Hemmed In: Responses to Africa's Economic Decline

Thomas M. Callaghy and John Ravenhill, editors

New York

Columbia University Press

1993

Bibliographic Data

11: Trading Places: Economic Policy in Kenya and Tanzania

M ICHAEL F. L OFCHIE

For more than a decade, scholars of African development have grappled with the elusive and complex relationship between what Thomas Callaghy has elsewhere identified as economic logic--"the policies needed to pursue economic efficiency"--and political logic--"the policies needed to assure domestic stability." ₁ The earliest scholarship on this topic emphasized the contradictory relationship between the two and the tendency for overriding political imperatives to yield poor economic policy. ₂ The key arguments in this literature can be quickly summarized. ₃ After independence, the need to create political support among urban pressure groups compelled political leaders to adopt policy frameworks marked by a sharp bias against agriculture. The purpose of these policies was to extract economic resources from the agricultural sector and shift them toward urban industries and clienteles that could provide stability for fragile regimes. Even though these policies have, almost everywhere, produced disastrous economic results, they have proven painfully resistant to change, for the constellation of interest group pressures and clientele demands that first led to their adoption has been sufficiently resilient to block major reforms.

The economic theme that dominates this literature is the notion of an ongoing, continent-wide policy bias against agriculture. Like any highly generalized social theory, this conception suffers from a number of striking shortcomings. ₄ Two of these are especially important. The first is the theory's inability to explain the great diversity of policy choices that were made following independence and the wide differences in agricultural performance that followed. In the midst of apparently universal agricultural decline, some countries have done far better than others and a few have even done sufficiently well to attract a designation as "success stories." The second shortcoming in the theory is its inability to explain policy change. Interest group approaches, such as that utilized by Bates in 1981, tend to suffer endemically from this weakness. They provide powerful explanations of the tendency toward the persistence of policy, but offer little guidance for an understanding of why some governments implement far-reaching policy changes, often within a short period of time. ₅

The theoretical point of departure for the first part of this chapter is the first of these weaknesses. It is taken up with an examination of the wide differences in the agricultural policies of Kenya and Tanzania throughout most of the post-independence period. The Ravenhill article cited above refers to the "West African wager" between the market oriented approach of Houphouët-Boigny's Côte d'Ivoire and the statism of Nkrumah's Ghana. ₆ East Africa also provides a variant of that wager in the contrast between the capitalist strategy of Jomo Kenyatta's Kenya and the socialist approach to development of Julius Nyerere's Tanzania.

The second part of this chapter takes as its point of departure the second weakness in the interest group approach, the inability to explain rapid changes in economic policy. It seeks to illuminate the reasons why Kenya and Tanzania have altered their approaches to the agricultural sector in recent years. For, during the 1980s, the two parties to the East African wager traded positions. Under President Daniel arap Moi, who succeeded to the presidency in late 1978, Kenya moved a considerable distance toward replacing policies of economic growth with those of redistribution. Under President Ali Hassan Mwinyi, who succeeded in the fall of 1985, Tanzania moved a considerable distance in the opposite direction, replacing policies of redistribution with policies of economic growth.

Comparing Agricultural Policy

The major economic differences between these two countries have probably been long familiar to most readers of this volume. They are quickly summarized in table 11.1.

Between 1965 and 1989, Kenya was able to attain an average rate of per capita increase in GDP of nearly 2%, compared with an average annual decrease for Tanzania of about 0.2%. Even this figure tends to understate the extent of Tanzania's economic decline. According to the World Bank, Tanzania had become the world's second poorest country, ranking only slightly above Mozambique. If Tanzania's GDP change were measured between 1965 and 1987, so as to rule out the economic gains caused by recent reforms, the average annual rate of economic decline was about 0.4%. Though both countries remained classified among the world's poorest nations, Kenya's GDP per capita at the end of the 1980s was more than triple Tanzania's.

Reduced to an elemental simplicity, the economic difference between Kenya and Tanzania was that between a country that enjoyed 25 years of agricultural growth sufficiently robust to afford important spillover benefits to other economic sectors, and a country that did not. The key to Kenya's positive performance in industry and manufacturing has been the productivity of its agricultural sector and especially the fact that agricultural exports generated sufficient foreign exchange earnings to permit the acquisition of the capital goods and other inputs that enabled these sectors to operate successfully. The key to Tanzania's inability to sustain positive growth rates in industry and manufacturing was the weak performance of its agricultural sector; the comparatively limited foreign exchange earnings it derived from agricultural exports meant a chronic and acute scarcity of capital and raw materials inputs for other sectors of the economy.

Kenya

Kenya must be reckoned among the most difficult countries in Africa to judge. By comparison with the economic performance of the vast majority of countries in independent Africa, it stands out as a rare example of successful economic development and constitutional stability. The distinctive feature of Kenya's economic record is its adoption, as early as 1965, of an economic strategy that anticipated many of the key ingredients in contemporary adjustment programs, including private ownership of the country's most productive economic assets, especially productive enterprises, a belief in individual incentives as stimulus for economic productivity, and a presupposition that the most important role for government was to provide a sound environment that would assure appropriate rewards for individual and corporate investors. ₇ Kenya's leaders worked assiduously to create a political climate in which entrepreneurship was regarded as socially beneficial and profit-seeking was considered an indispensable source of economic growth. ₈

Although Kenya's economic system offered great latitude for private entrepreneurship, it would be incorrect to conclude that the basis of its economic success was a laissez-faire approach. For alongside its deep commitment to the importance of private economic activity as an engine of growth was a heavy state involvement in economic management. This involvement could be observed at virtually all levels. The government set and rigidly controlled the country's exchange rate and other vital aspects of monetary policy including interest rates. It was heavily involved in the protection of domestic industrial enterprise through the management of international trade and accomplished this through a highly complex system of tariffs and other trade restrictions, all of which were implemented through a cumbersome system of trade licensing. Through various co-investment schemes, the government became an active partner and co-owner in many of the country's largest business firms, including banks. It fixed and regulated the prices of agricultural commodities at the producer, processor, and retail levels. And it monopolized the procurement and marketing of the country's major agricultural commodities through a system of parastatal corporations. ₉

The underpinning of government involvement in the economy, however, was a concern that its policies be sensitive to the importance of market considerations and a belief that government-controlled prices be as close as possible to those that might prevail under equilibrium conditions. This concern was especially clear in the case of Kenya's agricultural policy, which was fundamentally driven by the classic economic notion of comparative advantage. Kenya's political leaders believed that maximum economic growth and improvements in human welfare could best be attained by pricing policies intended to increase the production of the highest value exportable crops, coffee and tea.

This approach to development afforded Kenya such a strong record of economic accomplishment that it has sometimes been judged by different and more difficult criteria than are commonly applied to other African countries. Economists in the World Bank and the U.S. Agency for International Development (USAID), for example, have taken the view that, although Kenya's performance has been impressive by African standards, its growth trajectory does not begin to compare with those of the more rapidly developing countries in Asia. These economists argue that Kenya does not compete for foreign investment with Uganda or Tanzania but, rather, with Singapore, Thailand, and Taiwan.

Agriculture has been the cornerstone of Kenya's economic growth. The figures in table 11.2 help provide a more detailed portrait of the performance of this sector since independence.

This table reveals the results of Kenya's high priority on increasing the production of export crops. Between the late 1960s and the late 1980s, coffee production more than doubled and tea production increased almost fivefold. The importance of high-value agricultural commodities as the country's principal source of foreign exchange has become even greater in recent years because of rapid increases in the production of horticultural products for export. ₁₀ As Kenya's most lucrative exports, coffee and tea have been vital in providing the financial basis for infrastructural development in the agricultural sector as well as the wherewithal for high growth rates in industry and manufacturing. Kenya's average annual earnings from agricultural exports have recently climbed to about $600 million per year, a figure that has gone far toward ensuring that the country does not suffer from the acute scarcity of hard currency that has so constrained economic development elsewhere on the continent.

As the figures in table 11.1 also indicate, however, Kenya attained its greatest agricultural growth and highest rates of industrial growth during the decade immediately following independence. Not only did its agricultural growth rate decline during the 1970s and 1980s, but it is also now less than the rate of population increase, presently estimated at just under 4% per year. As agricultural growth has fallen, it has exerted a commensurate drag effect on the country's manufacturing and industrial sectors.

This decline can only be interpreted as a reflection of the Kenya government's reduced commitment to the policies that encouraged the maximum productivity of the agricultural sector. Under Moi, Kenya's commitment to the principle of comparative advantage has been diluted in at least two ways. First, there has been a tendency to treat export crop policy as a vehicle for economic redistribution. The present government has given greatly increased emphasis to encouraging the production of coffee and tea in nontraditional regions such as Rift Valley and Nyanza Provinces. Since these areas are less suitable for export crops and since the farmers there are less experienced in the cultivation of these crops, the result has been production of poorer quality commodities that do not realize high prices on world markets. The second change has been a tendency to place much greater emphasis on self-sufficiency in the production of food grains, especially wheat. Since grain production results in far less output value per unit of input, the emphasis on self-sufficiency in food has also resulted in a lowering of the agricultural growth rate.

The Moi government's emphasis on self-sufficiency in grain has been partly occasioned by the less impressive performance of Kenya's agricultural sector so far as food crops were concerned, as table 11.2 demonstrates. Although the production of corn, wheat and rice increased during the two decades from 1967-69 to 1986-88, these increases did not nearly keep pace with the growth of the country's population, which more than doubled during this period, from about 10 million to about 22 million. Indeed, only corn production came even remotely close to matching this population increase. As a result, grain imports, especially of wheat, began to escalate rapidly. These imports were relatively manageable because of a combination of factors, including the country's substantial hard currency earnings from the export crops and tourism and the willingness of the donor community to provide food aid. But, in recent years, Kenya has nevertheless been forced to allocate almost 12% of its total foreign exchange earnings, about $120 million/yr., for food imports.

The problem of greater and greater volumes of food imports had given rise to a debate within the Ministry of Agriculture as early as the late 1970s. Some agricultural planners continued to take a comparative advantage viewpoint and argued that Kenya should not alter its standing emphasis on agricultural exports since these continued to yield the greatest amount of economic output and foreign exchange per unit of resources expended. In their perspective, it would be economically unwise to shift scarce resources toward the production of grains. Not only is Kenya a relatively inefficient producer of corn, wheat and rice but, because of the glutted world grain market and the availability of heavily subsidized supplies through food aid programs, these crops are available from foreign suppliers at prices that are sometimes considerably lower than Kenya's cost of production. ₁₁ The position of this group was officially adopted in the Kenya government's public statement on food policy, published in 1981. ₁₂

The idea of comparative advantage has been increasingly challenged during the Moi administration by those who advocate a greater degree of food self-sufficiency. The proponents of this policy acknowledge that programs to increase domestic grain production would be economically costly. Since grain crops are import-intensive, increased food production would inevitably involve considerable costs in foreign exchange, as in the importation of energy, equipment, and chemical inputs. Proponents of food self-sufficiency, however, believe strongly that the net effect of increased grain production would be a considerable savings in foreign exchange through a higher and higher annual savings on expenditures for grain imports. By increasing grain production, they argue, Kenya would free many millions of dollars per year in hard currency for further agricultural and industrial investment and to finance badly needed educational and medical imports. Although the government's 1981 Sessional Paper has never been officially repudiated, the views of the self-sufficiency group have been increasingly decisive since the early 1980s.

The debate over comparative advantage versus greater self-sufficiency is politically important because it is part of a greater conflict over which Kenyans will benefit most from the government's agricultural policies, the traditional export crop producers of the Central Highlands, who are predominantly Kikuyu, or the grain farmers of the Rift Valley, who are predominantly Kalenjin. From an economic standpoint, however, the important aspect of this debate is the fact that it concerns which agricultural subsector merits highest priority, not whether to shift the country's national priorities away from agriculture. Even as Kenya has sought to use agricultural policy as a means of attaining greater regional equality in the distribution of the gains from agricultural growth, its overall goal has continued to be the well-being of the agricultural sector. This emphasis is vividly reflected in its exchange rate and producer pricing policies.

Exchange Rate Policy

Among proponents of structural adjustment, it has become axiomatic that a country's exchange rate policy provides the clearest litmus test of its policy toward the agricultural sector. An overvalued exchange rate imposes an implicit tax on the producers of export crops because it lowers the local price they receive for their commodity; it also imposes a tax on the producers of food staples by forcing them to compete with artificially cheapened imports. The tax on export production is of special importance to governments concerned with urban well-being. Since a large portion of this tax takes the form of a country's foreign exchange earnings, it can be used to finance the importation of capital goods, parts and raw materials for urban industries. Since these imports are artificially cheapened, overvaluation is a form of subsidy to both industrial entrepreneurs and industrial workers. Overvaluation also facilitates the acquisition and subsidization of consumer goods that today form a critical component of the urban lifestyle, further contributing to urban welfare at the expense of the countryside. It is not surprising that overvaluation is widely regarded as the first and most unmistakable symptom of urban bias.

There are two distinctly different viewpoints about overvaluation, however. According to economists who employ the concept of "public choice," a certain amount of overvaluation may be efficient at the early stages of a country's development. They argue that developing countries often lack the administrative capacity to collect more complex forms of taxation such as a sales tax or value added tax which would, in any case, involve considerable bureaucratic costs as well as a certain amount of corruption. ₁₃ Moreover, peasant producers are the most difficult segment of society to tax since they have low cash incomes and are widely scattered. Economists who take this position also argue that, since the wage sector of these economies is relatively small, it might not be possible to collect more complex taxes, such as income tax or value added tax, in sufficient amounts to finance necessary levels of governmental activity. ₁₄ In this perspective, a modest overvaluation may provide an efficient and necessary means of imposing a tax on a large, otherwise untaxable, segment of society.

Economists within the Bank, however, take the position that any amount of overvaluation inevitably introduces harmful distortions into an economic system since it lowers the production of tradable goods, discourages foreign investment and leads to a misallocation of scarce resources. They believe that the only appropriate exchange rate is one based on a policy of economic equilibrium; that is, a situation in which government controls are no longer required because the ratio between a local currency and the U.S. dollar is set strictly by market forces. For members of this group, the Bank's energies should be directed toward moving a country's official exchange rate upward until there is no longer any discrepancy between this rate and the informal currency market. ₁₅

Table 11.3 provides a comparison of official and unofficial exchange rates for the Kenya and Tanzania shillings from 1967, when they were created, through 1989. Kenya's exchange rate policy could be viewed in two ways. Compared with Tanzania, Kenya's exchange rate policy has been remarkably conservative. For most of the 23 year period covered by the table, overvaluation has generally been in the range of 20-30%, exceeding this amount only briefly during the early 1970s. Although this has imposed a price penalty on agricultural producers, those who view overvaluation as an efficient mechanism of taxation might not consider that this amount of overvaluation has been harmful to Kenya's development. It may also have been politically essential as a means of helping to enlist urban political support. This is a deeply controversial matter that is best treated as an open question: can Kenya's exchange rate policy be defended as a means of taxing the country's peasant population to generate resources for urban industries and services? Officials in Kenya hold this viewpoint and also point out that since exports of coffee, tea, and horticultural goods have generally increased during the 1980s, overvaluation cannot be criticized as a production disincentive. A recent study sponsored by the World Bank has supported that view affirming that Kenya had not greatly overvalued its exchange rate. ₁₆

Judged by the criterion of those who advocate an equilibrium exchange rate policy, however, Kenya's tendency toward overvaluation has always imposed a growth penalty on Kenya but only in recent years has this penalty begun to reach alarming proportions. For this group, the costs of overvaluation can now be readily discerned in the lowered growth rates of the 1980s. Those who hold this view also believe that Kenya's tendency toward overvaluation may also be important because it symbolizes a deeper economic malaise; namely, a decline in the quality of economic management.

Producer Pricing

The most useful point of departure for understanding agricultural producer pricing in Kenya is to recollect that the structural adjustment approach is deeply critical of the pricing policies of most African governments, holding that African countries have generally reduced production incentives by suppressing producer incomes. Kenya's status as independent Africa's clearest exception to this generalization has been documented by Cathy L. Jabbara, an economist with the U.S. Department of Agriculture. Summarizing the results of a study that surveyed Kenya's producer prices from 1972 to 1983, Jabbara drew the following conclusion: "Within the period covered by this analysis, Kenya has used agricultural pricing to create incentives for increased agricultural production and to meet its development goals of promoting smallholder production. This finding runs counter to the widely held notion that producer pricing in Africa is uniformly inimical to producer interests." ₁₇ Jabbara concludes that the welfare of producers, not that of urban consumers, was consistently the paramount consideration in the determination of prices for Kenya's principal agricultural commodities.

One of the most commonly accepted economic concepts for evaluating agricultural producer prices is the nominal protection coefficient (NPC), a standard that compares the domestic price of a given commodity to its world market price. The purpose of the NPC is to determine the share of a crop's value retained by producers. The World Bank has explained the NPC in the following terms: "The NPC measures production incentives in a relative sense, comparing producer prices with the maximum price . . . that could be offered to producers without subsidies. A low NPC means high taxation; an NPC of about one implies no taxation; an NPC greater than one indicates subsidies to farmers." ₁₈

This method of evaluating a country's producer pricing policies has certain imperfections. ₁₉ Since the world market price for agricultural commodities tends to fluctuate greatly from one year to the next, there are inevitably corresponding fluctuations in a country's NPCs without changes in domestic prices. And, in addition, NPCs do not reflect the various subsidies that agricultural producers sometimes receive. But its advantages far outweigh its shortcomings. NPCs not only provide an effective method for measuring changes in real producer prices over time but also are possibly the best means of comparing producer prices across countries that use nonconvertible currencies.

Export Crops Kenya's pricing policy for its principal export crops, coffee and tea, differed from that of the vast majority of African countries in that the government did not set fixed producer prices for these commodities. Instead, it adopted a pricing policy commonly referred to as a "throughput" system; that is, the world market price of these commodities was passed on directly to the producers after a certain percentage had been deducted to cover the operating costs of the statutory boards that handle the procurement, processing, and transportation of these goods as well as applicable export and local taxes. The critical question is whether these deductions have been so great as to constitute an onerous burden on the producer. The NPCs for coffee and tea from 1967 to1985 are presented in table 11.4, which reveals that Kenya's marketing and pricing system for coffee and tea has consistently passed a large proportion of the world market price for these products on to the producer. The producers' share of coffee and tea sales has averaged 70%. This helps to account for Kenya's success, as demonstrated in table 11.2, in boosting exports of these commodities. Between the late 1960s and the late 1980s, Kenya increased its coffee export volume by about two and a half times and its tea exports nearly five-fold. The high producer shares for these crops are all the more remarkable in light of the fact that the dollar prices for these products have been calculated on the basis of parallel market exchange rates, a factor that, by itself, accounts for a reduction of between 10% and 30% in the throughput to the producer in any given year.

Food Grains

Kenya's record in increasing grain production has not matched its performance with respect to the export crops. As the figures in table 11.5 reveal, however, the relationship between agricultural performance and pricing policy is less clear for these commodities than in the case of coffee and tea.

For the period from 1971 to 1986, the average NPCs for maize, rice and wheat have been 61%, 45%, and 95% respectively. The surprising element in this picture is the lack of any readily discernible correlation between the NPCs for grains and production trends. Maize, which had the lowest NPC had the greatest production increase whereas wheat, which had the highest NPC, registered the smallest. (See table 11.2.) These results undoubtedly reflect the extent to which the production of any agricultural commodity is heavily dependent upon a wide variety of nonprice factors including infrastructure, capital investment, the availability of extension and marketing services, and the suitability of the physical environment. The important conclusions sustained by the figures in tables 11.4 and 11.5 have to do with the extent and timing of the Kenya government's changeover from its emphasis on export crops toward a priority on grain production. The NPCs for food crops show that the increased emphasis on food production began in the late 1970s for wheat and in the mid-1980s for maize and rice. The striking feature of the NPCs for grains is the extent to which they have all increased markedly since 1983. As a result, the NPCs for rice, wheat, and maize now compare more than favorably with those for coffee and tea. Comparing the periods 1971-74 and 1983-86, the NPC for rice tripled; that for wheat nearly doubled and that for maize increased approximately 61%.

There could be an exogenous factor behind this trend. All three of these grains are in such acute oversupply in world markets that their international prices have dropped precipitously during this period. Even if the domestic prices set by the Kenya government had remained stable, falling world market prices would have resulted in increases in the nominal protection coefficients for affected commodities; (i.e., the local producer price would increase as a percentage of the falling world price). The magnitude of relative price shifts between exports and grains would be reinforced by the sharp drop in world prices of exportable commodities. This drop has been especially harmful to exporters of the beverage crops since the mid 1980s. Since Kenya's pricing system for coffee and tea passes on the shocks as well as the benefits of international price shifts for these commodities to the producers, it is to be expected that coffee and tea farmers would have suffered a loss in income relative to grain farmers in recent years.

The basic explanation, however, is that the administration of President Daniel arap Moi has made concerted efforts to increase the producer prices of grains. During the period from 1982 to 1986, the domestic producer price of wheat was increased 69%; that of rice, 121%, and maize, 45%. ₂₀ These percentages alone demonstrate the extent to which the government of Kenya has sought to balance Kenya's previous priority on export crops, with a new emphasis on providing high prices for grain producers. This policy shift has undoubtedly had much to do with the very high levels of grain imports and the foreign exchange burden of sustaining them. By the early 1980s, Kenya's grain imports had begun to average well over 400,000 metric tons per year. Another factor in the policy shift is that the political support base of the Moi administration consists overwhelmingly of the Kalenjin grain farmers of the Rift Valley.

Kenya and the International Lending Institutions (ILIs)

The broadest conclusion sustained by these tables has to do with the timing of Kenya's willingness to undertake structural adjustment. By the time Kenya began to negotiate policy reform with the World Bank in the mid-1970s, the essential lineaments of a growth orientation toward the agricultural sector were already in place. As a result, the dialogue between Kenya and the Bank that took place in the intervening period was, for the most part, one between two parties basically aligned on the purposes of economic policy. Although Kenya's relationship with the World Bank and International Monetary Fund has been subject to occasional moments of strain, these do not begin to approach the levels of the lack of consensus characteristic of the dialogue between the Bank and nonadjusting or late-adjusting countries such as Tanzania.

Kenya's formal dialogue with the international lending institutions can be dated from mid-1975, when it received its first policy-based loans from the IMF. Characteristically, the Kenya government had taken the initiative to set the stage for these loans by developing its own set of policy responses to the balance-of-payments crisis created by the oil price increase of fall 1973. In 1974, for example, Kenya began a program to boost its export earnings by providing a 10% subsidy on manufactured exports. ₂₁ This effort did not prove to be successful inasmuch as the cost of the export subsidies was well beyond the government's budgetary resources and because implementation was severely hampered by bureaucratic delays.

From the donor standpoint, however, the important feature of the attempt to boost exports was that it provided tangible evidence of Kenya's impulse toward economic reform. This impulse became even more apparent in mid-1975 in the government's official adoption of a comprehensive policy reform program designed to address the balance-of-payments problem in a manner that would ensure continued long-term economic growth. The basic strategy for accomplishing this objective was set forth in an important government sessional paper on economic policy published in early 1975. This document, which set forth a program of price, wage, and import restraints, had great importance as a direct signal to the ILIs of the government's willingness to take politically difficult steps, if these were necessary, to achieve a stable balance of payments. ₂₂ Sessional Paper No. 4 of 1975 sounds a theme deeply familiar to both proponents and detractors of structural adjustment in Africa; namely, the need to impose domestic austerity as the precondition for Bank assistance and long-term economic health.

During the next 12 years, the Kenya government successfully negotiated at least 10 policy-based loans from the IMF. The conditions attached to these loans are well known to observers of the structural adjustment process. The Fund required the government of Kenya to accept sharp limitations on its budget deficits so as to help reduce inflation; to continue to impose wage restraint on public sector employees; to limit the growth of public sector employment; to reduce public borrowing from the banking system so as to facilitate a heightened level of lending to the private sector; to increase interest rates so as to encourage a higher rate of savings and discourage consumption; and to take concrete steps to move away from import-substitution toward an export-oriented industrial sector.

In the early 1980s, the World Bank, which was then initiating policy-based lending programs in Sub-Saharan Africa, began to press actively for trade liberalization. It required the government to curtail its policy of controlling certain imports through quantitative restrictions and to substitute tariffs wherever possible so as to make its controls more transparent. In addition, the Bank's conditionalities were an important impetus toward increased producer prices for grains and toward reforms in the management of the agricultural parastatal corporations.

Bank officials readily acknowledge that Kenya has done much to comply with these conditions. ₂₃ Indeed, among some there is concern that precisely because of its deep commitment to structural adjustment, Kenya may have been subjected to conditionality stresses that overloaded its administrative capability. Kenya's record as regards exchange rate policy and agricultural producer pricing, for example, has already been shown to be exceptional. During the past decade, Kenya has also taken a number of steps to move away from a highly protected industrial sector based on import-substitution toward the production of manufactured goods intended for export. The Kenya government has complied with a number of the Bank's lending requirements: that bans on imports for protective reasons be abolished; that quantitative controls on imports be gradually replaced by tariffs; and that the tariff system be rationalized on the basis of an export orientation. To further reinforce the emergence of export oriented industries, the program for subsidizing exports of manufactured goods was improved and the subsidy level increased from 10% to 20%.

Kenya's principal failure has been its utter inability, under the Moi government, to maintain the prudent fiscal and monetary policies of the Kenyatta era. During the early 1990s, for example, the government's budget deficit grew to more than 5% of gross domestic product. To finance deficits of this magnitude, the government has resorted to increasing the money supply. But this has given rise to a high inflation rate, recently averaging about 20%, and seriously undermined efforts to move from negative to positive interest rates. ₂₄ Although the government has indicated its intention to reduce the budget deficit by controlling spending, there are serious doubts that it will be able to do so. For there are severe political pressures to use government employment as means of coping with the country's growing unemployment problem.

Kenya has also failed to make any appreciable progress in the divestiture of state-owned enterprises whose losses account for about one-fifth of the total budget deficit. In the spring of 1991, the government announced plans to divest approximately 140 state concerns. But implementation of this goal was painstakingly slow. Although there are a variety of perfectly valid economic reasons that can help explain this, among them that even in Kenya capital markets are not sufficiently strong to finance large-scale corporate acquisitions, the principal factors are undoubtedly political. The parastatal corporations have proven to be politically indispensable in the government's efforts to maintain a support base. The possibility that large numbers of parastatal employees could become unemployed if privatization were to occur poses huge political risks.

Thus, despite Kenya's generally favorable record of economic growth, it would be misleading to suggest that all is or has been well in the relationship between Kenya and the World Bank and IMF. Tony Killick has shown that Kenya's structural adjustment agreements have not always worked out successfully and that, on several occasions, these agreements have provoked severe tension. ₂₅ Structural adjustment credits were sometimes jeopardized because of disagreements over such matters as the extent or timing of currency devaluations, and on several occasions canceled because the Kenya government failed to comply with targeted ceilings for its borrowing from the banking system. In recent years, both the IMF and the Bank have become increasingly critical of certain aspects of Kenya's economic management. World Bank officials have been especially vocal in their criticism of Kenya's continued commitment to a parastatal system of grain procurement and distribution.

There is also sharp criticism of Kenya's rigid refusal to reform its bureaucratically cumbersome system of trade restrictions that classifies imports into various categories and imposes restrictions on various imports according to their classification. Bank officials also feel that although Kenya has made considerable progress in shifting trade policy toward a greater emphasis on exports, its reforms have proven insufficient to prevent an unacceptably high growth in the balance-of-payments deficits on trade. Kenya's international lenders are convinced that Kenya's system of trade management continues to give far too much control to state institutions such as the Ministry of Commerce and the Central Bank. In their view, Kenya's continued insistence on a bureaucratically managed system of trade is not only a serious constraint on industrial growth, but also an open invitation to bureaucratic and political corruption.

Tanzania

If the relationship between Kenya and the international lending institutions can at present be characterized as one of increasing strain between friends who once shared a common underlying commitment to growth-oriented economic policies, that between Tanzania and the lenders is its exact opposite. Between the late 1960s and the mid-1980s, a period when Kenya was able to negotiate and implement a series of policy-based loan agreements with the World Bank and IMF, the relationship between Tanzania and its international creditors was one of complete mistrust and fundamental dissensus over the goals of development policy. While Kenya's relationship with the ILIs ranged from cordial to mildly difficult, the pattern of Tanzania's relations with these institutions was a tortured one, with negotiations begun and broken off, agreements completed but never implemented, ceaseless bickering over technical matters such as the timing of Bank payments or the extent to which Tanzania was meeting its adjustment targets-all of this conducted in an atmosphere that was politically charged and often ideologically accusatory. ₂₆

The reasons for this strain are readily apparent. Between the mid-1960s and the mid-1980s, Tanzania identified and perceived itself as a socialist country whose principal developmental goals had to do with the creation of an egalitarian society and the elimination of class antagonisms. Because of the influential writings of President Julius Nyerere, Tanzania became a global exemplar of a development strategy whose primary objective was to build a communal society in which inequalities of wealth and power would be minimized. From Tanzania's standpoint, the World Bank and IMF were adversarial institutions. Their commitment to the implementation of growth-oriented strategies based on the operation of economic incentives and free market forces ran directly counter to Nyerere's belief in state-managed socialist equality.

From the standpoint of the Bank and the Fund, Tanzania offered an excellent example of why an internationally supervised program of structural adjustment was necessary. Tanzania's post-independence economic performance provided a perfect illustration of the Berg Report's contention that inappropriate economic policies produce ruinous effects. Bank officials believed that Tanzania should be judged not by the lofty idealistic goals of its philosophical president, but by its concrete economic performance and, by this criterion, everything had gone wrong that could possibly go wrong. Gross domestic product and per capita income had declined disastrously; production of exportable agricultural commodities had fallen sharply causing a major deficit in the country's balance of payments; excessive expansion of public services in the face of severe revenue contraction had led to excessive budget deficits and these, in turn, had contributed directly to spiralling inflation, declining per capita food production and corruption in the public sector.

Tanzania's relationship with the international lenders has been improving dramatically just as Kenya's has become more strained. The turnabout can be dated roughly to the presidential succession in the fall of 1985, when President Ali Hassan Mwinyi replaced Julius Nyerere, a change that signalled Tanzania's intention to initiate a major program of economic reform. The presidential succession was quickly followed, in the summer of 1986, by a structural adjustment agreement with the IMF. More important, it has brought about a gradual but extremely comprehensive process of economic reform. The difficulty Tanzania currently faces is that the economic crisis preceding its present reform efforts was so deep that restoration of a healthy, diversified, and growing economy continues to prove an elusive target.

Tanzania's Economic Crisis

The precise extent of Tanzania's economic decline is difficult to measure because of the extreme unreliability of statistical materials. For any number of reasons, it is especially important to treat statistics about Tanzania with great caution. There can be little doubt, however, that the Bank's basic image of Tanzania as having one of independent Africa's worst performing economies is substantially correct. The figures in table 11.1 of the Bank's 1992 World Development Report indicate that Tanzania's GNP/Capita in 1990 was the second lowest in the world, exceeding only that of Mozambique, which had been devastated by more than two decades of civil war. ₂₇

The aspect of Tanzania's economic decline that gave international lenders their greatest leverage was its acute balance-of-payments crisis, a factor that resulted in a growing debt crisis and extreme scarcities of foreign exchange. Tanzania's balance-of-payments deficits first began to escalate sharply in the early 1970s, at which time, however, they were of relatively manageable proportions averaging only about $120 million per year for the period 1970-74. During this time its debt service ratio (debt payments as a percentage of export earnings) averaged about 29%. By the early 1980s, however, the country's balance-of-payments deficits had skyrocketed out of all proportion to its capacity to service its debts, averaging more than $340 million/yr for the period 1980-84. The IMF estimates that Tanzania's debt service ratio in the mid-1980s had climbed to a staggering 66%. As a result of balance-of-payments deficits, Tanzania had begun to suffer acute scarcities of every commodity that had to be purchased abroad, ranging from vital necessities to everyday consumer items.

The major cause of Tanzania's balance-of-payments deficits was not declining terms of trade or falling world prices for its exports, but rather the problem of stagnation or decline in the country's production of exportable goods. ₂₈ Table 11.6 demonstrates the poor quality of Tanzania's agricultural performance with respect to its key exports. Of Tanzania's five most critical agricultural exports, only one, coffee, registered a slight production increase between the mid-1960s and the mid-1980s. This increase could be attributed largely to the vast inpouring of donor assistance to improve coffee production. Tanzania's other four key exports suffered decreases ranging from moderate to disastrous. By the early 1980s, cotton production, for example, had fallen to just below two-thirds of its 1970-72 peak; sisal production, Tanzania's key export in the 1960s, to only 28% of its 1964-66 peak; and cashew production, Tanzania's key export in the early 1970s, to only about 30% of its former level. Tea production had fallen slightly to about 95% of its late 1970s peak.

The contrast with Kenya could not be greater. By the mid-1970s, if not earlier, both countries had become primarily dependent upon coffee and tea exports as their principal source of foreign exchange earnings. Between the early 1970s and early 1980s, a period when Tanzanian tea and coffee export levels were flat, Kenya increased its exports of these crops by 300% and 200% respectively. Its tea exports during this period increased from about 43,000 metric tons per year to about 128,000, and its coffee exports from about 60,000 metric tons per year to about 120,000.

The ripple effects of declines of key exports touched virtually every sector of Tanzanian society. Industrial production, for example, plummeted because of acute scarcities of funds for the importation of capital goods, spare parts, and raw materials. According to the 1986 budget address by the Minister of Finance, the majority of Tanzanian industries were operating at 30% or less of installed capacity. ₂₉ Acute shortages of equipment and supplies also had profoundly negative effects on the country's infrastructure. By the early 1980s Tanzania's system of roads and railroads had deteriorated to the point where transportation bottlenecks alone constituted a major constraint on economic recovery. Indeed, labor and input shortages in both agriculture and industry could be accounted for by the sheer difficulty of transporting workers and vital inputs to different regions of the country.

Foreign exchange shortages also made a mockery of Tanzania's claim to have sustained a high physical quality of life index (PQLI) during this period. ₃₀ Throughout the 1970s, the notion that basic needs had to be provided for had constituted a major line of defense for proponents and supporters of Tanzania's sociopolitical system. But the country's unremitting balance-of-payments deficits badly undermined this argument. The scarcity of hard currency had drastically reduced the country's ability to import medical and educational supplies, which had a profoundly negative effect on the quality of services in its hospitals and schools. For all practical purposes, vast portions of Tanzania's medical and educational systems had ceased to function in all but name. Doctors at Tanzania's principal national hospital in Dar es Salaam were regularly reporting that even the most basic supplies were unavailable. Librarians at the university of Dar es Salaam reported comparable shortages in the acquisition of critically important journals and monographs. By the late 1970s, the day-to-day operation of clinics and schools in the countryside was even more severely curtailed and services had to be provided without even the most rudimentary of medical or educational materials.

To help resolve its balance-of-payments problems, Tanzania adopted self-sufficiency in food as an official policy. This policy was partly occasioned by the severe drought in 1974-75, which had resulted in the need to import huge volumes of food at considerable cost. In addition, there was a heightened concern that the only alternative to commercial food imports, food aid, would involve a heavy and unwanted political dependency on the principal food donors. Since the World Bank and IMF were already beginning to express reservations about Tanzania's approach to socioeconomic development, food self-sufficiency appeared to represent one way to avoid their early strictures about the need for policy reforms.

Since numerous other African countries have committed themselves to the same goal for approximately the same reasons, Tanzania's experience with this strategy has broad significance for the continent. There can be little doubt that the strategy significantly worsened the country's economic crisis by exacerbating its balance-of-payments difficulties. The reasons for this have to do partly with a widespread misconception about the nature of food production in modern Africa; namely, that food production requires relatively few imported inputs. While this may have been the case in traditional, subsistence-oriented village communities, the food systems of contemporary African nations are heavily import-intensive. Not only does the process of production itself increasingly rely upon such imported inputs as fertilizers and pesticides, but the supplemental processes of milling, packaging and distribution are also heavily dependent upon imports of equipment, energy, and materials. ₃₁ Self-sufficiency is thus, paradoxically, utterly dependent upon success in the export sector. And it was here, of course, that Tanzania had its worst failure.

Tanzania seemed, as a result, to suffer the worst of all worlds. Not only did its policy fail to halt the increase in food imports, but also the cost of food imports as a percentage of total imports climbed steadily. Table 11.7, which compares Tanzania's grain imports with Kenya's, demonstrates this problem.

This table makes it clear that although Kenya was paying a heavy and increasing price in grain imports for its emphasis upon an export oriented agricultural system, the burden of these imports on its national economy was less than Tanzania's. Tanzania's program of food self-sufficiency succeeded in lowering the volume of grain imports to a level substantially below that of Kenya, but the cost of food imports as a percent of total imports remained measurably higher. Despite efforts to achieve self-sufficiency, Tanzania's food imports as a percentage of total imports increased considerably between the two periods. ₃₂

There was, however, one critical difference: food imports had far greater political consequences for Tanzania than for Kenya. For Tanzania's dependency upon food aid gave the international donor community considerable leverage to influence its domestic policy orientation. And, by 1985, the vast majority of Tanzania's donors had made it clear that food assistance, like development aid generally, would not be easily forthcoming in the absence of a structural adjustment agreement. Although the financial leverage of the World Bank and IMF does not by itself constitute an adequate explanation for Tanzania's recent policy changes, it was an important element in empowering those Tanzanians who sought economic reforms.

Tanzania's economic crisis was so all-pervasive that it imperiled the country's major social objective, the creation of an egalitarian society. One of the most conspicuous symptoms of the country's economic decline was the emergence of an informal economy in which goods and services that could not be obtained in the official marketplace were openly traded. The most significant feature of the informal economy was that incomes earned there were neither regulated by the government's wage policy nor subject to the country's steeply progressive system of income taxation. Indeed, but for a few attempts by government agencies to compare the prices of goods in the formal and informal economies, the existence of an informal economy was barely acknowledged by the Tanzanian state. Hence, the traders, transporters, and independent producers whose incomes were derived from informal economic activity came increasingly to represent an economic class whose wealth was entirely exempt from the government's efforts to maintain social leveling.

It is impossible to determine with accuracy the absolute extent of Tanzania's informal economy, much less the level of wealth of the successful traders who operated within it, because the transactions that occur there are not recorded in official statistics. ₃₃ But there is little doubt that in certain major spheres of economic activity, the informal economy was of far greater dimensions than the legitimate or official economy. It became commonplace, for example, to speak of Tanzania as having two food delivery systems, only one of which was official. The official system operated principally through the National Milling Corporation (NMC), Tanzania's parastatal corporation for the procurement, processing, and marketing of grains, and provided food for the country's major institutional consumers. The informal system consisted of innumerable private traders and transporters who illicitly sold food staples such as maize meal to the overwhelming majority of Tanzanians who could not obtain them through official outlets.

Many of the transactions that Tanzanians had to undertake with the public sector were also illicit and involved that classic and irremediable source of informal sector wealth, corruption. By the late 1970s, Tanzanian informants often complained that virtually every interaction with a public sector official involved some sort of bribery: bribes were required to acquire passports, to purchase tickets on Tanzania Airways, to obtain school transcripts or other vital documents, and to have repairs undertaken at their state-owned apartments. ₃₄ Bribery was also a major factor in the process of obtaining government jobs, promotions, and special allowances. One of the great ironies, if not hypocrisies, of Tanzanian socialism was that some of the highest informal incomes in the country accrued to persons entrusted to administer the country's system of social equality.

The Policy Roots of Crisis

Tanzania's economic collapse followed directly from precisely the sort of inappropriate economic policies economic adjustment is intended to correct. Throughout the period of economic decline, Tanzania was a textbook example of such policy mistakes as currency overvaluation, a pattern of commodity pricing that suppressed incentives to agricultural producers, and a politically derived tolerance for poor management in its agricultural parastatal corporations.

Currency Overvaluation

Table 11.3 showed the extent of overvaluation of the Tanzanian shilling from the time of its creation in 1967, through May 1989. Throughout most of the 1970s, Tanzania's currency was overvalued by between 200% to 300%. In the early 1980s, however, the extent of overvaluation increased, and by 1984 the gap between the official and unofficial exchange rates was greater than 10 to 1.

The effect of overvaluation on Tanzania's economic development cannot easily be exaggerated. It was manifestly a factor in lowering the production of exportable agricultural commodities, while acting as a disincentive to local food production by cheapening the cost of imported foods. This helps explain the continuing high levels of grain imports even during a period when the government's official goal was food self-sufficiency. Overvaluation also contributed to the problem of balance-of-payments deficits since it reduced the cost of imported consumer goods, thus stimulating both demand and consumption. Overvaluation can also be held partly responsible for the constant hemorrhage of hard currency to external bank accounts in European countries and North America and hence for the country's chronic scarcity of foreign exchange. Overvaluation may also have retarded Tanzania's industrial growth by raising the real cost of wages, thus discouraging foreign investment in any enterprise that required a substantial labor force.

Price Suppression

Overvaluation of the exchange rate also contributed to the country's balance-of-payments problems by suppressing the real value of official producer prices for the country's key export crops. Table 11.8 compares the Tanzanian NPCs for coffee and tea with those of Kenya and also presents the ratio of Tanzania's producer prices to those of Kenya.

The figures in this table alone provide an adequate explanation for Tanzania's economic crisis. During the 1970s, the producer prices for the country's key export crops collapsed. By 1980, the local price being paid to Tanzanian coffee producers was only about one-eighth of the world market price for their crop and only about one-sixth that being offered to coffee growers in Kenya. Tanzanian tea growers did not fare significantly better, receiving less than one-sixth of the world market price and about 17% of the Kenya price. In the light of these figures, it is no mystery why Tanzania's export levels plummeted and the country experienced an ever worsening deficit in its balance of payments.

Parastatal Mismanagement

Until the initial implementation of its structural adjustment agreement in fall 1986, Tanzania was structurally identical to Kenya in administering its agricultural sector through a parastatal corporation that had a legal monopoly in the procurement, processing, and marketing of the country's principal commodities. Here, however, any similarity ceased. Whereas Kenya's agricultural parastatals generally performed well in passing on a relatively high proportion of the market price to their producer-clienteles, in expanding their capital base of transportation and storage facilities, and in providing a range of additional services such as the provision of inputs, Tanzania's did not. ₃₅ Indeed, the quality of management in Tanzania's agricultural parastatals was sufficiently poor to be ranked alongside currency overvaluation and price suppression as one of the principal causes of the country's agroeconomic decline.

The inadequacies in the performance of Tanzania's agrarian parastatals substantially corroborate the critique of these institutions articulated in the Berg Report. Official studies by the Tanzanian government demonstrated that the operating costs of these corporations tended to absorb a higher and higher proportion of their sales revenues. As a result, there were often no funds remaining to pay the farmers or else the payments were badly delayed or consisted of only a small fraction of the amount actually owed. ₃₆ The performance of Tanzania's agricultural parastatals was so consistently poor that one scholar has posited a "law" of rising parastatal marketing costs. According to Frank Ellis, the problem of nonpayment or late payment was so severe that Tanzanian farmers sought by any means available, including the informal market, to avoid the official marketing system. The result was that the parastatal corporations processed smaller and smaller volumes of the commodities for which they were legally responsible relative to their costs of operation which tended to remain stable or even to increase. The result was that the per unit costs of parastatal operations often rose precipitously, further aggravating the problem of nonpayments to farmers. ₃₇

Poor parastatal performance contributed to Tanzania's economic crisis in other ways as well, most notably through its impact on the government's budget deficits and, hence, on inflation. This occurs because the parastatals are governmental corporations and as such are entitled to call on the government to make up their annual deficits through budgetary appropriations. Since the Tanzanian government financed its deficits through a rapid expansion of the money supply as well as through heavy borrowing from its banking system, fiscal policy significantly intensified the country's steep rate of inflation.

World Bank personnel in Dar es Salaam generally considered the budgetary position of the parastatal corporations to be financially intolerable. The parastatal authorities were able to act as autonomous corporations with independent jurisdiction over their cost of operations.Therefore they had the latitude to raise their cost-base without any exercise of control by the central government. But, at the same time, they were able to take advantage of their official status as governmental agencies when it came to the funding of their perennial cost overruns. Before the signing of the IMF agreement, the parastatals' deficits became so huge that they accounted for a large proportion of the government's budget deficits.

The World Bank has sometimes used the metaphor of medical treatment to describe its relationship with ailing countries such as Tanzania: the Bank as doctor formulates a specific remedy to cure the illness of its country-patient. In this metaphor, there are clear and readily observed differences between the viruses that cause an illness and the symptoms that are its result. The daunting feature of Tanzania's economic crisis, however, was that each of its different facets was so closely interwoven with others that cause and effect could not always be clearly identified. Poor export performance, for example, contributed to the country's balance-of-payments difficulties but these, in turn, affected the production of exportable commodities by reducing the flow of imported inputs. Similarly, the balance-of-payments problems resulted in failing industrial performance that reduced the availability of capital goods and raw materials while also contributing causally to these deficits by increasing demand for imported goods.

This quality of interrelatedness makes the process of structural adjustment a difficult and prolonged one in Tanzania and other countries that have experienced severe economic decline. For it means that there is no single feature of the country's policy framework that a structural adjustment agreement can isolate as having special potential to cure the economic malaise. In this respect, structural adjustment does not lend itself to the medical metaphor. It not only involves a simultaneous assault on both the causes and the symptoms of a country's economic malaise, but also often proceeds without any certainty as to which is which.

Adjustment in Tanzania

Both the impetus toward structural adjustment and the implementation of a process of policy reform preceded Tanzania's August 1986 agreement with the IMF. An internal momentum favorable to liberalization of the country's economic policies had been discernible since the early 1980s and was a visible part of the Tanzanian political scene as early as 1982 when the Ministry of Planning and Economic Affairs published a document that committed it to an economic reform agenda. ₃₈ The first major steps toward structural adjustment took place as early as the summer of 1984, two years before the IMF agreement and about 18 months before President Julius Nyerere left office. Thus, it would be an oversimplification to suggest that the 1986 agreement was thrust upon an unwilling Tanzanian government as a condition of its grudging compliance in a process of policy change or that the process of structural adjustment had to await a presidential succession. Since the very beginnings of the country's socialist experiment, there have been a variety of political forces at work in Tanzania, some favorable toward and some bitterly opposed to a more liberal politicoeconomic orientation.

The difference in the mid-1980s was simply that leaders favorable to policy changes began to have a decisive influence over the policy process. This is by no means to suggest that the process of policy reform even now is smooth and systematic. Structural adjustment in Tanzania has had something of a lurching quality, moving ahead often abruptly in some areas while little progress is made in others. This is perhaps to be expected in a country in which the balance of political forces is shifting only gradually and in which powerful factions unfavorable to policy change continued through the early 1990s to hold important positions within the government.

The Tanzanian experience illustrates the most important and difficult aspect of structural adjustment; namely, that it is, in the final analysis, a supremely political process, one that requires the methodical building of a reform-oriented governing coalition. Because the economic benefits from policy reform sometimes require a considerable period of time to become apparent, this is often painfully difficult.

If a single date were to be chosen as the unofficial starting point of Tanzania's present program of structural adjustment, it would undoubtedly be the June 1984 budget address. That address announced a series of economic reforms intended to affect virtually every important facet of the country's economic life. Among the most significant policy reforms were a major devaluation of the Tanzanian currency, substantial real increases in the producer prices of agricultural commodities (particularly export crops), the phased elimination of agricultural subsidies both on food prices and for production inputs, the complete restructuring of the country's system of agricultural marketing through the reintroduction of producer cooperatives and a series of sweeping steps to liberalize international trade. Tanzania's policy changes have been so extensive that in some major areas such as grain marketing, foreign exchange management, and international trade, its policies are now substantially more liberal than those in Kenya. Although it is too early to determine the full effects of these policy changes on the country's economic performance, there can be little doubt that Tanzania has taken fundamental steps to set aside the statist system of economic governance it employed throughout most of the post-independence period in favor of a more market based economic approach.

The breadth of policy change in contemporary Tanzania is so great that it would be impossible to inventory all the changes that have taken place. It is more feasible to illustrate the pace and magnitude of changes taking place by focusing on selected policy reforms. Four areas of structural adjustment stand out as having special political and economic importance: exchange rate reform, budgetary reform, trade liberalization, and reform of the grain marketing system.

Exchange Rate Reform

Currency devaluation has been by far the most dramatic aspect of economic reform in Tanzania. The June 14, 1984 budget address announced a substantial devaluation of the Tanzanian shilling to take effect as of the following day. The devaluation was from TS 12.35 = 1$U.S. to TS 17.80 per 1$U.S., a drop of 44%. Although the Shilling was then frozen in value for 1985, the year of the presidential succession, it has since been devalued further. As the figures in table 11.8 reveal, Tanzania's exchange rate in April, 1989, was TS 135 per $1, less than one-tenth its 1984 value. In June of 1991, following a new IMF agreement, the exchange rate for the Tanzania shilling was further lowered to TS 230 per $1. While this meant a devaluation of nearly 2000% in only eight years, the gap between the official and unofficial exchange rates remained persistently large. With the dollar worth TS 365 in informal markets, Tanzania's currency remained about 60% overvalued throughout 1991.

To remedy this gap, the Tanzania government decided to alter fundamentally the mechanism for determining exchange rates. In June 1992, Tanzania followed the Ghanaian example and allowed the creation of private exchange bureaus that could buy and sell international currencies openly. This reform has had dramatic short-term effects and can be expected to have even more fundamental long-term consequences. Its immediate effect was to lower the Tanzanian exchange rate even further, to about TS 400 = $1. The long-term result may be even more consequential. The bureau system provides greatly increased insurance that exchange rates reform will be institutionally irreversible. For it not only takes exchange rate policy out of the hands of the Ministry of Finance and the Central Bank, but also creates an influential lobby of the owners, managers, and clients of the bureaus. This adds greatly to the long-term political credibility of the of the reform effort and can be expected, as a result, to stimulate entrepreneurial activity. ₃₉

Budget Reforms

Budgetary reform is a matter of vital political importance. For fiscal discipline gives a powerful signal of the government's willingness to sustain the politically unpopular austerity measures that are part of the adjustment effort. The Tanzanian government has performed less well in this regard than in devaluing the currency. The broad outlines of the problem are as follows. On the eve of Tanzania's experiment with socialism in 1966-67, government expenditures constituted approximately 20% of GDP and revenues, about 15.5%, thus producing a deficit of only about 4.5% of GDP. Twenty years later, when economic reforms began, the Tanzanian budget was about 28% of GDP, and revenues about 17%. This left a deficit amounting to about 11% of GDP. The first three years of economic reform did almost nothing to correct this problem so that Tanzania's budget deficit at the end of the 1980s remained at 11% of GDP. The government's principal problem did not lie with revenue collection, which increased about 335% during this period, but rather with its failure to curb expenditures, which increased by almost the same amount. ₄₀ As a result, inflation has remained high at approximately 20%.

Much of the government's budgetary difficulty was caused by its chronic inability to reduce subsidies to politically powerful parastatal corporations, especially the crop authorities. To underscore the depth of its commitment to structural adjustment, the Tanzanian government, in late 1991, announced plans for budgetary reform that if fully implemented would end the ability of the parastatal corporations to make seemingly limitless claims for governmental support. As of fiscal year 1992-93, expenditures for parastatal corporations must be approved as part of the government's annual expenditures estimates. ₄₁ This is a highly risky political strategy. Since parastatal corporations whose losses exceed the limits approved by the National Assembly can no longer count upon the Treasury to fund the difference, there is, for the very first time, the very real possibility of parastatal employees and suppliers going unpaid.

Trade Liberalization

The Tanzanian government has made significant progress in liberalizing its policies governing international trade. In early 1984, it introduced a program of "own exchange" imports, which has done much to ease the shortage of consumer goods in Tanzania and, thereby, to diminish the unofficial flow of privately held hard currencies to overseas accounts. Basically the plan allowed Tanzanians who held foreign exchange accounts overseas to import certain categories of goods on a no-questions-asked basis. While it was originally limited to a fairly small range of economically utilitarian goods such as pickup trucks and automobile spare parts, it proved so popular and so successful that it was gradually expanded to allow virtually any kind of import. Goods that have been imported under the own exchange scheme can be resold without any restrictions. Estimates of the value of own exchange imports vary considerably but governmental officials believe that between $300 million and $500 million worth of goods are imported annually through this program.

The government has also sought to ease the flow of imports by introducing a foreign exchange retention scheme, which allows exporters to retain control of a certain percentage of their foreign exchange earnings. The hard currency thus retained can be used to purchase inputs vital to continue the exporting process. In this scheme, the amount of foreign exchange that a given exporter might retain varied considerably, depending upon whether the export was a traditionally exported commodity such as coffee or tea, or a new export. The idea behind foreign exchange retention has been to encourage the development of new export markets by allowing successful exporting companies a high degree of flexibility in the management of their hard currency earnings.

Perhaps most important, the government has made major strides in eliminating its cumbersome and corruption-riddled system of quantitative import restrictions. As early as 1988, the government established an Open General License (OGL) system that allowed Tanzanians to import large volumes of permissible goods relatively freely. In 1991, this system was extended further by changing the OGL system from a list of permitted goods to a "negative" list, one that identifies only those items that are specifically excluded. By the end of 1991, the most significant deterrent to free trade confronted by Tanzanians was the difficulty of obtaining a foreign exchange allocation through the banking system. With the establishment of private currency bureaus, even this constraint has been largely eliminated.

Grain Marketing and Pricing

Structural adjustment in Tanzania has had perhaps its greatest achievements in the field of grain marketing and pricing. The Tanzanian government has both announced and implemented a whole series of measures to liberalize the procurement and marketing of agricultural commodities and to increase the price incentives for producers. Most of the major changes had to do with ending the monopoly status of the parastatal corporations. As early as 1984, the government announced that parastatal bodies would no longer enjoy monopsony status as the sole legal purchasers of specific agricultural commodities but, instead, that regional cooperative societies would be established and given permission to purchase directly from farmers. The government is striving to move toward an arrangement in which crops will be purchased from farmers by primary cooperative societies which will then transfer the crops to regional cooperative unions. These will then transfer the crops to marketing boards whose role in national and international marketing will be supplemented by private companies and individuals.

Similar measures have now been taken to end the parastatal monopoly over the purchase and distribution of agricultural inputs. By 1987, these functions had been opened to cooperative unions and private importers. To facilitate the entry of private actors into the agricultural services sector, such organizations were also made eligible for foreign exchange allocations. An even further curtailment of the role of the parastatal corporations had to do with their role as monopolistic exporters of the country's key agricultural commodities. As early as 1985, privately held sisal and tea estates were given permission to market their products independently, both to internal processors and to foreign purchasers. This reform was slower in being made available to coffee producers, but by the end of the 1980s it had been extended to coffee growers as well. Since producers of these crops are able to take advantage of the foreign exchange retention scheme, production has begun to respond not only to more favorable prices but also to the greater availability of imported inputs.

In the field of grain marketing, the government has basically legalized the dualistic system of purchasing and distribution that had already become a de facto part of Tanzania's economic reality. The most important change has been the government's official acceptance of the role of private traders in purchasing and marketing the country's basic food staples. The National Milling Corporation (NMC), formerly the monopsony buyer at official prices, has now become a residual purchaser obligated to purchase only those supplies of grain that are not channeled through private markets and cooperatives. To facilitate the participation of cooperative societies and private traders in the procurement and marketing of grains, the government has dismantled the system of restrictions that formerly prevented the private transportation of grains between regions. Although there has been some debate about whether private traders should be legally allowed to purchase directly from peasant smallholders or limited to purchases from village and regional cooperative societies, the fact is that private traders operate in virtually unrestricted fashion throughout both the countryside and the urban centers.

Tanzania has also committed itself to a policy of increasing the real producer prices of all of the country's agricultural commodities, both export crops and food staples. Although this commitment is as great as the commitment to structural reform, progress in this area has been less dramatic, and for many of Tanzania's key crops, real prices remained relatively stagnant during the early years of structural adjustment. There are a number of reasons for this: the world prices of Tanzania's key agricultural exports have fallen greatly in recent years and these pose an outer limit on the government's efforts to improve price levels internally. More important, the country's high rate of inflation has substantially eroded the nominal price increases offered by government purchasing agencies.

Of greatest concern, however, is that market forces now have a much greater role in setting producer prices in Tanzania. The nation's experience demonstrates that liberalization can reverse the pre-adjustment pattern in which a parastatal corporation, by suppressing producer prices, would gradually alienate its farmer-clientele and create an opportunity for private traders in the informal market. The difficulty with informal markets was that so much of the profit was made by the middlemen, the transporters and traders who bore the risk of bypassing the state marketing system. Although informal markets did deliver the goods, they provided little to help farmers while resulting in considerable price increases for retail consumers.

Legalization of informal markets has helped both farmers and consumers. As supplies of most of the country's agricultural commodities, especially grains, have increased, retail price levels have generally tended to fall in response. But legalization of private markets has, paradoxically, enabled farmers to benefit from the price-setting capabilities of state corporations. When the free market producer price falls in response to added supply, it is occasionally lower than the official price. As a result, the NMC, formerly avoided by farmers who regarded it as an agent of governmental exploitation, has occasionally become a preferred purchaser, especially in the more distant regions of the country where private traders are reluctant to operate.

In sum, Tanzania's economic reforms have made it, in certain major respects, a more liberal society than Kenya. Its currency is now traded freely whereas Kenya's remains governmentally managed. Its export and food crop producers are much freer to bypass government purchasing agencies and market their crops privately or through cooperatives, whereas Kenya's are not. Its parastatal corporations have been put on strict notice that losses must be contained within limits and that staff reductions are inevitable for those that cannot do so whereas Kenya's parastatals, especially in agriculture, continue to depend heavily on government support. And its trade policies have become far more open than Kenya's.

Trading Places: Political Change in Kenya and Tanzania

We will now examine the process of policy change in contemporary Kenya and Tanzania. Our theoretical instruments for this purpose are regrettably weak. As Grindle has suggested, the advantages that first attracted development scholars to political economy-its ability to explain why inappropriate economic policies are first introduced and why they then persist for so long-are critical weaknesses when it comes to explaining why policies change. ₄₂ No single theorem suffices. Policy change, like policy diversity, can at present best be understood as the product of a variety of factors.

One of these is contextual. The longstanding economic differences between these two countries have had a considerable bearing on the nature and course of economic changes during the 1980s as well as on the extent to which they have been prepared, in recent years, to undertake structural adjustment. As in so much of their political history, the two countries tend to mirror one another. The present government of Kenya has been profoundly aware of its international reputation as an African success story much favored by Western bilateral donors, international lenders, and multinational investors. The administration of President Daniel arap Moi has behaved as if this legacy could be taken for granted and has not been greatly preoccupied with improving upon the growth-oriented policies of the past. Instead, it has identified economic redistribution as its primary agenda. As a result, some of the more recent policy changes in Kenya have introduced a pronounced tendency toward economic stagnation.

The government of Tanzania, on the other hand, has been fundamentally motivated by the urgent need to improve upon the economic record of the past. The administration of Ali Hassan Mwinyi, painfully aware that twenty years of socialism produced a dismal economic record that had become increasingly unacceptable to the majority of the Tanzanian people, therefore signed a structural agreement with the IMF in the summer of 1986. This government has identified as its highest political priority the introduction of a set of policy reforms that will produce economic growth. It has sought so assiduously for ways to eliminate political and institutional blockages to economic growth that the pace of policy change outstrips that in Kenya.

The figures in table 11.1 provide a partial economic reflection of the changes in the two countries' approaches to development policy during the 1980s. The recent economic story is clearest for Kenya whose growth rates for agriculture, industry and manufacturing for the period 1980-90 have all declined substantially from the early post-independence period. Kenya's rate of agricultural growth fell by 34%, industrial growth, by almost 60%, and manufacturing growth, by 53%. Although Kenya's growth rates remain higher than Tanzania's, their trajectory suggests that this may not long remain so. The analytic challenge posed by these figures is relatively simple: identify the policy changes that have induced economy wide tendencies toward economic decline.

The economic picture for Tanzania is more complex. Agricultural growth has recovered dramatically during the 1980s and now substantially exceeds Kenya's. Indeed, because agriculture accounts for such a large proportion of the Tanzanian economy, improved agricultural performance during the second half of the 1980s contributed to a positive overall per capita growth rate for the first time since the economic decline of the 1970s. Tanzania's industrial and manufacturing sectors, however, have continued to perform poorly, suggesting that Tanzania, like so many other societies that have recently embarked on structural adjustment, is plagued by the seemingly pandemic problem of a slow supply response. The intellectual challenge posed by Tanzania is to provide a political answer for why industrial recovery has lagged so far behind improved agricultural performance.

Kenya

One important lesson to be derived from a comparison of Kenya and Tanzania is that the benefits of a market-friendly approach to development have much to do with the timing of its introduction and the duration of a government's commitment to it. After independence, Kenya adopted economic policies that gave prominent importance to individual economic incentives and remained basically committed to those policies for twenty years. The sheer longevity of the government's commitment to prudent policies that created a favorable macroeconomic environment has been a vital factor in Kenya's ability to sustain a positive record of economic growth. For it has meant that sound economic policies were by and large predictable and dependable. Unlike Tanzania, Kenya did not have to turn to structural adjustment 25 years after independence as a desperate means of restoring health to an economy badly undermined by a set of egregiously inappropriate economic policies. Nor has Kenya had to deal with the task of building a completely new political climate, one that treats entrepreneurship as economically beneficial rather than socially exploitive.

Despite its long record of economic growth, Kenya's policy framework no longer satisfies the structural adjustment metric of the World Bank whose growth model features a minimalist state that allows maximum latitude for free market forces. Judged by that criterion, the statist side of Kenya's economic system has always left a great deal to be desired. But worsening economic performance has given the Bank and other donors a stronger basis for insisting upon changes in areas where free market approaches are likely to produce better economic results. The reason has to do with the economic downturn that has occurred during the Moi administration. As long as Kenya's special mixture of statism and market-based approaches to development was economically successful, a period that roughly coincided with the Kenyatta years, its system was relatively immune to the Bank's criticism. Indeed, Kenya's economic successes gave the Kenyans a good deal of independent leverage in their negotiations with the Bank and this leverage was consistently used to resist its demands for structural changes. Now that Kenya's system has begun to perform poorly, its interventionist approach to economic management is far more vulnerable to the Bank's insistence upon economic reforms that will drastically reduce the government's economic role.

Kenya's critics in the World Bank point reproachfully to a whole range of economic policies they would like to see structurally corrected. The exchange rate, for example, is still set by the government and private transactions are strictly proscribed; hard currency is allotted to licensed importers through a cumbersome, time consuming, and corrupt bureaucratic process; the procurement, processing, and marketing of all of the country's major agricultural commodities are still conducted almost entirely through a system of monopolistic parastatal corporations; the producer, processor, and retail prices of the country's basic food staples (maize, wheat, rice, and sugar) are also set by the government and, in general, strictly enforced; the country's import and export trade are also strictly regulated and trade licenses are still required for all imports. Kenya's industrial system also contravenes the Bank's strong commitment to free trade inasmuch as it has been based largely on the principle of import-substitution with domestic manufacturers given varying degrees of protection from foreign competition.

There is little doubt that the economic costs of these heavily statist systems of regulation and control have become more and more visible. Since the early 1980s, the quality of Kenya's economic management has deteriorated visibly. As a result, World Bank economists now place even greater emphasis on the urgent need for timely reforms of its highly bureaucratic system of economic controls. No small part of the Bank's urgent insistence on privatization in Kenya stems from a concern it shares with numerous foreign donors that an economic system that performed well for the first 20 years after independence is being increasingly subordinated to political expediency in ways that conflict with economically sound management. There is a growing consensus that, under President Moi, Kenya is squandering its precious legacy of economic credibility.

The conspicuous falloff in Kenya's recent growth rates is partly related to a dramatic reversal in its ability to attract foreign capital. Whereas Kenya was among Africa's most popular countries for external investment during the 1960s and 1970s, such investment has now all but disappeared and, according to recent reports, Kenya has recently begun to experience critically serious problems of capital flight. ₄₃ That flight, however, is itself reflective of a deeper malaise.

To understand the contextual political roots of economic decline in contemporary Kenya, it is essential to begin with the fact that the benefits of the two decades of booming agricultural and economic growth following independence were not uniformly distributed throughout the country. ₄₄ Take, for example, the cultivation of coffee and tea. Because of a combination of ecological factors that included suitability of soils and the geographical distribution of its annual rainfall, as well as historical considerations that included colonial policy and the regional preferences of colonial settlers, those crops are grown principally in the Central Province. The major African beneficiaries, as a result, were the Kikuyu people or, more specifically, the land-owning stratum of the Kikuyu people, who benefited not only agriculturally but also in such spheres as education, government service, and business. ₄₅

The Kenyatta government did little to ameliorate the country's ethnic inequality. Its political record suggests, instead, that it viewed its principal task as one of ethnic management and cooptation: maintain political domination by the Kikuyu elite while, at the same time, allotting sufficient opportunity for political and economic advance to individual members of other groups to avoid severely destabilizing ethnic confrontations. This is not to suggest that the Kenyatta administration was able to avoid political pressures for a more equitable distribution of the country's wealth. Such pressures were sometimes severe. ₄₆ But it was able, through adroit manipulation of the rewards and punishments available to it, to contain these pressures within the broad political framework of a constitutional and relatively pluralistic single-party system.

Toward the end of the 1970s, however, the pressures for economic redistribution became especially intense. The principal cause of this intensification was an exogenous event-the beverages boom of 1975-77. Since Kenya's pricing policy for export crops was to pass on as much of the world price as possible to the growers, the benefits of booming world prices for coffee and tea during this period meant little to most of the country and became a huge economic windfall for export-oriented Kikuyu farmers. ₄₇ As a result, the widespread resentment among other ethnic groups of Kikuyu wealth and power was abruptly intensified on the eve of Moi's succession. Chege, commenting on Moi's succession, notes that "he inherited a country seething with popular discontent produced by the political and economic domination of national affairs by the Kikuyu." ₄₈

The demands for economic redistribution Moi inherited from his predecessor fitted perfectly with his personal political agenda. Recall that he had always been politically motivated by resentment of Kikuyu wealth and power. He had begun his political career in the early 1960s as national chairman of a political party called the Kenya African Democratic Union (KADU), an organization best understood as a loose alliance of Kenya's ethnic have-nots drawn together by fears of Kikuyu politico-economic domination. Moi joined KANU only after KADU's dissolution in 1964 and his long career within KANU is both a tribute to the Kenyatta government's powers of cooptation and to his own patient willingness to occupy the political sidelines.

Since Moi's personal resentment of Kikuyu preeminence was widely shared throughout the country, he could potentially have gathered a governing coalition that comprised an overwhelming majority of the population. The supporters of economic redistribution included not only the have-not regions of the country that had formerly supported KADU, including Coast, Rift Valley, and Western Provinces, but substantial elements of Kikuyu society itself, such as the large landless population in the rural areas, low-paid agricultural and industrial workers, and the massive Kikuyu urban underclass.

During its earliest years in office, the Moi government undertook a number of steps intended to create and solidify a broad multiethnic coalition for socioeconomic reform. Admission to the university, for example, was no longer to be strictly by merit--a system that had decidedly favored the Kikuyu graduates of the country's best high schools, which were overwhelmingly located in Nairobi and the Central Province--but by merit within regions, with each region entitled to a proportion of the entering class according to its population. High schools in the various provinces were also required to restrict admissions to children of indigenous residents, thereby debarring the children of more recent Kikuyu immigrants. Similar steps were taken to broaden the ethnic basis of recruitment to the civil service and promotion to the highest ranking government positions with each region given a proportional allocation. In addition, the government sought to locate new educational and medical facilities in historically disfavored regions. ₄₉

The Moi administration also took a number of steps to broaden ethnic participation in business enterprise. Through a variety of formal and informal means, including leverage over the country's banking system, the government has sought to deny important benefits such as credit, trade protection, and import licenses to Kikuyu-owned businesses in order to provide business opportunity to non-Kikuyu. A determination to reduce Kikuyu domination of the business sector has also been an important consideration in the government's unwillingness to divest state-owned enterprises. Members of the Moi government have been fearful that only Kikuyu entrepreneurs possess sufficient wealth to acquire these large assets from the government.

The government's interest in redistribution has also been extended to the agricultural sector where the Moi government has initiated programs to broaden the ethnic basis of cash crop cultivation. New projects for coffee and tea cultivation, for example, were initiated to disseminate the production of these crops to the Coast, Rift Valley, and Western Province. The Moi government also shifted its sectoral agricultural priorities away from the historic emphasis on exportable commodities toward an increase in grain production for domestic consumption. This, too, was a product of the redistributive political impulse. The principal beneficiaries of the pricing policies that currently favor wheat and rice are non-Kikuyu grain farmers, especially wheat farmers in the Rift Valley.

In implementing its shift toward grain production in non-Kikuyu areas, the government has treated the traditional export crop producers, especially Kikuyu coffee farmers, as its political and economic adversaries. It has intervened actively and destructively in the network of organizations that served the needs of the export sector, eliminating or undermining those that had been of vital importance to coffee farmers and creating new ones to represent non-Kikuyu grain growers. In 1985, for example, the government banned the Kenya Farmers Association (KFA), a hybrid organization that functioned as part interest group and part grain marketing cooperative for the predominantly Kikuyu mixed farmers of the Central Province. In its place, the government established a new organization, the Kenya Grain Growers Cooperative Union (KGGCU), which was dominated by the Kalenjin wheat farmers of Rift Valley Province. In 1989, the banning of the KFA was followed by the banning of the Kenya Coffee Growers Association (KGCA), another organization vital to the interests of coffee growers. The KGCA had functioned as a lobby for trade and tax policies favorable to coffee. Throughout this period, the government was also attempting to influence the selection of leaders of the Kenya Planters Cooperative Union (KPCU), an important producers cooperative. Its goal was to remove those who had long held the growers confidence and replace them with leaders more amenable to the new policies.

That these policies should have led directly to lower rates of agricultural growth and indirectly to lower rates of growth in industry and manufacturing requires little explanation. ₅₀ It is axiomatic among many scholars of economic development that when governments intervene on the basis of Callaghy's political logic in the allocation of economic resources, those resources are often used in a less than optimal manner. ₅₁ The political actions of the Moi administration caused economic decline because they alienated and demoralized some of the country's most skilled agriculturists and business leaders, disrupted organizations that had contributed greatly to the country's economic growth, and produced higher and higher levels of inefficiency in the government agencies whose function was to serve the economic needs of the society. From the standpoint of long-term economic growth, perhaps the most important casualty of the Moi policies was the atmosphere of economic trustworthiness and credibility that the Kenyatta government had nurtured as the cornerstone of its growth orientation.

The great puzzlement of the Moi administration, then, is not the wrenching downturn of economic growth caused by its redistributive policies, but the fact that its attempts to build a broad coalition of ethnic have-nots has also failed so miserably. Far from having created such a coalition, the government has become ethnically and politically isolated and forced to depend upon extreme levels of political repression to remain in power. Its basis of support has been reduced for all practical purposes to the members of one ethnic group, the Kalenjin. ₅₂ As a result, the stability of this increasingly unstable regime rests not on the sort of broad political underpinning that, for so many years, helped buoy the Tanzanian administration of Julius Nyerere despite its economic woes-that is, the widespread popular support inspired by socially egalitarian policies-but, rather, on the government's apparent willingness to go to any lengths to terrorize and intimidate its political opponents. ₅₃ Once a symbol of constitutional pluralism, Kenya today is best known for having one of modern Africa's worst records in the area of human and political rights.

The reasons for this failure are by no means clear. Journalistic observers attribute it to personality factors, the combination of ineptitude and venality that characterizes Moi and his closest followers. There is an element of validity in this perception. It seems painfully clear that at some point in the early years of the Moi administration, its progressive social agenda-the goal of redistributing the wealth and privilege of the Kikuyu few to benefit the non-Kikuyu many-became transformed into something very different: an opportunity to use the power of government to confiscate and personally acquire the wealth of an older but now politically marginalized elite.

The concept of prebendalism, used by Richard Joseph to describe Nigeria, seems perfectly suited to describe the behavior of Kenya's new political elite. ₅₄ According to this concept, a state is prebendal when its leaders employ political power to amass personal wealth. This is exactly what has been occurring in Kenya. The economic and social assets of the Kikuyu upper class have been diminished in value. But this diminution was not so much an aspect of social redistribution as it was a matter of treating the wealth of Kikuyu farmers and business entrepreneurs in ways that augmented the private assets of the new ethnic hegemon, the Kalenjin followers of President Moi. Joseph's concern for Nigeria applies to Kenya: once prebendalism has become institutionalized as a feature of elite behavior, successor governments, even those that are democratically elected, seem prone to engage in it.

Prebendal behavior on the part of the Moi administration accounts well for the extraordinary breadth of opposition it has aroused. The coalition against the present government is an extremely disparate one. It includes first and foremost, of course, the old Kikuyu establishment that surrounded the Kenyatta government. Their political agenda could be best described as a conservative one: a return to pure meritocracy in government, nondiscriminatory approaches to business opportunity, and a return to the principle of comparative advantage in agricultural pricing policy.

Moi's opposition further includes the lower strata of Kikuyu society, ethnically offended that the notion of redistribution has transparently become a mere excuse for engaging in economic warfare against the leading members of Kikuyu society. Their agenda is more radical, and includes the calls for redistribution of wealth that have been a part of Kenya's political system since colonial times. The Kenyan left has also taken the conspicuous personal enrichment of the new Kalenjin establishment as an occasion to reassert old demands for genuine redistribution of the agricultural land in the central highlands. The Kikuyu radicals are in an especially difficult position. They have been unable to align themselves with the Kikuyu conservatives because of ideological differences and unable to align themselves fully with radicals of other ethnic groups because of their strong Kikuyu identity. ₅₅

The anti-Moi coalition further includes a variety of important ethnic groups that should have benefited from the redistributive efforts of the Moi government but that feel they have not. Such groups as the Luo, for example, are deeply offended that such distribution as has occurred seems largely to have bypassed their communities.

Moi's opposition is by no means wholly ethnic in character; it also includes a considerable variety of other professional, religious, and ideological elements. Members of Kenya's legal profession, for example, have been outspoken critics of the government because of its willingness to violate the rule of law and the independence of the judiciary in its zeal to repress political opponents; members of the clergy have openly criticized the government for its transparent cynicism and amoral insensitivity to the needless suffering its policies have caused. Even members of Kenya's nascent environmental movement have become an active part of the opposition because of the government's willingness to disregard ecological considerations in the raw pursuit of personal enrichment.

Kenya's internal political opposition has been significantly reinforced by the pressure of Kenya's international donors who, now, with the end of the Cold War, find it politically easier to apply added pressure by threatening to withhold financial aid until major political reforms occur. Some formerly friendly governments such as Norway have broken openly with Kenya and others, including Britain and the United States, have been outspoken in their criticism of the government's human rights record. Indeed, donor pressure for political reform is now so great that it is widely credited with having forced the Kenya government to accept multipartyism in early 1992.

Kenya's principal opposition group, the Forum to Restore Democracy (FORD), reflected the pluralism of Moi's opposition. From the beginning, it was a fragile admixture of formerly powerful Kikuyu conservatives who felt excluded from the Moi administration, Luo populists who, in the immediate post-independence era, championed the interests of small farmers against the large estates, and political reformers from a variety of ethnic groups. It was divided not only by its ideological and ethnic diversity, but also by a generational split, raising the larger question of whether political power in any post-Moi era would be held by a new generation of Kenyans or by individuals who were active in the country's independence movement and the Kenyatta government that followed it. By the fall of 1992 FORD had split into two parties. The principal source of the fragile cooperation that existed among the highly disparate elements of the opposition was their determination to oust Moi personally from political power.

The power of Kenya's opposition groups brought the ultimate survival of the Moi government into question. Further political reform, including open elections to the National Assembly, seemed inevitable. In an abrupt move in early November 1992, Moi caught the opposition off guard by announcing elections for early the following month, but the opposition managed to get the High Court to provide more time to nominate candidates. Complaining of harassment and impending electoral fraud, the opposition considered boycotting the elections. The elections were held on December 29, accompanied by charges of serious fraud. The badly fragmented opposition split the anti-government vote, allowing Moi to retain the presidency and KANU to maintain control of the National Assembly, even if only marginally. The major opposition leaders charged that democracy had been hijacked. Nevertheless, the political openness Kenya enjoyed during the early Kenyatta years has been partially but ambiguously restored.

The more difficult question is whether Kenya will also be able to recapture the economic growth it enjoyed during the immediate post-independence period. The prospect of this happening is less favorable. Moi's spending on the elections seriously aggravated Kenya's already precarious economic condition, and his post-election cabinet failed to inspire confidence in his commitment to viable economic reform. After consultations with the Moi government in late February 1993, the IMF was unwilling to resume assistance to Kenya despite the government having effectively floated its currency and made other economic changes. Powerful older political logics still held sway. If it has done nothing else, the Moi government may have permanently discredited the notion of comparative advantage by associating it with the notion of wealth for a single privileged ethnic minority.

Tanzania

The process of structural adjustment in contemporary Tanzania raises questions of profound importance for students of economic reform. Among the most difficult of these is the very basic issue of why governments change policy. In the case of countries such as Ghana, where a puritanical military coup suddenly replaced decadent civilian leadership and swept aside an older generation of equally decadent senior officers, a superficial answer, at least, is available: reform takes place because a reform-minded group of leaders replaces one that is disinclined toward change. But in Tanzania, no such violent succession occurred and important reforms such as currency devaluation had, in fact, begun under the Nyerere regime, several years before the constitutional succession. Clearly, an intellectually persuasive explanation of policy change must rest on factors more intellectually persuasive than such idiosyncratic events as a change of regime.

Perhaps the best answer to the question of "why change?" is that there is no single answer. In countries such as Tanzania, which have made important change from statist economic systems to a greater reliance on the market mechanism, a variety of factors, both external and internal, have come into play and, at some moment, converged in their effects. Among the most powerful forces at work are the internal political consequences of acute economic decline, the preeminent intellectual influence of orthodox economics since the early 1980s, and the leverage enjoyed by international lending institutions.

Perhaps the most powerful source of policy change is the economic crisis that precedes it. As John Ravenhill has elsewhere suggested, "the greater the deterioration of an economy, the more ruptured the social fabric, then the greater tolerance that government and society will have for . . . adjustment programs." ₅₆ Because of its overriding importance, the relationship between economic crisis and policy reform is one that cries out for theoretical precision. But there is great difficulty in becoming more precise about this relationship because governments seem to vary so greatly with respect to how deep an economic crisis must become before they become willing to change their policies. Perhaps the greatest challenge for the next generation of scholars of structural adjustment is to shed greater light on exactly how an economic crisis launches a new political dynamic.

One possible way to begin is through the conceptual prism of the rent-seeking model. ₅₇ In this model, the tendency for inappropriate economic policies to persist long after they have inflicted great economic damage is explained on the basis of the material rewards these policies make available to politicians and bureaucrats. It suggests that political elites have no incentive to change policies that benefit them personally and that also provide them the resources to cement together powerful coalitions of clienteles. This model may help us develop a slightly better profile of the relationship between economic crisis and policy change. For it directs our attention toward an important political process: when an economy has declined to the point that it no longer generates an adequate supply of rents to bind together the component parts of the old policy coalition, change is imminent and likely.

Tanzania had reached that point by the end of the 1970s. High and growing levels of unemployment had forfeited the regime the support of industrial workers hitherto cushioned from economic pain by the country's rigid protectionism and the security of guaranteed employment in state-owned industries. Virtually all urban wage earners, ranging from middle class civil servants and parastatal officials to low paid service sector workers, had become profoundly disaffected because of the precipitous decline in the purchasing power of their salaries and the extreme unavailability of even the most basic goods. Even university graduates, the country's presumptive elite, had begun to question the economic system. Long accustomed to a government that had always acted as employer of last resort (and often as employer of first resort), they increasingly found that the government's impossibly bloated ministries had finally reached the point where they simply could no longer add additional staff. ₅₈ When all these groups faced the collective prospect of long-term and growing economic hardship under the statist regime, the stage was set for policy reform.

A government that is perched on the edge of political transformation can also be decisively influenced by shocking events that so jar the existing economic system as to make its further continuation impossibly difficult. Between 1979 and 1981, Tanzania suffered three of these; two were exogenous and one was internal. ₅₉ The first exogenous shock was the oil price increase of late 1979. This had a variety of effects beginning immediately with an inflationary impact on urban consumers and, because fuel purchases consumed a larger portion of foreign exchange earnings, an additional impact in constricting the country's ability to import the equipment and materials necessary to operate its industrial plants, transportation networks, and medical and educational supplies. The second external shock was the sharp decline in the international terms of trade suffered by primary agricultural exporters between 1979 and 1981. The decline drastically reduced Tanzania's foreign exchange earnings and, like the oil price increase, reduced the country's capacity to maintain an operating economic system. These events worsened the economic circumstances of urban Tanzanians and, by themselves, could well provide an adequate explanation of the policy changes that began to take place in the early 1980s.

The political effect of external shocks, however, is difficult to assess and has probably been exaggerated by most observers. Even the most restive social groups tend to be aware that they cannot hold a government accountable for events that are beyond its control. In addition, the explanatory usefulness of external shocks is weakened by their very universality: the same exogenous shock (e.g., an oil price increase) that launches an economic revolution in one struggling country may produce few tangible political effects in another. The shock events that are unique to a specific country, therefore, are of greater explanatory utility.

It is likely, therefore, that the shock event that contributed most to Tanzania's change of political course in the early 1980s was internal, the government's decision to conduct the enormously costly 1979 war that led to the overthrow of the Idi Amin regime in Uganda. Whatever meager resources of hard currency remained to Tanzania at the end of the 1970s seemed to be devoted almost entirely to the war effort and, as a result, virtually every civilian sector of Tanzanian society seemed to suffer visibly from the effects of the government's need to supply a massive and prolonged military operation more than a thousand miles from the capital city. ₆₀ Many Tanzanians did not understand or accept the reasons for the government's decision to engage in a war with Uganda and, as a result, came to feel that the economic deprivations it occasioned could have been avoided. ₆₁ For the very first time since independence, there was widespread and open criticism of Nyerere's leadership and a public outcry for an end to the economic system he had personally sponsored. ₆₂

The hardships suffered by Tanzanians during the late 1970s and early 1980s produced an all-pervasive belief that the old system had failed. While some political leaders remained vocally supportive of the country's socialist approach, their reasons for doing so were transparently self-interested. The vast majority of Tanzanians were not only deeply frustrated by the chronic shortages of goods, inadequacies of government services, and unfulfilled promises of better things to come, but also increasingly prepared to endure the uncertainties of a wholly new strategy of economic development.

Economic crisis, then, induces a willingness to change and a determination to support leaders who will bring it about. But it does not provide a satisfactory answer to the question of why policy reform has, virtually everywhere, become synonymous with economic liberalization. To deal with this question, it is essential to consider intellectual influences. For ideas inspire and empower political choice and, on a global basis, the most consequential and persuasive ideas of the past decade have been those that embraced and adumbrated the superiority of market-based economic systems.

Why these ideas should have attained such seemingly universal ascendancy at this time is a question far beyond the scope of this chapter. But part of the explanation undoubtedly lies in real-world economic trends. And the most conspicuous of these, in recent years, has been the glaring contrast between the dismal performance of statist economic systems that have disregarded the vitally important role of free markets and the superior performance of economies where governmental interventions have supplemented and reinforced the growth possibilities inherent in production for the international marketplace. By the end of the 1970s, innumerable Tanzanians, including countless students, scholars, journalists, bureaucrats and politicians, had been personally exposed to this contrast and would have drawn obvious conclusions about the policy changes necessary to rekindle a process of economic growth in their society.

Ideas also attain an impact on policy when they become the official doctrine of powerful financial institutions. The role played by the World Bank, IMF and bilateral donor agencies such as USAID has been widely referred to and, in many cases, is probably greatly exaggerated. Such organizations are undoubtedly important as the institutional transmission belts of orthodox economic analysis and they can apply great pressure for the adoption of orthodox economic reforms through their capacity to provide or withhold financial assistance. But it would be wholly incorrect and completely unfair to Tanzania and to a host of other African countries to explain policy change on the basis of the power of international lenders. Such explanations completely ignore the importance of internal political changes. The most accurate description of the role of the international lenders is that they become involved only when governments are themselves actively considering policy changes. Their role is then to lend support to groups and leaders favorable to change, thereby strengthening the internal political impulse toward economic adjustment. ₆₃

The search for causal factors, whether internal or external, misses a major part of the drama of political reform. Social science has yet to capture, even descriptively, a portrait of the way an existing policy coalition disassembles itself and gives way to a new one. Recent scholarship on Tanzania is especially weak on this subject and, as a result, little is yet known about the exact sequence of internal political changes that have fundamentally transformed the policy framework of the Tanzanian government. ₆₄ All that can be said with utter certainty is that, by the early 1980s, a growing number of Tanzania's leading politicians had become supportive of policy reforms that would involve a major increase in the role of market forces.

The outward signs of this receptiveness were everywhere. The most important ministries, such as Agriculture and Finance, had come under the jurisdiction of ministers favorable to reform. And, within these ministries, officials who favored reform had risen to highly influential positions and were engaged in a visible effort to transform their country's most important economic policies. Tanzania's rapidly changing intellectual climate could also be discerned at the University of Dar es Salaam, where a growing number of members of the Economics faculty and economists associated with the university's highly influential Economics Research Bureau had begun to take public positions favoring reform. Indeed, from the very beginning, Tanzanian economists were among the most important actors pressing for political changes. Through the intellectual strength of their research, through the force of their presentations at professional conferences and seminars, through their influence as consultants to government ministries, and, perhaps most importantly, through personal friendships with some of the country's highest ranking politicians, they had a profound effect in bringing about a greater openness to economic change.

One of the most frustrating aspects of policy reform in societies such as Tanzania has to do with the tendency for economic recovery to lag behind policy change. There is an implicit assumption among many proponents of structural adjustment that, once significant reforms have been introduced, private sector actors that have been patiently waiting on the sidelines will become actively involved and buoy the process of economic recovery. But this has rarely occurred, producing frustration among economists who tend to view structural adjustment as principally a matter of getting the policies right. Important as that is, it overlooks the supremely political aspect of the connection between policy change and economic recovery, the factor of policy credibility.

The timing of official reform efforts may be something very different from the moment at which those efforts become sufficiently credible to energize the involvement of segments of the population whose participation is economically indispensable. The distinction between policy change and credibility is especially salient in formerly socialist countries, such as Tanzania, where private entrepreneurs have been subjected to decades of harassment and official political disparagement. Here, mere policy changes may not suffice to provide assurance that those who engage in business activity are no longer personally or economically at risk.

Political economy has also been regrettably weak in dealing with this issue. The disturbing fact of the matter is that very little attention has been directed toward the question of how long a reform effort must be in place before it becomes credible or whether some institutional arrangements for reform can inspire more confidence than others. ₆₅ The factor of credibility is of overriding importance in areas of economic liberalization that require individuals and organizations to invest resources and skills in the reform process and in sectors of the economy where whole institutions must be reformed or created in order to generate an economic response. There are numerous examples of areas where credibility is absolutely vital to a reform effort, including divestiture of state enterprises, which requires that private investors risk capital to acquire ownership. Where the credibility of a reform effort is uncertain, it is to be expected that such sectors will exhibit an especially slow response to a reform program.

Perhaps the best example in Tanzania has to do with the restoration of cooperative unions which, the present government hopes, will replace parastatal corporations in the local procurement and marketing of agricultural commodities. The difficulty lies in the fact that cooperative unions are such enormously complicated organizations; they require highly sophisticated accounting skills to deal with complex financial matters and equally sophisticated managerial skills to handle such daunting tasks as the management of large inventories of physical equipment, crop storage facilities, and rolling stock. Cooperatives must also be able to administer such widely assorted activities as credit provision and delivery of farm inputs. Under the best of circumstances, then, the creation of an effective cooperative society requires years and sometimes generations of laborious and painstaking effort.

In contemporary Tanzania, those with the experience and skills required to perform these tasks are in desperately short supply. The most likely candidates for the rebuilding of cooperative societies are those with previous cooperative experience. But, among members of this group, the government's sudden and arbitrary decision, in 1976, to ban the producer cooperatives and assign their functions to governmental crop authorities remains a bitter memory. In such agriculturally vital regions as Kilimanjaro, West Lake, and Mwanza, the banning of cooperatives continues to evoke pervasive feelings of suspicion and mistrust. It thus provides an important component in an explanation of the slowness in the recovery of the cooperative movement.

The most visible source of Tanzania's on-going credibility constraint has been the political difference between the government, which is dedicated to economic liberalization, and the governing party, which officially remains a socialist organization. During the years immediately preceding the structural adjustment agreement of 1986, and throughout the period since, it has been an obvious, sharp division within Tanzania's political leadership between those in favor of policy reform, who generally identified themselves as "pragmatists," and those who remained committed to the old statist system of economic management, generally referred to as "socialists." In this respect, Tanzania manifested its own version of the classic "reds" versus "experts" confrontation. It would be an oversimplification to suggest that this division conformed entirely to the institutional division between the country's socialist governing party, the Chama Cha Mapinduzi (or Party of the Revolution [CCM]) on the one hand, and its state apparatus on the other; for pragmatists and socialists were to be found in both the government and within the CCM. But by and large, the CCM has been the most important source of resistance to the process of economic reform.

Not only did the party remain officially committed to a socialist vision of Tanzanian development, but, until early 1992, it also remained the supreme organ of government. For these reasons, there was a tendency for the country's leading socialist politicians to congregate within the CCM and use its policy-making authority in ways that frustrated the country's economic reformers. Not the least of Mwinyi's difficulties was the brooding presence of his predecessor, President Nyerere, who, while never explicitly denouncing the government's reform policies, took numerous occasions to remind Tanzanians, many of whom continued to revere him personally, of his own devotion to a socialist outlook and distaste for market-oriented approaches to development.

Nyerere's highly visible role as implicit critic of economic reform was an acute source of unease to reform-minded Tanzanians and a constant reminder of the potential fragility of the reform process. Since the ultimate triumph of the pragmatists could not, even until recently, be taken for granted, the prudent course for a rational economic actor would have been to refrain from making a potentially risky investment of personal skills and resources in the reform effort.

The first political challenge for Tanzania's economic pragmatists, therefore, was to dislodge the party's most prominent socialists from important ministerial positions without inducing a destabilizing rupture between the party and the state. The early years of the administration of President Ali Hassan Mwinyi were much taken up with this process and were characterized by the increasing political ascendancy of the pragmatist group, with socialist conservatives being gradually displaced from virtually all key governmental positions. But the process of pragmatic consolidation was a frustratingly slow one and the removal of conservatives was not basically completed until the end of 1991. ₆₆ The process of reform proceeded, however, because the party's power was ultimately more symbolic than real: it could pass resolutions that implicitly criticized the process of liberalization, it could conspicuously embarrass the government by refusing to renounce its commitment to a socialist ideology, and it could continue to insist on some representation of the socialist position within the cabinet. But, by the early 1990s, the party's socialist stalwarts no longer controlled the key organs of government within which economic policies were formulated and implemented.

The party's ability to maintain a high political profile, even as major policy reforms were proceeding without its formal assent, was, however, a source of unease to those awaiting signals of credibility from the reform effort. For it meant that the process of economic reform was largely a matter of administrative changes in policy implementation rather than legal changes legislatively authorized by a majority of the National Assembly. So far as the credibility of the reform effort was concerned, the difficulty for contemporary Tanzania was that the power of the CCM created the bewildering anomaly of a reformist government coexisting in the same political space with a socialist party whose representatives continued to dominate the country's most important representative body.

The political changes that might begin to resolve this anomaly did not occur until the beginning of 1992, when the governing party approved a series of constitutional reforms that effectively removed two of the most important pillars of the old Tanzanian system. In the spring of 1992, the CCM officially accepted political pluralism and also agreed to end its status as the supreme organ of government. The CCM must now contend with a vast proliferation of more than a dozen opposition organizations seeking to position themselves for Tanzania's next general election, presently scheduled for late 1995. Since some of these parties seem certain to oppose economic reform, the credibility constraint posed by a reformist government coexisting uncomfortably with a socialist party may not be fully resolved until that time.

Multiparty Democracy and Structural Adjustment

It would be wholly premature to assess the long-term political sustainability of economic reform in Tanzania but the country's present political situation typifies the dilemma confronting a number of African nations undergoing far-reaching economic liberalization: is economic reform compatible with political democratization? To address this issue, it is useful to begin with the conventional wisdom on the topic: the view that structural adjustment cannot readily coexist with democratic politics. The reasoning generally offered for this position holds that because structural adjustment imposes painful and prolonged austerity on large numbers of people, it creates a ready-made opportunity for would-be political entrepreneurs to mobilize opposition sentiment. Under democratic conditions, the parties formed by these entrepreneurs could be expected to enjoy sufficient levels of electoral support to capture political power. The politically reformed government would then find itself electorally committed to repeal major components of the economic reform package. Therefore, it is argued, only authoritarian regimes capable of repressing or withstanding adverse public opinion can sustain reform efforts.

The weaknesses in this position are glaring. The first is its uncritical assumption that neither ordinary citizens nor political elites learn from experience. This would be completely untrue in the Tanzanian context. One of the most powerful political forces in Tanzania today is the memory of the severe day-to-day hardships experienced during the Nyerere regime. The shortages and deprivations of that period remain a living memory for vast numbers of Tanzanians who, as a result, would be deeply loathe to see a return of a socialist or statist approach to development. An additional memory of the socialist period has to do with the constant need to bribe government officials for even the most trivial services. The memory of pandemic official corruption continues to foster a vast amount of social rage toward politicians who enriched themselves while articulating the phraseology of socialist egalitarianism. Today, these memories constitute a formidable barrier to politicians who base their appeal on a return to the old system.

A second shortcoming of the conventional wisdom is its tendency to overlook the fact that large and potentially influential segments of society do well under adjustment even in the short run. The list of "winners" includes not only export-oriented farmers but also a wide variety of other important social groupings. Since currency devaluation, insofar as it increases the domestic price of imported goods, constitutes a natural form of protectionism, it can provide considerable stimulation for a host of import-competing industries. Food-producing farmers, for example, benefit as the prices of imported foods increase and as the increased incentives for agricultural exports provides a more generalized buoyancy for the agricultural sector. ₆₇ Structural adjustment can also provide small-scale productive enterprises since these also benefit from the protective cocoon of devaluation. Examples include small-scale machining firms that remanufacture truck, bus, and automobile parts, tailoring and clothing shops that benefit as imported clothes increase in price, and furniture manufacturing firms that can provide local substitutes for imported goods. Indeed, one of the ironies of structural adjustment is that devaluation provides natural protection for some of the same industries that formerly received protection through quantitative restrictions.

A third shortcoming of the conventional wisdom lies in its repeated assertion that structural adjustment is merely a developmental version of "trickle down" economics and that, as a result, it provides few benefits for lower socioeconomic strata. The fact is that it provides substantial and immediate benefits for large segments of the working class. Workers in trade-related sectors such as plantation workers, mineworkers, and transportation and dock workers, for example, benefit quickly as increased international trade stimulates increased labor demand in these areas. And workers in the construction trades also tend to benefit because adjustment tends to stimulate immediate investment in the rehabilitation of infrastructure, especially road building and railroad reconstruction. Workers in building construction have also benefited with a boom in commercial real estate and housing. And as government revenues improve, civil service salaries may also be allowed to rise, easing the inflationary pressure on middle- and lower-level public servants. As the incomes of these groups improve, there are often ripple effects that include a heightened demand for labor-intensive services such as automobile and truck repair, dry cleaning, and food preparation.

Since there has been so much misunderstanding and so much misrepresentation of the socioeconomic effects of economic reform, it is vitally important to be as precise as possible about the question of which members of society are, indeed, the losers. There are only two groups that obviously fall into this category. The first and most obvious consists of those who were the most conspicuous winners under the old system; that is, the rent-seeking politicians and bureaucrats who were able to benefit from the governmentally induced scarcities that were integral features of an economic approach that justified trade restrictions as a means of attaining economic growth. About the declining fortunes of this group, little need be said except possibly that this group of economic losers should never have been winners in the first place.

The second group of losers is more elusive but consists generally of the workers, managers, and suppliers of formerly protected enterprises. The effect of adjustment on this group is complex and, because of severe information scarcity, difficult to assess. The little that is known with certainty can be quickly summarized. Both the size and the plight of this group have probably been considerably exaggerated. Since rates of capacity utilization in protected industries had already fallen to abysmally low levels well before structural adjustment began, unemployment and underemployment were already severe. Indeed, as production levels fell, many of the workers in these firms had already begun to drift away, to seek employment elsewhere in the system. Moreover, even before the period of severe economic decline, a certain proportion of the labor force in the protected industries had always consisted of "ghost workers," the relatives, friends, and political clients of politically influential persons, placed there purely for the purpose of patronage.

The relationship between structural adjustment and real industrial unemployment, then, is unclear. Undoubtedly, some proportion of the industrial workforce has suffered economic misfortune as a result of the withdrawal of protections and subsidies. But, given the extent and rapidity of recovery in import-competing industries, construction activity, and the extractive sectors, it is by no means clear that industrial workers as a whole are worse off under adjustment.

The fact of the matter, then, is that the relationship between democracy and adjustment may be exactly opposite to that stated in the conventional wisdom. It is wholly plausible to this observer that the number of those benefiting from economic reform substantially exceeds those who have suffered from it and that their improved fortunes, combined with even broader social memories of the inequities of the earlier rent-seeking society, would provide a solid electoral basis for further reform efforts.

---

Note 1: Thomas M. Callaghy, "Toward State Capability and Embedded Liberalism in the Third World: Lessons for Adjustment," in Joan M. Nelson, ed., Fragile Coalitions: The Politics of Economic Adjustment (New Brunswick, N.J.: Transaction Books, 1989), p. 115. The author wishes to thank Michael Kurtzig and Margaret Missiaen of the Economic Research Service of the U.S. Department of Agriculture for their willingness to make available the agricultural data used in this chapter. Back.

Note 2: The "classics" in this literature are widely familiar and include most notably The World Bank, Accelerated Development in Sub-Saharan Africa: An Agenda for Action (Washington, D.C.: World Bank, 1981), referred to hereafter as the "Berg Report" and Robert H. Bates, Markets and States in Tropical Africa: The Political Basis of Agricultural Policies (Berkeley and Los Angeles: University of California Press, 1981). Back.

Note 3: For a full discussion of this scholarly genre, see John Ravenhill, "Africa's Continuing Crises: The Elusiveness of Development," ch. 1 in Ravenhill, ed. Africa in Economic Crisis (New York: Columbia University Press, 1986). Back.

Note 4: One recent article that compiles an inventory of these shortcomings is Robert H. Bates, "Agricultural Policy & the Study of Politics in Post-Independence Africa" in Douglas Rimmer, ed. Africa 30 Years On (Portsmouth, N.H.: Heinemann, 1991), pp. 115-129. Back.

Note 5: For an excellent discussion of this issue, see Merilee S. Grindle, "The New Political Economy: Positive Economics and Negative Politics." (Cambridge, Harvard Institute for International Development, Development Discussion Paper No. 311, 1989). Back.

Note 6: Ravenhill, "Africa's Continuing Crises.," p. 3. Back.

Note 7: Republic of Kenya, Sessional Paper No. 10: "African Socialism and its Application to Planning in Kenya" (Nairobi: Government Printer, 1965). Back.

Note 8: Ibid. Back.

Note 9: The Kenya case provides strong support for Robert Wade's argument that neoliberalism does not provide an adequate explanation for economic growth in developing countries. See his article "East Asia's Economic Success," World Politics  44, 2 (January 1992): 270-320. Back.

Note 10: Between 1985 and 1990, horticultural exports increased 225% in volume and 375% in value. By the early 1990s, horticulture nearly equaled coffee exports in total annual earnings. In 1990, tea exports earned approximately $257 million; coffee exports, $151 million; and horticultural exports, $148 million. Economist Intelligence Unit, EIU Country Profile 1991-92, Kenya (London: Economist Intelligence Unit, 1992), pp. 16-17. Back.

Note 11: In interview, one agricultural planner claimed that the foreign exchange component alone of wheat produced in Kenya was higher than the world market price of this commodity. Back.

Note 12: Republic of Kenya, Sessional Paper no. 4 of 1981: "National Food Policy" (Nairobi: Government Printer, 1981), esp. pp. 20-21. Back.

Note 13: The author is indebted to Professor Charles Hickson, School of Business, University of Belfast, for this observation. Back.

Note 14: For a theoretically generalized treatment of the political efficiencies of different tax systems, see Margaret Levi, Of Rule and Revenue (Berkeley and Los Angeles: University of California Press, 1988). Back.

Note 15: The terminology is sometimes confusing. Overvaluation of a currency consists of offering too few units of local currency per $ U.S. Devaluation, therefore, consists of increasing the number of units of local currency per $, an upward movement in the exchange rate. Back.

Note 16: Uma Lele and Richard Meyers, Growth and Structural Change in East Africa: Domestic Policies, Agricultural Performance, and World Bank Assistance, 1963-1986 (Washington, D.C.: World Bank, 1987), p. 14. Back.

Note 17: Cathy L. Jabbara, "Agricultural Pricing Policy in Kenya," in World Development  13, 5 (May 1985): 624. Back.

Note 18: The World Bank and the UNDP, Africa's Adjustment and Growth in the 1980s (Washington, D.C.: World Bank, 1989), p. 22. Back.

Note 19: For an excellent discussion, see Michael Westlake, "Nominal Protection Coefficients and the Measurement of Agricultural Price Distortion in Developing Countries" (Cambridge: Harvard Institute for International Development, Discussion Paper, No. 214, January, 1986). Back.

Note 20: Since the World Bank calculates NPCs using official exchange rates, its figures show even higher increases for grain crops than table 11.5. Figures generously supplied by the Bank place the NPCs for maize, wheat and rice at 131, 115 and 113 respectively, indicating considerable "subsidies" for these crops. Back.

Note 21: This legislation was the Local Manufactures (Export Compensation) Act. A detailed discussion of this law and its effects can be found in Patrick Low, "Export Subsidies and Trade Policy: The Experience of Kenya," World Development 10, 4 (1982). Back.

Note 22: Republic of Kenya, Sessional paper No. 4 of 1975, "Economic Prospects and Policies" (Nairobi: Government Printer, 1974), p. 4. Back.

Note 23: See, for example, The World Bank, Kenya: Growth and Structural Change, pp. 97-99. Back.

Note 24: Financial Times  (London), January 8, 1992. Back.

Note 25: Tony Killick, "Kenya, the IMF and the Unsuccessful Quest for Stabilization," in John Williamson, ed., IMF Conditionality, (Washington, D.C.: Institute for International Economics, 1983). Back.

Note 26: For a different view that considers Tanzania's negotiating position as substantially closer to the conditionality requirements of the IMF, see Reginald Herbold Green, "Politico-Economic Adjustment and IMF Conditionality : Tanzania 1974-1981" in Williamson, IMF Conditionality . Back.

Note 27: World Bank, World Development Report 1992 (New York: Oxford, 1992), p. 218. Back.

Note 28: John de Wilde has also suggested that part of Tanzania's foreign exchange crisis may have been caused by the declining quality of agricultural exports and the fact that, as a result, the prices it received were increasingly at the low end of the world market scale. See his Agriculture, Marketing and Pricing in Sub-Saharan Africa (Los Angeles: African Studies Center and Crossroads press, 1984), p. 35. Back.

Note 29: Daily News (Tanzania), June 20, 1986. Back.

Note 30: The physical quality of life index is a United Nations standard of measurement that is a statistical composite of such indices as infant mortality, adult literacy, life expectancy, and availability of food. Back.

Note 31: Jennifer Sharpley has shown that, as a percentage of market value, food crops in Tanzania are twice as import-intensive as export crops. See "External Versus Internal Factors in Tanzania's Macro-Economic Crisis" (unpublished manuscript), table 11.5, p16a. Back.

Note 32: If the drought year 1984 is dropped from the calculation, and grain imports are calculated on the basis of 1981-83, and 1985-86, Kenya's grain imports averaged 337,000 metric tons and its grain imports as a percentage of total imports were 9.0. Tanzania's grain imports averaged 263,000 metric tons and its food imports as a percentage of total imports were 12.9. Back.

Note 33: According to one estimate, the informal economy GDP ranged between 21% and 31% of the official economy GDP throughout most of the 1980s and did not begin to decline significantly until 1988. See M. S. D. Bagachwa, "The Nature and Magnitude of the Second Economy in Tanzania," Tanzanian Economic Trends 2, 3 (October 1989) and 2, 4 (January 1990): 32. Back.

Note 34: According to one informant, the bribe to be assigned a preferred place on the waiting list for a state-owned apartment was equal to 20 years rent. Back.

Note 35: For a detailed survey of Kenya's experience, see Barbara Grosh, "Agricultural Parastatals Since Independence: How Have They Performed?" (Nairobi, Institute of Development Studies, Working Paper No. 435, 1986). Back.

Note 36: See, for example, The United Republic of Tanzania, Analysis of Accounts of Parastatal Enterprises (Dar es Salaam, Bureau of Statistics, Ministry of Planning and Economic Affairs, 1983), and United Republic of Tanzania, Ministry of Agriculture, Crop Authorities: Financial Position and Financial Performance 1977 to 1981/82 (Dar es Salaam: Project Preparation and Monitoring Bureau, 1983). Back.

Note 37: Frank Ellis, "Agricultural Pricing Policy in Tanzania 1970-1979: Implications for Agricultural Output, Rural Incomes and Crop Marketing Costs," (University of Dar es Salaam, Economic Research Bureau, 1980), p. 38. Back.

Note 38: The United Republic of Tanzania, Structural Adjustment Programme for Tanzania  (Dar es Salaam: Ministry of Planning and Economic Affairs, 1982). Back.

Note 39: For an article on the vital importance of credibility in a reform effort, see Dani Rodrik, "Credibility of Trade Reform-A Policy Maker's Guide," World Economy 12, 1 (March 1989). Back.

Note 40: Economist Intelligence Unit, EIU Country Report 1991-91, Tanzania (London: Economist Intelligence Unit, 1992), pp. 19-20. Back.

Note 41: This reform is intended to replace the system that grants parastatal corporations guaranteed lines of credit with the nation's banking institutions. Initially intended to allow parastatals to conduct business on a month-to-month basis before crop sales produced an end-of-year surge in revenue, the line of credit system was an open invitation to abuse. Back.

Note 42: Grindle, "The New Political Economy," p. 45. Back.

Note 43: Roger Thurow, "Capital Flight Strains Kenyan Economy," Wall Street Journal, August 17, 1989. Back.

Note 44: The following observations are much indebted to Michael Chege and especially to his unpublished ms., "The Search For Optimal Sequence Between Production and Distribution: Kenya and Tanzania After Their Founding Presidents." This ms. is intended for publication in David E Apter and Carl G. Rosberg, eds., Political Development and the New Realism in Sub-Saharan Africa (Charlottesville: University of Virginia Press, forthcoming, 1993). Back.

Note 45: Income inequality in Kenya became the subject of extensive research by scholars and international organizations. The most well known and carefully documented study was International Labour Office, Employment, Incomes and Equality: A Strategy for Increasing Productive Employment in Kenya (Geneva: International Labour Office, 1972), see esp. ch. 5. Back.

Note 46: The assassinations of Pio Gama Pinto, Tom Mboya and J. M. Kariuki indicate that the Kenyatta government was fully prepared to deal severely with the most outspoken proponents of economic redistribution. Back.

Note 47: This 1970s beverages boom and the differences in the way Kenya and Tanzania responded to it are the subject of an excellent volume. See David Bevan, Paul Collier and Jan Willem Gunning, Peasants and Governments: An Economic Analysis (Oxford: Clarendon Press, 1989). Back.

Note 48: Chege, "The Search For Optimal Sequence," p. 10. Back.

Note 49: The best example is the creation of Moi University in western Kenya. Back.

Note 50: Not surprisingly, the effects on coffee production have been calamitous. Total production fell from an all time high of 129,000 metric tons in 1986-87 to 104,000 metric tons in 1989-90. Back.

Note 51: Mancur Olson has further suggested that distributional governments "slow down a society's capacity to adopt new technologies and to reallocate resources in response to changing conditions." See his The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities (New Haven: Yale University Press, 1982), p. 65. Back.

Note 52: The designation Kalenjin does not refer, in fact, to a single ethnic group but to a language family that includes such diverse groups as Nandi, Kipsigis, Elgweyo, and Tugen. The president himself is a member of the Tugen subgroup of the Kalenjin people. Back.

Note 53: For an excellent analysis, see Joel D. Barkan, "The Rise and Fall of a Governance Realm in Kenya," in Goran Hyden and Michael Bratton, eds., Governance and Politics in Africa, (Boulder: Lynne Rienner, 1992). Back.

Note 54: See Richard A Joseph, Democracy and Prebendal Politics in Nigeria: The Rise and Fall of the Second Republic (New York: Cambridge University Press, 1987). Back.

Note 55: The most obvious example of this group is the famous Kenyan novelist, Ngugi wa Thiongo. Back.

Note 56: John Ravenhill, "Reversing Africa's Economic Decline: No Easy Answers," World Policy Journal 7, 4 (Fall 1990): 716. Back.

Note 57: The classic article by Anne O. Krueger, "The Political Economy of the Rent-Seeking Society," American Economic Review 64, 3 (1974). Back.

Note 58: Lele and Meyers estimate that public sector employment in Tanzania had grown at a rate of nearly 16% annually between the late 1960's and late 1970s, Growth, p. 21. Back.

Note 59: See R. H. Green, D. G. Rwegasira and B. Van Arkadie, Economic Shocks and National Policy Making: Tanzania in the 1970s (The Hague: Institute of Social Studies, 1980). Back.

Note 60: The country's transportation system, for example, suffered greatly as its railroads, trucks, and even busses from the Dar es Salaam municipal transport authority were devoted to transporting troops and equipment to the Uganda border. Back.

Note 61: One of the worst after-effects of the war was a sudden, sharp rise in violent crime as demobilized soldiers, many of whom had weapons that had been captured or not handed in, returned to a depressed economy that was unable to provide employment. Back.

Note 62: At the University of Dar es Salaam, some scholars spoke of "Tanzania's Viet Nam" pointing to opposition to certain forms of conscription, the social demoralization occasioned by battlefield casualties, the adjustment difficulties experienced by demobilized and unemployed soldiers, and the budgetary squeeze on social programs that had been the pride of Tanzanian socialism. Back.

Note 63: The author is indebted to Professor Anne O. Krueger of the Department of Economics, Duke University, for this observation. Back.

Note 64: See, for example, John Loxley, "The Devaluation Debate in Tanzania," ch. 1 in Bonnie K. Campbell and John Loxley, eds. Structural Adjustment in Africa  (New York: St. Martin's Press, 1989). While comprehensive as an exposition of the arguments for and against devaluation, Loxley's argument says little about which groups of Tanzanians took either position and why, in the end, the pro-devaluation group won. Back.

Note 65: For example, an auction method of setting exchange rates is not credible because it can be canceled or suspended so easily by a central bank. An exchange bureau system is more credible because the bureaus and their clienteles become an institutional lobby. Back.

Note 66: In October 1991, one of the major socialists, Joseph Warioba, a former Prime Minister and relative of President Nyerere, was removed from his position as Minister for Regional Administration and Local Government. Another, Rashid Kawawa, retained a position as Minister without Portfolio, an assignment that reflected the party's continuing ability to influence cabinet-level appointments. Back.

Note 67: Before economic reforms began, overvaluation was sufficiently great that coffee production had become so unprofitable that a number of coffee farmers had abandoned this crop to engage in the cultivation of uncontrolled food crops. As real coffee prices rose with devaluation, these coffee farmers returned to the export crop, thereby creating an important niche for traditional food crop cultivators. Back.