IV Macroeconomic and Other Nonfiscal Policies and Their Implications for Poverty
- International Monetary Fund
- Published Date:
- May 1988
All of the programs included a combination of policies to limit absorption, expand supply, and promote structural adjustment. For demand restraint, limits on credit expansion and fiscal adjustment were common policy elements, with the additional objective of limiting the government’s crowding-out of the private sector. The combination of these two policies implied a reorientation of domestic credit expansion in favor of the private sector. In most cases, these policies were complemented by pricing, wage, and exchange rate policies. The pricing and exchange rate policies aimed at promoting the production of specific groups of commodities (mostly tradables) and at containing the domestic demand for these products, whereas the labor market policies sought to restrain wages and increase employment. Structural measures sought a medium-term improvement in the functioning of product, factor, and foreign exchange markets.
Obviously, the precise mix and emphasis on policies differed across programs, reflecting the nature of the problems facing each country at the time of the program. In particular, the program for Ghana focused on radical changes in the severely misaligned exchange rate and domestic price structure, supported by a reduction in pressure from the fiscal sector. Exchange rate policy played an essential role also in the programs with Chile, the Dominican Republic, and the Philippines. Restrictive wage policies in the public and private sectors accompanied a depreciation of the exchange rate in Chile and the Dominican Republic. In the Philippines (1984-85 program), tight monetary policies were a key element of the program. Fiscal adjustment in the programs of Kenya, Sri Lanka, and the Philippines relied more on restraining expenditure than on increasing revenue, while the program with Thailand relied more on revenue measures. For the Dominican Republic and Sri Lanka, the reduction of consumer subsidies was a major element of the programs.
Macroeconomic and other nonfiscal measures constituted an important part of all adjustment programs examined in the case studies. Most programs aimed at moderate growth in real GDP along with external adjustment (Table 4). Relative to the preprogram growth rates, all programs targeted a higher or similar rate of growth of GDP, except the 1983-84 program with the Philippines. In particular, the programs with Ghana and Chile aimed at reversing the fall in economic activity that had preceded the program period. At the same time, all programs except those with Ghana and Thailand targeted a significant fall in the external current account deficit.
|Growth in real GDP|
|(Annual percentage changes)|
|Current account deficit including grants|
|(In percent of GDP)|
|Rate of increase in consumer price index|
|(Annual percentage changes)|
Rate of increase in GDP deflator.
Rate of increase in GDP deflator.
Thus, the growth rates of GDP did not reflect a reduced growth of domestic absorption aimed at external adjustment. This was due not only to supply-side measures but also to the recovery of external demand from the worldwide economic recession in 1982-83. Accordingly, none of the programs incorporated a large immediate fall in the rate of inflation (Table 4). In fact, the programs with Ghana and Chile envisaged a significant acceleration of inflation, which partly reflected the inflationary effects of a depreciation of the exchange rate.
All programs contained ceilings on credit expansion by the banking system and emphasized flexible interest rate policies. The following subsections discuss the monetary aspects of the programs and their possible implications for poverty, examine price liberalization measures aimed at improving the efficiency of domestic resource allocation, explore market and wage policies pertaining to the private sector, discuss exchange rate policy, which is often regarded as the single most important policy instrument of structural adjustment employed in Fund-supported adjustment programs, and finally, evaluate the effects on poverty groups of other external policy actions.
Money and Credit Policy
Fund-supported adjustment programs generally seek to limit aggregate credit expansion. Within this overall constraint, they limit the access of the public sector to credit from the banking system in order to ensure that an adequate share of total credit is available to the private sector. In a repressed financial system, credit controls and the associated credit rationing may strengthen the position of larger firms and farms relative to their smaller counterparts, with the latter being forced to turn to curb markets in which higher interest rates are charged. This process may prevent the poor engaged in either small-scale enterprises in the informal sector or small farms in the rural sector from expanding their productive capacity, thus limiting their participation in the process of economic development and structural adjustment. Also, interest rate ceilings on bank deposits may limit the income of some groups of the poor.
Stabilization packages often combine credit restraint with financial reform, which involves raising interest rates, decreasing market fragmentation, and increasing competition. These policies of financial liberalization may also have beneficial effects for the poor. By encouraging domestic saving, discouraging capital flight, widening the coverage of the financial system, and removing credit rationing policies that tend to favor the well-off, an increase in the supply of loanable funds available to the poor may be observed.
The objective of limiting monetary growth—when supported by credible fiscal adjustment—is to reduce inflation. One expects the poor to benefit from reduced inflation because they typically pay most of the inflation tax. They are least able to protect their real incomes and assets against inflation, lacking both effective indexation of their wages and assets that maintain their real value, such as foreign or real assets.
On the other hand, overall credit restraint may adversely affect employment and real output in the short run, especially if nominal prices and wages are rigid. Empirical evidence suggests that monetary contraction tends to exert a two- to three-year deflationary effect on output and employment.21 The adverse short-run employment effects associated with these deflationary effects may seriously hurt some poverty groups, because in lower-income groups the least skilled tend to be the first to become unemployed when aggregate employment falls.
Stabilization packages, however, generally complement monetary restraint by supply-side measures aimed at maintaining or raising real activity during the adjustment period and improving the trade-off between inflation and employment. Once the influence of all relevant policies, including the supply-side policies, on the growth rate is recognized, there is no clear presumption that Fund-supported programs adversely affect growth.22
Many programs restructured credit expansion in favor of the private sector (Table 5). In all countries except the Dominican Republic and the Philippines (1984-85 program), the programs allowed an increase in real terms in domestic credit expansion to the private sector. Private sector credit was programmed to rise at about the same rate as in the year prior to the program in Kenya, Sri Lanka, and the Dominican Republic. In Sri Lanka, Thailand, and the Dominican Republic, credit to the private sector exceeded the program targets.
|Net domestic credit expansion to public sector|
|Net domestic credit expansion to private sector|
Changes as percent of liabilities to the private sector at the beginning of the period.
Changes as percent of liabilities to the private sector at the beginning of the period.
Monetary policy in the Philippines remained quite accommodating during the stand-by arrangement in 1983-84. It led to both rising inflation, which eroded real wages, and increasing capital outflows, which contributed to a foreign exchange crisis later in the year. During the stand-by arrangement in 1984-85, the authorities tightened monetary policy significantly to restore confidence in the domestic financial system. In the short run, the tightening of credit expansion contributed to declining real output and rising unemployment in commerce and service activities. Although these measures may have restored some confidence in the financial system, it is likely that these adverse short-run developments hurt urban poverty groups.
Only a few programs contained specific provisions to allocate credit among various elements of the private sector. The program for the Dominican Republic, however, explicitly supported an expansion of credit to the agricultural sector in general and to small-scale agriculture in particular. The program for Sri Lanka specifically aimed at providing adequate credit to the export sector, where most of the poor are employed. Although no data are available on the share of loanable funds received by the poor, the programs with Ghana and Kenya may have enhanced the access of the poor to loanable funds by reducing the role of rationing in allocating credit and by raising the overall supply of credit in formal markets.
In the short run, higher interest rates in official markets probably had an adverse impact on labor employed in capital-intensive firms. In the Philippines, for example, high real interest rates contributed to a substantial decline in manufacturing output in the short run. In the medium to longer term, higher interest rates may have encouraged large firms in the formal sector to adopt more labor-intensive production techniques over time, which would have increased the demand for labor with presumably some impact on the poor, particularly in the urban informal sector.
The program with Chile provided support for measures that the authorities designed to strengthen the financial system, such as the introduction of a deposit insurance scheme, a rescheduling of domestic debt owed by the productive sectors in the economy, and a preferential exchange rate for debt-service payments on foreign loans. The distribution of the benefits associated with these measures is difficult to determine. On the one hand, the preferential exchange rate benefited the higher-income groups, who had been in a position to assume foreign currency debt. On the other hand, small depositors with lower incomes, who did not move their capital abroad, took advantage of the deposit insurance scheme. Moreover, the recovery of the financial system encouraged capital to return to or to stay in Chile, which benefited other factors of production.
The liberalization of prices is often part of a Fund-supported adjustment program. These measures aim at transmitting exchange rate movements throughout the economy, improving incentives, and encouraging a more efficient resource allocation. Administered prices tend to benefit both traders in illicit markets and bureaucrats who administer the controls at the expense of low-income producers. Furthermore, although price controls are supposed to protect poor consumers, they often reduce the supplies available to the poor, thereby raising the prices that poor consumers actually pay in parallel markets.
How the poor may benefit from price liberalization in countries suffering from serious imbalances is most clearly illustrated by the program with Ghana. Extensive price controls had shifted transactions away from official markets toward parallel markets. In these markets, the poor consumers, who did not have access to goods at official prices, paid market-clearing prices that far exceeded official prices. These high prices not only reflected insufficient supplies and the scarcity rents that had been extracted by illicit traders and other politically powerful groups but also the real resource costs associated with rent-seeking behavior and the inefficiencies associated with trading on black markets. In addition, the decline in trade transacted through formal channels had eroded the public revenue base, forcing the public sector to cut services provided to the poor.
The removal of most price controls reversed these negative developments to some extent, thereby positively affecting the urban and rural poor alike.23 The increases in official prices for agricultural products also directly benefited the rural poor engaged in the production of these goods. Moreover, market-clearing prices in formal markets removed the need for producers to engage in costly trading through illicit channels or smuggling.
An extensive liberalization of commodity prices, which was supplemented by trade and marketing liberalization policies, also benefited the poor in the Philippines. These policies helped the rural poor, from the sources of income side, by enhancing efficiency and incentives in the agricultural sector. At the same time, these policies appear not to have hurt the urban dwellers, from the uses of income side, because most commodities already commanded high market prices through the rent-seeking activities of various intermediaries.
The programs with Sri Lanka and the Dominican Republic also emphasized flexible pricing policies. These policies are discussed in Section V because they relate directly to public expenditures on price subsidies.
Labor Market Policies
Labor market policies are generally aimed at maintaining or increasing employment by reducing rigidities in the labor market and by supporting the movement of labor to the production of tradable goods. These policies include wage restraint and measures to increase the flexibility of average and relative wages through a reduced role for indexation.24 In the short run, these policies impose hardship on organized labor and those who have a secure job in the formal sector, which tends to make them politically difficult to implement. Moreover, lower employment in the formal sector generally raises the supply of labor to the informal sector, thereby reducing the real earnings of the poor engaged in the informal sector. Unemployment not only reduces the incomes of the poor in the short run but also reduces their longer-term potential for employment.25
In the medium run, such flexible wage policies may contain the negative employment effects of a fall in domestic absorption or of adverse external shocks, such as a decline in capital inflows or in the external terms of trade. This indirectly benefits the vulnerable groups, because the poorest segments with the least skills are often the first to be laid off when aggregate employment falls in response to unsustainably high real wages.
Wage restraint was an important element in the programs with Sri Lanka, Kenya, and Chile. The program with Sri Lanka included measures to replace a monthly indexation scheme for government employees with a semiannual indexation mechanism. A larger-than-anticipated inflation rate caused minimum wages to decline in real terms. The decline in minimum wages in the industrial and commercial sectors exceeded the corresponding decline in the agricultural sector, which employs the poorest workers.
Similarly, in Kenya, the wage reductions in the private sector, which contains most of the poor, were smaller than those in the public sector. On the one hand, the overall wage restraint hurt middle- and low-income workers with.secure employment and may also have adversely affected the poor indirectly through a reduction in both remittances and demand for products supplied by the poor. On the other hand, by contributing to a strong employment performance, wage restraint may have assisted other vulnerable groups, notwithstanding an initial fall in the terms of trade. Whereas the short-run overall poverty effect of wage restraint is difficult to determine, many of the poor are likely to have benefited in the medium run as the wage restraint expanded employment opportunities.
The program in Chile supported the abolition of wage indexation. This indexation mechanism had contributed to an unsustainably high level of real wages, which had become incompatible with internal and external balance. It had forced the economy to adjust to adverse external shocks through a severe stagnation of economic activity. Moreover, the rigid wage policy had rationed labor out of the formal urban sector, thereby increasing labor supply in the rural and informal urban sectors. Consequently, this policy had widened the wage gap between the formal urban sector and the other sectors. The more flexible wage policy, which was supported by the program, contributed to a narrowing of this wage gap by stimulating rapid growth in formal employment during and after the program period, thereby reducing the downward pressure on wages in the other sectors.
The program in Chile also included emergency employment programs. These programs sought to contain the damage to the human capital of the poorest caused by the combination of adverse external developments and the inappropriate policies that had preceded the program. Despite some shortcomings,26 the programs helped to maintain the employment record and the productive capacity of the poor.
In contrast to Chile, economic circumstances in Ghana and Thailand did not necessitate a sharp decline in real wages. In fact, in Ghana and Thailand higher real wages were compatible with internal and external balances. In Ghana, measures aimed at improving the supply performance of the economy warranted wage increases that directly benefited the poorest wage earners. The real minimum wage in Ghana, which at the start of the program had declined to only 13 percent of its level in 1975, more than doubled between 1983 and 1985. In Thailand, favorable external circumstances allowed the program to increase minimum wages in real terms without negatively affecting employment. These developments benefited the low-income groups employed in the organized sector.
Exchange Rate Policy
Most adjustment programs arise from a chronic imbalance in the external sector, which is often addressed through a depreciation of the exchange rate. Such action, when supported by appropriate monetary and fiscal policies,27 raises the official relative prices of exports, imports, and import substitutes. These price changes generally will affect various groups differently, particularly in the short run when production factors are rather immobile.
This subsection first discusses those structural features of an economy that, when combined with the structure of poverty in particular countries, determine the immediate effects of a currency depreciation on households in poverty. Based on the programs in the case studies, it appears that exchange rate policies in Kenya and Ghana benefited the dominant poverty groups in the short run. In contrast, in the Dominican Republic, the Philippines, and particularly Chile, the devaluation appears to have been costly to important poverty groups in the short run. Counterfactual arguments are then employed to show that maintenance of the exchange rate at an overvalued level would have hurt the poor even more.
Whether the immediate effects on relative prices benefit the poor depends on the structure of the economy.28 Disadvantaged groups are positively affected from the sources of income side if the production of tradables is in the hands of poor small-scale farmers or is intensive in the labor resources supplied by the poor. However, if the production of tradables is capital intensive or concentrated in a few large firms while the poor are primarily engaged in nontradable activities or in activities using imports as intermediate inputs (which may include some small-scale farmers), poverty groups generally suffer from the immediate effects of depreciation. Effects from the uses of income side may magnify these adverse effects if the consumption basket of the poor contains a large share of tradable goods or goods that use tradable goods as intermediate inputs.29 In the longer run, however, the improved economy-wide efficiency associated with an exchange rate depreciation generally offers the poor the opportunity to share in the benefits derived from better growth and employment performance. This will be more likely if the depreciation is accompanied by policies that ensure that poverty groups participate in the structural adjustment toward tradable activities.30
All programs considered in the case studies relied on a flexible exchange rate policy.31 A sharp depreciation of the cedi in Ghana corrected a grossly overvalued exchange rate. The program in the Dominican Republic aimed at a more competitive exchange rate by expanding the scope of the parallel market for foreign exchange. The exchange rate depreciated in real effective terms during the program period in Kenya and Chile as the terms of trade declined. In Chile, and in the Philippines in 1983, a sharp fall in net capital inflows from abroad also played a major role. In Thailand and Sri Lanka, the real exchange rate appreciated owing to an improvement in the terms of trade.
The poor appear to have benefited most from the exchange rate policy in Ghana. The most important poverty groups in Ghana are landless agricultural workers and smallholders in rural areas. These groups are engaged in the production of export crops, particularly cocoa. The fall in the exchange rate assisted the poor smallholders, primarily by reducing their need to engage in smuggling and other costly trading in parallel markets. At the same time, the landless laborers benefited from expanding rural employment.
Another important poverty group in Ghana is the urban poor. Although this group faced higher official prices for tradable goods in formal markets, it tended to benefit from the fall in the official exchange rate for at least three reasons. First, official prices had become almost irrelevant for most of the poor before the program came into effect because tradable goods were mainly traded in black markets against prices that reflected the scarcity premiums of foreign exchange. The devaluation reduced the prices that the urban poor were actually paying, not only by reducing the productive resources involved in rent-seeking activities, but also by limiting the opportunity for politically powerful groups to appropriate scarcity premiums on tradable goods. The devaluation promoted overall equity by reducing rents based on political privilege rather than on economic performance.
The second reason for the improved position of the urban poor in Ghana was that the depreciation in the exchange rate revived the bases of taxation of international trade, which were the primary revenue source of the Government. This allowed the authorities to expand the public services provided to the urban poor without widening the fiscal deficit. Finally, many of the urban poor retained strong working and family ties with the agricultural sector, and they, therefore, shared in the benefits accruing to the rural sector.32
In Kenya, where a real depreciation was a major element in the adjustment program, smallholders are also an important poverty group. The declining exchange rate directly assisted smallholders by raising the real prices of some of their crops. Furthermore, it improved employment opportunities for impoverished landless workers. However, the short-run positive effect on the rural poor was smaller in Kenya than in Ghana for several reasons. First, the poorest smallholders in Kenya are basically subsistence farmers and thus have limited contact with the market and are little, if at all, affected by changes in market prices. Second, many poor farmers specialize in nontradable crops whose prices are largely determined in the domestic market rather than by world prices. Production of the major export crops—tea and coffee—is mostly concentrated in large plantations. Third, whereas the depreciation was intended to give a clear signal to the poor to expand production in tradable activities, the poorest farmers may have faced difficulties in doing this because they tend to be located in areas with a lack of agricultural support services and infrastructure.
Thus, although improving agricultural prices may be a necessary condition for enhancing the earning capabilities of poor farmers, such policies need to be complemented by measures to ensure that these farmers are in a position to take advantage of improved price incentives. In this respect, the sharp cutback in public expenditure on agriculture in Kenya may have hampered poverty alleviation through structural adjustment.33 On the positive side, however, the targeted expansion in real credit to the private sector may have helped poor farmers to expand their production.
Although the Dominican Republic, Chile, and the Philippines are more urbanized, a large share of the poor live in impoverished rural areas. The rural poor are mainly landless farm laborers, while the ownership of land is concentrated. Whereas many of the rural poor are engaged in the production of tradable crops and took advantage of increased rural employment, the immediate benefits to these groups from the depreciation were generally rather limited.34 At the same time, the inequities within the rural sector tended to widen. The rural laborers had to pay higher prices for the food they bought in the market. Furthermore, the adverse short-run effects of the depreciation on non-tradable activities in the urban sector increased the rural labor supply because of substantial urban-rural labor mobility. These developments, while allowing the urban poor to share in the improved employment opportunities in the rural sector, mitigated the positive effect of the devaluation on rural wages and allowed the landowners to absorb most of the immediate benefits associated with the exchange rate depreciation.35 In the Philippines, however, the liberalization of trade and marketing in the agricultural sector helped to shift the benefits of the depreciation away from intermediaries to the rural poor.
In the Dominican Republic and the Philippines, the urban poor are an important poverty group. A depreciation aimed at relocating resources toward tradable activities imposed immediate costs on the urban poor engaged in the production of both nontradables and goods using imports as an intermediate input.
An appraisal of these costs must recognize the likely costs that would have been borne in the absence of a devaluation. For example, to prevent these costs, the authorities could have adopted alternative policy instruments in response to the unsustainable external imbalances. In particular, instead of depreciating the exchange rate, they could have maintained a higher exchange rate by borrowing abroad (or by reducing their official reserves),36 by imposing trade restrictions and trade taxes, or by allowing a rise in the unemployment rate.37 However, it is likely that these alternative policy responses would have been even more damaging to poverty groups, particularly in the urban sector.
The preprogram experience in the Philippines during the period preceding the financial crisis in 1983 illustrates how delaying the exchange rate adjustment by borrowing abroad or depleting reserves tends to hurt the urban poor. The delay provided the better-off with the opportunity to move their assets abroad, thereby escaping both the implicit taxes imposed by future depreciations and inflation and the explicit future taxation on capital incomes that might be implied by the build-up of external public debt. Such options were unavailable to the poor, since their only assets are their own human capital.38 They, therefore, bore the burden of both the adverse external shocks and the debt-service obligations associated with net external public borrowing.
The experience in Chile in mid-1984 also illustrates that the effects of a delay in adjusting exchange rates in the face of an unsustainable external position may benefit higher-income groups at the expense of others.39 In particular, before the authorities raised import duties and devalued the exchange rate, imports temporarily increased in mid-1984. This suggests that the higher-income groups were able to escape part of the impact of higher prices for imports by borrowing to increase their imports in anticipation of the measures.40
An overvalued exchange rate may hurt the poor in other ways as well. It tends to raise the unemployment rate and reduce the real value of labor supplied by the poor by encouraging the substitution of cheap imported capital for local labor, particularly in the medium run. During the program period in Chile, the declining exchange rate, in combination with the abolition of wage indexation, moderated the rise in unemployment that would otherwise have resulted from a decline in capital inflows and the terms of trade.
An overvalued exchange rate also misdirects the productive potential of poverty groups into unsustainable nontradable activities. When the currency over-valuation lasts for an extended period of time, it may cause serious structural damage to the ability of the poor to supply tradable goods. Thus, maintaining an unsustainable exchange rate by borrowing abroad shifts the burden to the future not only by raising future debt service but also by aggravating the costs imposed on the poor when the inevitable restructuring of production toward tradable activities eventually occurs.
Other alternatives to depreciation are trade restrictions and trade taxes. Fund-supported programs generally reject trade restrictions. In fact, the programs for the Dominican Republic and Kenya included provisions liberalizing imports. The preprogram developments in Ghana illustrate how trade restrictions tend to impose a heavy burden on the poor.41
The counterfactual arguments presented above indicate that, over the medium term, maintaining an overvalued exchange rate is generally an inefficient way of protecting the poor when external conditions call for a reallocation of resources toward efficient tradable activities. Therefore, given the objective of poverty alleviation, other policy instruments should be used to provide adjustment assistance to those vulnerable groups suffering the immediate adverse effects of restructuring of production. These policy interventions should guard against removing incentives to reallocate resources toward tradable sectors, and they should instead encourage the poor to move their productive resources into tradable activities.
Whereas compensatory targeting measures are generally not an explicit part of Fund-supported stabilization programs, the authorities in Chile and the Dominican Republic took some measures aimed at protecting the poor. These measures illustrate the kinds of adjustment assistance possible, including increased public expenditures on labor-intensive investment programs.42 In Chile, the authorities expanded emergency employment schemes and targeted food subsidies and health to protect the most vulnerable urban groups.
Other External Policies 43
In other external policy actions, the programs in Kenya and the Dominican Republic liberalized imports and at the same time raised the average level of import duties.44 These policies shifted the rents associated with scarce imports away from well-established importers to the public sector (and possibly from others as well).
To reduce the fiscal deficit, the average rate of tax on international trade was also raised in the programs with Sri Lanka, Chile, and Thailand, and in the 1983–84 program with the Philippines. In these cases, however, the tax increases were not accompanied by trade liberalization and, therefore, raised the real exchange rate consistent with the external balance. Thus, they created a bias against factors engaged in the production of both exports and goods using imports as intermediate inputs, and in favor of factors engaged in the production of import substitutes. In Sri Lanka and Thailand, these tax increases tended to hurt the poor smallholders producing export crops. In the Philippines, higher rates on imports, together with a temporary intensification of import controls, contributed to higher prices for tradables consumed by the urban poor. These policies strengthened the market power of producers of import substitutes, reduced static and dynamic economies of scale achieved by producing for world markets, and encouraged inefficient import substitution.