/15-zo PSD Occasional Paper No. 16 September 1995 Governance and Structural Adjustment 16 in Ghana Charles D. Jebuni - The World Bank Private Sector Development Department Private Sector Development Department Occasional Paper No. 16 Governance and Structural Adjustment in Ghana Charles D. Jebuni September 1995 The author would like to thank Abena D. Oduro, Department of Economics. University of Ghana, for his contribution to this paper. This paper has been cleared for inclusion in the occasional paper series by R. Shyam Khemani, Manager, Competition & Strategy. The views expressed are those of the author(s) and should not be attributed to the World Bank. This paper forms 1part of a forthcoming volume, Leila Frischtak and Izak Atiyas (eds.), "Governance, Leadership and Communication: Building Constituencies for Economic Reform." The World Bank Private Sector Development Department it Contents 1. Introduction ...............................................1 2. Background to Economic Decline and Successful Reform .......................................3 3. The First Reform Episode, 1966-71 ...............................................5 Initial Conditions ...............................................5 The Program ...............................................6 Response of the Economy to Reform .............................................8 Explaining What Happened .............................................I 1 4. The Current Reform Program: 1983 to the Present ............................................. 15 The Deterioration of the Economy Prior to Reform ............................................. 15 The Decision to Change ............................................. 18 The Reform Program ............................................. 20 Trade and Exchange Rate Reforn .............................................. 21 Fiscal Reform .............................................. 22 Public Sector Reforms .............................................. 24 The Response of the Economy to Reform .............................................. 25 Fiscal Response: 1983-89 ............................................. 26 Fiscal Response, 1990-94 ............................................. 29 5. Governance Capacity and the Adjustment Program ............................................. 33 The Initial Phase, 1984-89: Commitment and Credibility .............................................. 33 Program Articulation and Interest Groups ............................................. 35 The Bureaucracy and Adjustment ............................................. 38 6. The Transition, 1990-94 ............................................. 41 7. Conclusions ............................................. 45 Appendix ..47 Principal Political Dates and Regimes .47 Bibliography ..49 . . Al 1 Introduction 1.1 The structural adjustment program in Ghana has been one of the most enduring programs in Africa, and it has been viewed as the model of success. It began as a stabilization program, and has evolved into an effort to change the development strategy of the country. Prior to 1983, Ghana, like most other African countries imbued with the dirigiste dogma of the 1960s and 1970s, pursued a state-led, import-substitution development strategy. In a thirty-one-country study, Ghana was classified as having the highest degree of price distortions (Agarwala 1983). 1.2 Within a decade Ghana has succeeded in dismantling and eliminating most of the controls and distortions in the economy, and it now boasts one of the most liberal trade and payments regimes with some of the lowest tariffs on international trade in Africa. Ghana now pursues an outward-oriented, export-led development strategy, with a reduced role for the state in directly productive economic activities. 1.3 Perceptions of the success of this program vary. To the international community, Ghana is a model of success. All the traditional indicators used to measure the success of economic reform-growth in GDP and exports, inflation, government budget deficits, and balance of payments-have shown improvement. Within Ghana, however, the assessment is mixed. Most Ghanaians will admit that the program has succeeded in removing the distortions in the economy and moving to market-based policy instruments. At the household level, however, the impact of the program is less visible. Agricultural production, the mainstay of the majority of the population, has lagged behind population growth. Cost recovery measures in health and education are biting hard. 1.4 In spite of these doubts, the program as a whole has brought substantial progress, even if the pace is slow. This is the first time that the Ghanaian economy has grown consistently for a decade, but the growth has been uneven. For the first six years of the program, real GDP growth was steady and averaged about 5 percent annually. Between 1990 and 1994 the growth rate decelerated, budgetary and balance of payments deficits reemerged, the rate of inflation increased, and the economy ran into unresolved difficulties. A number of fundamnental questions are raised by this experience. Why has Ghana appeared to succeed where other African countries have failed, and reversed these policies? Even within Ghana, why has the program endured this time, when similar policies were tried between 1967 and 1971 and were reversed? Why did the I 2 Governance and Structure Adjustment in Ghana program run into difficulties after six years? The difference could lie in governance capacity, where governance capacity is defined as the ability of the state to exercise power over society.' 1.5 The objective of this study is to examine the relationship between governance and economic reform. The next section offers a brief review of the political economy of decline and successful reform. The first attempted reform of 1967-71 is then examined, followed by a look at the current program, starting from 1983. In the last section we draw together the evidence from the two economic reform experiences and offer some conclusions. ' For definitions of governance see Boeninger 1991 and Landell-Mills and Serageldin 1991. 2 Background to Economic Decline and Successful Reform 2.1 Several reasons may be adduced to explain economic decline and the inability of governments to effect change. The new political economy theory identifies capture of the state by interest groups as a major reason for such failure. These interest groups have "a weight or influence that is disproportionate with reference to the economic gain that accrues to them" (Seabright 1994, p. 9). In some cases these interest groups were initially created by governments to provide support for their policies; they, in turn, captured the government. These interest groups became so entrenched that they gained the upper hand over rulers and blocked the implementation of appropriate economic policies (Frimpong-Ansah 1991). In the African context, Grindle concludes: "Relatively autonomous in their choice of policy at the outset, they [African leaders] soon become captive of the beneficiaries they have introduced and lose their capacity to alter policy" (quoted in Leith and Lofchie 1993, p. 245). There is a tendency for such a state to increase expenditures beyond the socially optimum levels; it then becomes revenue- starved. 2.2 According to Leith and Lofchie (1993), the Ghanaian postindependence policy regime both rewarded urban interest groups and led to their growth. Urban interest groups thus became powerful forces in the maintenance of these policies. Both Frimpong-Ansah (1991) and Leith and Lofchie (1993) identify the interest groups, and thus the beneficiaries of these policies, as the politicians, urban classes, workers and managers of the protected state industries, traders, the military, and bureaucrats. 2.3 Why were the losers, primarily small-scale farmers, unable to lobby for appropriate policies? According to the theory of collective action, this may be because of the small size of their holdings, the large number of farmers, their geographical distribution, and the cost of organizing. When a market contains many small agents, and therefore large fixed costs of organizing, the benefits that each can realize by securing appropriate policies may not exceed the costs of lobbying. Rational behavior may thus lead the members of this group to seek a free ride, and there is little incentive to engage in collective action (Bates 1990). Government's ability to offer selective incentives enables it to reward those who conform and punish those who do not. According to Leith and Lofchie, selective incentives were also used to buy-off large-scale 3 4 Governance and Structure Adjustment in Ghana farmers. The supply of inputs and equipment to large-scale farmers at subsidized prices reduced their incentive to organize the other farmers. 2.4 When the exploitation of these groups has gone to extremes, the losers from the policies of the ruling elite exit from the economic system through parallel markets, smuggling, and movement toward subsistence production. The rents then disappear, and the costs of inappropriate policies are passed back to the ruling coalition. The response of the peasant farmers by the late 1970s- particularly cocoa farmers-was to exit from the economic system through the parallel market. For Lal (1987), economic reform is an attempt by the state to regain control of such an economy. 2.5 Reforming the economic situation described above requires conviction in favor of radical economic reform and the desire to bring about fundamental change in the ideological underpinnings of domestic politics (Frimpong-Ansah 1991). This requires both political and economic resources. Political leadership is needed to form coalitions or create associations to tackle social problems. The just exercise of authority, a capacity for problemsolving and conflict resolution, and the efficient performance of its functions based on the involvement of a societal coalition are important in Boeninger's (1991) conception of such good government. 2.6 All economic reform involves winners and losers. It encompasses changes in the distribution of gains and losses in society. The creation of a vision that will bring about a change in the distribution of interests in such a way that the gaining group is able to compensate the losers, and everyone will be better-off, may be important. If such a vision is perceived as a positive influence for all groups, and not for just one sector of society, it will promote voluntary compliance and commitment. Support can be marshalled for that vision if it is well articulated, if a new support group is developed to counterbalance detractors by indicating clearly whom the potential winners will be, and if the previous set of economic policies can be proven not to have served the interests of all. 2.7 Given the presence of a vision, success will depend on implementation capacity, which requires adequate human resources for analysis, planning, and implementation, as well as institutions able to meet the requirements of the new policy environment. Major institutional reforms and innovations may be required in some cases; in others, minimal institutional reforms and motivation may be all that is needed. Whatever the surrounding conditions, implementation capacity requires financial resources. 2.8 For a captured state that is also revenue-starved, an immediate increase in government revenues is essential for institution building. Increased revenues are also crucial to the provision of a credible system of incentives to motivate appropriate behavior and reduce dependence on rents. 3 The First Reform Episode, 1966-71 Initial Conditions 3.1 The first attempt at economic reform took place between 1966 and 1971. It was initiated by a rnilitary regime that handed over power to a constitutionally elected government in 1969. The military government, the National Liberation Council (NLC) had considered its tenure temporary from the start. 3.2 The NLC government inherited a heavily controlled economy in decline, with high budgetary deficits and inflation. The balance of payments situation was critical, reserves had run down, and there was a general shortage of goods and services (as will be seen later, these conditions were nothing compared with the devastation of the economy inherited by the time of the structural adjustment program in 1983). 3.3 The development strategy adopted by Ghana in the immediate postindependence period was based on state-led import-substitution. The vision was the eventual establishment of an industrial socialist state. Starting with the initiation of the five-year development plan in 1959, there were massive increases in government expenditures to create both the physical and human capital required to attain this objective. These investments were high in import content and brought about a deterioration in the balance of payments. At the initial stages, both the resulting budgetary and external payments deficits were financed by running down the huge external reserves that Ghana inherited at the time of independence. 3.4 With the foreign exchange reserves declining sharply and the budget deficits rising rapidly, the government resorted to a series of strong measures. Taxes were raised and exchange controls introduced in July 1961, and by December 1961 comprehensive import licensing was introduced. 3.5 In spite of the deteriorating external payments and fiscal deficit, a more ambitious development plan was launched in 1963-64 to boost the import-substitution industrialization process. Both trade and payments controls were intensified, and the government refused to use the exchange rate as a policy instrument. 3.6 The result of these policies was that in spite of the high rates of investment- principally public investment-the growth performance of the economy declined. In the immediate postindependence period (1957-61), when the liberal trade regime was maintained, 5 6 Governance and Structure Adjustment in Ghana real GDP grew at an annual average rate of about 5 percent. With the introduction of the controls following the big push, GDP growth slumped by more than half. Between 1962 and 1966 the real GDP growth decelerated to an annual average of 2.4 percent. 3.7 The annual rate of inflation increased from an average of 6 percent between 1961 and 1963, to 16 percent in 1964, and 23 percent in 1965. With increasing inflation and a fixed nominal exchange rate, the real exchange rate appreciated and the currency became overvalued. Frimpong-Ansah (1991) estimates that the currency was overvalued by 7 percent in 1963. By 1965 the overvaluation had increased to 45 percent, and to 60 percent in 1966. This overvaluation of the currency, combined with the taxation of exports implied by the restrictive regime, caused the external payments situation to deteriorate. The current account deficit increased from $33.7 million in 1962 to $222.9 million in 1965. 3.8 The main sources of government revenues were cocoa and imports. The cocoa tax base declined from 16 percent of GDP in 1958, to 9.2 percent in 1965, and 7.7 percent in 1966. The import tax base declined from 35 percent in 1961 to 22.9 percent in 1964; it then increased to 26.7 percent in 1965, but declined to about 20 percent in 1966. 3.9 As a result, government revenue as a percentage of GDP stagnated at around 14 percent between 1961 and 1963. It increased to 18.7 percent and 19 percent in 1964 and 1965, respectively, but declined to 15 percent in 1966. While the revenue base declined, expenditures increased throughout the period, which led to severe budget deficits that fueled inflation. The Program 3.10 The economic situation gave rise to popular dissatisfaction. Combined with political repression, this led to a military coup that overthrew the Nkrumah government in February 1966 and the formation of the National Liberation Council (NLC) government. The military government immediately set about consulting chiefs and the opposition leaders to gain insight into how to tackle the economic problems of the country. This could be interpreted as an attempt to shift the base of the ruling coalition. 3.11 While these groups were important, the more politically active were the urban workers who benefited from the controls. The immediate economic hardships of the latter part of the Nkrumah regime must have turned urban workers against him, but his socialist policies created a sense of nationalism. Reconciling nationalism based on state-led industrialization and controls with the immediate needs of the economy was the major problem confronting the military government. Its approach was to portray itself as transitional, playing only a stabilization role. This presentation, however, did not resolve the problem of Ghanaian attachment to Nkrumah's ideals. The government recognized the need to break this attachment. It set up a civil education department, led by Professor Busia, a Ghanaian sociologist, to do this, but it failed. It seemed that the NLC itself was attached to these ideals and that its commitment to liberalize was only "skin-deep." As Killick (1978, p. 301) put it, "In fact, one suspects that the The First Reform Episode, 1966-71 7 NLC's public commitment to complete liberalization was intended chiefly to mollify the IMF and aid donors." 3.12 Without an alternative vision to rally support for its program, the NLC had to be sensitive to worker reaction to the immediate hardship it would impose. This sensitivity implied that when it had to choose between reducing real wages and cutting back on capital expenditures, the NLC chose to cut back capital expenditures, while increasing nominal wages. It could not close down inefficient state enterprises, and controls remained in place throughout. 3.13 The NLC was content to leave the actual management of the economy to civil servants and other professionals led by Omaboe, a former government statistician under the Nkrumah regime. "It set out to restore 'efficiency' to the economy and was not much concerned with the elaboration of a coherent body of economic ideas" (Killick 1978, p. 54). 3.14 Under the auspices of the National Economic Committee, the main problems of the country at the time of the coup were identified as the following (Killick 1978, p. 55): * correction of the present imbalance in the country's foreign payments position - arrest the inflationary pressures to which the economy had been subjected during the past few years * provision of more job openings for the growing population * the restoration of balance to the government's budget. 3.15 The government's overall strategy for dealing with these issues, together with the advice of the IMF, led to the pursuit of orthodox disinflationary fiscal and monetary policies. At the same time, the government sought to avoid adverse effects on production by improving capacity utilization and preparing the economy for another development effort to be launched as soon as stabilization had been achieved. 3.16 In pursuit of this strategy, a number of measures were undertaken. First, government expenditures were cut from C361.1 million in 1965 to C268.9 million in 1966 to reduce demand pressures on prices and the balance of payments. The NLC did not put much emphasis on improving the fiscal balance by increased taxation. It was thought that a substantial increase in the tax effort had already been achieved in the preceding years. Second, a program of liberalization intended to eventually abolish import, price, and exchange controls was implemented. Emphasis was placed on the role of private enterprises. A few state enterprises were sold to private investors, some state farms were closed down, and the Ghana Industrial Holding Corporation (GIHOC) was formed to improve the economic performance of the remaining state enterprises. The currency was devalued in July 1967 as part of a package to obtain external assistance. 3.17 The NLC economic program was, in general, an attempt to stabilize the economy and create an atmosphere conducive to private investment. But before the soldiers left office, they began to worry about the effects of their stabilization policies, as expressed in the subheading of the development plan they prepared, "From Stabilization to Development." 8 Governance and Structure Adjustment in Ghana 3.18 The NLC kept its promise to hand over power to an elected government. It organized elections on the Western model, which were won by the Progress Party (PP), led by Professor Busia. While committing itself to continue with the reform program, the civilian administration that took over from the NLC government nevertheless came into power without an economic program. What economic policies could be identified in the PP's discourse were not credible in view of their ambiguities. In the words of Killick (1978, p.56), "In spite of an apparently positive attitude to the freer operation of market forces, a large role was also envisaged for the state-an ambiguity neatly illustrated by the co-existence on the same page of pledges to move away from trade controls and to improve the enforcement of price controls." 3.19 The elections did not give the government enough political power to continue the economic reforms begun in 1966 with the same authority the NLC had demonstrated, despite its parliamentary majority. The two parties that contested the elections were ethnically based. The Convention Peoples Party was proscribed, and its members refused to join other parties. The 1969 elections thus did not generate adequate political competition. As Afrifa, one of the leaders of the 1966 coup was to indicate later, "the irony of the present situation in Ghana is that it is quite possible that President Nkrumah and the CPP would command the support of a majority of the electorate.... in genuinely free elections.... Busia as leader of government was, in political economy terms, at the centre of an essentially Nkrumahist state that he did not identify with, and therefore could not control" (quoted in Frimpong-Ansah 1991, p. 100). 3.20 The government's support base was rural, while most of the urban masses, created by the policies of the Nkrumah era, were Nkrumahist. Busia needed mass support for his program. He favored a shift of resources to the rural areas, but with limited financial resources, he could not do this without adversely affecting the welfare of the urban masses. He had to pay attention to these urban pressure groups, even though he could not incorporate them into his power base. 3.21 Growth was to be achieved through an increase in government expenditures in line with Nkrumah's strategy. The growth strategy began during a period of increasing world cocoa prices and rising terms of trade. There was increasing dependence on external financial resources to finance the budget. Revenues were buoyant, primarily because of the increase in cocoa prices. Response of the Economy to Reform 3.22 As a stabilization measure, the program was successful even though the expected balanced budget was not achieved. The overall budget deficit as a percentage of GDP generally declined between 1967 and 1970 compared with the preprogram period (Tables I and 2). From a high of 8.8 percent in 1966, the budget deficit/GDP ratio declined to about 3 percent in 1970, although there was a slight increase in 1968. 3.23 Expenditure targets were met in the first two years by restraining development expenditures to meet recurrent expenditure considerations. To placate urban workers, wages and The First Reform Episode, 1966-71 9 salaries were increased. As a share of total recurrent expenditures, worker compensation rose from 28.5 percent in 1965 to 34.6 percent in 1968 (Central Bureau of Statistics 1976). It rose by 83 percent between 1966 and 1969, while total recurrent expenditures rose by 54 percent over the same period. Development expenditures rose rapidly during 1969/70. Central government spending on new capital goods declined in 1967, and again in 1969. As table 2 indicates, a decline in the rate of inflation was achieved. From a high level of 34 percent in 1965, the rate declined continuously to about 3 percent in 1970, but it rose to about 9 percent in 1971. 3.24 The constraints on aggregate demand expansion and the import licensing system that was still in place ensured a decline in the volume of imports in 1966 and a considerable reduction in the current account deficit (Table 3). The devaluation of the cedi in 1967 had a further dampening effect on the demand for imports, with import volumes declining by 6 percent. 3.25 To ameliorate the impact of the devaluation of the currency, customs duties and some sales taxes were reduced or removed, and the export duty on diamonds was removed. Table 1: Budget Estimates and Expenditures, 1966-71 (million cedis) Item 1966/67 1967/68 1968/69 1969/70 1970/71 Budget estimates Expenditures 334.3 393.2 429.3 468.7 520.2 Development 124.6 89.4 104.6 122.6 137.5 Recurrent 209.7 303.8 324.7 302.7 382.7 Ordinary revenue 250.05 288.7 340.0 375.7 426.7 Capital receipts 47.05 104.7 88.9 95.3 112.0 Current account surplus/deficit 40.35 (15.1) 15.7 73 44 Overall surplus/deficit (37.2) 0.2 0.0 2.3 18.5 Budget out-turna Expenditures 302.301 370.54 382.43 457.65 488.4 Development 75.42 72.41 65.29 83.95 110.11 Recurrent 226.881 298.13 317.14 373.7 378.27 Revenue 241.46 300.24 291.17 369.23 501.44 Current account surplus/deficit 14.65 2.11 -25.17 -4.47 123.17 Overall surplus/deficit -60.84 -70.3 -91.26 -88.42 13.04 Notes: a: Excludes below-the-line receipts and payments. Source: Central Bureau of Statistics, Ouarterly Digest of Statistics, 1970 and 1971. Adjustments were made to the income tax regulations to reduce the burden of the devaluation for some companies. These measures caused a decline in import duties between 1966 and 1969, thus partially explaining the inability to achieve revenue targets. The net effect of the tax measures accompanying the devaluation was to reduce the average import tariff rate from 30.8 percent in 1966 to 25.8 percent in 1967 (Leith 1974). The cocoa tax rate declined between 1966 and 1967, but rose in 1968. The average export tax rate fell between 1966 and 1968. The effective A1111|1|"|1 1AM . 111 1 Niiiiiii tit A 1|1,1|iv1Il V ll 10 Governance and Structure Adjustment in Ghana exchange rate for imports thus rose by some 36 percent between 1966 and 1967, while the effective exchange rate for exports rose by 84 percent. The bias in the trade regime, however, persisted. Table 2: Major Budgetary Indicators, 1965-72 (percent) Item 1965 1966 1967 1968 1969 1970 1971 1972 Overall budget/GDP ratio 7.7 8.16 5.34 7.58 3.84 3.28 5.54 5.7 Share of domestic credit Central government 51.1 50.4 55.3 58.6 54.2 48.5 39.2 43.2 Other government 12.9 18 19.7 12.9 12.9 19.1 16.3 20.2 Change in credit to central government 45.9 15.1 29.7 18.4 -0.63 -7.02 -0.68 23.8 Inflation 34.9 1 1.7 -8.98 8.78 7.13 3.68 9.28 10.04 Source: Central Bureau of Statistics, Economic Survey of Ghana, 1969-1971; IMF, Intemational Financial Statistics Yearbook 1992. Table 3: The External Account, 1965-71 Item 1965 1966 1967 1968 1969 1970 1971 Trade balance (US$ million) -118.2 -40.4 18.8 39.8 49.7 51.9 -33.6 Current account(US$ million) -222.9 -127.8 -84.9 -56.1 -60.1 -67.7 -145.8 Current account deficit/GDP ratio (percent) 10.8 6.11 4.91 3.36 2.94 3.06 5.99 Percent change in export volumes 2.5 -15.7 -8.63 2.3 -17.3 10.86 -29.7 Percent change in import volumes 27.26 -11.4 -5.87 32.5 -6.45 9.5 -15.99 Commodity terms of trade (1975 = 100) 76 75 89 93 111 121 91 Source: IMF, International Financial Statistics Yearbook 1992; Central Bureau of Stalistics, Ouarterlv Digest of Statistics, 1964- 1972. 3.26 The current account deficit and the current account-GDP ratio fell continuously until 1970, largely because of the rising cominodity terms of trade, the constraint on domestic aggregate demand, suspension of interest payments on suppliers' credit in the 1966/67 budget, and subsequent rescheduling of loans. Export performance, however, was disappointing. Export volumes were below their 1966 levels at the end of the stabilization period. An export promotion package was introduced in 1969 to encourage nontraditional exports in particular, but was not implemented until 1971. The shortcomings of the trade regime during the stabilization period persisted, and explain the poor performance of the export sector. 3.27 Not much progress was achieved in divesting the state of productive enterprises. Of twenty enterprises that were listed for divestiture in 1967, seven were put up for outright sale. The NLC government succeeded in having only one outright sale, and transferred two state- owned enterprises into joint ventureships. No headway was made in the divestiture of state enterprises, and by the end of 1971 five new state enterprises had been created. Although the results of this effort were encouraging in the reduction of the external deficit, urban unrest increased because of growing unemployment, reduced real incomes, and declining expenditures on education and health (Stryker 1990). Both the balance of payments and budgetary situations The First Reform Episode, 1966-71 11 remained precarious, but were masked by rising cocoa prices. Table 1 illustrates the improve- ment in the terms of trade from 1967 to 1970. 3.28 In 1971, when world cocoa prices dropped sharply, and the government envisaged a large budget deficit for 1971/72, it devalued the currency by 44 percent, cut government expenditures, and increased taxes. Prices of most consumer items increased substantially. The resulting urban unrest provided justification for a military coup led by Colonel Acheampong in January 1972. The immediate action of the new government, the National Redemption Council (NRC), was to partially revalue the currency and reintroduce controls. Explaining What Happened 3.29 This first attempt at economic reform failed because of a number of interrelated governance issues, including the failure to articulate an alternate and credible economic vision, capture by interest groups, weakness of the leadership, and inconsistent policies. 3.30 There was a failure to articulate an alternate vision to replace Nkrumah's, and it was thus difficult to make a dramatic shift from previous policies. The NLC government presented itself as temporary, in power only to perform a stabilization role. It therefore probably did not consider the need for the creation of a different strategy. The new civilian government that took over in 1969 suffered from the same disadvantage. According to Killick, "Busia and his colleagues came to office with little by way of an economic policy. It was, as those advising them quickly discovered, impossible to derive from the election manifesto of the Progress Party a meaningful economic program" (Killick 1978, p. 56). 3.31 Because of this lack of an alternate economic policy strategy, no new interest groups emerged that exhibited support for the policies. The potential losers from the new economic strategy were therefore more likely to exert significant pressure on government and achieve their objectives. 3.32 Both the NLC and the Busia administration succumbed to pressure and were captured by interest groups. It would appear that both governments succumbed to labor demands to increase wages to reduce the possibility of unrest. Wage and salary increases were provided to some public sector employees in the July 1967 budget statement to "alleviate the immediate hardships due to the . . . devaluation of the currency" (Ghana 1968, p. 4). An indication of the lack of autonomy of the government was the growth in the share of compensation of employees as a proportion of recurrent expenditures. By 1971 this line item was about 45 percent of recurrent expenditures, rising from 34 percent in 1968. Wages and employment were used as a ploy to get the support of the vocal urban workers, a group that gave the government little support. 3.33 In the 1969/70 budget statement, expenditure cuts, including a ban on spending on minor capital items, was to be imposed because of "the relatively large share of personnel emoluments in the total recurrent budget . . . and a decision not to reduce the army of government employees by any substantial numbers" (Ghana 1969, p. 23). A redeployment 12 Govemance and Structure Adjustment in Ghana program was put in place at the same time that expenditures were to be constrained. In 1968, public sector employment expanded by 10 percent, despite the policy of retrenchment. 3.34 Both governments sought to control the leadership of the Trade Union Congress, but failed. This led to several incidents of unrest in the 1967-71 period, and the Busia regime dissolved the TUC in September 1971. An examination of the number of disputes by sector between 1968 and 1971 reveals that they increased during this period, with an escalation in 1971 (see Figure 1). The trade unions and their management had formed lobbies to resist some of the government measures, and this may explain the failure of the divestiture program, as well as the skewed nature of the cuts in central government spending. There was a lack of state autonomy, with private interests able to influence government policies. One reason the import surcharge scheme introduced by the Busia administration was not as effective as it should have been was because "the tax effort was far lower than would have been required to make the liberalization viable and protect the balance of payments in the circumstances of an overvalued currency " (Ofori-Atta 1975, p. 18). This was explained by a deputy minister of finance of the Busia administration: "when succumbing to political pressures, the government exempted a number of importers from surcharges and granted stretch-out payment terms to others" (Ofori-Atta 1975, p. 18). Figure 1 WORK DAYS LOST DUE TO STRIKES (1963 - 1972) 160 140 120 (A 100 # 'g 60 so 40 20 0 1963 1965 1967 1969 1971 1964 1966 1968 1970 1972 YEARS DAYS LOST The First Reform Episode, 1966-71 13 3.35 Another reason for the failure of the economic reform program was the weakness of the Busia government. The regime was unsure of how to respond. An assessment of the structure of the Busia government by the governor of the Bank of Ghana notes that Busia was "running his cabinet more like a college tutorial. Decisions were based on careful study and lengthy arguments. Individual members who were eloquent or persuasive tended to create and develop their own lines of thought and action which were not necessarily those of the Cabinet, the Party or the Prime Minister" (Frimpong-Ansah 1991, p. 100). Libby's (1976) discussion of the decisionmaking process leading up to the December 1971 devaluation shows that there was a contest within the cabinet for the control of economic policy. The conflicts within government and the indecisiveness of the prime minister resulted in a sluggish response to the crisis as it emerged. These conflicts were accentuated because some members of the cabinet were of the Nkumahist tradition. This may explain the lack of an articulate strategy with clear objectives that were different from Nkrumahist ideas. 3.36 The reform process was also reversed because the import liberalization component had become macroincompatible. Import liberalization had begun cautiously with a gradual relaxation of quantitative restrictions and a reduction of import duties on selected goods. As it progressed, it had to be supported by a currency devaluation, an increase in indirect taxes, or a slowdown in the growth of the money supply to ensure that the balance of payments would not deteriorate. Unfortunately, the required devaluation did not occur until December 1971. Restraints on the growth of the money supply were relaxed in 1969. Concern about generating the support of the urban workers and capture by some interest groups reduced the effectiveness of the import surcharge scheme introduced in lieu of devaluation. The policy inconsistency of the program is evident in the increase in the size of the current account deficit in 1970 (Table 3). Short-term debts increased by $79.2 million between September 1969 and June 1971 (Frimpong- Ansah 1991, p. 105). There was no effective export promotion package to counteract the effect of the import control system still in place, despite the liberalization of imports and the appreciation of the real exchange rate. The improvement in cocoa prices as the import liberalization program progressed initially masked the domestic policy inconsistencies. 3.37 Busia's economic program was dependent on aid flows that did not materialize to finance the excess of government expenditures above domestic revenues. The 1969/70 budget included expected external budgetary support of C55 million, but only C39 million was achieved. A similar shortfall occurred in the 1970/71 financial year, when only C37.2 million of the expected C50 million was received. It has been estimated that gross total assistance (that is, debt relief, net credit from the IMF, and long-term aid receipts) declined between 1969 and 1971 (Killick 1978). At the same time, import revenues had fallen continuously between 1966 and 1969. In 1969 they stood at C82.8 million, compared with Cl 11.2 million in 1966. The increase in the import surcharges in 1970 reversed the decline. 3.38 Policymakers failed to realize that since export performance was poor and external aid flows generally fell below requirements, there was a need to slow down the import liberalization process, restrain expenditures, and possibly change the growth strategy to encourage increased private investment through an improved incentive structure. The decline in 14 Governance and Structure Adjustment in Ghana cocoa prices exposed the weakness of a strategy that depended on increasing government expenditures and a buoyancy in the revenue base that depended on artificially high cocoa prices. 4 The Current Reform Program: 1983 to the Present The Deterioration of the Economy Prior to Reform 4.1 Acheampong, the leader of the military government that overthrew the Busia administration, set out what was to be the guiding principle of economic policy in an address to the Ministry of Finance and Economic Planning (quoted in Killick 1978, p. 299): "The political frame of reference which has guided your actions and your advice especially in the last two years must be cast into the rubbish heap of history. This means a departure from the 'laissez-faire' so called free market economy and the institution of effective planning in the allocation and utilization of resources." 4.2 Whether laissez-faire or a free market economy represented the correct interpretation of the Busia period or not, this statement implied a rejection of the previous policies. The government partially revalued the currency and unilaterally repudiated some of Ghana's debts. Comprehensive import and payment controls were reintroduced, and the scope and coverage of price controls were widened. The overall principle of government was self- reliance and a reduction in dependence on external resources. Economic policy had moved full circle to the command economy of the Nkrumah era.2 Except for brief periods between 1979 and 1980, when some attempt was made at liberalizing the trade regime with a devaluation of the cedi in 1978, these controls remained in place until 1983. 4.3 As these policies began to take effect, the rate of inflation almost tripled within four years, from 11 percent in 1972 to 30 percent in 1975 (Table 4). The real depreciation in the currency achieved with the devaluation of 1971 began to erode. In subsequent years the currency appreciated sharply. The index of the real effective exchange rate increased continuously from 22 in 1972, to 46 in 1976, and 278 by 1982.3 The ratio of the parallel market to the official exchange rate, which stood at 0.96 in 1971, increased continuously, to 2.53 in 1976 and 22.4 by 1982 (Table 4). 4.4 With the highly overvalued exchange rate discouraging exports, the intensification of controls worsened the situation. In spite of the highly favorable terms of trade between 1972 2 For full details of economic policies during this period, see Jebuni, Oduro, and Tutu 1994 or Rimmer 1992. 3 See Jebuni, Oduro, and Tutu 1994. The index is calculated in such a way that an increase indicates appreciation. 15 Table 4: Selected Economic Indicators, 1979-83 _ Item 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 a Domestic indicators GDPgrowth 6.78 5.56 -2.49 5.55 6.85 -12.43 -3.52 2.26 8.47 -3.77 -0.23 -3.17 -5.85 -4.33 3 Revenue ratio, percent of GDP 19.3 18.0 14.9 10.9 12.5 15.5 13.4 10.5 6.6 9.9 6.9 4.5 5.6 5.6 e Expenditure ratio, percentofGDP 20.7 21.0 19.3 15.7 16.2 21.5 22.9 19.1 15.1 16.4 10.9 10.6 11.0 8.0 ca Fiscal balance / GDP ratio 0.1 -0.4 -5.4 -4.4 -7.4 -11.8 -13.3 -13.2 -9.1 -5.8 -6.1 -6.1 -5.9 -2.5 c Investment / CD GDPratio 13 13 9 8 12 12 9.8 9.4 6.4 9.7 6.1 4.7 3.5 3.7 > Inflationrate 3.68 9.28 11 17 19 30 55 116 73 55 50 116 22 123 c Credit billion' (cedi) 0.291 0.289 0.358 0.376 0.584 0.924 1.57 2.78 4.52 4.90 6.51 10.64 11.05 28.05 3 Credit b billion (cedi) 0.115 0.12 0.168 0.231 0.338 0.295 0.393 0.423 1.11 1.63 1.96 3.39 6.08 1.34 3 External indicators Export volume index (1975=100) 220.9 155.2 186.6 169.2 139.5 142.8 149.7 123.3 101.6 96.8 100.0 103.1 116.5 84.0 Import volume index (1975=100) 256.9 215.8 111.5 132.8 179.1 155.4 149.6 155.4 128.8 97.9 100.0 112.7 64.4 58.2 Offical exchange rate (CedisperUS$)' 1.02 1.82 1.28 1.15 1.15 1.15 1.15 1.15 2.75 2.75 2.75 2.75 2.75 30.00 Ratio of parallel to official exchange rate 1.6 1.0 1.3 1.3 1.5 1.7 2.5 8.0 5.9 5.7 5.8 9.5 22.4 22.2d Notes: ': Net credit to the central government by the banking system, billions of cedis. b: Credit to the nonfinancial public institutions by the banking system, billions of cedis. X End of period. d: October 1983. Source: Statistical Services, Ouarterlv Digest of Statistics, various issues: IMF, Intemational Financial Statistics Yearbook 1992 The Current Reform Program: 1983 to the Present 17 and 1979, both import and export volumes declined (Table 4). Exports as a proportion of GNP decreased from 20 percent in 1972, to 10.5 percent in 1977, and 3.3 percent in 1982. The proportion for imports followed a similar pattern, declining from 13.5 percent in 1972, to 10 percent in 1977, and to 2.5 percent in 1982. Even though trade and exchange controls were intensified to contain the adverse balance of payments effects, by 1981 the current account deficit before official transfers stood at $508 million. 4.5 With an expanded role for the state government, expenditures increased. To finance these expenditures the government instituted high excise duties, export taxes, import duties, and an array of complex income and profit taxes. Nevertheless, with the general controls and overvalued exchange rate, both rent-seeking activities and parallel markets in commodities and currency developed. The official economy shrank, and with it government revenues. The tax base on both imports and exports declined sharply, and tax compliance decreased. Government revenues as a proportion of GDP declined throughout 1975-83, from 15.5 percent in 1975 to 5.6 percent in 1982 (Frimpong-Ansah 1991, p. 160). Although occasional attempts were made to cut expenditures, they remained high, largely because of recurrent expenditures. One lesson most Ghanaian governments had learned was that political success depended on the ability to minimize the discontent of the politically active urban classes. As the rate of inflation accelerated and the control system failed to supply commodities either in sufficient quantities or at controlled prices, urban unrest developed. To placate these groups, beginning in 1975 the government increased borrowing from the banking system to finance expenditures. Government spending had no relationship to budgetary plans. For the 1976/77 fiscal year, the original estimated deficit on the current account of the budget was about C20 million, but the year ended with a deficit of C233 million. In 1977/78, a surplus was originally envisaged, but the government incurred a huge deficit of Cl. 1.19 billion. Against the large surplus of C287 million planned for 1978/79, a deficit of C346 million was incurred. The resulting deficits and the underlying trend in components of government expenditures are captured by the Economic Survey of 1980. "Even though the deficits were incurred on account of large increases in expenditures by the government an overwhelming proportion of the increases were on account of recurrent expenditures: development expenditures moved up slowly and far behind the rate of inflation. Thus investment in the government sector in real terms recorded a declining trend" (Central Bureau of Statistics 1980, p. 22). 4.6 The situation was made more difficult because the government budget did not cover expenditures by state-owned enterprises, which generally were funded by "below the line" items. Although it was expected that the budgets for these enterprises would be submitted to the Ministry of Finance and Economic Planning for approval, according to Ministry sources this was not done. Data to assess expenditures by these enterprises, and thus total public expenditure, were generally unavailable because they did not keep records. Nonetheless, some indication of their effects on expenditure control can be obtained by looking at their sources of funding. The first source was direct subvention from the below-the-line item of the budget. The extent of this funding depended on the minister or commissioner of state, and the relationship between the managing director of the enterprise and the minister. A second source was internal borrowing from the banking system. This was done directly by the enterprises themselves, by instructions 18 Governance and Structure Adjustment in Ghana from ministers to banks to lend to these enterprises, or by government acquisition of loans on behalf of the enterprises. The third source was external borrowing by the state enterprises and the receipt of plant and machinery donated by foreign governments, particularly governments of the former Eastern Bloc. The final source of funding was the capitalization of the tax, dividend, and Social Security funds of workers. Some indication of financing may be found through an examination of bank credit to nonfinancial public institutions. According to the monetary survey, credit to these enterprises increased from Cl 15 million in 1970 to a peak of C6 billion in 1982 (Table 4). 4.7 Thus, by the start of the structural adjustment program Ghana had experienced a continuous decline in growth of GDP since 1975 as a result of these policies. In every year after this period, except for 1977 and 1978, real GDP declined. The external position of the economy was precarious. Gross external reserves as a ratio of GDP stood at 0.7 percent by 1982. 4.8 The deterioration in economic conditions created fertile ground for another military intervention. On 31 December 1981, t,e government was overthrown by a military coup led by Flight-Lieutenant Rawlings, who formed the Provisional National Defence Council (PNDC) to rule the country. The Decision to Change 4.9 It is unclear when the commitment to change occurred within the PNDC. What is clear is that tne PNDC did not come to power with a program of economic reform based on a market system. If they had a vision, it was based on the notion that the economic woes of the country were the result of exploitation of the masses and the corruption of the system- essentially a socialist philosophy. Both Frimpong-Ansah (1991) and Leith and Lofchie (1993) argue that the fundamental reason for the change in direction by the PNDC was the failure of its original philosophy and the increasing threat to the survival of the nation arising from the policies-or lack of policies-associated with this philosophy. 4.10 During its fifteen months in power, from January 1982 to March 1983, the PNDC government adopted a radical populist approach to economic recovery. The objective was to create a democratic society free of corruption and inequality.4 The majority opinion among government functionaries favored attempts to rid the country of corruption and stamp out smuggling. The strategy was to implement a Libyan-style revolution, while relying on domestic resources. Where external assistance was required, this was to be obtained from Libya or the friends of Africa, essentially the former USSR and Eastern Bloc countries. These measures had disastrous consequences for the economy and society. The economic situation deteriorated sharply, government revenues plummeted, and the authority and governance capacity of the state diminished. There was no improvement in the tax base, because parallel market activities continued, and estimated output declined. Opppsition to the government grew with the worsening 4 This was despite the actions of the regime in coming to power by overthrowing a democratically elected govemment. The Current Reform Program: 1983 to the Present 19 economic conditions. There were two coup attempts in 1982, and several groups registered their protests against the government-TUC, Christian Council, Ghana Bar Association, and groups on the political left. Visible signs of suffering, social chaos, and political disintegration were everywhere. These developments, combined with the limited assistance from Libya compared with the capital requirements for reform and the negligible prospects of substantial assistance from the Eastern Bloc donors, must have convinced some members of the PNDC of the need for change. 4.11 A second fundamental reason for change was prior adjustment. One of the crucial elements in reform programs that had held back most Ghanaian governments was the devaluation component. Possible inflation and the resulting political turmoil were the main concerns. This is why the Nkrumah government rejected the IMF package in 1965. The Busia regime was concerned about its political base, and when it did eventually devalue the currency it was overthrown. By 1983, however, the booming parallel market had pushed most prices to a level that reflected the scarcity value of foreign exchange. With virtually no supplies flowing through official channels, most Ghanaians adjusted to the prices. In these circumstances a devaluation would not affect domestic prices of importables. Thus, in the short run it could be expected that domestic prices would not be affected by the devaluation. This de facto price adjustment would imply that official prices would be adjusting toward what Ghanaians already paid. 4.12 In addition to this prior price adjustment, the rents that held the ruling coalition together had disappeared as the groups that generated the rents, the cocoa farmers and small-scale farmers, exited from the formal economic system through parallel markets and smuggling. According to Leith and Lofchie, the extraction of rents had gone beyond the optimum. This explains why the PNDC could expect that social groups that ordinarily would have opposed the reforms would probably not do so now. For the urban sector, much of the suffering and deprivation associated with economic decline had already taken place. Urban wage earners had suffered a decline in real incomes as inflation outpaced the increases in wages. 4.13 Moreover, the main organized opposition outside the government believed in the need to devalue the currency and liberalize the system. The government of the Third Republic that was overthrown by Flight-Lieutenant Rawlings had actually agreed to an IMF program of reform involving a devaluation, demonetization, and liberalization of the system. The minister of finance and economic planning was expected to travel to Washington, D.C., in 1982 to sign the agreement. The other main organized political group included members of the government that had attempted the first reform. They still believed in the kind of reform measures that were to be undertaken in 1983. If there was opposition, it was to the extent and degree of the reform program. 4.14 The main obstacle to change was within the PNDC government itself. The left wing of the government was opposed to discussion of market-based policy and involvement of the IMF and World Bank. The dilemma within the government was summarized by Kwesi Botchwey, Secretary for Finance and Economic Planning (quoted in Callaghy 1989, p. 115): 20 Govemance and Structure Adjustment in Ghana We were faced with two options, which we debated very fiercely before we finally chose this path. I know because I participated very actively in these debates. Two choices: We had to maneuver our way around the naiveties of leftism, which has a sort of disdain for any talk of financial discipline, which seeks refuge in some vague concept of structuralism in which everything doable is possible. Moreover, [we had to find a way between] this naivete and the crudities and rigidities and dogma of monetarism, which behaves as if once you set the monetary incentives everybody will do the right thing and the market will be perfect. 4.15 A third reason was the realization by the leadership-as demonstrated by a group of Ghanaian professional economists led by Dr. Abbey-that the source of the economic problem was the overvaluation of the currency and the producer price of cocoa. It was recognized that the correction of these distortions would be beneficial to the well-being of the cocoa farmer and would reduce rents going to the corrupt (Leith and Lofchie 1993). The exchange rate and cocoa price policy were interlinked, because the ability to increase the producer price of cocoa without compromising cocoa revenues was dependent on a devaluation of the cedi. It was a mark of the strength of the leadership of the PNDC that, unlike those in office during the previous reform attempt, it was not paralyzed by internal debate. 4.16 When the leadership was convinced of the need to adopt market-based reform policies, it made the decision. Given that the PNDC had come to power to reform a system it believed was corrupted by the exploitation of capitalists, it is doubtful that its attitude toward the private sector could change. For the private sector the philosophy of the PNDC prior to the reform program created doubts about its commitment to market reform. The debate, the struggle for power, and attempts to overthrow the government forced the PNDC into a security-centered policymaking process that did not allow wider national debate or participation. As will be discussed later, this approach led to exclusion of the bureaucracy from policy formulation. The Reform Program 4.17 There were three immediate requirem1ents for reforming the Ghanaian economy in 1983. First, it was essential to reduce the rate of inflation from the three-digit level to a reasonable rate. Second, there was a need to increase production, particularly of traditional exports, both as an input into reducing the balance of payments deficit and reducing inflation. Finally, it was critical to increase government revenues sharply. As discussed earlier, the previous policies had depleted government revenues, and increased revenues were essential for the restoration of the state's governance capacity. 4.18 In addressing these immediate concerns, and extending through almost all of the adjustment period, the strategy was to address simultaneously both issues of policy design and implementation. Not only would the policy initiative require institutional changes, but also arrangements aimed at addressing the issue of adequate compensation or incentives to induce the relevant response at each stage. The Current Reform Program: 1983 to the Present 21 Trade and Exchange Rate Reform 4.19 Fear of the political consequences of an outright nominal devaluation delayed the government's use of this instrument to correct the exchange overvaluation. Instead, a multiple exchange rate system was introduced. The nominal exchange rate remained fixed at its 1978 level, but a system of surcharges and bonuses was introduced that increased the actual exchange rate. As a result of the system of surcharges and bonuses, traditional exports and imports of crude oil, essential raw materials, basic foodstuffs, and capital goods were subject to an actual rate of C23.75 to the U.S. dollar. Nontraditional exports and other imports faced a rate of C25.975 to the U.S. dollar. The exchange rate was unified in October 1983 to C30 to the U.S. dollar. These policy measures were then followed within the same year by fiscal retrenchment and disinflationary monetary policies. 4.20 While these macroeconomic policy initiatives were taking place, and providing the relevant incentives for production, government also took measures to remove the immediate supply constraints to the traditional export sector. Roads and railways to cocoa, timber, and mineral producing areas were rehabilitated. Foreign loans were on-lent to timber and mineral producers. 4.21 Having achieved a certain level of stabilization in the initial years, the program moved to address microeconomic distortions. 4.22 Significant changes in trade and exchange rate policy have been undertaken since 1983. Following the initial unification of the exchange rate, incentives were maintained through periodic devaluations. These announcements became increasingly difficult for government to make. In September 1986 a multiple exchange rate system based on an auction system (Window II) and a fixed exchange rate (Window I) was introduced. The objective of the new system was to allow the exchange rate to be determined by domestic demand and supply factors, promote trade liberalization, and divert foreign exchange held outside the banks into the banking system. Three months later the exchange rate was unified and all transactions took place at the auction rate. Liberalization of the exchange rate system went a step further with the introduction in 1988 of the forex bureaux. Since then, an interbank market in foreign exchange has been introduced and exchange rate regulations relaxed. These measures resulted in a substantial depreciation of the real exchange rate, especially in the first three to four years of the program (see Table 5). The incentive structure is now more favorable toward exporting compared with the strong anti-export bias that prevailed prior to the introduction of the reforms. 4.23 Previous experience in Ghana had shown that serious balance of payments consequences could follow trade liberalization. At the same time, it was thought that imports should be used to relieve the critical import needs of the production sector. In keeping with this view, during the initial stages of the program, import liberalization through reduction of tariffs occurred within the context of import programming. This policy was pursued until 1986, when import programming ended. Import tariffs were reduced further after 1986. At the same time the negative import list was reduced, and in 1989 import licensing was abolished. Further trade reform involved the streamlining and lowering of nominal tariff rates to provide uniform 22 Governance and Structure Adjustment in Ghana incentives for industries. An export promotion package was put in place and the Ghana Export Promotion Council was revitalized to encourage nontraditional exports. Fiscal Reform 4.24 Critical to the governance capacity of the state was the generation of financial resources. As noted earlier, the tax base had contracted as a result of the previous policies. At the time of the introduction of the Economic Recovery Program (ERP), it was recognized that there was a large potential for revenue generation. One of the principal elements of the Ghanaian success story was the ability of the government to raise tax revenues substantially. Rather than focus on cuts in expenditures to deal with the budgetary crisis, revenue generation was emphasized. The objective of the tax reform system was to improve the efficiency of tax administration, broaden the tax base, reform the punitive system of taxation, and shift the burden of taxation from the wage earners and cocoa farmers to the consumption of luxury goods by the better-off. 4.25 The devaluation of the currency and trade reform could be expected to increase explicit tax revenues. Compliance and an increase in efficiency were essential if these gains were to be realized. To increase the revenues from these measures a number of institutional reforms were made in the tax collection agencies. Extreme measures were taken initially to ensure that individuals and organizations that were required to pay their taxes did so. The restructuring of the tax-collecting organizations and the activities of investigating committees to check for tax evasion, as well as the requirement that evidence of tax payment be produced before important documents (for example, passports) would be released, were important measures in the improvement of tax generation. 4.26 In accordance with the policy to improve revenue generation, the National Revenue Secretariat was formed in 1985. The purpose of the secretariat was to improve the revenue generation and collection process through a review of existing laws and measures with a view to needed changes, proposal of new revenue measures, removal of bottlenecks in revenue collection and administration, and the preparation of revenue budgets for the approval of the Ministry of Finance and Economic Planning. These functions had previously been performed by a unit in the Ministry of Finance and Economic Planning, headed by a director. Given the importance of revenue generation, the unit was upgraded and headed by a secretary of state who had direct access to the cabinet. The National Revenue Secretariat was thus given a greater degree of autonomy than its predecessor in the pursuit of its functions. The advantages of the autonomous unit were that the chain of command to the minister of finance was shortened considerably, and policy issues could reach the cabinet more quickly. The secretariat was also responsible for the 1986 restructuring of the Customs and Excise Department and the Central Revenue Department, which were formerly part of the civil service. These departments were reorganized into autonomous units headed by commissioners. At the time of the restructuring the border guards were absorbed by the Customs and Excise Department to reduce the number of tax collection units, and thus reduce costs. The Customs and Excise Department became the The Current Reform Program: 1983 to the Present 23 Customs, Excise, and Preventive Services (CEPS). The Office of the Revenue Commissioners was established, with units in each regional capital staffed by people committed to the program. 4.27 The issue of incentives became relevant. It was recognized that there was a need for improved quality of personnel, as well as personnel committed to revenue collection. It was unlikely that the graduates and professionals required by these agencies could be recruited given the unattractive pay conditions in the civil service, and it was impossible at the time to provide improved earnings for all civil service personnel. By removing the tax collection units from the civil service, this problem was solved at least cost. The CEPS used a bonus system to reward employees with a set proportion of the additional revenues collected. 4.28 To encourage compliance there was a reduction in income tax bands from nine in 1985 to six (including exemptions). There was also a reduction in tax rates from 55 percent to 35 percent. 4.29 Novel methods were devised to generate taxes from the informal sector, which normally escaped the tax net. The informal sector consists largely of illiterate and semi-illiterate entrepreneurs, who do not usually keep accounts. They could not be expected to complete tax forms for their incomes to be assessed and taxed. To overcome this difficulty standardized tax rates for the trades were established. The secretariat solicited the assistance of artisanal groups to collect taxes on behalf of the Internal Revenue Service, and paid them a commission for their efforts. One of the largest such organization is the Ghana Private Road Transport Union of TUC (GPRTU), which collects daily taxes on transportation and retains a predetermined percentage as commission. A paramilitary unit was established within this organization, with training provided by the govemment, to be responsible for the inspection of daily transport payments by drivers. Although it has been successful in collecting taxes on a daily basis, because the GPRTU was involved in the determination of the tax rate, it became a regressive system. Government was unable to increase the rates over a four-year period, and it was not until 1994 that the rates were reviewed. 4.30 Govemment expenditure policy was not geared toward cutting expenditures. Instead it was designed to achieve levels of outlays or expenditures consistent with macroeconomic stability. The budgetary process in Ghana usually consists of four interrelated stages: formulation, authorization and approval, implementation, and evaluation. This process had fallen into disuse during the pre-ERP period, but it was reactivated and strengthened as a basis for control and rationalization of expenditures. In addition to this procedure, there were a number of units to deal with audits and the control of expenditures. These included the Expenditure Monitoring Unit, established within the Ministry of Finance and Economic Planning in 1989. The Financial Administration and Regulations, 1979, provides the framework for expenditure control, disbursements, and accountability. 4.31 Several other policy initiatives were undertaken to achieve expenditure objectives. The wage determination process was rationalized, and govemment made it clear that large wage increases would not be granted. Beginning in 1986, wage increases in subvented institutions were to be based on inflation, productivity, and profitability. In subsequent years wage payments were 24 Governance and Structure Adjustment in Ghana computerized to remove the substantial number of nonexistent workers on the payrolls. Several allowances that had been granted in response to political pressures in the previous period were consolidated into a basic wage or salary. 4.32 Although these measures could substantially reduce the growth of government expenditures, several subsidies remained in the system, the public service was overstaffed, and nonfunctioning state enterprises constituted a drain on the budget. Public Sector Reforms 4.33 The policy toward state enterprises had several dimensions. The first was divestiture and outright liquidation. There was also the reorganization component, which initially involved ,he signing of performance contracts with the government. The laws that established state enterprises were repealed, and they were eventually replaced with the Statutory Corporations Conversions Bill, which allows them to register as limited liability companies with boards of directors. Their budgetary proposals have to be part of a three-year, rolling plan. If deficits were incurred for three years in a row, the companies would be liquidated. 4.34 Progress in the state enterprise divestiture program has been slow. Government's ambiguous attitude toward privatization in general, and foreign direct investment in particular, may explain why the program has not met with much success. There is a credibility gap because of previous government action-for example, the early support of the takeover of factories by workers, as well as statements by the leadership. Inadequate staff capacity to prepare corporate plans and performance agreements also initially slowed the process of monitoring firms that remained state-owned, and there have been difficulties in identifying the assets and liabilities of the corporations. 4.35 It was recognized that there was a need to increase salaries in the public sector to improve productivity, as well as to retain qualified personnel. Real average monthly earnings had been declining since 1974, and by 1983 they were about 14 percent of their 1974 levels. Increases in salaries, given the number employed, would have disrupted the government's effort to meet expenditure targets. It was established that there was a significant amount of overstaffing among the lower grades and a substantial amount of underemployment. To achieve the objective of improving productivity and remuneration, it was decided that shedding of labor in the lower grades was the solution. Labor would be laid off, with full benefits paid. It was expected that the laid-off workers would be actively employed in other sectors of the economy. It was decided that 45,000 workers would be laid off between 1987 and 1990. The actual figure separated from service was 29,000 by August 1990, and an additional group of 12,000 workers was let go in 1990. It has been much easier to lay off workers in the civil and educational service than in other areas. The layoffs in the state-owned enterprises have been slow because of the problems these enterprises face in paying the end-of-service benefits. It was expected that by 1988, 63,000 workers would have been laid off in the state-owned enterprises; by 1990, 45,000 workers had lost their jobs. There has been concern that recruiting procedures have not been strictly followed. Despite this concern, public sector employment fell from 464,300 in 1985 to 186,300 in 1991. The Current Reform Program: 1983 to the Present 25 Provisions were made for the separated workers in the budget under the special efficiency heading, which was established to finance the redeployment program, among other efforts. 4.36 In addition, attempts were made to improve civil service analytical, implementation, monitoring, and administrative capacity. As far as back as 1987, the government had offered a compensation package to attract Ghanaians resident abroad into the service. At the same time, donor support was available to attract high-caliber Ghanaian professionals to positions in the civil service at differential salaries. This attempt failed, to a large extent because of civil service rejection of differential salaries for work that they considered similar to their own. Funded positions remained unfilled because of the attitude of civil service personnel. In a large number of cases, government resorted to employment of foreign consultants. This, however, did not improve the capacity of the civil service, nor did it improve the efficiency of the work. As in the earlier reform, ministries were expected to set up policy, monitoring, and evaluation units to be staffed by top-level professionals. Computerization of the ministries and the provision of office equipment and vehicles was arranged to facilitate the operations of the ministries. While donors were supporting these reforms, they were also fragmenting the civil service through the establishment of special units that offered substantially higher pecuniary and nonpecuniary benefits for the implementation of donor programs. Each program or project had a secretariat staffed by some of the best individuals from civil service. It seemed that there was competition among donors to establish such units for their individual projects. It is understandable that each donor wanted to effectively implement its own projects or programs, but there were two problems with this approach. The first was the fragmentation of the service. The second was that the processes and methods used in the projects did not filter through to the rest of the service, and thus were not internalized. The Response of the Economy to Reform 4.37 The data in Table 5 suggest that the response of the economy to the reform program can be divided into two periods: the initial years, 1984-89, and the period of political transition, 1990-93. 4.38 In the 1984-89 period, real GDP grew an average of 5.7 percent a year. The rate of inflation fell to 40 percent in 1984 from 123 percent in 1983, and fell further, to 10 percent, in 1985. Export volumes increased by 2.02 percent in 1984, compared with a decline of 27.8 percent in the previous year. The ratio of the current account deficit to GDP declined, and there was a sharp drop in the ratio of the parallel and the official exchange rate. 4.39 The significant relative price changes caused by the devaluation, the existing excess capacity, the large parallel economy, and the good rains of 1984 brought about an immediate, substantial response to the program. The existence of the parallel market meant that there was a shift of supply from the parallel market outlets to the official market outlets, so that there were no significant costs involved in responding to the change in relative prices. 26 Governance and Structure Adjustment in Ghana 4.40 Between 1990 and 1993, however, the rate of GDP growth decelerated. From an average of 5.7 percent between 1984 and 1989, it dropped to an average of 4.3 percent a year between 1990 and 1993. The rate of inflation also increased sharply in 1990. The external current account deficit, which had remained below 2 percent of GDP, increased sharply to 4.4 percent of GDP in 1990, and has remained relatively high. 4.41 A similar pattern of dichotomy between the earlier years and the 1990-93 period can be observed in the fiscal response to the adjustment program. Table 5: Selected Economic Indicators, 1984-92 (percent) Item 1984 1985 1986 J987 1988 1989 1990 1991 1992 GDP growth 8.6 5.1 5.2 4.8 5.6 5.1 3.3 5.0 3.9 Agriculture growth 9.7 0.6 3.3 0.0 3.6 3.6 -2.0 4.7 -0.6 Industry growth 9.1 17.6 7.6 11.5 7.2 2.6 6.9 3.7 5.8 Services growth 6.6 7.5 6.5 9.4 7.8 6.7 7.9 6.3 7.7 Per capita income growth 5.9 2.4 2.5 1.2 3.4 3.3 -0.5 2.0 0.9 Investment ratea 6.9 9.6 9.7 13.4 14.2 15.5 16.0 15.9 12.6 Public sector financial balance" b -2.3 -3.0 -3.3 -2.4 -2.8 -2.2 -2.4 Private sector financial balanceab 1.3 0.6 1.8 0.3 1.1 0.4 -2.0 Export volumec 85.7 103.7 114.9 124 139 166.6 170.9 182.5 188.5 Import volume' 74 82 94 106 111 119 122.1 131.8 134.8 External current 3ccount balancea - 1.0 -2.5 -1.5 -2.1 -1.7 -1.8 -4.4 Inflation 40 10 24 40 31 25 37 18 10 Credit to central government (net)d 37.46 48.05 77 157 148 110 119 87 Credit to nonfinancial public 4.29 16.32 20 20 8 41 30 47 institutionsd Real effective exchange rate 72.2 52.5 30.2 23.2 22.1 20.8 20.8 21.6 19.3 Notes: a: Percent of GDP. b: The public sector financial balance is public savings minus public investment. The private sector financial balance is private savings minus private investment. c; Base year is 1980. d: Million cedis. Source: Statistical Services Ouarterly Digest of Statistics, various issues; IMF Intemational Financial Statistics Yearbook 1993. Fiscal Response: 1983-89 4.42 The initial period, 1983-89, is characterized by an improvement in the ratio of the budget balance to GDP and an improvement in the ability to achieve budget balance and revenue targets. Total expenditures were within the target area for most of the period, but this was achieved at the expense of restraints on capital spending (see Table 6). 4.43 Implementation of the ERP saw an almost immediate reversal of the poor state of budgetary affairs. The budget deficit/GDP ratio fell to 1.8 percent in 1984, with a surplus in 1986. The improvement in the ratio was the result of success in keeping the size of the budget The Current Reform Program: 1983 to the Present 27 deficit below the 1982 levels and the rapid increase in output. Until 1988 the size of the actual budget deficit was smaller than had been presented in the budget proposals, and the surpluses achieved were larger than expected (Table 6). 4.44 The ability to achieve tax revenue targets improved considerably between 1984 and 1988, with an increase in the probability of achieving targets, as well as a widening of the sources of taxes brought about by the improved efficiency in tax collection. 4.45 In 1987 and 1988 actual tax revenues were larger than had been budgeted (Table 6). Grants have been important in contributing to the budget surpluses; if they had not been included in the calculation of the budget balance, it would have registered a deficit in each year. 4.46 Tax revenues as a percentage of GDP increased from 4.6 percent in 1983 to 12.78 percent in 1987, and they remained at approximately 12 percent (Table 6). The tax base increased because of the increase in output, the improved efficiency in tax collection, and the widening of the tax base. The substantial aid inflows not only allowed an increase in imports, but also permitted the government to embark on rehabilitation of the economic and social infrastructure, which would contribute to building its support base in the rural areas. Nominal and real revenues grew continuously between 1983 and 1988, and the rate of growth of nominal revenues decelerated in each year between 1986 and 1990. Revenues had declined to such low levels in 1983 that the sharp increases in 1984 and 1985 are to be expected. 4.47 The improvement in the budgetary situation occurred within the context of increases in government expenditures. The substantial increases in tax revenues, combined with foreign grants, permitted increases in discretionary expenditures and improved governance capacity. This is the one reason-unlike the experiences of other countries-that the Ghanaian reforms proceeded without substantial reductions in government expenditures and their resulting adverse effects. Nominal and real expenditures grew faster between 1983 and 1989 than in the seven years prior to reform. Nominal revenues grew an average of 55 percent between 1983 and 1989, compared with an average of 36 percent between 1975 and 1982. The rate of increase in nominal expenditures slowed between 1985 and 1989 after the initial increase of 81 percent between 1983 and 1984 (Table 6). 4.48 Expenditures would have grown more rapidly if restraints had not been imposed. In most years this restraint was achieved through cuts in capital spending. In 1985, for example, total expenditures were limited because of insufficient foreign and other noninflationary inflows. Planned net lending to state enterprises had to be reduced significantly, and development projects with a low priority ranking were not implemented. Despite the restraints on the expansion of capital expenditures, its share of total expenditure increased from 8 percent in 1983 to 21.9 percent in 1989 (Table 7). Development spending has constituted a significant proportion of this category of expenditure. 4.49 The wage and salary increases, as well as the restructuring of the pension scheme, have affected expenditures and can explain some of the difficulties in restraining current expenditures during the first four years. Nominal wages in the public sector increased substantially during the period of economic reform. The increase in monthly average earnings Table 6: Budgetary Performance, 1983-94 S Item 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 19940 C) Budget balance (million cedis) Actual -4.582.6 -5,053.1 -5,453.0 2,985.8 8,920.4 6,451.5 -549.9 -11,517.4 50,428.4 -144,557.0 -118,175.0 42,951 Proposed -10,811 -5,878.7 -8,610 -3,448 1,081 3,855 10,000 22,625 43,921 47,413 15,257 111,709 Actual, as proportion of target Revenues Tax 0.728 0.907 0.965 0.864 1.025 1.059 0.932 0.856 0.89 1.05 0.931 0.955 c' Grants 0.356 18.280 8.778 0.699 0.765 1.006 1.243 0.778 1.079 0.829 0.992 0.662 o Expenditures c Total 0.627 0.937 0.943 0.799 0.937 1.039 1.002 0.992 0.995 1.301 1.166 1.077 > Current 0.882 1.04 1.08 0.936 1.003 1.118 1.095 1.08 1.056 1.321 1.226 1.059 Capital 0.147 0.546 0.499 0.382 0.874 0.802 0.307 0.28 0.276 0.391 0.377 1.129 Ratio of budget CD balance / GDP -2.49 -1.86 -1.58 0.58 1.12 0.06 -0.04 -5.67 1.95 -4.8 -2.9 Ratio of tax revenue /GDP 4.58 6.59 9.28 12.04 12.78 11.96 12.64 10.54 11.28 11.17 13.12 C5 Ratio of expenditures /GDP 8.01 9.86 13.34 13.81 13.68 13.6 13.86 12.52 13.21 16.7 19.34 Annual Percentage Changeb Revenues Total 110.91 112.87 78.60 90.59 50.82 35.39 30.30 24.63 46.10 1.39 67.73 Tax 11140 78.57 93.42 54.90 31.84 42.42 19.60 35.60 15.68 53.56 Expenditures Total 60.03 80.91 71.44 54.40 44.54 40.89 36.53 29.53 33.71 46.60 52.54 Current 51.91 72.04 70.58 56.48 31.76 41.47 32.19 31.12 33.39 40.40 62.5 Capital 22.49 182.65 78.08 40.65 139.37 38.57 54.57 23.49 32.64 69.94 20.97 Notes: a' Provisional estimates obtained from the 1995 budget statement. b: Nominal value. Source: Budget Proposals obtained from various budget statements. Actual values obtained from Statistical Services, Ouarterly Digest of Statistics, various issues. The Current Reform Program: 1983 to the Present 29 during the period from 1983 to 1986 was higher than increases in the five years prior to economic reform, and between 1984 and 1986 it was higher than the rate of inflation5 (Jebuni, Oduro, and Tutu forthcoming). Salary-related expenditures increased from 23.3 percent of total expenditures in 1983 to 38 percent in 1986, and fell to 34.5 percent in 1989 (Table 7). As a share of current expenditures, however, this item increased from 27 percent in 1983 to 46 percent in 1987, declining to 44.3 percent in 1989. Table 7: Composition of Expenditures, 1983-94 (percentage) Current Employee Captial Interest Year expenditures compensation expenditurs Transfers payments 1983 91.9 23.3 8.08 28.58 14.94 1984 87.3 19.79 12.62 17.39 12.83 1985 86.8 31.74 13.06 15.47 11.11 1986 88.1 38.04 11.46 15.60 15.46 1987 80.3 37.53 19.70 17.40 10.36 1988 80.6 34.21 19.38 19.20 8.31 1989 78.1 34.57 21.94 16.16 9.18 1990 79.1 34.62 20.91 18.48 10.34 1991 79.2 31.47 20.75 17.36 12.18 1992 75.9 34.87 24.05 15.29 12.23 1993 80.9 32.94 19.07 16.77 17.86 1994 72.9 32.6 27.1 20.02 Source: Statistical Services, Ouarterly Digest of Statistics, various issues; Republic of Ghana, The PNDC Budget Statement and Economic Policy for 1992. Provisional estimates for the year 1994, Budget Statement and Economic Policy of the Government of Ghana for the Financial Year 1995. 4.50 These salary increases may be considered a means of gaining support for the economic reform program. They represented an attempt not only to gain the support of the general populace, but also to reduce the discontent of the bureaucracy, which had benefited from the rent-seeking system that was to be abolished. This support was achieved in the short run at the cost of the failure to implement fully the capital expenditure program. The redeployment program thus provided some relief. Recorded public sector employment declined by more than half between 1985 and 1990. Fiscal Response, 1990-94 4.51 Unlike the earlier period of adjustment, the last four years have seen the emergence of large budget deficits, a decline in the ability to attain tax revenue targets (in contrast to the improvements of the previous period), and large expenditure overruns (Table 6). These developments may be attributed to the political developments of the time-the transition 5 The minimum wage rose from C35 in April 1984 to C52 in December 1984 and to C70 in January 1985. 30 Governance and Structure Adjustment in Ghana to democracy, institutional restructuring, laxity as the aid inflows increased, and the changing structure of government spending. 4.52 In 1988 the secretary of state of the National Revenue Secretariat was given additional responsibilities, and by 1991 the secretariat had lost its ministerial status. The secretariat was brought back under the Ministry of Finance and was headed by a director responsible to the minister of finance and economic planning. It maintained a certain degree of autonomy, and it did not have to go through a long chain of command to reach the minister. This could be interpreted as complacency on the part of decisionmakers, who may have considered that the favorable tax performance of the previous years would continue. An alternative explanation is that the targets became too ambitious given past achievements. It is unlikely that this is a major reason, because the rate of increase in proposed revenues was much lower after 1988 than it had been earlier. Another possible explanation may be laxity because of substantial foreign aid inflows. Between 1988 and 1991, foreign grants tended to be larger than expected (Table 6). The movement into indirect monetary control through Treasury bills made it easier for the government to finance its deficit by borrowing from the nonbanking public. This may have reduced pressure on government to generate resources through taxes. 4.53 The capital expenditure budget has had to be more severely constrained since 1988 than in previous years; at the same time, the overrun of the current expenditures has worsened. Expenditure slippages have become more pronounced since 1992 (Table 6). In that year salary-related payments in the budget increased by 62 percent as government responded to pressures from the civil service to raise salaries during an election year. Capital expenditures increased by 69.9 percent at the same time, largely because of the quadrupling of the Special Efficiency Fund, which makes payments to redeployed workers. 4.54 Interest payments on public debt have been rising since 1989, both in absolute terms and as a share of total expenditures. The rise in interest payments on domestic debt has been quite dramatic. In 1989 the interest on external debt exceeded domestic debt interest payments. By 1991, domestic interest payments had quadrupled in nominal terms and were twice the value of payments on external loans. In 1993 interest on public debt had risen to 18 percent of total expenditures, and provisional estimates for 1994 indicate that they have risen to 20 percent (Table 7). The increase in nominal interest rates since 1990-as government financed its deficit by borrowing from the general public-can explain the growing significance of this line item. The interest payments may be perceived as introducing rigidity into expenditures, because these are obligations the govemment has to meet. 4.55 A review of public expenditures in 1993 indicated that the expenditure control mechanisms have not been operating as they should. It was found that ministries were hot following the established procedures for the purchase of supplies (Ministry of Finance and Planning 1994, p. 19). There have also been problems in enforcing the rules governing the authorization of transactions financed by the government's accounts. Prior approval from the controller and accountant-general's office is required, but this does not happen. In the 1989 budget statement the expenditure overruns were attributed to the depreciation of the cedi and the policy of improving the quality of repairs. It is not entirely clear how the latter can explain these The Current Reform Program: 1983 to the Present 31 overruns, since it would be expected that the projected costs would have incorporated the required quality standard. In the 1990 budget, concern was expressed again about the problem of excessive spending. The measures suggested to control this problem-ensuring that tender values do not exceed programmed expenditures, strict enforcement of regulations concerning variation orders, and ensuring that contractors conform to the investment schedule and financing plan assumed in the public investment program-may be indicative of the underlying causes. Most of the laxities in expenditure control and the budgetary system were in existence prior to 1989. 4.56 The question then is why the problem did not emerge during the years from 1983 to 1988. It is possible that complacency, which has been suggested as the explanation of the revenue situation, may be relevant here. The substantial inflow of foreign financing between 1989 and 1991, which was equivalent to about 10 percent of revenue during those years, may explain both the lack of pressure to generate domestic resources and the pressure on ministries and departments to spend. 4.57 The problem may have been compounded in 1987, when monthly average earnings deflated by the consumer price index increased only 1.2 percent, and fell by 0.14 percent in 1988. The dismantling of the control system had removed rent-seeking activities as a supplement to slowly increasing real incomes. The alternative may have been to take advantage of the pressure to spend brought about by the inflow of foreign grants and loans. Ghana's involvement in peace-keeping operations in Liberia and the hosting of the NAM conference in 1991 also put pressure on expenditures. The transition to democracy in 1992 had no small effect on the size of government expenditures, and resulted in the largest expenditure overrun since the economic reform program in 1983. 4.58 In 1994, despite the failure to achieve revenue targets, disappointment in aid inflows, and the expenditure overruns, a fiscal surplus was recorded. I'his is explained by the inclusion of divestiture funds in above-the-line items. 32 Governance and Structure Adjustment in Ghana 5 Governance Capacity and the Adjustment Program 5.1 This study has defined governance capacity as being determined by the ability to articulate an alternate vision and present a credible program, the commitment and credibility of the implementing government, the ability to raise government revenues substantially, and the ability to develop alternate supportive interest groups while remaining autonomous. This section examines the extent to which Ghana's program and implementation were consistent with these criteria. The Initial Phase, 1984-89: Commitment and Credibility 5.2 Recent literature based on structural adjustment experience suggests that commitment and credibility are essential for successful adjustment. 5.3 The PNDC government, unlike the Busia government, showed its leadership by the decision to change course when it realized that its initial program was worsening the economy. Once it had made the decision, it committed itself to the new course of action. This commitment was indicated in a statement made by the chairman in the aftermath of the coup attempt two months after the launching of the Economic Recovery Program (Rawlings 1983, p. 25). We may anticipate that soon the floodgate of protests against human rights violations will be opened as the next tactical move of organized hypocrisy, but for the defence of our revolution we will run that risk. If we must use our strength to crush actions which threaten the process so be it. 5.4 The repressive nature of the regime, with its intolerance of opposition, was indicative of purpose of intent. The departure of original members of the PNDC who held views contrary to those of Rawlings was also a positive signal of commitment to the chosen economic strategy. The Rawlings regime had earned a reputation for carrying out its intentions by handing over power to a civilian regime in 1979. 5.5 A second indication of the commitment of the government is the speed of implementation of policy measures crucial to the achievement of its stated objectives. A major objective of the program was to improve the incentives for production and export and to increase the availability of foreign exchange. The immediate exchange rate policy actions implemented in 33 34 Governance and Structure Adjustment in Ghana April 1983 and the subsequent, frequent devaluations between October 1983 and September 1986 must have sent strong signals that policy reversal was unlikely. No previous government had undertaken such a dramatic devaluation of the currency, and continued with a series of devaluations despite the dissatisfaction expressed by the populace. 5.6 A decline in the size of the budget deficit was achieved in 1983, despite the increase in nominal government expenditures. This was managed by the more than doubling of revenues that accompanied the exchange rate adjustments. Another indicator of the improvement in the fiscal effort was adherence to established targets. The actual deficit in 1983 was less than half that proposed in the budget. This was achieved primarily through restraint in the expansion of expenditures, particularly capital expenditures. Actual recurrent and capital expenditures in 1983 were 88 percent and 14.7 percent, respectively, of their proposed levels. 5.7 It is unlikely that a problem of time inconsistency emerged during the initial three years of the program. It was clear that the policies pursued prior to 1982 had failed, and there was a need to adopt and implement a new approach. At the Paris donor conference, financial resources had been pledged over a three-year period, thus committing the government to the program over that interval. The lack of alternative sources of funding at the time was another important determinant. 5.8 As a result of this commitment, Ghana achieved a 58 percent compliance with the conditionalities agreed upon with the World Bank and International Monetary Fund during the 1983-88 period. This is considerably above the compliance efforts of Kenya, at 38 percent, and Ecuador, at 13 percent (Mosley, Harrigan, and Toye 1991, p. 136). Trade reform was extensive and rapid; according to Levy (1993), complete compliance was achieved. 5.9 The government's commitment to implement the program and its technical soundness made it credible to foreign donors. This generated a substantial inflow of the external resources that were crucial to the success of the changes introduced. As previously noted, these resources were important in the rebuilding of the dilapidated infrastructure, the rehabilitation of the traditional export sector, budgetary support, and the financing of the foreign exchange auction system introduced in 1986. 5.10 On the domestic front, however, there was a credibility gap. The government's ideological stand prior to reform did not endear it to the private Ghanaian business community. Even when the government had shown its changed perspective by introducing market reform, many private sector operators remained skeptical. Their doubts were increased by the government's ambivalence toward the private sector. In part, this attitude must have been the product of a deep ideological conviction that economic reality forced it to change. The flow of external assistance contributed to this ambivalence, because it meant that development of the private sector was not essential to the initial implementation and success of the program. Governance Capacity and the Adjustment Program 35 Program Articulation and Interest Groups 5.11 Figure 2 shows that the program was implemented in Ghana without the kinds of protests that usually accompany such reforms. Compared with the previous reform attempt, when numerous industrial strikes arose to protest the reforms, there have been no significant labor protests against the current program. Labor unrest has been nowhere near the levels of the 1970s or 1980s. Indeed, strikes have never been as uncommon during Ghana's postindependence experience. It is argued in this section that this was, in part, the result of the manner in which the program was presented to the Ghanaian public and the ability of the government to organize alternative interest groups while escaping capture. Figure 2 GHANA: WORKING DAYS LOST (1973 - 1991) 250 200 o ' 150 (n 1 < 100/ 50 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1974 1976 1978 1980 1982 1984 1986 1988 1990 YEARS DAYS LOST 36 Governance and Structure Adjustment in Ghana 5.12 Although the program was drawn up without the broad consultations that had taken place in the previous reform program, it was presented to the nation as the only viable option for national survival. In its presentation, the government portrayed itself as placing the national interest above its own survival. This was made clear in radio and television broadcasts to the nation after the launching of the program in the 1983 budget. Only a government which places the interest and the very survival of the nation before its own interests would have come out with such an austere budget. If the PNDC Government was concerned with cheap popularity, we would not have presented you with such a budget. The medicine is bitter, I agree, but it is necessary [Rawlings 1983, p. 16]. 5.13 Opponents of the reform program were portrayed as opportunists seeking their own self-interest: "They have suddenly and hypocritically become defenders of the poor workers, but only as long as this will help them regain their lost power and privileges" (Rawlings 1983, p. 16). 5.14 As discussed earlier, one of the main expected losers in the reform program was the urban population, particularly the labor movement, which is also the most politically vocal group. Fear of urban unrest had led previous governments to tax rural farmers to subsidize urban consumption through the imposition of a controlled trade and price regime. The urban population had also benefited from subsidies on imported commodities made possible by the overvaluation of the currency, as well as explicit subsidies. For instance, while it was popularly believed that the subsidy on kerosene would be of greatest benefit to rural farmers, the major portion of the subsidy went to urban consumption. Estimates in 1983 showed that 52 percent of kerosene was consumed in urban areas, together with industrial and commercial users. 5.15 The PNDC's initial conception of the economic woes of the country was that they were the legacy of the exploitation of the masses and the corruption of the system. This belief was important to the extent that it recognized that reform of the system required the establishment of alternate interest groups. This was the basis of the formation of the Workers Defence Committees (WDCs), the Peoples Defence Committees (PDCs), and the Civil Defence Organization (CDO). These were voluntary organizations, funded by the state. Formed to mobilize the people for development, these groups were important in organizing on the shop floor and neutralizing the possible protests and threats to the program from the Trade Union Congress (TUC). The laws and regulations governing public boards and institutions, as well as state enterprises, were changed to ensure that members of the WDCs were included in their boards of directors. The WDCs increasingly became the center for the expression of worker interests; some became managers of state enterprises. Their survival depended on the survival of the regime itself. 5.16 Government repression also played a major part in the relative ease with which Ghana's program was implemented. The philosophy of the government was that individual rights were of less importance than the collective rights of the people. One official observed of the PNDC: "This government was prepared to take action. It also had a strong constittiency among Governance Capacity and the Adjustment Program 37 those who hold the gun. The population knows that if you complain, you will be silenced. If you did misbehave you would be taken care of'(quoted in Herbst 1991). 5.17 In a critique during his 1988 Danquah Lectures, the role of repression is made clear by Adu Boahen: "We have not protested or staged riots not because we trust the PNDC but because we fear the PNDC! We are afraid of being detained, liquidated or dragged before the CVC or the NIC or being subjected to all sorts of molestation" (quoted in Jeffries 1991, p. 169). 5.18 This repression, compounded by state control and monopoly of both the print and electronic media, made it difficult to determine the extent of support for the program among workers and other interest groups, especially the urban masses who have been the traditional enemies of market-based economic reforms. The increase in real wages that occurred in the first three years of the program suggests that it must have favorably affected the economic conditions of the urban masses-or at least did not worsen them. As indicated earlier, the economic situation before the reform program was such that most Ghanaians had to purchase commodities in the parallel market. Very few, if any, were benefiting from the controlled system. Furthermore, to cope with the hardships most people had taken second jobs, so that a decrease in wages or outright loss of a job would not have meant as much as it may have seemed (Herbst 1991). 5.19 From 1987, however, when employment rationalization in the public sector and the divestiture of state-owned enterprises were introduced, things began to change. Redeployment had a checkered history. It appeared unpopular at the beginning; the first group of workers who were laid off did not receive their terminal benefits or compensation for a long period of time. The payments system was improved, however, and the prompt payment of substantial accrued benefits persuaded some workers to volunteer to be laid off. By 1990, budgetary pressures on the government again created payments difficulties. Payments to workers of divested state enterprises also became a problem. These enterprises were making losses, and yet had money to award themselves several entitlements. The divestiture of these enterprises transferred the burden of the payment of entitlements to the state. 5.20 While the state was laying off workers, the private sector was also rationalizing employment in an environment that was increasingly competitive following the introduction of trade liberalization in 1986. The actions of the two sectors led to rising unemployment. Combined with cost recovery in education and health, workers felt betrayed by the government. Although the government argued that it had kept faith with workers and that the unemployment was short term, it is clear that these arguments were not convincing to those adversely affected. 5.21 Beneath the apparent calm generated by repression was a restive labor force held in check. The transition to political liberalization, as will be illustrated below, was to provide a vent and expose the government's vulnerability and failure to win the confidence of the urban masses. 38 Governance and Structure Adjustment in Ghana The Bureaucracy and Adjustment 5.22 The bureaucracy affects the implementation of economic reform programs in two ways: (1) through its technical and administrative capabilities and (2) through its role as a power block and interest group. 5.23 Pervasive administrative weakness hampered policy formulation and implementation in Ghana. In part, this arose from the general deterioration in the economy discussed earlier. The bureaucracy was preoccupied with regulating the system and issuing licenses and permits. Basic data gathering and analytic and accounting skills seemed to have disappeared. The one reason the Ghana's reform effort did not collapse is that the same senior professionals who had convinced the government of the need to change continued to do the relevant programming and policy formulation. 5.24 To compensate for this administrative weakness, according to Callaghy there was Ila real and quite visible resurgence of expatriate influence in Ghana-the nearly constant presence of IMF and World Bank personnel, visiting missions, hired consultants and seconded bureaucrats and managers. The whole recovery effort is a high-conditionality process, and the Fund, the Bank, and the donor countries believe that the expatriate personnel and their skills are necessary to ensure that their finds are used wisely" (Callaghy 1989 p. 127). 5.25 World Bank and the Overseas Development Agency (U.K.) financing of US$11 million and US$4 million, respectively, was provided for the Structural Adjustment Institutional Support project (SAIS), which was to run from 1987 to 1993. One of the project's objectives was to provide capacity building for the Ministry of Finance and Economic Planning staff to manage the public investment program, expenditure forecasting and control, and to establish a sound administration and personnel management system. The project achieved none of its objectives. An assessment of the project by the World Bank is that it did not succeed because Ghanaian administrators did not participate in its identification and preparation. 5.26 Callaghy estimates that the Bank alone sent more than forty missions to Ghana in 1987. The resentment of this expatriate personnel and their methods of operation clearly identified the program with external actors, and weakened its legitimacy. Even more important, the efficacy of expatriate staff and consultants in improving the capacity of the bureaucracy through learning-by-doing is questionable unless it is combined with training. 5.27 The attempt to reform and improve the capacity of the bureaucracy from 1987 onward did not succeed. This technical and administrative weakness must be partly responsible for Ghana's inability to absorb the increased external financing after 1989. 5.28 In the initial phases, program design and implementation were carried out by task forces and several committees. The result was that the top administrators of the policies were not involved in their design. This created implementation problems. As one top official of the Ministry of Finance and Economic Planning put it: "To the extent that they did not feel related to the formulation you could not count on them for implementation" (personal interview by the author). Governance Capacity and the Adjustment Program 39 5.29 As an interest group, the bureaucracy was opposed to the adjustment program. Their opposition was to be expected, because they were the main beneficiaries of the rents arising from the controlled system. Although these rents had disappeared with the virtual collapse of the economy, it was thought that any resumption of growth in the economy would see the reemergence of the rents. Up until 1989, when import-licensing was abolished, the bureaucracy still argued in favor of the imposition of controls. 5.30 The immediate increase in government revenues as a result of the reform program allowed for annual salary increases for the entire public service between 1983 and 1986. This provided partial compensation for the lost rent. The basis of the increases changed after 1986 to reflect productivity and efficiency gains. The proliferation of foreign funds and projects provided nonpecuniary benefits to the bureaucracy, and in some cases explicit wage subsidies. These activities gained the acquiescence of the bureaucracy, if not their voluntary compliance. 40 Governance and Structure Adjustment in Ghana 6 The Transition, 1990-94 6.1 The transition in Ghana is associated with a general deterioration in all indicators of economic performance. The rate of growth of the economy, which was above 5 percent annually between 1984 and 1989, dropped to 3.3 percent in 1990, and has averaged about 4 percent in the five-year period of 1990-94. The period of 1992-94 was supposed to be one of accelerated growth, with expected growth rates of between 6 percent and 8 percent. Instead, as indicated earlier, the rate of growth decelerated. Agricultural production decreased by 2 percent in 1990, and has since alternated between decline and positive growth. Wide divergences occurred between government program targets and outcomes. For all years since 1989, except for 1991 and 1994, huge budgetary deficits were recorded after surpluses had been projected. The 1994 surplus was recorded because of the sale of part of the government share in the Ashanti Goldfields Company (AGC). Rates of inflation were much higher than had been programmed. What happened? Was this a reflection of the general characteristics of transition to democracy, or was this a special Ghanaian phenomenon? 6.2 At the time of the adjustment program in Ghana, capacity utilization was very low and stocks were depleted. Capacity utilization in manufacturing was estimated at 18 percent in 1984. The initial response was thus based on utilizing this capacity. Flows of foreign aid allowed stocks to be built up and industrial capacity utilization to increase. In view of the depth of the crisis in Ghana, this process could have been expected to take between three and four years, followed by substantial investment to sustain the growth. This did not occur. There was a domestic credibility gap born out of the government's previous ideological flirtation with socialism, its attitude toward the private sector, and the use of repression. This manifested in the failure of domestic private investment to increase substantially. In a survey of manufacturing enterprises in Accra in early 1994, Asante (1994) found that 77 percent of the firms interviewed regarded the government attitude toward business as a problem for investment: 27.3 percent thought it was a major problem, and for 50 percent it was a moderate problem. Only 23 percent of those surveyed did not consider government attitude to be a problem. As shown in table 6, overall investment increased, but it remained low compared with other, less-developed countries. The expected private sector investment did not occur. The program was therefore largely driven by public investment, funded by external resources. When the level of external support expected did not materialize in the 1990s, the program ran into a crisis. 41 42 Governance and Structure Adjustment in Ghana 6.3 This credibility problem not only led to the nonresponse of private investment, but also to substantial increases in expenditures and an inability to regain control of expenditures after the elections held during the transition. Wide disparities emerged between government expenditures and revenues, which led to budgetary deficits. These, in turn, put pressure on prices and the balance of payments. 6.4 To compensate for the widening disparity between expenditures and revenues, government resorted to short-term borrowing on the domestic market. As discussed earlier, this resulted in substantial fixed interest obligations on the budget in subsequent years. Interest payments as a proportion of GDP, which were less than 1.5 percent between 1984 and 1990, rose sharply to an average of 2.9 percent of GDP between 1991 and 1994, with 1993 recording 3.4 percent and 1994 above 4 percent. These interest payments are dominated by interest on the domestic debt, which accounted for about 70 percent of total interest payments in the last two years. 6.5 The transition to democracy started in earnest in 1990. Some might argue that it started with the district elections in 1987, but although these elections were important in generating grassroots participation in local development efforts, they were not conceived in the context of a national transition to democracy. The national effort started when the Chairman of the PNDC, Flight-Lieutenant Rawlings, charged the National Commission for Democracy with collating views from all ten regions of Ghana on the evolution of a democratic system for Ghana in July 1990. This involved holding regional forums and seminars to solicit reactions on the way forward. Fo;lowing the submission of its report in March 1991, which recommended the establishment of a multiparty democracy, a committee of experts was established to draft proposals for the Fourth Republican Constitution. This was followed by a Consultative Assembly that considered the proposals and drafted a Constitution, which was subjected to a national referendum in 1992. National elections were then held and won by Flight-Lieutenant Rawlings and his National Democratic Congress (NDC). 6.6 The elections themselves generated political competition among the three main contending parties, the National Democratic Congress (NDC), the New Patriotic Party (NPP), and the People's National Convention (PNC). They did not generate substantially new economic ideas or alternatives to the program the PNDC government was pursuing. The main issues involved human rights and questions of accountability and probity. The result of the elections was a virtual one-party Parliament. 6.7 The stages involved in this process put pressure on government expenditures, even though some external assistance was available to finance the effort. All through the process, it was clear that the incumbent government was interested in retaining power. This self-interest exposed the government to two kinds of pressures to increase expenditures: pressures from wage demands and the need to increase discretionary expenditures for political gain (pump-priming). 6.8 There is always a tension in adjustment programs between the need to hold down public sector expenditures through wage restraint and increased wages as an incentive for greater productivity and compliance. Although wages must be related to productivity, there is also a The Transition, 1990-94 43 moral hazard in expecting honesty and increased productivity from underpaid workers. The Ghanaian program was never able to resolve this tension. In the initial phases the government functionaries were expected to receive their salaries from their initial places of work. Allowances to compensate for the increased responsibility were not fixed, at least for the first three years. The fear was that there would be spillover effects from such action on general wage demand. The same dilemma is indicated by the government's inability to fix wages, salaries, and allowances for its own ministers of state, parliamentarians, and other functionaries managing the institution established under the 1992 Constitution, almost two years into the Fourth Republic. In the past, government had responded to wage demands through threats to leadership of the labor movement, repression, or selective compromise. As indicated earlier, this did not reduce the government's vulnerability to the urban masses. When it became clear that the government was vulnerable because of its self-interest, demand mounted for wage increases. This resulted in a 60-100 percent salary increase for the civil service in 1992, which had an immediate spillover effect in the public sector as a whole. 6.9 Expenditures for political gain are undertaken by all governments, from the national democracies such as the United States, to transitional democracies such as Ghana. These expenditures usually occur under the capital account of the budget, particularly development expenditures. While one can document the increases in these expenditures, it is impossible to infer the motive. While ministry officials have indicated that the sharp depreciation in the cedi played a major role in budgetary overruns in development expenditures, they also indicate that part of the problem of expenditure control in this component is political. Implementation of politically sensitive projects that were not given priority, and then switching back to priority projects, has given the government considerable difficulties in monitoring these expenditures. 6.10 The government initially established credibility for taking action against revealed cases of corruption or abuse of office. The special organizations developed by government were important in exposing such corrupt behavior and malpractice. With political liberalization, both the emerging private press and the state-owned media started publishing stories of impropriety and malfeasance in public places. This undermined the moral authority of some of the bodies responsible for implementing the program, such as the Committee for the Defence of the Revolution, the Tribunals, the Citizens Vetting Committee (CVC), the National Investigation Committee (NIC), and the Civil Defence Organization (CDO). Some of the stories were certainly false, but they increased the public perception of corruption in the system. Failure of the government or Parliament to take public action on alleged instances of abuse of public office intensified the negative public perception of government credibility. The auditor-general's report on the public accounts of Ghana for 1992 revealed substantial evidence of embezzlement of funds and failure to comply with established procedures in the purchase of supplies and equipment. No court examination or official investigation of these allegations has been made; if any inquiries have been conducted, they have not been made public. The failure to take public action on allegations of abuse created a credibility problem for government and removed an essential aspect of governance-public trust of the behavior of officeholders. 44 Governance and Structure Adjustment in Ghana 6.11 The moral authority of the special interest groups, having been challenged by allegations of malfeasance, was damaged, and their legal foundation was removed by their lack of recognition by the Constitution. They were now perceived by the public as powerless, and a passing phase of Ghanaian history. The government was saddled with the problem of finding ways to absorb these groups into the system at a time when employment in the public sector was frozen and workers were being retrenched, to maintain them, or make some financial settlement with them. All three methods were tried. 6.12 Adjustment of revenue-collecting agencies and the bureaucracy to declining incentives was also important in explaining the slippages in the transition. This borders on the reciprocity aspects of governance. According to this concept, " Actors contribute to the welfare of others in the hope that, at some unspecified time in the future, they will receive some appropriate consideration in return" (Bratton and Rothchild 1992, p. 271). As discussed earlier, to motivate the revenue-collecting institutions, government had removed them from the civil service and provided improved salary and benefit conditions. These conditions were not maintained. For four years prior to the 1994 fiscal year, these incentives were gradually eroded by inflation. In 1994, however, salaries were increased. In the interim, the declining real wages led to declining staff morale and efficiency. This problem was acknowledged by government, although belatedly, when in the 1994 budget the minister for finance and economic planning stated: "Misdescription and misclassification as well as blatant collusion between importer and officials have sadly become rather pervasive" (Ghana 1994, p. 24). 6.13 Under the control regime, the bureaucracy extracts its rents mainly through the revenue side of the budget. When the system is liberalized and these revenue-side rents disappear, there is a tendency to adjust through the expenditure side. This could take the form of inflating expenditures or simply refusing to follow financial regulations. The slippage in adhering to financial regulations in the acquisition of supplies and the embezzlements recorded in the auditor-general's report could be interpreted as the adjustment of the bureaucracy to declining real wages in a liberalized economy. 6.14 This credibility gap was not balanced by a motivated and efficient bureaucracy. As indicated earlier, the attempt to build the capacity of the bureaucracy was a complete failure. 7 Conclusions 7.1 Ghana has gone through two experiences of economic reform programs: 1967-71 and 1983 to the present. The first program was initiated by an unelected military government that handed over power to an elected civilian government two years into the program. The second was implemented by a military dictatorship that evolved into an elected civilian government. Both programs involved movement away from a controlled economy to a more liberalized one. In both instances economic reform was necessitated by the slowdown in economic growth and internal and external disequilibria. The extent of the disequilibria, however, differed significantly between the two cases. 7.2 A certain pattern of performance can be observed for both programs. In each case, the initial phases of stabilization appeared to be successful. For both reform programs the shift to accelerating the growth rate was unsuccessful. While the first reform attempt was reversed, however, the current program has been sustained. 7.3 The differences, it is argued in this paper, arose because the first reform lacked both external and domestic credibility. This was the result of inconsistencies in the program, lack of a well-articulated alternative, the lack of commitment of government, and a weak leadership. The current program was largely successful in the initial phases, exhibiting capacities in all these areas. The program was technically sound, and the government was committed to its implementation. This made it credible to the external donor community and ensured substantial flows of the external assistance crucial to the initial phases. 7.4 The government was able to develop new interest groups that appeared committed to the program: the Committee for the Defence of the Revolution (CDR), the Civil Defence Organisation, 31st December Womens' Movement (31 DWM), Office of the Revenue Commissions, and the like. These groups were instrumental in articulating and "selling" the program to the Ghanaian public as well as raising substantial government revenues. They were also used to suppress the traditional interest groups that could have opposed the program. Nevertheless, the existence of these iiterest groups as the main source of government support could not be sustained during a democracy. There was thus a domestic credibility gap. 7.5 The creation of organizations and agencies to implement specific aspects of the program allowed government to effectively bypass the bureaucracy. The substantial and rapid increases in foreign inflows in the initial five years allowed the reform program to proceed 45 46 Governance and Structure Adjustment in Ghana without any special overtures to the private sector to participate in the program. Because of. the substantial foreign inflows, the problems arising from the existence of a domestic credibility gap did not emerge in the initial years. 7.6 The failure to restructure the bureaucracy to improve efficiency and commitment and to fully integrate the private sector into the decisionmaking process later emerged as problems that could have derailed the program. This was reflected in the increased expenditure overruns, the stagnant investment rate, and the slowdown in average growth rates. As the rate of growth of foreign inflows slowed, the weakness of the strategy of nurturing close links with the external donors without developing a broad-based domestic support network emerged. With a weak support base among the vocal urban groups, particularly because of the retrenchment program, the government's strategy of maintaining a support base through expenditure overruns threw the economy off balance. 7.7 The PNDC government under Flight-Lieutenant Rawlings mortgaged its political capital to carry out a far-reaching reform program in Ghana. It was largely successful in the initial years. In the transition, as the government's electoral survival became uncertain, it lost control of its expenditures. Getting back on track will depend on whether the NDC government under Rawlings and subsequent democratic governments consider that they have enough political space to exert some leadership and place long-term economic considerations ahead of short-term political gains. Appendix Principal Political Dates and Regimes March 6, 1957 Independence, with Nkrumah as prime minister and the Convention People's Party (CPP) as the ruling party. July 1, 1960 Ghana becomes a republic within the Commonwealth and Nkrumah becomes president. February, 1966 Military coup d'etat overthrows Nkrumah and establishes the government of the National Liberation Council. 1969 (August & October) Elections under new Constitution won by Progress Party, led by Busia as prime minister. January 13, 1972 Military coup d'etat overthrows Busia government and establishes the National Redemption Council (NRC) led by Colonel Acheampong. June 4, 1979 Military insurrection overthrows Acheampong and establishes the Armed Forces Revolutionary Council led by Flight-Lieutenant Rawlings. 1979 (June & September) Elections held under new Constitution and won by Peoples National Party led by Limann as president. December 31, 1981 Military coup d'etat overthrows Limann's government and establishes Provisional National Defence Council (PNDC) led by Flight-Lieutenant Rawlings. November 1992 Elections under new Constitution won by National Democratic Congress (NDC) led by Flight-Lieutenant Rawlings as president. January 7, 1993 Fourth Republic is inaugurated. 47 48 Governance and Structure Adjustment in Ghana Bibliography Agarwala, R. 1983. "Price Distortions and Growth in Developing Countries." World Bank Staff Working Paper 575. Washington D.C. Asante, Y. 1994. "Determinants of Private Investment Behaviour in Ghana." An Interim Report presented at the African Economic Research Consortium Workshop, Nairobi, Kenya. Bank of Ghana. 1971. Report of the Boardfor the Financial Year Ended 30th, June 1971. Accra. Bates. R. H. 1990. "Macropolitical Economy in the Field of Development." In J. E. Alt and K. A. Shepsle, eds., Perspectives on Positive Political Economy. Cambridge, U.K.: Cambridge University Press. Boeninger, E. 1991. "Governance and Development: Issues and Constraints." Proceedings of the World Bank Annual Conference on Development Economics: 267-87. Washington, D.C.: World Bank. Bratton, M., and D. Rothchild. 1992. "The Institutional Bases of Governance in Africa." In G. Hyden and M. Bratton, eds., Governance and Politics in Africa. Boulder, Colo., and London: Lynne Rienner. Callaghy, T. M. 1989. "Toward State Capability and Embedded Liberalism in the Third World." In J. M. Nelson and others, Fragile Coalitions: The Politics of Economic Adjustment, U.S.-Third World Policy Perspectives 12. Washington, D.C.: Overseas Development Council. Central Bureau of Statistics (CBS). Economic Survey of Ghana 1977-1980. Accra. V. Various issues. Quarterly Digest of Statistics. Accra. Collier, P., D. Greenaway, and J. Gunning. 1994. "Evaluating Trade Liberalisation: A Methodological Framework." Framework Paper, AERC Project Meeting on Regional Integration and Trade Liberalisation in Sub-Saharan Africa. Corbo, V., and J. de Melo. 1987. "Lessons from the Southern Cone Policy Reforms." The World Bank Research Observer 2(2):111-42. Fischer, S. 1986. "Issues in Medium-Term Macroeconomic Adjustment." The World Bank Research Observer 1 (2):163-82. Frimpong-Ansah, J. 1991. The Vampire State in Africa. The Political Economy of Decline in Ghana. London: James Currey. Ghana, Republic of, Ministry of Finance and Economic Planning. 1994. Public Expenditure Review, 1993. Accra. . Various years. Budget Statements. Accra. Herbst, J. 1991. "Labour in Ghana under Structural Adjustment. The Politics of Acquiescence." In D. Rothchild, ed., Ghana: The Political Economy of Recovery. Boulder, Colo., and London: Lynne Reinner. 49 50 Governance and Structure Adjustment in Ghana IMF (International Monetary Fund). 1992. International Financial Statistics Yearbook. Washington D.C. Jebuni, C. D., A. Oduro, and K. A. Tutu. Forthcoming. Trade and Payments Liberalisation and Economic Performance in Ghana. AERC Research Paper, Nairobi. Jeffries, R. 1991. "Leadership Commnitment and Political Opposition to Structural Adjustment in Ghana." In D. Rothchild, ed., Ghana: The Political Economy of Recovery. Boulder, Colo., and London: Lynne Reinner. Killick, T. 1978. Development Policy in Action. A Study of Economic Policies in Ghana. London: Heinemann. Lal, Deepak. 1987. "The Political Economy of Economic Liberalization." World Bank Economic Review 1(2): 273-99. Landell-Mills, P., and I. Serageldin. 1991. "Governance and the External Factor." Proceedings of the World Bank Annual Conference on Development Economics: 303-20. Washington, D.C.: World Bank. Leith, J. Clark. 1974. Foreign Trade Regimes and Economic Development: Ghana. New York and London: Columbia University Press. Leith, J. Clark, and M. F. Lofchie. 1993. "The Political Economy of Structural Adjustment in Ghana." In R. H. Bates and A. 0. Krueger, eds., Political and Economic Interactions in Economic Policy Reforrn. Oxford, U.K.: Blackwell. Levy, B. 1993. "An Institutional Analysis of the Design and Sequence of Trade and Investment Policy Reform." The World Bank Economic Review 7(2):247-62. Libby, R. T. 1976. "External Co-optation of a Less Developed Country's Policy Making: The Case of Ghana, 1969-1972." World Politics 29(1):67-89. Mosley, P., J. Harrigan, and J. Toye. 1991. Aid and Power. The World Bank and Policy-based Lending, Volume 1. London: Routledge. Ofori-Atta, J. 1975. "Devaluation-The Prelude to Self-Reliance in Ghana." The Economic Bulletin of Ghana (Third Series) I(1/2). Rawlings, J. J. 1983. Forging Ahead. Selected Speeches of Flt.-Lt Jerry John Rawlings, Chairman of the PNDC. January Ist 1983-December 31st 1983, Volume 2. Accra: Information Services Division, Ministry of Information. Rimmer, D. 1992. Staying Poor. Ghana's Political Economy, 1950-1990. Oxford, U.K.: Pergamon. Sachs, J. 1987. "Trade and Exchange-Rate Policies in Growth-Oriented Adjustment Programmes." In V. Corbo, M. Goldstein, and M. Khan, eds., Growth-Oriented Adjustment Programmes. Washington, D.C.: International Monetary Fund and World Bank. Bibliography 51 Seabright, P. 1994. "Governance and Structural Adjustment. Transactions Costs, Agency Theory and State Capture." Churchill College, Cambridge University, Cambridge, U.K. Statistical Services. Various issues. Quarterly Digest of Statistics. Accra. Stryker, J. D., and others. 1990. Trade, Exchange Rate, and Agricultural Pricing Policies in Ghana. Washington, D.C.: World Bank. PRIVATE SECTOR DEVELOPMENT DEPARTMENT OCCASIONAL PAPERS No. I Rhee, Katterbach, Belot, Bowring, Jun and Lee, Inducing Foreign Industrial Catalysts into Sub-Saharan Africa No. 2 Mody and Wang, Explaining Industrial Growth in Coastal China: Economic Reform ... and What Else? No. 3 Biddle and Milor, Institutional Influences on Economic Policy in Turkey: A Three-Industry Comparison No. 4 Lanjouw and Mody, Stimulating Innovation and the International Diffusion of Environmentally Responsive Technology No. 5 Tan and Batra, Technology and Industry Wage Differentials: Evidencefrom Three Developing Countries No. 6 Navarro, Reversal of Fortune: The Ephemeral Success of Adjustment in Venezuela, 1989-93 No. 7 Morales, Bolivia and the Slowdown of the Reform Process No. 8 Ibrahim and Lofgren, Governance and Structural Adjustment: The Egyptian Case No. 9 Tan and Batra, Enterprise Training in Developing Countries: Overview of Incidence. Determinants, and Productivity Outcomes No. 10 Sundaram, Teik and Tan, Vision, Policy and Governance in Malaysia No. 11 Tzannatos, Labor Policies and Regulatory Regimes No. 12 Kessides and Willig, Competition and Regulation in the Railroad Industry No. 13 Nagaoka, Antidumping Policy and Competition No. 14 Atiyas, Bankruptcy Policies No. 15 Brandao and Feder, The Case of Land Markets