structure to finance agriculturalinvestment would seem a majorloss. The banks arguedthat they were constrainedby feelings toward land and forced sales: since neighborswould not buy land sold at distressedsales, it was not valuableas collateral.6This view was statedby T. W. Delaney, rep- resentativeof the County LongfordCommittee of Agriculture:"I consider the difficulty which banks . . . have on a forced realisation of their security,makes it practicallyimpossible to borrow a sixpence on land security... the timidityof the people... aboutinterfering in a forced sale, naturallyprevents banks or othersfrom advancingmoney."61 Dela- ney did not foresee mortgagebanking until people viewed debtsto banks as like any other obligation.62 The problem did not disappearwith partition;the 1926 Banking Commission, which took a generally critical attitudetoward bankers, blamed the lack of mortgagefacilities on the people: "It will never be possible for any countryor its inhabitantsto borrowfreely, unless there be full recognitionof the justice of the debt so incurred,and full readi- ness to have the creditortake possession of the propertyand convertit 606 EconomicDevelopment and CulturalChange to his own use..,. in the event that the debtorfinds himself unwilling or unableto meet the terms of his agreement."63The banks' unwilling- ness to extend mortgagecredit robbed Irish tenantpurchasers of one of the majorbenefits of ownership,the ability to use land as collateralfor loans. Yet here the culpritwas not Parliament,which could not legislate ruralattitudes; the problemlay in the banks' perceptionsof what would happenshould they have to sell out a failed borrower.To the extent that these perceptionswere accurate-and a profit-makingbank had no reason to abstainfrom a particularline of business withoutgood reason- then ruralnorms formed one importantlimit to the abilityof land reform to improveagricultural efficiency in Ireland.

Mortgage Creditand the Commoditizationof Land Another possible benefit to be had from the redefinitionof property rightsin land may come frommaking land more like any othercommodity-relatively simpleto purchaseand to sell. If land can easily be traded in open markets,then the usual marketmechanism ensures that land is allocatedto its highest-valueuse. Ireland'sland reform had mixed impli- cationsfor the commoditizationof land. Criticsof the prereformtenancy system often complained,bitterly, that landlordswould make tenants compete for the right to rent farms,thus increasingrents. To the extent this practicewas general-and more recent researchshows that it was not-tenants were simplycomplaining that such competitionraised rents by allocatingland to its highest-valueuse.64 We will now considerhow tenantright, whetherin its customary form or in the form establishedby the 1870 and 1881 LandActs, altered the ability to buy and to sell land. Legally it did not do so at all; there were no provisionsagainst selling a holdingthat had a judicialrent fixed, and the rentremained with the holding,so a new tenantcould be guaran- teed of the judicial rent. The reformof tenancymight have actuallyim- proved the land marketbecause it removed all uncertaintyabout com- pensationfor improvementsand the futurecourse of rents.Previously, a purchaserwould have to satisfyhimself as to the landlord'spractices and intentionsbefore purchasinga holding. With regulation,however, indi- vidual landlordpractices became nearlysuperfluous. Crottyhas arguedthat the tenantpurchase program was a step back- ward in this respect, in that "it substantiallydestroyed the competitive marketfor land."65He creditsthe land purchaseacts with saddlingIre- land with a large numberof small, uneconomicholdings. Crotty claims that once a reducedannuity payment had replacedrent payments,Irish farmersbecame less sensitive to the true cost of owning and using their land; "as an owner-occupier,a farmerneed not worrylest anotherman, by his willingnessto invest more capitalat a lower rate of return,might be able to pay a higher rent and cause his own eviction."'66This posits eithera remarkabledegree of irrationalityon the partof the new owners TimothyW. Guinnaneand RonaldI. Miller 607 or simply a preferencefor land ownership.An owner-occupiercould sell his land at a profitto anyonewho believed thathe could more efficiently farm it. If the owner chose not to do so because of a strong desire to hold land then this would indeed cause productionto be lower thanoth- erwise. But it would not affect economic efficiency per se since the owner would gain utility from simply owning the land: land would remain in the hands of those to whom it had the most value. While the acts had little direct impact on the commoditizationof land, theirindirect effects throughthe limitationof mortgagecredit may have createdimportant obstacles to the consolidationof holdings. The acts' direct effect on the patternof landholdingwas close to neutral. While the acts containedprovisions that limited subdivisionand sublet- ting, no seriousattempt forcibly to amalgamatesmall holdingswas made in the reforms,apart from the somewhatlimited efforts of the Congested DistrictsBoard. But the limitationson mortgagingdiscussed above cre- ated majordifficulties in raising sufficientmoney to purchaseholdings. As was reportedto the DepartmentalCommittee, the purchaseof holdings that had been obtainedthrough tenant purchase was made signifi- cantly more difficultthrough the limitationson mortgaging.67Some land transactionsthat could have improvedthe efficiency of land use would not have been possiblebecause of the inabilityto borrowfunds sufficient for purchase.This meantthat only those in possessionof sufficientinde- pendentresources could actively engage in consolidatingholdings and employingland in its highest value use. IV. The Limits to Land Reform Agitation over such issues as tenantright and tenant purchasewere a centralfeature of the several nationalistmovements of late nineteenth- centuryIreland. Yet interestin land and land reformwas, as historians now appreciate,less a matterof genuine concern over ruralconditions than an attemptto use a popular,bread-and-butter issue to advancethe nationalistcause. Parnell himself, probablythe most successful nineteenth-centuryIrish politician outside of O'Connell,rode to prominence largely by harnessingagrarian distress; "It was the land issue that sup- plied, in 1879, the politically mobilizing factor that made Parnell."68 Gladstone'sinterest in Irish land stemmednot from any interestin economic developmentbut from his desire to pacify Ireland.Earlier economic historiansaccepted the critiqueof land tenurein Irelandas pro- moting inefficiency, and with that view also acceptedthe idea that the Land Acts must have promotedincreased efficiency. Revisionist economic historians,starting with Solow, effectively challengedthe view that the LandActs were a cure for any disease the Irish were really suf- fering. This articleprovides further analysis to strengthenthe revisionist position. Yet this is not to say thatland reform in Irelandcould not have been 608 Economic Development and Cultural Change a helpful step toward agricultural efficiency. Rural credit in Ireland was problematic and formed an impediment to efficiency; and several Conti- nental countries offered models that might have been profitably imitated. Yet when Parliament regulated, and then abolished, Irish landlordism, it did so in a way that did little to improve agricultural conditions, and missed a major opportunity to place rural credit on a more secure foot- ing. The limitations to mortgaging imposed by the acts may have even reduced rural credit availability by eliminating the ability to borrow against tenant right and thus have helped to lock in a pattern of ineffi- ciently small holdings. This is not to say that land reform in Ireland was a mistake: "To show that [the Land Act of 1881] was widely realized on economic grounds to be either wrong or irrelevant is to miss the whole point."69 The same can be said for the land purchase program. The Land Acts in Ireland reflect the outcome of a political struggle, a struggle between landlords and tenants, between tenants and their creditors, and between Irish Parliamentaryrepresentatives and English political parties. Land reform in Ireland accomplished what the people of the day in fact wanted. Nonetheless, it should be kept in mind that, within the bounds of its political objectives, a land reform can still be effective economic policy: the Irish reforms missed important opportunities to create real economic benefits.

Notes * We thankTimothy J. Besley, WilliamEnglish, and ChristinaPaxson for commentson an earlierdraft. 1. B. Solow, TheLand Question and the IrishEconomy, 1870-1903 (Cam- bridge,Mass., 1971). 2. D. Collison Black, "The IrishExperience in Relationto the Theoryand Policy of EconomicDevelopment," in EconomicDevelopment in the LongRun, ed. A. J. Youngson(London: Allen & Unwin, 1972). 3. Timothy Besley, "PropertyRights and InvestmentIncentives: Theory and Evidencefrom Ghana,"Journal of Political Economy103 (1995): 909-37. 4. Evidencefrom earlier in the centuryis more sketchybut shows the same patternof a predominanceof tenancy at will among smaller farmers.Strictly speaking,tenancy at will and yearly tenancywere different(see Solow for discussion).Yet the distinctionis usuallynot maintainedin the Irishhistoriography, and the term "tenancyat will" is used throughoutthis article. 5. J. Mokyr, WhyIreland Starved:A Quantitativeand AnalyticalHistory of the Irish Economy,1800-1850 (London,1985), chap. 4. 6. RaymondCrotty, Irish AgriculturalProduction: Its Volumeand Struc- ture (Cork:Cork University Press, 1966), p. 51. 7. K. H. Connell, "PeasantMarriage in Ireland:Its Structureand Develop- ment since the Famine," EconomicHistory Review, 2d ser., 14, no. 3 (1962): 502-23, quote on 521. Connell's view of pre-Land Act tenuresechoes a long line of Irish economic historians, the most notable of which was George O'Brien,whose TheEconomic History of Irelandfrom the Unionto the Famine (London:Longman, Green, 1921) was quite influential."Elastic rent" and its Timothy W. Guinnane and Ronald I. Miller 609 deterrentto propertyaccumulation were the centerpieceof Connell's explanation of prefaminepopulation growth. 8. Solow; W. E. Vaughan,Landlords and Tenantsin Mid-VictorianIre- land, (Oxford:Clarendon Press, 1994). 9. Solow; and Vaughan,Landlords and Tenantsin Mid-VictorianIreland discuss the basic institution.In our "Bonds withoutBondsmen: Tenant-Right in NineteenthCentury Ireland," Journal of EconomicHistory 56, no. 1 (1996): 113-42, we arguethat tenantright was, from the landlord'sviewpoint, a work- able alternativeto securitydeposits, thus helpingto explainwhy some landlords were willing to tolerate the practice.Historians have rightly insisted that the three F's do not exhaustthe significanceof tenantright. See, W. E. Vaughan, Landlordsand Tenantsin Ireland, 1848-1904 (Dublin:Economic and Social History Society of Ireland, 1984), p. 20, and Landlordsand Tenantsin Mid- VictorianIreland. 10. In "Bonds withoutBondsmen," we arguethat it would generallynot be in the landlord'sinterest to raise rents to the Ricardianlevel even in the absence of rent control.The judicial rentsmay have been lower than was optimal from the landlord'sviewpoint, however. 11. That is, ?4 capitalizedat 3% is ?133. 12. Of the 588,000 acresof landresold by the CongestedDistricts Board as of 1919, approximatelyhalf had been eitherimproved, enlarged, or reorganized significantlyby the board. This is a relatively small fractionof the total land affectedby the acts (House of Commons,"Twenty-Seventh Report of the Con- gested DistrictsBoard for Ireland,"Cmd. 759 [1920], p. 61). 13. E. R. Hooker,Readjustments of AgriculturalTenure in Ireland(Chapel Hill: Univeristyof NorthCarolina Press, 1938), p. 80. 14. The 1903 act requiredtenants to pay at 3.25%;the 1909 act raisedthe interestrate to 3.5% (ibid., p. 90). 15. Ibid., p. 92. 16. These figuresexclude sales to tenantsof the Churchof Ireland,which were relatively few and did not come under the legal definitionsof the Land Acts. The data are from House of Commons, "Irish Land PurchaseActs: Re- turn," Cmd. 6930 (1913). 17. House of Commons, "Reportof the Irish Land Commission,"Cmd. 572 (1920), p. vi, and "Report of the Estates Commissioners,"Cmd. 577 (1920), p. iv. 18. House of Commons,"Reports from Poor Law Inspectorsin Irelandas to the Existing Relationsbetween Landlordand Tenantin Respect of Improve- ments on Farms,"Cmd. 31 (1870). 19. Vaughan,Landlords and Tenantsin Ireland,1848-1904 (n. 9 above), p. 22, and Landlordsand Tenantsin Mid-VictorianIreland (n. 8 above). 20. Connell (n. 7 above), p. 522. 21. AlfredMarshall, Principles of Economics,8th ed. (New York:Macmil- lan, 1979), pp. 535-36, outlines the basic objectionto sharecropping.Marshall exaggeratedsharecropping's undesirable effects; for a morerecent discussion of the institution,see JosephE. Stiglitz, "EconomicOrganization, Information, and Development,"in Handbookof DevelopmentEconomics, ed. H. Cheneryand T. N. Srinivasan(New York:North Holland, 1988), 1:93-160, andthe literature he discusses. 22. CormacO Grida, "IrishAgricultural History: Recent Research,"Ag- riculturalHistory Review 38, no. 2 (1990): 165-73, esp. 165. 23. In Irelandthe Land PurchaseActs forbadthe subdivisionof holdings boughtwith state aid. But, given thatthe tendencyafter the faminehad been for the averagefarm to become larger,it seems unlikely that these provisionshad 610 Economic Development and Cultural Change much effect on the distributionof land. Some Irish tenantsdid lose common rights,such as access to turf, as partof the land purchasearrangements. 24. Underthe provisionsof the 1903 act some untenantedlands were sold by directpurchase; in value these amountedto aboutone-half of 1%of the land sold by directpurchase through the EstatesCommissioners. After the 1909 act untenantedland had to be purchasedinitially by the EstatesCommissioners, re- packagedand resold. As of 1919 this amountedto less than 200,000 acres, of which some fractionwas accountedfor by wastes, mountains,and turbarylands (House of Commons,"Report of the EstatesCommissioners," p. xi). 25. An older Irishhistoriography sometimes claimed that the regulationof tenancydiscouraged subdivision because only with the threeF's did Irishtenants care enoughabout a holdingto resist subdivision.This view is centralConnell's explanationof postfaminechanges in marriagepractices, e.g. 26. On the confiscationof land in revolutionaryFrance, see JeromeBlum, The End of the Old Order in Europe (Princeton, N.J.: Princeton University Press, 1978), p. 394. Regardingthe reorganizationof Prussiantenures, see RobertA. Dickler, "Organizationand Changein Productivityin EasternPrussia," in Eu- ropean Peasants and Their Markets, ed. W. N. Parker and E. L. Jones (Princeton,N.J.: PrincetonUniversity Press, 1975), pp. 269-92, esp. p. 277. 27. See Russell King, Land Reform: A World Survey (London: Westview, 1977) for one descriptionof the Japanesereforms. 28. GeneralMcArthur himself wrote in 1964 that the main thrustof the reformwas anticommunism(as cited in JamesPutzel, CaptiveLand: The Poli- tics ofAgrarian Reform in the Philippines [London: Catholic Institute for Inter- nationalRelations, 1992], p. 237). 29. The productivitynumbers are derived from table 5.4 in Michael Turner,After the Famine: Irish Agriculture, 1859-1914 (Cambridge: Cambridge UniversityPress, 1996), based on the "nonstandard,"changing weight, agricultural price index. The capital growth figures that follow are from the second panel of the same table. 30. The sourceis ibid., table 5.2. 31. Fromcalculation based on ibid., table 5.3, panel B, col. 2. 32. John P. Huttman,"The Impactof Land Reformon AgriculturalPro- ductionin Ireland,"Agricultural History 46 (1972): 353-68, esp. 355. 33. This mechanismis describedin MarkGersovitz, "Land Reform: Some Theoretical Considerations," Journal of Development Studies 13, no. 1 (1976): 79-91; and MarkR. Rosenzweig, "RuralWages, LaborSupply, and LandRe- form:A Theoreticaland EmpiricalAnalysis," AmericanEconomic Review 68, no. 5 (1978): 847-61. In the more generalmodel presentedby Rosenzweigthe effect of land reformon the ruralwage is indeterminate;but a negative effect would requirethat landlordsbe insufficientlycompensated for the land and for landlordsor their families to supply significantamounts of farm labor,circum- stances that do not apply to the Irish case. Of course, changes in labor supply have nothingto do with agriculturalefficiency. 34. ParthaDasgupta and Debraj Ray, "Inequalityas a Determinantof Mal- nutritionand Unemployment:Policy," EconomicJournal 97 (1987): 177-88; Karl Ove Moene, "Povertyand Landownership,"American Economic Review 82, no. 1 (1992): 52-64. 35. Periodicharvest shortfalls did occurthroughout the late nineteenthcen- turybut are not strictlyrelevant to the mechanismwe are discussing.If the Land Acts did improvenutrition in economicallymeaningful ways, the improvements would have been concentratedamong the very poor landholdersof the Atlantic fringe. 36. TimothyW. Guinnane,"A FailedInstitutional Transplant: Raiffeisen's Timothy W. Guinnane and Ronald I. Miller 611

Credit Cooperatives in Ireland, 1894-1914," Explorations in Economic History 31, no. 1 (1994): 38-61, argues that althoughthe Raiffeisen system of credit cooperativesdid not work well in Ireland,the collapse of the creditcooperatives does not suggest that loans were not needed. The DepartmentalCommittee report on agriculturalcredit (House of Commons, "Reportof the Departmental Committee on AgriculturalCredit in Ireland," Cd. 7375 [1914], Q9152) is surely a polemical document,but many of the faults they point to in Ireland's system were both admittedby representativesof the banksand commonto other Europeancountries. 37. House of Commons,"Reports from Poor Law Inspectorsin Irelandas to the ExistingRelations between Landlordand Tenantin Respect of Improve- ments on Farms" (n. 18 above). 38. W. Bence Jones, The Life's Work in Ireland of a Landlord Who Tried to Do His Duty (London:MacMillan, 1880), p. 182. 39. M. A. Maguire, The Downshire Estates in Ireland, 1801-1845: The Management of Irish Landed Estates in the Early Nineteenth Century (Oxford: Clarendon,1972), p. 145, gives this concernas one reasonwhy LordDownshire, who accededto tenantright on his northernestates, tried nonetheless to regulate the practice. 40. Devon Commission, Digest of Evidence Taken Before Her Majesty's Commissioners of Inquiry into the State of the Law and Practice in Respect to the Occupation of Land in Ireland, 2 vols. (Dublin: Thom, 1847), p. 290. For additionaldiscussion of tenant right mortgages,see Maguire;or K. O'Neill, Family and Farm in Pre-Famine Ireland: The Parish of Killashanda (Madison: Universityof WisconsinPress, 1984). 41. O'Neill, p. 67. 42. Guinnaneand Miller (n. 9 above). 43. The landlordwould not make large loans to the tenant because that would eliminatethe incentiveeffects that were the motivationfor tenantright. 44. House of Commons, "Reportby Mr. W. F. Bailey, Legal Assistant Commissioner,of an Inquiryinto the PresentConditions of TenantPurchasers under the Land Purchase Acts," Sessional Papers, 1903, Land Purchase (Ire- land) Acts, March 25, 1903, vol. 57, p. 19. 45. Ibid., p. 24. 46. House of Commons, "Reportof the DepartmentalCommittee on Ag- riculturalCredit in Ireland"(n. 36 above). 47. The Royal Commission on Congestion heard similar evidence and came to similarconclusions. Liam Kennedy, "A ScepticalView on the Reincar- nation of the Irish 'Gombeenman,'" Economic and Social Review 8, no. 3 (1977): 213-22, sharestheir view of ruralcredit conditionsat the turn of the century. 48. House of Commons,"Bill to Amendthe Law Relatingto the Occupa- tion and Qwnershipof Landin Irelandand for OtherPurposes Relating Thereto, and to Amend the Labourers(Ireland) Acts," Sessional Papers, 1903, Irish Land,July 8, 1903, para.54.3, p. 157. 49. Note, however,that Hooker (n. 13 above) interpretsthe rule as includ- ing the state debt (p. 80), in which case no mortgageat all could be raiseduntil many years aftertenant purchase. 50. House of Commons, "Reportof the DepartmentalCommittee on Ag- riculturalCredit in Ireland,"app. 17. 51. Ibid., sec. 822. 52. Ibid., sec. 825. 53. House of Commons,"Report of an Inquiryinto the PresentConditions of TenantPurchasers under the Land PurchaseActs," p. 23. 612 Economic Development and Cultural Change

54. Bailey's concernabout excessive borrowingis difficultto squarewith his generalimpression that land purchase had led farmersto take a greaterinter- est in theirholdings. He even claims that tenantpurchasers were more likely to be responsibleabout debt. (ibid., p. 10). 55. House of Commons,"Report of the Irish Land Commissioners,"pp. viii-ix. 56. House of Commons,"Report of the DepartmentalCommittee on Ag- riculturalCredit in Ireland"(n. 46 above), sec. 782. 57. Ibid., secs. 782-93. 58. J~iger,Hypothekenbuch und Grundbuch: Lehrbriefe fiir den mittleren Archivdienst,Nr. 10 (Munich, 1975), describesboth the land recordsand the mortgageregistry. On this pamphlet,see TimothyGuinnane, "Notes on Archi- val Sourcesfor Creditand Land-Holdingin Bavaria,1807-1914," workingpa- per (Yale University,New Haven, Conn., 1995). 59. House of Commons,"Report of the DepartmentalCommittee on Ag- riculturalCredit in Ireland,"app. 3. 60. Althoughwe do not have any accountsof violence at particulardis- tressedsales, fear of violence againstthose biddingwas likely partlyresponsible for the difficultyof such sales. 61. House of Commons, "Reportof the DepartmentalCommittee on Ag- riculturalCredit in Ireland,"Q12403. 62. Ibid., Q12409. 63. Banking Commission,(Irish Free State), Banking Commission:2nd, 3rd, and 4th InterimReports (Dublin: Stationary Office, 1926), pp. 20-23. 64. Competitiverents are, in fact, incompatiblewith tenant right. See Vaughan, Landlords and Tenants in Mid-Victorian Ireland (n. 8 above), for a discussionof rent-settingpractices. 65. Crotty(n. 6 above), p. 88. 66. Ibid., p. 90. 67. House of Commons,"Report of the DepartmentalCommittee on Ag- riculturalCredit in Ireland"(n. 46 above), p. 356. 68. R. A. Foster,Modem Ireland, 1600-1972 (New York:Viking Penguin, 1989), p. 402. 69. Solow (n. 1 above), p. 155.