a. The 1977 liberalization and the economic response through 1980

In late 1977, the authorities initiated comprehensive economic reform, reversing the policies of the previous two decades. This reform was aimed at rejuvenating the agricultural sector, boosting industrial production, increasing employment, raising domestic saving and investment, and strengthening the balance of payments in the medium term. The basic thrust of the reform was to reduce administrative control and intervention over the allocation of resources and to establish realistic relative prices. The reform was supported by arrangements with the IMF: a one-year stand-by arrangement (December 1977–December 1978) was immediately followed by an extended arrangement (January 1979–December 1981).

The initial phase of the reform involved:

• dismantling trade and payments restrictions, notably through the removal of most import licensing requirements;

• unifying the exchange rate and allowing it to reflect developments in the balance of payments (initially, the rupee was devalued by over 50 percent in foreign currency terms);

• restructuring agricultural prices and export taxes to reduce dependence on imported food grains and to increase resources for the essential tree crop sector;

• adjusting the prices of essential consumer goods, production inputs, and public services in preparation for eventual liberalization;

• restructuring government priorities to promote investment (especially in social infrastructure) while substantially reducing widespread subsidies for food and fertilizer; and

• liberalizing interest rates and curtailing central bank lending to promote saving and a more rational allocation of credit.

These measures had a pronounced effect on economic activity. GDP growth, which had been about 3 percent during 1971–77, doubled in 1978–80 as agricultural and industrial production responded to improved incentives. At the same time, domestic demand increased sharply as import liberalization facilitated public investment and private consumption. This surge in demand was manifested in (i) a sharp rise in inflation, from an average of less than 6 percent in 1971–77 to 16 percent in the subsequent three years; and (ii) the emergence of a large external current account deficit, which averaged about 9 percent of GDP during 1978–80, compared with an average of 1 percent of GDP in the preceding years.

The proximate causes of the growing inflation and balance of payments pressures may be found in accelerating monetary and credit growth and a burgeoning fiscal deficit. In the three years after 1978, the pace of monetary expansion doubled from its previous seven-year average to 34 percent, despite considerable leakage through the loss of external reserves. This expansion reflected the rapid growth of credit, which averaged 43 percent during this period. Almost half the increased demand for credit came from the central government and was the direct result of the large investment-induced increase in its overall deficit.

The structure of the balance of payments changed considerably during this period. Despite the much-enlarged current account deficit, gross international reserves continued to grow through 1980, reflecting a variety of factors that included rising aid flows, an increase in direct foreign investment from previously negligible levels, other financial transfers by residents, and substantial drawings from the IMF. External debt as a proportion of GDP doubled during the late 1970s to about 48 percent in 1980, but because much of this borrowing came from official sources on concessional terms, the debt service ratio rose only gradually, reaching 13 percent in 1980.

b. Economic policies during the early 1980s

Economic policy in the first half of the 1980s began to focus more on maintaining macroeconomic stability and less on extending the reforms. In support of its efforts to reduce macroeconomic imbalances, in 1983, the government introduced a stabilization package supported by a standby arrangement with the IMF. However, the arrangement lapsed when growing ethnic conflict weakened the government’s capacity to implement sound economic policies and military expenditures rose (from 1 percent of GDP in 1980 to 3–4 percent in the mid-1980s).

A variety of factors were involved in this slowdown in reform. First, the civil unrest caused private investors to lose confidence in the reforms, so few efforts were made to transfer public enterprises to private hands. Second, the initial liberalization of the trade and payments system was not extended in subsequent years. Third, from 1980 through 1984, the real effective exchange rate was allowed to appreciate gradually, undermining both the competitive edge of Sri Lanka’s exports and efforts to diversify its economy.

Although defense expenditures were rising, the authorities managed to reduce the scale of fiscal operations considerably during this period. Expenditures were reduced relative to GDP as public investment projects were completed, private investment rose, and current expenditures were contained by freezing (in nominal terms) such items as the fertilizer subsidy and the national food stamp program. Although an inelastic tax system hindered the government’s revenue-raising efforts, the overall deficit was reduced from 18 percent of GDP in 1980 to 10 percent in 1986.

Monetary and credit growth also slowed markedly. Overall credit growth fell progressively between 1981 and 1985 as credit to the public sector declined, reflecting a tighter fiscal policy and improvements in the performance and pricing policies of public enterprises. In 1986, credit and monetary expansion slowed further—to 8 percent and 6 percent, respectively—primarily because of sluggish private sector activity and reduced bank borrowing by the government.

The net effect of these policies was to achieve some moderation in the rate of inflation and in external imbalances. Inflation averaged 12 percent during 1981–85, falling to only 2 percent in 1985. However, numerous imbalances remained: inflation rose to 8 percent in 1986, the external current account deficit remained at 7 percent of GDP, and external reserves fell to ten weeks of import coverage.³ The developments in the financial sector were associated with a sharp deceleration in domestic demand during the early 1980s and a modest reduction in GDP growth. In 1986, GDP growth fell while domestic demand recovered, suggesting that structural rigidities remained in a number of areas and that sustainable growth was unlikely to be achieved without a revival of the reform effort.

c. Economic policies in 1987

The authorities had hoped to make further progress in stabilization and adjustment during 1987. However, their fiscal goal—to reduce the overall deficit by 1.5–2 percent of GDP—became unfeasible in the face of unanticipated spending in a number of areas, including security, drought relief, and infrastructural investment. Regional spending related to the ethnic conflict also increased.⁴ The severe drought caused agricultural output to decline and overall GDP growth to slow to 1.5 percent. Inflation, which had slowed early in the year in response to cautious monetary policies, accelerated in the second half of 1987, climbing to 10 percent by year’s end.

The authorities initiated a number of policies to adjust to these new conditions. Two of these policies were especially important. First, the authorities began actively to use exchange rate policy as a tool in their stabilization program. In 1985, they had begun to reverse the appreciation in the real effective rate of the previous several years. In 1987, depreciation was accelerated to over 10 percent (in real effective terms), and there was further liberalization of external trade. Second, the authorities began to use market-related instruments of monetary policy and to reduce reliance on credit controls, reserve requirements, and administered interest rates as the main tools of policy. Open market operations were conducted, with Treasury bills as the primary instrument, and reserve requirements were lowered from 14–18 percent to a uniform 10 percent.

The external current account deficit fell during 1987 by about 1.5 percent of GDP to 5 percent. Although export volume remained flat, import volume fell by 5 percent, reflecting import substitution in food and a decline in investment imports. Net capital inflows fell because of rising scheduled debt payments and a modest decline in Official Development Assistance (ODA). Although the government did some commercial borrowing, reserves continued to decline, and by the end of 1987 official reserves had fallen to less than eight weeks of imports.

d. Economic policies in 1988: the start of a three-year SAF arrangement

Based on renewed hopes of social and ethnic stability, the government developed a medium-term adjustment framework that began in January 1988 and was aimed at reducing domestic and external macroeconomic imbalances as well as structural impediments to growth. This program was supported by a three-year arrangement under the IMF’s Structural Adjustment Facility (SAF).⁵

The key macroeconomic objectives of the program were (i) to restore real GDP growth to 5 percent annually; (ii) to lower inflation to the average of Sri Lanka’s major trading partners (about 5 percent); and (iii) to reduce the external current account deficit to 6 percent of GDP (excluding official transfers) by 1990 in order to restore gross international reserves to at least eight weeks of imports. Because the government expected to invest heavily in infrastructure (especially for reconstruction in war-damaged areas) and also anticipated that foreign saving would decline, the program envisaged a stronger domestic saving effort.

At the core of the program to reduce domestic imbalances was a medium-term fiscal strategy designed to cut the government’s domestic borrowing by half and to eliminate bank borrowing altogether by 1990. This strategy required that the overall budget deficit be reduced from 12 percent of GDP to 9 percent by 1990 (excluding grants). The brunt of the adjustment fell on the expenditure budget: reductions were planned in defense, support for public enterprises, other subsidies and transfers, and normal investment expenditure. The tax reform was designed to increase the revenue system’s elasticity in the medium term but did not envisage any increase in the ratio of revenue to GDP.

The program’s structural policies were based on two key precepts: that public sector claims on resources (including those of public enterprises) would be reduced; and that the industrial sector, with government support, would become more efficient and outwardly oriented. The strategy for public enterprises had three key elements:

• the conversion of some enterprises to public liability companies (with the government holding all the shares initially) that would operate on commercial principles, without the benefit of preferential government policies;

• the privatization of selected enterprises; and

• the promotion of private sector access to activities previously reserved for the public sector.

Steps were also to be taken to strengthen the tree crop sector through a medium-term investment program supported by the World Bank. However, at this juncture, the government did not envisage any changes in the ownership of the sector, which remained dominated by two large state corporations.

Exchange rate policy was a key component in developing an export-oriented industrial sector: the specific objectives were to preserve competitiveness and to meet the program’s reserve targets. Other key components of the industrial strategy included:

• a reform of the tariff system designed to lower the level and dispersion of effective protection, to be carried out in three stages during 1988–90;

• the lifting of most remaining import licensing requirements by 1990;

• a reduction in the tax burden on exports, notably tea; and

• a reform of the tax system (implemented with the 1988 budget) to lower exemptions and reduce marginal tax rates.

The authorities were also committed to increasing the role of market forces in the system of monetary control by improving the Treasury bill auction system. Specifically, the Central Bank (which participated in the auction as a bidder) would be gradually removed from the tender system and market demand allowed to determine prices. The program did not specifically address the ownership issue in the banking system, which remained dominated by two large state-owned banks.

In the event, performance during the first year of the SAF fell considerably short of expectations. Real GDP grew by only 3 percent (half the program target), inflation accelerated to 15 percent (double the program target); and the overall balance of payments recorded a substantial deficit (instead of a modest surplus). Underlying this performance were a number of noneconomic considerations, notably a rapid deterioration in national security and a number of policy decisions (in particular, to expand welfare programs considerably) made ahead of the presidential and parliamentary elections of December 1988 and early 1989, respectively.

The fiscal deficit rose by 4 percent of GDP to 13 percent; about half of the overrun was related to spending on security. Other key factors included wage increases for all public sector workers, which involved large transfers to public enterprises, and higher subsidies as the food stamp program was expanded.

Credit and monetary growth were affected by similar considerations. The expansion of credit to the public sector was especially marked, with the government providing the chief impetus. Rising inflationary expectations (reflected in a further decline in the growth rate of quasi-money holdings), combined with declining net foreign assets, led to relatively slow monetary growth.

The expansionary stance of domestic economic policy was reflected primarily in accelerated inflation. The external current account deficit increased only slightly in 1987, possibly because much of the additional demand came late in the year. Although imports grew only slowly, the greatest slowdown came in aid-related imports and resulted in a corresponding reduction in capital inflows. Consequently, gross official reserves, which had increased to well over eight weeks of imports in the first quarter of 1988, fell sharply through the rest of the year to six weeks at the end of 1988.

e. Economic policies in 1989: decline and rebound

Financial imbalances persisted into 1989 as the southern insurgency intensified, unsettled conditions continued in the north, and the policy decisions made prior to the presidential elections in late 1988 began to jeopardize the adjustment effort. The government continued to borrow heavily from the banking system, particularly from the Central Bank, and bank credit to public enterprises also rose rapidly. With domestic confidence at a low ebb, the impact on prices was relatively muted (inflation rates actually declined through much of 1989), but external reserves came under severe pressure. By August 1989, gross official reserves had fallen to three weeks of imports, and even this low level was possible only because of short-term external borrowing by the Central Bank.

The authorities’ eventual response, which was supported by the second-year SAF arrangement, was a pronounced tightening (after August 1989) of fiscal and monetary policies, accompanied by a rupee depreciation of 7.5 percent in real terms.⁶ With the sharp change in expectations that followed on these credible economic policies, macroeconomic performance improved dramatically.

Fiscal policy aimed to reduce the overall deficit below that of the original 1989 budget.⁷ Most of the measures concentrated on reducing expenditures. Subsidies were phased out on fertilizer and rice by the end of 1989 and on flour in mid-1990, and petroleum prices were increased substantially; an ambitious new poverty alleviation program was designed to hold expenditures to 3.5 percent of GDP; and sharp across-the-board reductions in other nonwage, non-interest-bearing current expenditures were implemented. When revenue shortfalls seemed likely, the government instituted further controls on spending. In the event, revenue improved, so that the actual deficit was held to 9 percent of GDP (well below the program target), or 4 percentage points less than in 1988. Instead of borrowing 1.6 percent of GDP from the banking system, as had been planned, the government made repayments amounting to 1.3 percent of GDP.

Monetary policy was also tightened. During September 1989, Treasury bill interest rates rose by 3.5 percent to almost 20 percent, and most other interest rates also rose by 1–2 percentage points. The Central Bank was able to sell a substantial proportion of its holdings of Treasury bills, almost completely sterilizing the impact of an unexpectedly strong improvement in bank net foreign assets (NFA). Thus, the expansion of domestic credit was contained at just 5 percent (compared with 29 percent in 1988) and of total liquidity at 11 percent (compared with 17 percent in 1988).

The external position responded very strongly to the coincidence of strong economic policy and improved security conditions. After the devaluation in September, foreign reserves showed an immediate and striking improvement. Short-term capital outflows were reversed, export growth improved, imports slowed, and service receipts (notably from tourism) began to improve. The current account deficit, which at mid-year had seemed set to rise to 10 percent of GDP, actually fell by 1.5 percentage points from its 1988 level to 4.5 percent of GDP. Gross foreign reserves rose to about six weeks of imports by the end of 1989, and a significant proportion of earlier short-term borrowing was repaid.

The uncertain macroeconomic and security situations during 1989 meant that the focus of policies was on stabilization, and relatively few steps were taken to realize the structural goals of the SAF. It is also noteworthy that the summary economic data for 1989 as a whole concealed not only the severity of the economic crisis that had developed by the middle of the year but the speed with which macroeconomic variables began to improve once expectations, reinforced by tighter policies, shifted.

f. Economic policies in 1990: a third-year SAF and an anticipated ESAF

The improved economic performance of the second half of 1989 continued into 1990, reflecting relatively restrained policies. As the year wore on, policies were gradually relaxed, but the authorities’ commitment to achieving stable economic conditions was confirmed in the third quarter of 1990 by the adoption of a program supported by a third-year SAF arrangement. Preliminary consideration was also given to the preparation of a new medium-term adjustment program that could be supported by the IMF’s Enhanced Structural Adjustment Facility (ESAF).

Real GDP grew by 6 percent, and there was a steady strengthening of the external position that reflected higher export volumes, higher tea prices, and moderate import growth. The external current account deficit declined by 2 percent of GDP from its 1989 level to 5 percent of GDP. Gross official reserves continued to climb, reaching eight weeks of imports by the end of the year. Controlling inflation proved to be a more difficult task owing to several factors: (i) the lagged effect of the 1989 rupee depreciation; (ii) increases in the controlled prices of wheat, sugar, fertilizer, and petroleum; and (iii) monetary and credit policies that proved insufficiently tight in the light of the improved balance of payments performance. Inflation rose to 20 percent, double the rate expected under the second annual SAF program.

Domestic financial policies were intended to contain demand but not choke off an incipient recovery in production. The overall budget deficit was held to 10 percent of GDP in 1990 (a reduction of 1 percentage point from 1989), mainly through stronger revenue efforts and reduced capital expenditures. Domestic financing was sharply reduced and bank financing virtually eliminated. But even though the government halted consumer subsidies, current spending remained high (22 percent of GDP) because of higher defense spending and new welfare programs.

The credit restraint of 1989 gradually gave way to more relaxed policies. By the end of 1990, Treasury bill interest rates had turned negative in real terms, and monetary expansion had almost doubled (to 20 percent). Underlying this performance was an unexpectedly large surge in private sector credit, which grew by 24 percent (up from 5 percent in 1989); substantial concessional refinancing by the Central Bank; and a sharp increase in the banking system’s NFA that (unlike the 1989 increase) was not sterilized.

Until 1990, the authorities viewed exchange rate policy as an integral element of their anti-inflation strategy. After the substantial depreciation of September 1989, the rupee remained almost constant against the U.S. dollar. But as inflation accelerated in the first half of 1990, the real effective exchange rate appreciated by 8 percent. During the second half of the year, the authorities intervened to stem the appreciation by making large net purchases of foreign exchange.

Some progress was made on structural measures during 1990. Significant developments included:

• a reduction in civil service employment of 10 percent during 1990 and early 1991;

• the privatization of some small manufacturing enterprises and the partial restructuring of the state-owned bus company;

• tariff reform that reduced the maximum tariff to 50 percent, accompanied by an extension of the coverage of excise duties;

• the liberalization of ocean freight and airline services to allow private sector competition; and

• increases in foreign exchange allocations for education and travel.

However, some commitments were not implemented on schedule, including reform of the Treasury bill auction, complete privatization of the bus company, and the deregulation of fares for public transportation.

g. A synopsis of policy instruments

The Sri Lankan authorities have relied on a range of instruments to conduct monetary policy, principally direct bank-by-bank credit ceilings, changes in administered interest rates (including bank and refinancing rates), and reserve requirements. Open market operations have been used to a limited extent, first in the form of transactions in Central Bank bills and, after auctions were established in 1986, through Treasury bills. The Central Bank has been involved extensively in these auctions, absorbing all Treasury bills required by the central government that did not draw bids at the bank’s cut-off price. From 1986 until 1990, the share of Central Bank holdings fell from 84 percent of the total to 45 percent.

Fiscal policy consisted of ad hoc adjustments to tax rates (direct and indirect), usually intended to compensate for the inelasticity of the system. As noted above, the authorities made frequent efforts to contain expenditures, either through targeted and across-the-board reductions or by tightening controls on spending.

Exchange rate policy became marginally more market oriented in late 1990. From September 1989 through August 1990, the rate had been held at SL Rs 40 to US$1. After August 1990, a new system was adopted under which a reference rate for the rupee was calculated daily; this rate was the weighted average of the previous day’s rates in the interbank market, including dealings with the Central Bank. Central Bank rates were determined by negotiations and varied according to the monetary authorities’ external objectives and market conditions.

The market forces released by liberalization made the economy more vulnerable to inflation and external reserve pressures in the absence of sound macroeconomic management. The 1980s were characterized by large external current account deficits, sizable fiscal deficits (largely attributable to an upsurge in public investment), and an accommodative monetary policy. Although the fiscal and external current account deficits declined in the late 1980s, there was no commensurate improvement in the overall balance because capital inflows diminished, increasing pressure on external reserves and domestic budgetary financing.

h. Structural issues at the end of 1990

Sri Lanka’s economic restructuring efforts originated with the far-reaching trade liberalization initiated in the late 1970s. Liberalization provided a powerful stimulus for economic activities in the private sector. However, after a strong initial response, Sri Lanka was unable to capitalize on the changes, for four reasons: (i) failure to expand the reforms into the public sector; (ii) the country’s continued vulnerability to exogenous shocks (including adverse terms of trade and unfavorable weather conditions); (iii) relatively lax financial policies that continued for extended periods; and (iv) persistent civil conflict.

In 1990, Sri Lanka retained many of the structural impediments with which it had entered the 1980s. Many of these impediments can be traced to the size, inefficiency, and pervasive influence of the public sector. The central government employed almost 10 percent of the labor force and maintained a high level of expenditure—about 30–32 percent of GDP—because of perpetually high subsidies and transfers to households and corporations; a large and costly public administration; and a public investment program that expanded to very high levels in the early 1980s. A major part of this investment program was a large-scale integrated rural development and irrigation project (Mahaweli), which absorbed significant budgetary resources throughout the decade.

Liberalization largely bypassed the public enterprises, which remained large and inefficient: in 1990, some 130 corporations in Sri Lanka employed 600,000 people (about 10 percent of the labor force), even though the sector’s manufacturing output had stagnated or declined. Attempts to improve the financial performance of a few of these enterprises met with some success, but the imbalance within the sector lingered. A significant part of the problem lay with the three large state-owned plantation corporations (tea, rubber, and coconut), which incurred heavy losses in the 1980s and were generally producing at a much higher unit cost than smaller estates in the private sector.

The pervasive influence of Sri Lanka’s public sector has long been reflected in the array of regulations and bureaucratic hurdles faced by potential investors or exporters. Over the years, many efforts were made to remove these distortions. Import restrictions and protective tariffs were reduced and investment regulations diminished. Some steps were taken to liberalize financial and insurance markets and, more recently, ocean shipping. Nevertheless, many minor distortions and regulations remained in 1990, which, although individually inconspicuous, together constituted a significant hindrance to private sector activity.

These developments were aggravated by repeated waves of civil conflict, which at times became the dominant influence over economic activity. A further threat to economic stability came from developments in the Middle East in mid-1990 that not only raised the cost of imported oil but also threatened to weaken tea exports, reduce transfer payments from migrant workers, and create additional expenses for repatriating returning nationals.