Theories of Institutions and Economic Reform in Africa

HOWARD STEIN* Roosevelt University, Chicago, USA.

Summary. - The paper presents and critically examines the theories of institutions that can be used as a guide to economic reform in Africa Three approaches to institutions are initially discussed the model of structural adjustment, new institutional economics and the old institutionalist tradition. Five institutions related to the development of markets in Africa are examined from the three perspectives. The institutions include property rights (including their role in agricultme), money and financial institutions, prices and markets, firms and industrial organization, and markets and states. The main conclusion of the paper is that structural adjustment due to its neoclassical roots is basically ainstitutional and therefore ill-equipped to promote the development of markets in Africa. While new institutionalism provides interesting insights, it is also limited by its neoclassical economic foundations. Thus if African governments &sire to promote market reform they would be best advised to consult the old institutionalistliterature, pattlcularly the very interestingmcent work of proponentsof this tradition.

The neglect of other aspects of the system has been ma& agriculture, money and linancial institutions, markets easier by another feature of modern economic theory - and prices, Iirms, and markets and states. The nature of the growing abstraction of the analysis, which does not each institution is examined from the perspective of seem to call for a detailed knowledge of the actual eco- the neoclassical/adjustment model, new institutional- nomic system or at any rate has managed to proceed with- ism, and the old institntionalist tradition. out it...What is studied is a system which lives in the minds of economists but not on earth. I have called the This paper argues that structural adjustment, result “black” board” economics...The ex-communist because it is derived from neoclassical economic countries are advised to move to a market economy and theory, is basically ainstitutional and therefore ill- their leaders wish to do so, but without the appropriate equipped to promote the development of market institutions no market economy of any significance is institutions in Africa.’ If African governments are possible. If we knew more about our economy, we would interested in economic reform that develops market be in a better position to advise them (Ronald Cease, institutions then they would be best advised to consult 1992, p. 174). the institutionalist literature. At this point, however, the purpose of this paper is not to systematically devise concrete programs (although policy examples are given), but to present and contrast the alternative 1. INTRODUCTION theories of institutions that can be used as a guide to developing policies. Central to the process of economic reform is the The paper begins with a discussion of the role of Institutions in the formation of markets. The Walrasian general equilibrium roots of neoclassical paper examines the theory of institutions embedded in economics which underlies structural adjustment. In three different traditions in economics, the neoclassi- this view markets are seen as a product of the sponta- cal/structural adjustment viewpoint, the new institutional perspective found in the work of Douglass North, Oliver Williamson and Ronald Coase, and the *An earlier version of this paper was presented at the old institutional approach that was generated by Conference on Public Choice Theories and Third World Commons and Veblen and mom recently in the writ- Experiences, Third World Economic History/Development ings of people such as Geoffrey Hodgson and William Group,London School of Economics, September, 1993. The Laxonick. For this purpose the paper looks at a series author wishes to thank Bill Dugger, Catherine Boone, of institutions central to the development of markets Leonardo Vera, Nit Van de Walle, Gareth Austin, Deborah in Africa including property rights, pmperty rights in Brautigam, peter Lewis and Randall Wray for their helpful comments. Final revision accepti July 11,1994.

1833 1834 WORLD DEVELOPMENT . neous interaction of atomistic self-seeking individu- rights are also transferred in exchange and therefore als. As Coase has aptly put it in the quote at the begin- require some external guarantor (such as the state). In ning of the paper this model “only lives in the minds of addition, while monetary institutions have never been economists but not on earth.” While there are no rea- adequately explained in a general equilibrium frame- sons for institutions in the original model, in the more work (since one must have reasons for holding money relaxed version of structural adjustment there is a need which requires an assumption about uncertainty), to legally ensure property rights and for monetarist- there is some recognition that money is needed as a type guarantees of the stability of the currency. means of payment (Hodgson, 1992, p. 753). This then In the new institutionalist model institutions are sets preconditions for a monetary institution such as a more broadly defined as a means to reduce transaction central bank which, like the guarantor for property and information costs. The new institutionalists still rights, would play only a neutral role (or in Friedman rely, however, on the “choice theoretic approach that terms would ensure that money expansion does not underlies [neoclassical] microeconomics” (North, cause inflation or deflation by putting it on automatic 1993, p. 1). pilot so it only expands at rate of real growth in the The old institutionalists reject the emphasis placed economy - 3 to 4% per annum in the long run). on rational maximizing, self-seeking behavior of indi- The aim of structural adjustment in Africa (and viduals which is at the heart of both neoclassical economics and new institutionalism. They believe that elsewhere) is to remove the impediments caused by institutions are less instrumental and more “as settled state interference in the operation of these markets. habits of thought common to the generality of man” Capitalism would be promoted in Africa by removing (Veblen, 1919, p. 239). Let me begin with a more the distortions that have disrupted prices from equaliz- detailed examination of the relationship between ing supply and demand. The nature of these reforms institutions, neoclassical economics and structural are well documented so I will only briefly summarize adjustment. them (see Quarco, 1990 for a summary). Tax and tariff concessions need to be removed or at a minimum lowered and equalized so that firms can choose inputs 2. INSTITUTIONS, ADJUSTMENT AND based on prices that reflect the relative scarcity of the NEOCLASSICAL ECONOMICS factors of production in the country. The govem- ment needs to scale back by reducing social The neoclassical model, which provides the theo- subsidies (introducing user fees), deregulating the retical underpinning of structural adjustment, is a seri- conditions of private sector operation and privatizing ously flawed representation of how markets operate. or closing state run public enterprises. Private Exchange in the neoclassical model arises sponta- property rights need to be carefully defined and guar- neously from the atomistic interaction of self-seeking anteed so that there is no risk of state nationalization at individuals. Goods traded in every market are a later date. assumed to be homogenous so that prices provide the Exchange rate controls need to be removed and only information needed to make the decisions on pro- currencies should be permitted to float so that the duction and purchasing. No individual has sufficient exchange rate reflects supply and demand conditions market power to affect the market price. Markets must and permits the free-flow of investment in and out of exist for all goods and services for now and in the the country. Financial reforms focus on the need to future so that individuals can make completely introduce real positive interest rates to attract savings informed rational decisions based on perfect informa- and to ensure that only projects with a high rate of tion. Finally, to ensure that equilibrium is reached return will be undertaken. Overall credit constraints in neoclassicals posit the existence of a Walrasian auc- the banking system are necessary to reduce price tioneer who gathers and processes the information from all these markets so that individual agents levels and to lower balance of payments deficits. through a tdtonnement or groping process can adjust Reducing government expenditures in the manner dis- their decisions to remove excess demand and supply cussed above helps to lower credit expansion and from all markets. The result will be that pareto optimal ensure that credit to the private sector can expand with conditions will be reached thereby maximizing the the real growth in the economy (Khan et al., 1990).2 In welfare of society (no one will be able to be better off general, once price distortions and other impediments without making someone worse off). are removed the private sector driven market economy In the strict model no institutions are necessary will naturally occur and prosper. since exchange is simply driven by utility considera- As we can see the model of structural adjustment tions or as von Mises has put it “...an attempt to sub- mirrors the neoclassical view of how markets operate. stitute a more satisfactory state of affairs for a less Their view of state institutions is also limited to satisfactory one” (Mises, 1949, p. 97). In the more guarantors of the rights of private property and the relaxed versions there is the recognition that property money supply. THEORIES OF INSTITUTIONS 1835

3. FLAWS IN THE NEOCLASSICAL/ If we move from a world of zero transaction costs to one ADJUSTMENT VIEW OF INSTITUTIONS AND of positive transaction costs what becomes immediately CAPITALISM: TWO PERSPECTIVES clear is the crucial importance of the legal system in the new world...what are traded on the market are not, as is often supposed by economists, physical entities, but the As indicated in the introduction, there are two rights to perform certain actions and the rights which somewhat competing views one can use to criticize individuals possess are established by the legal sys- neoclassicals and their model of structural adjustment tem...As a result, the legal system will have a profound from an institutional perspective. New institutional- effect on the working of the economic system and may in ism (NIE) which derives from the work of Ronald certain respects be said to control it (1992, pp. 717-718). Coase, Douglass North and Oliver Williamson does not fundamentally challenge the precepts of neoclassi- NIE takes this one step further to emphasize the cal economics but criticizes it for failing to explain need to design a system so that “. . .these rights should the nature of institutions and the role they play in sup- be assigned to those who can use them most produc- porting the existence and operation of markets. tively with incentives that lead them to do so...” Institutions exist as a means of reducing transaction (Coase, 1992, p. 718). The question of who is allo- and information costs so that markets can operate with cated property rights has implications to economic the kind of fluidity and efficiency projected in the neo- reform in Africa. Toye (1993) is critical of the neoclassical model. To quote North, “Information pro- classical/adjustment view of privatization. In Africa, cessing by the actors as a result of the costliness of where both market and state failures are present, the transactions is what underlies the formation of institu- decision should be based on whether the act of reas- tions” (North, 1990, p. 107). signing property rights will lower transaction costs. A second older institutional tradition (OIE for old To Toye, NIE can be used as a tool for designing institutional economics) arises from the work of institutions. Citing the work of D. K. Leonard, he dis- Thorstein Veblen, Commons and others which rejects cusses the case of privatizing veterinary services in much of the neoclassical tradition with its emphasis on Africa. He argues that openly privatizing would likely rational maximizing atomistic agents. Instead they lead to services being run by qualified veterinarians in focus on economic outcomes as a product of entities urban areas where transaction costs would inhibit the like large corporations operating in a complex histori- use by herders, increase the likelihood of diseases and cally specific environment of social, economic and probably lead to a reversal of privatization. Instead the legal institutions.3 Institutions are seen as much less state could semi-privatize by handing out contracts to instrumental and more as “settled habits of thought veterinarians to patrol the known routes of herders at common to the generality of man” (Veblen, 1919, specified times while allowing them to practice pri- p. 239).4 vately. Thus to NIE one should embed a concept of efficient property rights in designing reform.5 This emphasis on legalism and efficiency is what 4. PROPERTY RIGHTS differentiates NIE from the old tradition of institutionalism. It is what makes NIE an addendum to neo- It is easiest to understand the weakness of the classical economics rather than an alternative. OIE neoclassical/adjustment view of reform in Africa by emphasizes the need to differentiate the legalization of focusing on a number of issues deemed vital by insti- property rights emphasized by NIE from the institu- tutionalists to develop capitalism in Africa. Perhaps at tionalization of property rights (Koslowski, 1992, the center of adjustment from an institutional perspec- p. 684). In particular, property rights must not only tive is the question of property rights. The World Bank become established but legitimate. NIE shares the in principle recognizes the need to protect property same weakness with the neoclassical model in the and contract rights to build African entrepreneurship sense of having a common belief in the naturalism of (World Bank, 1989, p. 134). In practice, however, the markets, In the neoclassical case with their assump- emphasis has been on specifying conditions of de- tion of zero transaction costs, market exchanges will regulation and privatization in structural adjustment naturally arise due to the inherent actions of self-seek- programs which is consistent with the neoclassical ing individuals. To NIE, however, positive transaction notion that impediments created by the state are the costs provide an impediment to these actions. Once single most important factor inhibiting the expansion the state properly designs and enforces contract rights of the private sector. Thus scaling back the state will then transaction costs can be reduced and markets can allow capitalism to flourish. naturally proliferate. NIE is critical of this perspective since it does not To OIE property rights are much more than legally put sufficient stress on the role of a systemic state- recognized entities. They are part of a whole w&an- sponsored legal system in encouraging and enforcing schauung that involves a particular mode of thinking market exchanges in a world of positive transaction that is historically specific. Their view, then funda- costs. As Coase has succinctly put it: mentally questions the naturalism of neoclassical 1836 WORLD DEVELOPMENT theory. To OIE the concept of property inherent in more democratic forms of control. If nothing else, this the functioning of capitalism is very different than the will enhance the identification of the general popula- ideas of what constitutes legitimate property under tion with the reform process in a manner which is socialist regimes both in Africa and Eastern Europe. In more consonant with the prevailing ideology. The dis- parts of Africa, private property and accumulation is tinction between structural adjustment, NIE and OIE seen as evidence of exploitation which often entails can also be seen in the question of pmperty rights and the enrichment of visible minorities at the expense of agriculture. the local African population. In Europe, the reelection of members of former communist regimes in Hungary and Poland, is partially a reaction to the overt accumu- 5. PROPERTY RIGHTS AND AGRICULTURE lation of wealth by a small elite. For capitalist-type exchanges to operate and Structural adjustment policies emphasize the need become widely acceptable both the polity and the to provide farmers with the permanent right to culti- society must reconceptualize the legitimacy of mar- vate and bequeath their land. To the World Bank kets. The expansion of market activity and the encour- secure rights will provide incentives for individuals to agement of investment and accumulation requires improve their land and “help rural credit markets to stability in the concepts which represent property develop, because land is good collateral” (World rights. Once a particular mode of thinking becomes Bank, 1989, p. 104). Once again, implicit in this analy- habitual, markets will operate with greater fluidity. sis is the universal neoclassical notion that efficiency Both the society and the polity will then be committed can only be achieved by ensuring that impediments to an acceptable form of property rights which will are removed to the rational decision making of self- ensure their reproduction. seeking individuals. In the atomistic world of neoclas- Structural adjustment largely misses this crucial sical economics the right to decide what, when and dimension of reform. Since the superiority of the pri- how to produce must be vested in individual produc- vate sector is axiomatic to the neoclassicals, resistance tion decisions. to privatization, which has been widespread in Africa, NIE also supports reforms that redistribute land to is deemed to be the product of entrenched interests individual owners, based on slightly different reason- where parastatal and government officials do not want ing. The notion of efficient pmperty rights in agricul- to forego opportunities for patronage and pilferage.6 In ture arises from the need to reduce transaction costs countries such as Tanzania, however, state ownership and to avoid principle-agent problems that can arise in and control have been part of the prevailing ideology other forms of agricultural relations such as hiring (ujumaa in this case). While the ruling elements have labor (Newbery, 1989, p. 288). utilized the ideology to enhance their hegemony over OJE, on the other hand, is more skeptical of the civil society, they have also legitimized state forms of capacity of shifts in landownership to have the effi- ownership. In Tanzania, the problem with inefficient ciency properties suggested by both the neoclassi- parastatal enterprises has not been perceived as being cals/World Bank and the NIE school. In particular inherent in the form of the property right but in the there is a skepticism that there is a singular relation- exercise of property rights. This is why the anti-cor- ship between the security of property rights and the ruption campaigns have been so popular in Tanzania patterns of the use of that property. and why the government has introduced so many In Africa, there is a variety of coincidental and minor organizational shifts to try to deal with the prob- competing claims based on clientage and kinship that lems. do not disappear after a shift toward private property Moreover, private ownership has been presented as rights. The literature from anthropology is replete with the antithesis of the national ethos. During colonial examples.* More. recently some economists have been times, legal barriers impeded the ability of Africans to deriving similar conclusions. Barrows and Roth engage in commerce.’ After independence, the choice (1990) draw on a variety of sources of illustrate that was often to recognize and extend the dominance of contrary to the neoclassicals, in Kenya, security of groups such the Asians in the economy by guarantee- tenure was not increased with titling, there was little ing their property rights or to utilize newly formed evidence of a correlation between tiding and overall bureaucracies to usurp those rights on behalf of the long-term investment, and a “well-functioning” land general African population. Thus, encouraging private market did not develop after titling with customary property rights has become synonymous with the law still determining sales and successions. negation of the economic rights bestowed on the local When looking. at land and property rights, one population by nationhood. All of this points to the needs to distinguish between the right to ownership, need in many African countries to consider less alien- the right to claim ownership, the iw urendi er obutendi ating paths of property right reform including (right to use and dispose of land) and the usufruct (the employee stock ownership, distribution of stock right to use and enjoy the fruits of the land). In the vouchers, or retaining formal state ownership with African countryside merely shifting to titled owner- THEORIESOF INS’IITUTIONS 1837 ship will have little impact on the other rights which operateto produceefficient prices. are a product of a complex web of social interaction. NIB is critical of this view of building markets. In The unification of ius #endi et abutendi and the particular, markets mqulre mom than simply the usufruct under a generally recognized singular right to absence of any hindrances to individual maximization ownership is the sin qua nob of the institution of decisions; markets require an instltutlonal stmctum private property. This, however, implies an entirely that supports the exchange process. As North puts it: different set of societal notms,values and st~ctures which entails much mom than new categories of Institutious provide the structure for exchange that legally detined property rights. The other fotms of (togetherwitb tbe technology employed) determinesthe socially defined pmperty rights are in many cases cost of transactingand tlte cost of transformation(North, 1990, p. 34). more legitimate than the new definitions of private property rights superimposed in the rural areas. As economies become more sophisticated the lnsti- Economic reform efforts need to fully comprehend the mtional structure of markets must also evolve. North basis of the existing legitimacy and the transformative distinguishes between three levels of market develop- prerequisites (and implications) of moving toward ment: personal&d exchange involving small scale new forms of legitimacy. Structural adjustment, with production and local aade, impersonal&d exchange its neoclassical concepts is not well-equipped for this that involves some long distance and crosscultural task. There are other problems with the neoclassi- trade; and the impersonal exchange of modern cal/World Bank view of the measures to develop mar- economies. The institutions of the first type focus on kets in Africa. repeat dealings and cultural homogeneity (common values). In the second case the exchange requires kln- ship links, bonding, the exchange of hostages or mer- 6. MARKETS AND PRICES chant codes of conduct. Finally, modern economies require third-party enforcement. Central to the success The discussion above focused on the role of of modern markets is the creation of a set of rules that property rights in the reform process. Central to the make a variety of informal constraints operational strategy of structural adjustment is the promotion of otherwise continual enforcement would make trans- efficiency through the encouragement of market actions too costly (North, 1990, pp. 34-35). prices which reflect their scarcity value: This view has a number of implications for the model of structural adjustment. North emphasizes the If the economy is producing efficiently, scarcity values gradual nature of the evolutionary process and the must be equal to opportunitycosts, and their common impediments that exist to rapid change: . value is the efficiency price...An economy is efficient, as opposed to just productionefficient, if it is impossible to ...institutions typically change incrementallytamer than make anyone better off without making someone else in a discontinuous fashion. How and why they chaage worse off. In addition to producingefficiently, the final incrementally and why even discoutim~ouschanges (such consumersmust have exhausted all possibilities of mutu- as revolution and conquest) are never completelydiscen- ally beneficial exchange. This in turn requires they all tinuous are a result of the embed&&e ssaadinformal face tbe same market prices and tbat tltese are equal to constraints in society (North, 1990,p. 6). efficiency prices...Tbe case for removing distortionsand moving market prices closer to efficiency prices rests on tbe argumentthat prices influence productionefficiency Resistance to change is even mom acute in the first and tbe nform will increase production efficiency type of exchanges which would characterize many (World Bank, 1983,p. 42). African markets where “[t]mnsaction costs are low but because specialization and division of labor is rudi- This revealing quote from the 1983 World mentary, transformation costs are hiw (North, 1990, Development Report, illustrates the World Bank view p. 34). What this suggests is that the time horizons for of the role of prices in reform. Markets will operate reform are much longer than those typically embed- with efficiency as detined by neoclassical pamto crite- ded in the targets of structural adjustment/stabilization ria, if certain distortions am removed. The source of programs. Second, different policies need to be the distortions are the state: designed for different market structures. In Africa, one often has the three types of markets North describes in In most iastaaces...price distortious are iutrcduced by a parallel existence.g If higher levels of growth are to govermnent directly or indimcdy in pursuit of some be sustained then one needs to examine not only how social or economic objective, sometimes deliberately, policies can assist markets at the three levels, but how sometimes incidentally (World Bank, 1983, p. 57). one can design policies that will expedite the transformation of type one and two markets into the third type The thrust of structural adjustment then, is to which would assist in the expansion of growth and remove the state’s interference so that markets can accumulation. 1838 WORLD DEVELOPMENT

Where NIE and the neoclassicals/World Bank cov- no impact on the real side of the economy and there- erage is on the importance of prices and their impact fore simply determines the price level. on the choices of individuals. This should not be sur- As we saw above a central focus of adjustment is to prising since both share the same microfoundations, provide price stability through proper regulation of the e.g. exchange is a product of the atomistic interaction money supply. The focus of financial liberalization of self-seeking individuals. In the case of NIE, how- should be aimed at strengthening the institutions that ever, institutional change is also the result of individ- can be used to control the rate of expansion of the ual preferences reacting to shifts in relative prices: money supply. This can best be achieved by allowing the commercial banking sector to operate free of gov- Institutions change and the fundamental change in rela- ernment intervention while building institutions such tive prices are the most important source of that as bond markets to control the money supply through change...relative price changes alter the incentives of open market operations. Beyond open market opera- individuals in human interaction (North, 1990,p. 84). tions, the monetary authorities should focus on regular audits and enforcing reserve ratios. State intervention As a result, NIB and structural adjustment would should be reduced by privatizing banks, prohibiting agree that “getting prices right” would assist eco- the government allocation of credit and subsidizing of nomic reform. In NIE’s case, however, the conduit for interest rates to prioritized sectors, and curtailing the improving conditions would include not only prices use of the commercial banking sector to finance the for current decisions on what to produce and consume government debt (World Bank, 1989, pp. 170-173). but prices that could be used to encourage more effi- In general the private sector should be used to funnel cient types of institutional transformation. investment funds to credit worthy individuals while OIE relies on different microfoundations which the state is restricted to properly expanding the money doubts the impact of getting prices right.‘OOIE rejects supply: the notion that prices are the ex-post product of the equilibrium of supply and demand and perfectly Banks or informal savings and credit associationsshould reflect scarcity value once hindrances are removed. be entrusted with the task of assessing the commercial Instead prices are seen as providing norms or conven- risks attached to individual requests. Monetary authori- tions. They are the product of historical time and ties will need to ensure that the pace of money creation is “depend in part on expectations and the legitimizing consistent with broader economic objectives. This equi- and informational functions of institutions” (Hodgson, librium should ideally be reached with interest and foreign exchange rates that clear markets and avoid the 1988, p. 187). need for rationing (World Bank, 1989,p. 171). Prices are only one aspect of markets. To OIE, markets are social institutions which structure, organ- NIE largely rejects the neoclassical view of money ize and legitimate contractual agreements and the as a “veil” pointing to the central role that credit and exchange of property rights. They not only provide finance play in firm investment decisions. Building on price conventions but are a means to communicate his critique of Modigliani and Miller, who argued, in information regarding products, quantities, potential very neoclassical terms about the unimportance of a buyers and potential sellers (Hodgson, 1988, p. 187). firm’s financial structure to their investment decisions, Thus OIE would concur with the NIB argument for Stiglitz (1992a) argues that the financial structure policies that are more broadly defined than the narrow affects the probability of bankruptcy, the perception of focus of structural adjustment on distortions. Since the potential profitability of the firm by possible investors, causal movement is, however, from institutions to the managerial incentives, tax liabilities, how the firm’s formation of price conventions, and not the reverse, managers were monitored, to some extent who con- OIE rejects the notion that a change of relative prices trolled the firm and the flow of funds under different will lead to some predictable more efficient economic exigencies (p. 17). outcome. NIE does not accept, (I priori, however, the superiority of market over nonmarket forms of capital and financial allocation. Stiglitz (1992a) rejects the notion 7. MONEY AND FINANCIAL INSTITUTIONS that the move historically to more market forms of finance (such as junk bonds) has necessarily provided Closely linked to the notion of prices and markets is a more efficient way of providing funds, lowered the concept of money and financial institutions. We transaction costs and increased the potential for risk saw above that money in the more relaxed neoclassi- diversification. Since privately motivated historical cal model is needed as a means of payment. To the changes have not necessarily been efficiency enhanc- neoclassicals, however, the real sector should be dis- ing government direction may be needed: tinguished from the monetary sector. Following Say’s Law, production is determined by the supply side of If, as we suggested, the evolution of financial institutions the economy. Money then is “a veil” which can have has entailed a movement from more to less control of bor- THEORIES OF INSTITUTIONS 1839

rowers...andif markets are not necessarily efficient...this as a unit of accotmt and appeared when private prop- may suggest a potential role for government interven- erty was loaned with the expectation of a return of a tion...(Stiglitz, 1992a,p. 30). sum exceeding the original loan. Thus unlike the neoclassical view of barter arising naturally out of the util- NIB rejects the treatment by neoclassicals of capital ity-maximizing behavior of individuals and money and financial markets as mere auctions. Banks do not simply evolving to facilitate exchange, we find that simply allocate credit to those that are willing to pay money was in existence well before the development the highest interest rate. Them are real questions con- of markets. Barter was never an important economic cerning the type of institutional arrangements that are activity nor did barter exchange lead to the develop- best suited to enhance the climate of investment. ment of markets. Thus money could not have devel- Stiglitz (1992b, pp. 181-183) views the Japanese sys- oped out of markets. This is quite clear from the tem with its interlocking directorships between banks research of the economic anthropologist Karl Polanyi: and other companies as one way of dealing with the multiple-principle agent problem (the manager is the Broadly, coins spread much faster than markets. While agent, the banks and the workers principles). Banks as trade was abounding and money as a standard was com- both the lender and shareholder are more likely to mon, markets were few and far between...Coinage had pursue actions that will improve the overall return to spread like wildtire, but outside of Athens the market capital to the group. habit was not particularly popular...As late as the begin- Stiglitz also warns of relying on indirect mechan- ning of the fourth century, the Ionian countryside pos- sessed no regular food markets. The chief promoters of isms such as market operations to control the volume markets were at the time the Greek armies, notably the of credit particularly in reforming economies with mercenary troops now more and more frequently high levels of uncertainty. He suggests the use of more employed as a business venture (Polanyi, 1971, pp. direct forms of allocation such as the central bank 83-85 quoted in Wray, 1990, p. 5). issuing rights to loans to commercial banks who would then have the option of interbank trading With the collapse of private property after the (1992b, pp. 175-176). Thus the government’s role as demise of the Roman empire, money and markets suggested above by the World Bank, of simply ensur- almost completely disappeared and did not reappear ing monetary expansion and market-clearing equilib- until the development of private property after the rium interest rates, ignores broader questions such as breakup of feudalism. To quote L. Randall Wray: the appropriate design of financial structures. OIE like NIE questions the noeclassical treatment Money as a medium of exchange becomes important of money as a veil: when workers are paid in medium of exchange rather than in wage goods directly. Since wage goods constitute ...the core of classical and neo-classical economic theory the majority of goods produced (particularly in the early has been the economics of neutral money. Like God in stages of development), it is tempting (but misleading) to Unitarian theology, money is there, but it does not do focus on exchanges and on money as a medium of very much. This is ironical since capitalism is above all a exchange [like the neo-classicals]. Once capitalist pro- monetary economy, yet it was presumedto behave as if it duction dominates the economy, money becomes univer- were a barter economy (Dillard, 1980, p. 256). sally important: money operates as a medium of exchange and money hoards provide a measure of secu- OIE goes beyond NIE’s focus however, on the role rity. However, production involves goods and services now in exchange for a promise to pay in the future. That of money and credit in the lirm’s investment deci- is, money is involved in the production process because sions. To OIE money is “the strategic institution of production is time-based and involves debt commit- modem capitalism” (Dillard, 1980, p. 255) and “is so ments. If one only focuses on the use of money in central to the determination of output that it might be exchange or as a store of value, one ignores how money represented as an institutional factor in the functional creation is inextricably related to time-based production relation between factors of production and output” in private property economies (1990, p. 10). (Dillard, 1980, p. 265). While the neoclassicals and NIE concentrate on To understand OIE’s concept of money one needs defining money in terms of its uses, OIE focuses on to focus on the evolution of financial institutions asso- the “source of money: how is money created, and how ciated with the rise of capitalism. What is absolutely does it enter a capitalist economy?“*1 Money to OIE is crucial to capitalist development is the movement to “any balance sheet item which transfers purchasing fiat money which is “currency issued by the state power across time” and “was created when private whose value is purely nominal” (Wray, 1990, p. 27). property arises and an individual becomes a creditor One of the important steps to developing fiat money or potential debtor” (Wray, 1990, pp. 2,s). was the evolution of giro banking institutions. A giro Money largely predated markets. Dating back to was a payment society whose members agree to the Greek cities, the earliest form of money was used accept credit issued by other member{ s 1 as a medium 1840 WORLD DEVELOPMENT of exchange and a means of payment. Typically an willingness of a lirm to enter into debt while money English exporter would sell wool to a Flemish supply is representative of the bank’s acceptance of importer for one-third cash and a bill of exchange cov- the IOU and to issue liabilities to purchase the IOU of ering the other two-thirds. The Flemish importer the firm. These are simply two sides of the balance would sell the wool and purchase a bill of exchange sheet. In the United States if it is profitable, the bank issued by a Flemish merchant who exported goods to will find the reserves to cover the additional loans by London. The bill of exchange could be retired and the using asset and liability management, the Fed funds London wool exporter could be paid once these market, international sources, or the discount window Flemish good were sold in the London fairs. (Wray, 1990, pp. 73-74). The central bank can make The early expansion of banks was linked to their this more costly by raising the discount rate or it can role as a guarantor of bills of exchange and as a trans- influence reserves via open market operations or fer point between debtors and creditors in giros. Once refusing to loan sums through its discount window. It bank debt became generally accepted as payment, the can hardly set the rate, however, at any specific target expansion potential of the giro was greatly enhanced (as monetarists would have it in their exogenous and trade would not be limited by the circulation of money-supply concept) and is likely to need to flood commodity money.‘* the system with reserves as lender of last resort if The demand of the state for revenue (in the early interest rates become too high. stages to finance wars) and the restrictions of relying Moreover, there is not likely to be any predictable on finance through commodity money led them in- relationship between money supply and inflation exorably toward entry into the giro network. Once the rates. To OIE, money cannot be neutral in a credit state accepted bank liabilities for tax payments from economy only influencing prices via the monetarist the general public, the liabilities issued by banks real balance adjustments. The concern for nominal became acceptable to citizens outside the giro, greatly monetary values does not arise out of money illusion expanding economic activity inside and outside the but the fact that credit-debt relations at the heart of state. The key to enhancing state power and modemiz- economic growth are denominated in money. ing the banking system was by organizing a central Neoclassicals tend to confuse money with the medium bank and enforcing a monoreserve system. Once fiat of exchange. In their view, an exogenously deter- money replaced commodity money the state could mined stock of money as a medium of exchange leads issue currency and increase spending without fear of a to a particular spending flow. One needs to distin- depreciation of the value of state money relative to guish, however, between money and the medium giro money. The transition to a modem financial sys- of exchange, money and spending, and the medium of tem based on credit creation and fiat money was exchange and spending. Money is created in the absolutely central to the development of modem process of facilitating flows and is representative of capitalism.13 the conditions of debt generation. The medium of This was not the product of the spontaneous evolu- exchange allows one to spend without incurring a tion of the private sector as some neoclassicals would debt. Money as a medium of exchange permits one to like us to believe but the product of conscious state- use someone else’s debt to make a purchase. Money directed policy intervention.14 Just as the state was can also be used by another as a means of payment to critical in the development of financial institutions settle debt, reduce balance sheets and destroy other during the rise of capitalism, it will be argued below money. This has absolutely nothing to do with the role of the state in finance must go well beyond spending. simply “guaranteeing” the money supply. Similarly, many types of transactions have little to As we indicated above, a core element of structural do with the broadly accepted medium of exchange adjustment, following the monetarist doctrine, was to (such as demand deposits and currency). Credit card ensure that the money supply expands at the rate of purchases involve the generation of debt which might real growth in the economy so that prices remain or might not be settled at the end of the month by the stable. A second important element was that invest- payment of demand deposits (Wray, 1990, pp. 14-15). ment cannot expand without an increase in savings Overall given the complexity of money (even where which will only rise if interest rates are greater than financial institutions are less developed as in Africa) zero in real terms. OIE questions both of these policies., there is likely to be a variable relationship between a At the heart of the OIE theory of money and finan- given stock of money as a medium of exchange and a cial institutions is the idea that money is basically particular level of prices and nominal expenditures.15 endogenous which means that loans make deposits, OIE is also critical of the structural adjustment/neo- deposits expand reserves and money demand induces classical view of interest rates and savings. First, as money supply. From the perspective of a firm, money mentioned above, behavior in economies is not inher- demand represents the inducement to go into debt ent as neoclassicals would suggest, but learned. while money supply is the IOU which it issues. From Neoclassicals would have us believe that banks the bank’s perspective money demand represents the evolved to act as the intermediaries to channel the THEORIESOF lNSlTHI”HONS 1841 deposits of savers to investors. From the OIE perspec- in spite of the rather problematic results of financial tive, Wray disagrees: reform under structural adjustment programs. In Nigeria, the deregulation of the fmancial system, [Tlhe true order of events show that orthodoxy clearly which was part of the structural adjustment package has reversedthe processthmugh which investmentis fund- (SAP) in 1986, led to a proliferation of new banks ed. Banks do not begin as intermediarieswhich acceptthe which tripled in number by 1991 (to around 120). deposit of “savers” and then make loans to “investors”, Most new banks undertook little or no lending to the for this would assume that the public has almdy devel- real sector with most loans financing interest rate and oped the “banking habit”. This habit is the end result of foreign exchange speculation and arbitrage. There public experience with short term bank liabilities which have been created as banks extend short term credit to was little or no relationship between savings and lend- finance working capital expenses (Wray, 1990,p. 58). ing rates. In response, the central bank in 1990 reintro- duced a spread between savings and lending rates. In Africa, the institution of banking, is not particu- Subsequently in 1991, maximum lending rates and larly well developed and in many areas is restricted to minimum deposit rates were imposed. With the demg- large urban centers. Arguing that raising interest rates ulation of the rates in 1992, interest rates on the inter- to positive real levels will lead to some predicted bank market soared to mote than 100% leading to a increase in savings is untenable to Oil even in the reimposing of interest rate levels in January 1994. context of more ubiquitous banking habits in devel- Deregulation in a poor supervisory climate, weak oped countries. Recently, the OIE position has been licensing procedures and undeveloped institutional buttressed by a number of empirical studies. arrangements (such as open market operations), have Dombusch and Reynoso (1993) surveying the litera- pushed the Nigerian financial system to the brink of ture on developed and developing countries conclude disaster with more than 50% of the banks either tech- “that there is virtually no evidence that higher interest nically insolvent or in distress by June 1994.” In contrast, OIE emphasizes the development of rates mobilize increased saving” (p. 71).r6 financial institutions which will ensure that credit The causal direction between savings and invest- flows into productive activities. As most of the Asian ment is reversed in the view of OIE. Spending on countries have illustrated, this requires more not less investment goods financed by credit creates profit intervention in financial markets with a careful main- income which becomes the basis of savings. Just as tenance of spreads between savings and borrowing capitalism requires that it not be constrained by com- rates, and the use of interest subsidies to priority modity money, savings cannot impede the production sectors. In Korea, interest rates were subsidized by as of investment goods (Wray, 1990, p. 58). What is much as 75% on loans to exporters. Both Korea and required is well-developed financial institutions that Taiwan carefully directed the flow of funds to targeted can provide credit for growth and accumulation. industries. In Taiwan, during the 197Os, banks Once again recent empirical studies have confirmed extended 75% of all loans to industries targeted by the OIE view of finance, savings and growth. government planners (Stem, forthcoming). As we Although Asia has had enormous success in raising have seen freeing up banks will do nothing to ensure savings, the evidence now indicates that it was not that credit flows into the priority sectors.‘* savings that led to the phenomenal investment and On the savings side, what is required is secure growth rates in the past few decades, but the rise of (government-backed) financial intermediaries with income that led to the very high level of savings. The low transaction costs for small savers. In East and World Bank (1993) in a test of the savings-growth Southeast Asia this has been accomplished through causality in Asia, found that in five countries the cau- everything from the widely dispersed postal pavings sation was clearly from growth to savings (Indonesia, system (Japan) to government workplace funds Japan, Korea, Thailand and Taiwan), in two it was (Singapore). The pattern successfully promoted sav- ambiguous (Hong Kong and Malaysia) and in one it ings and banking habits in the population which are was due to other factors (Singapore where the govem- important to the functioning of a modem banking sys- ment-sponsored provident fund was central). tem ‘(Stein, forthcoming). In contrast deregulation The new Bank study on adjustment in Africa recog- seems to have little impact in Africa. In Nigeria, nizes that low savings does not lead to low investment although there was a significant increase in the num- and slow growth and that the opposite is probably the ber of banks under deregulation, there was little case (World Bank, 1994, p. 156). This realization does improvement in the quality of banking services nor an not, however, seem to change their standard analysis increase in branches in rural areas (Ojo, 1991, p. 117). and policy recommendations. The Bank is still cap tured by the neoclassical view of financial repression 8. THEORY OF THE FLRM: which means deregulating financial markets and rais- STRUCTURE AND INNOVATION ing interest rates to positive teal levels (World Bank, 1994, pp. 114-l 15). The position has been maintained The firm in neoclassical theory has often been 1842 WORLD DEVELOPMENT described as a “black box” (Coase, 1992, p. 714). strain the ability to achieve global objectives. Since prices are the only element necessary to make To Williamson, both concepts are necessary to pre- production and consumption decisions, there was no sent the analytical complexity of his transaction costs apparent reason for the existence of the firm in main- construct. Opportunism is not sufficient to explain stream theory. Similarly, we have seen that the focus transaction costs since in a world of unbounded in structural adjustment includes price reform. In the rationality the information about which participants case of firms the primary aim is ensuring publicly were opportunistic would be known. Similarly with- owned corporations are privatized. The key is to make out the opportunism of participants in exchanges, the sure that production decisions are put into private costs of entering transactions with limited cognitive hands, or in other words, that the public “black box” capacity would also be minimized. becomes a private “black box.” According to Williamson, there are two branches of In the real world of modem capitalism corpora- transaction costs, the “governance” branch and the tions, not individuals, play the central role in the “measurement” branch. The former is concerned with production and distribution side of the market. the capacity of firms to organize transactions to adapt Understanding the nature of the firm and the role that to disturbances in the external environment of the it plays in innovation and growth would seem to be an firm. The second focuses on the capacity to bring important part of designing a strategy for Africa’s goods and services to the market at a cost which is jus- future development. tified by the price. Thus what links opportunism and To NIE corporations exist as a means of reducing unbounded rationality is uncertainty created by both transaction costs. Coase in his classic 1937 article uses the cognitive limitations of corporations and the a marginal cost approach in describing the decision of unforseen disturbances which create opportunities for using the market vs. the fhm in undertaking an addi- one party to the exchange to take advantage of the tional exchange transaction: other. The focus here is to create an organization which The question always is, will it pay to bring an extra can respond rapidly and efficiently to the shifting exchange transaction under the organizing authority? At external environment. This is complicated by the the margin, the costs of organizing within the lirm will be degree of “asset specificity” which refers to the extent equal either to the costs of organizing in another firm or to the costs involved in leaving the transaction to be “orga- that physical and human assets are tied to particular nixed” by the price mechanism. Business men will be transactions in economic organizations. The more constantly experimenting, controlling more or less, and transaction-specific assets in the organization the less in this way equilibrium will be maintained. This gives the the capacity of the firm to respond to uncertainty in position of equilibrium for static analysis (Cease, 1937, the face of opportunism and bounded rationality. P. 404). Williamson also recognizes that the “frequency” of transactions is also important since it allows one to Williamson (1985), takes the transaction cost con- spread fixed-cost governance structures over a greater cept one step further by formulating a theory to number of units (Lazonick, 199 1, pp. 206-2 13). explain the factors responsible for transaction cost dif- OIE is very critical of the focus on firms as transac- ferences. For this purpose, he draws on the work of tion cost minimizers. First, OIE rejects the concept of Herbert Simon and Kenneth Arrow. Arrow, a strong the firm as a calculating subject adding a structure on proponent of general equilibrium theory argued that the margin when the transaction cost warrants it (i.e. market failure can arise from the problem of informa- Coase’s quote above). Again this illustrates the neo- tion. In particular, since the price cannot capture all classical roots of NIE. Second, while pricing norms the information relevant to transacting parties, when are available to calculate transaction costs in a market the integrity of one of the parties is suspect transaction setting, it is much more difficult to calculate the costs costs will arise. Williamson terms this source of trans- of organizational structures in the nonmarket setting action costs as “opportunism.” Various governance of the firm. How can one make the kind of rational cal- structures of firms have the capacity to contain oppor- culus implied in NIE? tunism thereby economizing on transaction costs. To OIE firms are seen as social institutions which Without opportunism there would be no reason for the provide a refuge from the vicissitudes of the market. existence of internal organizations in firms. They embody the habits and routines that allow corpo- Simon, on the other hand, emphasizes that decision rations to deal with the complexity of production and making given the limited computational .capacity of exchange and to develop expectations of the future in the human brain and incompleteness of knowledge is a world of uncertainty: undertaken in a world of bounded rationality. Unlike the neoclassical maximizer including all possible The nature of the firm is not simply a minimizer of trans- information, Simon argues that decisions are taken action costs, but a kind of protective enclave from the from a small set of prerogatives. These cognitive lim- potentially volatile and sometimes destructive, ravaging itations also create transaction costs since they con- speculation of a competitive market. In the market the THEORIES OF INSTITLJTIONS 1843

rational calculus depends upon the fragile price conven- chanced upon or where the nature and application of tion which can often depend on ‘whim or sentiment or the project will be unknown in advance?) one cannot chance’. Habits and traditions within the firm are neces- point to the transaction costs that would be saved by sarily more enduring because they embody skills and using an internal corporate structure. He points to the information which cannot always or easily be codifiedor made subject to a rational calculus. what the tlrm role of the firm as a refuge for research and develop- achieves is an institutionalization of these rules and ment and the importance of the scale of operation: routines within a durable organizational structure. In consequence they arc given some degree of permanence The firm as a relatively durable organizational structure and guarded to some extent from the mood waves of is able to deal with the lack of knowledgeabout the future speculation in the market (Hodgson, 1988, p. 208). fruits of research and development and innovation. Its relative internal stability means that it can carry unqusn- tifiable risks which would be eschewedin the volatility of OIE also rejects the notion of hierarchy implied in the market. ln particular large tlrms arc able to set up and Williamson’s corporate structures of governance. The sustain R&D departments with their own funds. It is problem is that human behavior from Williamson’s widely recognized that atomized, small-scale private perspective utilizes a neoclassical view of individuals enterprise is not well able to make such long-term com- as acquisitive and self-seeking. Thus at all points one mitments (Hodgson, 1988, p. 213). will have to guard against the natural tendency toward opportunism. Lazonick (1991) also doubts the capacity of OIE points out that while opportunism exists, the Williamson to explain innovation in his framework. successful corporation would not rely on hierarchy To Lazonick, Williamson provides a theory of adap- and supervision but on the capacity to encourage other tive organization, one where firms react to a given human traits such as loyalty and trust. In simple terms economic environment as opposed to an innovative it would be impossible to supervise and monitor every environment where the firm attempts to change its activity in a corporate structure. This loyalty and trust economic environment. A crucial difference in the also spills over to the market place. In places such as two approaches to organization is that asset specificity Japan long-term arrangements between suppliers and in Williamson’s adaptive organization creates diffi- producers have avoided the uncertainty (and therefore culties in dealing with uncertainty (p. 218). In the transaction costs) of open market interactions more dynamic innovative organization, however, (Hodgson, 1988, pp. 209-210). This is one of the rea- asset-specificity is created by the organization and is a sons for the relative success of some forms of indus- symbol of success not failure. trial organization and arguably helps explain the Through the augmentation of fixed costs associated relative rise of Japan over the United States, where with asset-specificity, the innovative organization there is less reliance on trust and loyalty inside and chooses to create uncertainty with the knowledge that outside of the firm. it could produce a superior product at a competitive In Africa loyalty and trust have also played an cost (product innovation) or an existing product at important role in enterprise success. A case in point lower cost (process innovation). Higher fixed costs are can be found in the development of the auto-parts taken in order to reap the potentially higher generation manufacturing in the Nnewi Township in Eastern of revenue. The firm organizes its operations to deal Nigeria during the 1980s. Built along Hong with the productive and competitive uncertainty Kong/Chinese lines, cooperation was encouraged created with a potential innovation. Productive un- internally by incorporating family members into the certainty is internal to the firm and is linked to the management of the companies. Long-term relations, unknown impact of the innovation on new products developed in trading with Taiwanese businessmen and and methods of production. Competitive uncertainty among themselves, led to the trust and cooperation is external to the organization and is associated with needed to import Taiwanese equipment (avoiding the incapacity of the firm to know the availability of adverse selection) and the organization of a mutually factor supply and the demand for their products, both reinforcing interactive factory cluster (Brautigam, of which are necessary to reap financial returns from 1993).19 The failure to recognize the importance of the fixed costs. broader human traits and their role in the organization To reduce productive uncertainty the successful of firms is one of the weaknesses of NIE and neoclas- innovative organization will invest in a managerial sical economics. bureaucracy that is capable of the planning and coor- The organization of firms also has implications to dination of physical and human resources. In addition the question of innovation which is central to eco- the organization must develop not only the technical nomic growth and development. Williamson and the skills to deal with the innovation but also a “collective NIE have no adequate explanation for why innovation force” that “permits the planned coordination of the occurs in the firm. Hodgson (1988), emphasizes that horizontal and vertical division of labor required to since innovation cannot occur in a market setting (e.g. generate an innovation” (Lazonik, 1991, p. 203). This how could you design a futures market for something is best accomplished with the reduction of barriers to 1844 WORLD DEVELOPMENT

the mobility of labor within the firm, the free flow of pure neoclassical model as represented by Walrasian ideas between all levels of the company and incentives equilibrium, there is no need for a state since society’s that ensure that the participants receive the benefits of welfare is maximized. In the less extreme model of innovations (such as through long-term employment structural adjustment, the state is the guarantor of guarantees). The Japanese companies with their property rights and the money supply. Implicit in this absence of craft unions, the ringi system which notion is that the state will benignly intervene in these encourages the two-way movement of ideas up and matters. State intervention in any other matters sets up down the corporate ladder and the permanent employ- the opportunity for predation (following the public ment guarantees provide a quintessential example of choice literature) and is less superior than the opera- the “collective force” (Lazonik, 1991, pp. 39-43). tion of the market. To reduce competitive uncertainty the innovative NIE defines the state in more consistent terms. firm needs to push forward and backward integration. North (198 1) sees the state as an organization with a production facilities must be sufficiently large to comparative advantage in violence. This is important lower unit costs. Thus the firm will be tempted to if it is to enforce property rights since “the essence of expand into mass marketing to ensure that there are property rights is the right to exclude and an organiza- sufficiently large sales. It will also need to move into tion which has a comparative advantage in violence is material supplies to ensure a high quantity and quality in the position to specify and enforce property rights” flow of inputs (Lazonik, 1991, p. 204). Again the (P. 21). Japanese have been most successful at reducing com- OIE sees the state as playing a much broader role in petitive uncertainty with their vertically integrated support of markets than either the neoclassicals or enterprise groups which have evolved from the family NIE. To OIE markets did not arise out of the sponta- run zaibatsu of the prewar period. Large banks have neous activities of utility maximizing individuals but played an important role in reducing financial un- from direct state intervention. To quote Polanyi: certainty at the production and marketing stages of innovations. [W]e have evidence of organizational and financial activ- As African governments search for models of ities initiated by kings, general or governments respon- economic reform, they would be well advised to study sible for the military undettakings...Go-ahead generals the experience of institutions that have successfully devised up-to-date methods of stimulating local market spawned innovation and growth models that are con- activities, financing sutlers to wait upon the troops and engaging local craftsmen in improvised markets for the spicuously absent from structural adjustment. Lall supply of armaments. They boosted market supply and (1992) writing in the African context and drawing market services by all means at their disposal, however heavily on the experience of the newly industrializing tentative and hesitant local initiative sometimes may countries (NICs) in Asia, illustrates how inadequate have been. There. was, in effect, but little reliance on the adjustment is to the task of building a network of spontaneous business spirit of the residents (1971, p. 85 dynamic industrial firms in Africa. Along with incen- quoted in Wray, 1990, p. 5). tive structures (which are not necessarily those associated with free trade) Africa will require a broad array OIE believes that focusing on violence defines the of capabilities and institutions. Capabilities include state too narrowly. In particular it not only defines the ability to launch new investments, the provision of and enforces rights, but broadly supports the whole human capital and the capacity to collect and assimi- process of exchange and is thoroughly involved with late new technical knowledge. Institutions in the the social distribution of goods and services while also industrial context include, finance, information, ser- monitoring performance. Dugger (1992) argues that vices, standards, export assistance etc. Most of these the state and exchange as they developed through the capabilities and institutions are poorly developed or city-state to the territorial-state to the nation-state even absent in many African countries. Structural have evolved together in a mutually enforcing man- adjustment, however, with its narrow and flawed ner. To Dugger and the OIE “laissez-faire is a myth” emphasis on incentives misses this broader network of which has no historical basis (1992, p. 101). support. As La11puts it “this network is not thrown up The state has also not lived by coercion. It pos- automatically by the free market” but as with the sesses and uses its authority not due to force but due to Asian NICs it will require the broad intervention of the legitimacy. Following Commons’s emphasis on state (1992, p. 114). A final section will focus on the sovereignty OIE defines the state as “an agent that view of the state and the role it can play in supporting exercises sovereignty” (Dugger, 1993). the development of market institutions. We saw above that markets, according to OIE, are a social institution and require broadly defined forms of intervention in support of its operation, Even the so- 9. THE MARKET AND THE STATE called free-market experiments of the Thatcher and Reagen era involved continual juridical, political Strictly speaking, as we have seen above, in the and institutional intervention by the government. As . THEORIES OF IICXITUTIONS 1845

Hodgson, puts it “the main argument is not really tutions are discussed. The new institutionalists, who between intervention and non-intervention, but which use the same theoretical precepts as the neoclassicals, type of intervention is to be carried out and for see institutions reducing transaction and information which ends” (Hodgson, 1988, p. 253). costs. The old institutionalists detine institutions more Innovative organizations, vital to growth and broadly as “settled habits of thought common to the development, have thrived under a fostering environ- generality of man” and generally reject the neoclassi- ment supported by the state. In Japan the state has pro- cal emphasis on maximizing, atomistic agents. tected home markets to permit innovative industries to The paper looks at a number of institutions deemed attain competitive advantage. It has also maintained vital to the development of markets. NIE sees legally high levels of employment, and an equal distribution guaranteed property rights as vital to reducing transac- of income to encourage a market for manufactured tion costs. OIE interprets property rights in broader goods. It has created incentives for individuals and terms as a mode of thinking which is historically spe- companies to purchase goods that embody new tech- cific and not guaranteed merely through shifts in nology. It has limited the number of enterprises in legally defined terms. This is particularly the case in each industry to permit a sufficient market size for the African countryside where there am a companies to incur the fixed costs to make them com- variety of coincidental and competing claims to land petitive internationally. It has also encouraged cooper- based on clientage and kinship. ative research efforts among competitors and provided Both NIE and OIE are critical of the adjustment cheap sources of financing. Finally, it has also claim that markets will operate efficiently by simply invested heavily in educating its labor force which has removing hindrances to individual maximization deci- prepared it for the internal generation of innovations sions. Both see markets as broad institutional struc- (Lazonick, 1991, p. 37). Other Asian countries have tures that support the exchange process. OIE is, how- followed similar routes (Stein, forthcoming). ever, more skeptical of the impact of a change in In Africa, the state must be called on to build up the relative prices on the evolution of institutions as capabilities and institutions (broadly defined) needed suggested by NIE, arguing instead that the causal to promote market reform. This not only includes gen- movement is from institutions to the formation of eral investment in technology, education and infra- price conventions. structure but in the capacity of the state to selectively NIE and OIE reject the neoclassical notion that intervene to promote specific sectors or subsectors. If money is a veil and has no impact on the real side of Asia is any indication, the state has used a mix of guar- the economy. NIE focuses on the role of finance in antees, credit flows, subsidies, ownership and disci- investment decisions and questions the neoclassical pline to encourage the movement up the technological assertion that the movement to more market forms of ladder. While some existing coordinating bodies in finance is necessarily efficient. OIE goes beyond the Africa have been subject to rent-seeking and ineffi- role of money and credit in firms’ investment deci- ciencies, the focus should be on their reform and sions arguing that money is the most central institution transformation not the wholesale abandonment of of modem capitalism. While neoclassical economics the concept of intervention which is embedded in and NIE focus on the uses of money, OIE focuses on programs of adjustment. This is a long way from its sources. structural adjustment and its attempts to create com- While the neoclassicals focus on the evolution of petitive markets by removing the “distortions” created money as a medium of exchange arising sponta- by the state. neously to assist utility-maximizing barter activities, OIE points to the role of credit creation which has nothing to do with barter. To OIE money is endoge- 10. CONCLUSIONS nous, banking habits must be learned and do not come naturally, and investment generates savings. Thus The paper began with a discussion of the model of structural adjustment’s focus on the monetary rule and structural adjustment and its neoclassical economic real interest rates is likely to prove ineffective. To OIE roots. In the strict neoclassical model, there are no rea- the problems of finance in Africa are more deeply sons for the existence of institutions. In the relaxed structural and will require the state to find ways to version there is some need for a central authority to integrate its fiat money into national and international guarantee property rights and the stability of the giros. money supply. Following the neoclassical model, While NIE focuses on a transaction-cost explana- structural adjustment focuses on removing state-spon- tion of firms, OIE is more critical of the assumptions sored impediments to the private sector-driven market of the behavior of individuals as opportunistic in economy. State institutions should also be limited to describing the rationale for the hierarchical structure guarantors of the rights of private property and the of the firm. OIE emphasizes the role of trust and loy- money supply. alty in the organization both in internal and external Two competing perspectives on the nature of insti- company relations. Links based on encouragement of 1846 WORLD DEVELOPMENT these attributes are vital to the creation of the innova- ture the debate over the direction of policy in Africa tive firm which should be one of the prime considera- must be based on an historical understanding of the tions of reform in Africa. institutions underlying the development of markets. Finally, states play a vital role in the support and African countries should choose their future economic development of markets and other capitalist institu- strategy fully informed of the institutional options that tions. The state is the primary agent of institutional exist. Structural adjustment with its neoclasssical intervention. It has the capacity to stablize or trans- roots is rather ill-equipped to meet this challenge. form institutions including markets. At this vital junc-

NOTES

1. Elsewhere I have discussed some of the pitfalls of bas- might be some problems with elements of the microfounda- ing structural adjustment on neoclassical economic theory tions of institutionalism (e.g. no unified theory of price for- both in terms of the impact on industry (Stein, 1992), basic mation), the problem with NIB is that it is captured by the needs (Stem and Nafziger, 1991) and through the misinter- precepts of neoclassical theory which, we will argue below. pretation of the policy lessons from Asian development limits its understanding of how capitalism operates and by (Stein, 1994). This paper extends this critique by focusing on implication how to design institutions to build markets in the institutional implications of relying on neoclassical eco- African countries. nomic theory. 5. It should be noted that not all proponents of new institu- 2. Khan er al. in their 1990 article lay out the explicit mon- tional economics agree that efficient property rights will be etarist assumptions at the heart of the IMP’s model of stabi- created in practice. North (1989) argues that there is a ten- lization. In particular the rise in the money supply is a product dency for the state for political reasons to produce inefficient of the aggregration of the growth in foreign reserves (a positive rights. He refers to transaction cost and competitive con- balance of payments), the rise in private sector credit and the straints on polities which often produce inefficient rights. increase in public sector credit. Using the strict monetarist Transaction costs are related to principle-agent problems assumption that velocity is constant (or at least predictable), insofar as rights need to be defined in a manner which will the demand for money rises with the increase in money in- augment revenue collection by the state. Competitive con- come, and the money market is in flow equilibrium then it is straints arise due the need by political parties to avoid defin- easy to show that the balance of payments will improve with ing property rights in a manner which would antagonize con- a fall in private sector and public sector credit growth. The stituents (p. 665). Still, embedded in this framework is some model completely falls apart with any real world adjustments hypothetical measurement of efficient property rights which to the assumptions including the endogeneity of money, a would lead to maximizing behavior and higher levels of eco- lack of constancy or predictability of velocity, the intmduc- nomic growth. This concept is not found in OIE. tkm of other masons for demanding money (e.g. Keynes’s specu- lative motive), the existence of partitions between reserves 6. To quote a 1990 World Bank discussion paper observ- and the money supply (when you relax the assumption that ing the entire gamut of parastatal reform: the central bank is the only financial institution), etc. There am of course limits to the extent to which public sector institutions can be reformed. Established traditions 3. In discussing the limitations of the neoclassical concept and ways of doing business, vested interests of powerful of marginal utility, Veblen draws out the difference between groups and lack of pressure from users/clients for greater his and the mainstream approach: public accountability are factors that tend to delay the To any modem scientist interested in economic phenom- process of reform (Samuel, 1990). ena, the chain of cause and effect in which a given phase of human culture is involved as well as the cumulative changes 7. The extent of private property and markets in the pre- wrought in the fabric of human conduct itself by the habitual colonial period has of course been subject to extensive activity of [hu]mankind, are matters of more engrossing and debate. Authors such as Hopkins (1973) have compiled more abiding interest than the method of inference by which impressive data to present a complex economy with exten- an individual is presumed invariably to balance pleasure and sive trading networks, capital markets and general purpose pain under given conditions that are presumed to be normal currencies. While much of this places doubt on a strict sub- and invariable (Veblen, 1919, p. 240). stantivist interpretation of the period (that exchange was culture-bound) it hardly supports the opposite formalist extreme 4. The new institutionalists like to project themselves as of the universality of economic behavior (along the lines of being superior both to neoclassic& and the old institutional the neoclassical “horn0-economicus”). Iliffe (1983, pp. tradition. In a recent article on new institutionalism and 53-54), for example, argues that Robert Harms’s observation development economics the authors quote Langlois to but- that the Bobangi traders of the middle Congo perceived the tress their point “the problem with many early institutional- acquiring of wealth as a zero-sum game (it could only be ists is that they wanted an economics with institutions but acquired by typically sacrificing one’s relative to witches) without theory; the problem with many neoclassicists is that was fairly widespread in Africa. He also details the colonial they want economic theory without institutions; what NIB period’s disarticulation of indigenous accumulation, particu- tried to do is to provide an economics with both theory and larly in the rural areas. McCarthy (1982) systematically doc- institutions” (Nabli and Nugent, 1989, p. 1336). While there uments colonial policies such as stiff graduated licensing, THEORIES OF lNSTITUTIONS 1847 credit restrictions, controls on itinerant trading and the direct- is generally preferred. As a result external sources via aid or ing of commerce toward official markets which inhibited the balance of payment surpluses have provided much of the growth of indigenous accumulation in British Tanganyika. finance. The focus of economic reform in the financial sector The impact of these measures can be seen in the relatively should not be on interest rate or money supply targets as the low levels of participation of Africans (as opposed to Asians) neoclassicals would like but on developing endogenous in the ntail and wholesaling sectors at independence banking institutions that will lead to the acceptance of (Hawkins, 1%5). What one Sees at independence is generally domestically based fiat money. This will occur where state a weak indigenous private sector, without widespread money gets integrated into domestic and international giros legitimacy, which can do little to resist the policies of the and banking habits become more universally adopted. postindependence governments. 14. Wray (1990) has a detailed discussion of how the state 8. See. for example Shipton’s (1987) study .of the failed intervened with a heavy hand to centralize monetary control attempts to impose private property through titling in the Luo via the Bank of England (p. 54). areas of Kenya. 15. As Wray (1990) points out, even if one is to believe 9. See for example Meillassoux (197 1) for a rich depiction Friedman’s statistical correlations between money supply of the complexity and heterogeneity of markets in Africa. and spending, it says nothing about the direction of causality (e.g. endogenous theory would argue that spending generates 10. In the context of Asia, there is growing evidence that it money (p.82). was not market prices that were responsible for the “miracle” but the use of incentives and institutions. In Korea, for 16. For example, in the developing country context, example, the export boom was a product of such measures as Donibusch cites Giovannini (1985) and Fry (1988). tariff exemptions, income tax reductions, state financing Giovannini (1985) finds no significant relationship between of imports, favorable depreciation allowances, tax rebates on savings rates and real interest rates in Asian coqntries. Fry imports, subsidized interest rate loans, foreign currency loans (1988) finds in the Asian context, a significance which is and export insurance. The government set annual targets and quantitatively meaningless (an increase of 10 to 25 percent- used a combination of moral incentives and variations in sub- age points in the real deposit rate would raise savings by a sidization when levels fell below their targets. Agencies set mere one percentage point). up trade fairs, dispatched trader missions and hosted foreign business visitors. Private trade conglomerates were pro- 17. The Central Bank of Nigeria has estimated that it will moted to improve international marketing (see Stein, forth- cost 86 billion naira to reorganize the financial system. If the coming). In contrast adjustment in Africa has emphasized experience of the United States is any guide (the savings and massive devaluations to promote exports with rather Poor loan crisis of the 198Os),the final figure is likely to be many results. During 1980-90, the decade of adjustment, annual times the initial estimate. The information on the Nigerian export gmwth in sub-Saharan Africa was only 0.2%/ammm financial system is based on interviews in Lagos and Abuja compared to 6.1% in 1965-80. East Asia and the Pacific during June 1994 with officials in the Central Bank of exports grew by a very impressive 9.8% Per year over the Nigeria, the Nigerian Deposit Insurance Corporation, private 1980s (World Bank, 1992, p. 245). banks, state banks and accounting firms.

11. According to Siglitz (1992a. p. 20) the use of money in 18. Whether banks are state run or privately run is of less transactions is what distinguishes it from all other types of significance. In countries such as Korea there has tended to assets (“The transactions motive for holding money [is] the be a mix of private and state ownership. What is crucial has only motive that distinguishes money from other short term been the flow of funds to productive sectors. In the past, assets”). Neoclassicals with their emphasis on money as a African governments have not done a particularly good job in medium of exchange would have little difficulty with this this area. The answer is not, however, the deregulation of the functional focus. banking sector, but improving the state capacity to set priori- ties and ensure that incentives are designed so that credit 12. Commodity money is defined as “circulating currency flows to the priority sectors. These capacities are needed with whose value is determined by an objective measure - usu- a private or state owned system. ally weight” (Wray, 1990, p. 27). 19. These kinds of spontaneous developments in Africa 13. African countries, due to their colonial Period, inher- would seem to support the laissez-faire approach (although ited serious structural impediments. Prior to independence arising not from atomistic actions along neoclassical lines African countries relied on banks created by the colonial but group activities). Paradoxically, the macroeconomic power which typically maintained 100% reserves against any instability after 1988 brought on partially by structural currency issued. Currency could expand only if they were adjustment (e.g. devaluing naira made it difficult to plan a able to run a trade surplus (or if they were given grants or large investment based on imported capital goods, inflation loans from the home country). Money was thus largely in a was eroding savings, credit constraints made it difficult to commodity form. (Nafziger, forthcoming, provides a good obtain working capital from banking sources, etc.) was summary of the banking system in colonial Africa, contrast- threatening the growth of the cluster. Overall, given the pecu- ing it to the system under Meiji Japan.) In many ways in the liatities of the case and the threat to its sustainability this is postcolonial era, African countries have had difficulty bmak- unlikely to be the main route to industrialization. Brautigam ing out of this particular mode of money. Wbile the state can (1993) concludes there is no substitute for the state capacity freely issue currency, the fiat money of developed countries to reduce transaction costs and foster entrepreneurship. 1848 WORLD DEVELOPMENT

REFERENCES

Barrows, Richard and Roth, Michael, “Land tenure and Frances Stewart, Sanjaya Lall and Samuel Wangwe investment in African agricultum: Theory and evidence,” (Eds.), Alternative Development Strategies in Africa (New Journal of Modern African Studies, Vol. 28, No. 2, (1990). York: St. Martin’s Press, 1992). Brautigam, Deborah, “Substituting for the state: Explaining Langlois, Stephen, “The new institutional economics: An industrial development in Eastem Nigeria,” Paper pre- introductory essay,” in Stephen Langlois (Ed.), sented at the African Studies Association Annual Meeting Economics as a Process (Cambridge: Cambridge (Boston, MA: December, 1993). University Press, 1986). Coase, Ronald, “The institutional structure of production,” ’ Lazonick, William, Business Organizations and the Myth of American Economic Review, Vol. 82, No. 4 (1992). the Market Economy (Cambridge: Cambridge University Coase, Ronald, “The nature of the firm,” Economica, Vol. 4, Press, 1991). No. 16 (November 1937). Meillassoux, Claude (Ed.), The Development of Indigenous Dillard, Dudley, “A monetary theory of production: Keynes Trade and Markets in West Africa (Oxford: Oxford and the institutionalists,” Journal of Economic Issues, University Press, 1971). Vol. XIV, No. 2 (June 1980). McCarthy, D. M. P., Colonial Bureaucracy and Creating Dombusch, Rudiger, and Reynoso, Alejandro, “Financial Vnderdevelopment 1919-1940 (Ames: Iowa State factors and economic development,” in Rudiger University Press, 1982). Dombusch @cl.), Policy Making in the Open Economy, Mises, Ludwig von, Human Action, A Treatise on Economics Concepts and Case Studies in-Economic Performance (New Haven: Yale University Press, 1949). (Oxford: Oxford Universitv Press 1993). Nabli, Mustapha, and Jeffrey Nugent, “The new institutional Dugger, William M., “Transaction cost economics and the economics and its applicability to development,” World state,” in Christos Pitelis (Ed.), Transaction Costs, Development, Vol. 17, No. 9 (1989). Markets and Hierarchies: Critical Assessment (Oxford: Nafziger, E. Wayne, “Japan’s industrial development, Blackwell, 1993). 1868-1939: Lessons for sub&haran Africa,” in Howard Dugger, William M., “An evolutionary theory of the state Stein (Ed.), Asian Industrialization and Africa: Studies in and the market,” in William M. Dugger and William Policy Alternatives to Structural Adjustment (Basing- Waller @Is.), The Statijied State (New York M.E. stoke: Macmillan, forthcoming). Sharpe, 1992). Newbery, David, “Agricultural institutions for insurance and Ebohodaghe, J. U., “Regulatory regimes: An evaluation of stabilization, ” in Pranab Bardhan (Ed.), The Economic past and present,” in FITC (Financial Institutions Training Theory of Agrarian Institutions (Oxford: Oxford Centre), Deregulation in the Banktng btdustry: University Press, 1989). Directions, Challenges, Problems and Prospects, 1991 North, Douglass, “The New Institutional Economics and Bank Directors Seminar Papers and Proceedings (Lagos: Development,” Paper presented at the Conference on FITC, 1991). Public Choice Theories and Third World Experiences. Fry, Maxwell, Money, Interest and Banking in Economic London School of Economics,(London~ September 17-19. Development (Baltimore: Johns Hopkins University 1993). Press, 1988). North, Douglass, Institutions, Institutional Change and Giovannini, Alberta, “Saving and real interest rates in Economic Performance (Cambridge: Cambridge LDCs,” Journal of Development Economics, Vol. 18, University Press, 1990). Nos. 2-3 (1985). North, Douglass, “A transaction cost approach to the histori- Harms, Robert, River of Wealth, River of Sorrow, the Central cal development of polities and economies,” Journal of Zaire Basin in the Era of the Slave and Ivory Trade (New Institutional and Theoretical Economics, Vol. 141, No. 4 Haven: Yale University Press, 1981). (1989). Hawkins, H.C.G., Wholesale and Retail trade in Tanganyika: North, Douglass, Structure and Change in Economic History A Study of Distribution in East Africa (New York (New York: Norton, 1981). Praeger, 1965). Ojo, M. O., “Deregulation in the banking industry,” in FITC Hodgson, Geoffrey. Economics and Institutions (Cambridge: (Financial Institutions Training Centre), Deregulation in Polity Press, 1988). the Banking Industry: Directions, Challenges, Problems Hodgson, Geoffrey, ‘The reconstruction of economics: Is and Prospects, 1991 Bank Directors Seminar Papers and them still a place for neoclassical theory?,” Journal of Proceedings (Lagos: FITC, 1991). Economic Issues, Vol. 3 (September 1992). Polanyi, Karl, “Aristotle discovers the economy,” in K. Hopkins, Anthony, An Economic History of West Africa Polanyi et al. (Eds)., Trade and Market in the Early (London: Longman, 1973). EmpireslChicago: Regnery Co., 1971). Iliffe, John, The Emergence of African Capitalism Quarco, Philip, “Structural adjustment programmes in sub- (Minneapolis: The University of Minnesota Press, 1983). Saharan Africa: Evolution of approaches,” African Khan, Mohsin S. et al., “Adjustment with growth: Relating Development Review, Vol. 2, No. 2 (December 1990). the analytical approaches of the IMF and the World Samuel, Paul, “Assessment of the private sector, A case Bank,” Journal of Development Economics, Vol. 32 study and its methodological implications,” World Bank (1990). Discussion Paper, No. 93 (Washington, DC: The World Koslowski, Rey, “Market institutions, East European reform, Bank, 1990). and economic theory,” Journal of Economic Issues, Vol. 3 Shipton, Parker, “The Kenyan land tenure reform: (September 1992). Misunderstandings in the public creation of private prop- Lall, Sanjaya, “Structural problems of African industry,” in erty,” Harvard Institute for International Development, . THEORlES OF JNSTITUTJONS 1849

Development Discussion Paper, No. 239 (Cambridge, Blackwelt, 1992b). MA: Harvard University, 1987). Toye, John, “The new institutional economics and ita impli- Stein, Howard (Ed.), Asian Indusrrialization and Africa: cations for development theory,” Paper presented at the Studies in Policy Alternatives to StnrcrUral Aa&stment Conference on Public Choice Theories and Third World (Basingstoke: Macmillan, forthcoming). Experiences, London School of Economics (Londom Stein, Howard, “The World Bank and the application of September 17-19,1993). Asian industrial policy to Africa: Theoretical considera- Veblen, Thorstein, The Place of Science in Modern tions” Journal of International Development, Vol. 6, No. 3 Civilization and Other Essays (New York: Huebsch, (1994). 1919). Stem, Howard, “Deindustrialixation, adjustment, the World Williamson, Oliver, The Economic Institutions of Capitalism Bank and the IMF in Africa,” World Development, Vol. (New York: Free Press, 1985). 20, No. l(1992). pp, 83-95. World Bank, World Development Report (New York: Stein, Howard, and Wayne Nafziger, “Structural adjustment, Oxford University Press, various years). human needs and the World Bank agenda,” Journal of World Bank, Adjustment in Mica. Reforms. Results and the Modern African Studies, Vol. XXIV, No. 1 (March 1991). Road Ahead (New York: Oxford University Press, 1994). , Stiglitx, Joseph, “Banks versus markets as mechanisms for World Bank, The East Asian Miracle. Economic Growth and allocating and coordinating investment,” in J. Roumasset Public Policy (New York: Oxford University Press, and S. Barr @is.), The Economics of Cooperation, East 1993). Asian Development and the Case for Pro-Market World Bank, Sub-Saharan Afrca. From Crisis to Intervention (Boulder: Westview Press, 1992a). Sustainable Growth (Washington, DC: The World Bank, Stiglitz, Joseph, “The design of financial systems for the 1989). newly emerging democracies of Eastern Europe,” in Wray, L. Randall, Money and Credit in Capitalist C. Clague and G. Raussere @is.), The Emergence of Economies. The Emiogenous Money Approach Market Economies in Eastern Europe (Oxford: Basil (Aldershot: Edward Elgar, 1990).