This chapter examines the composition of Turkey’s numerous fiscal adjustments over the last decade in the context of academic literature that analyzes the cross-country experience of what constitutes a successful fiscal adjustment. The comparison shows that the composition of Turkey’s most recent fiscal adjustment may need to be rebalanced if it is to be sustained for a prolonged period.
Emerging market countries often have to undertake sustained fiscal consolidation in quite difficult circumstances. In contrast with industrial countries, where adjustments are usually driven by the needs of demand management, emerging market countries often have to adjust because of debt sustainability concerns. As a result, the fiscal consolidation effort generally needs to be maintained for a longer time to achieve debt sustainability.
The composition of fiscal adjustment is critical to the success of longer-term fiscal consolidation. Short-term improvements in fiscal balance can be achieved by a wide variety of policy actions on revenues or expenditures. But when the objective is to maintain the improvement over a longer period of time, the composition of the adjustment can increase the chances of sustaining the consolidation. Alesina and Perotti (1996) studied the composition of fiscal adjustments in countries in the Organization for Economic Cooperation and Development (OECD) from the early 1960s to the mid-1990s to explain the sustainability of the adjustments.1 They found that countries with the most durable fiscal improvements were those that had placed the most effort toward reducing expenditures (especially sensitive items such as wages and transfers). Adjustments in countries that had implemented revenue increases and capital expenditure cuts tended to be shorter lived.
For countries with debt sustainability problems, the benefits are greatest if the fiscal effort is seen as permanent. When debt is on a potentially unsustainable path, this adds to the urgency of adjusting. The costs of failure are increased, but at the same time these countries can benefit the most from any fall in real interest rates that results from consolidation. Increasing revenues through tax hikes does not seem to have the same positive effects as expenditure cuts because tax increases can be quite easily reversed. Tax increases can also have adverse effects on the economy and weaken the resolve to maintain the adjustment. Also, expenditure cuts that fall on investment are less likely to be seen as permanent and may affect long-term economic prospects.
However, some caution is needed in applying results from the industrial country experience to the problems of emerging market economies. Industrial countries usually have large welfare states, high levels of public employment, and high tax rates. Nonindustrial countries tend to have smaller welfare states and generally lower tax rates. Thus, it is possible that what determines the success in an industrial economy may not apply to emerging markets. For example, a recent study by Gupta and others (2002) found that revenue-increasing measures could play a useful role in middle-income countries. Even so, emerging countries that reduce the wage bill and other current spending were more likely to be successful in sustaining their consolidations.
Fiscal Adjustments
Turkey has undertaken a number of large fiscal adjustments over the past decade, with varying degrees of success. In 1993, Turkey had a general government fiscal primary deficit of more than 5 percent of GNP (Table 8.1). Over 1994–95, the government undertook an adjustment of 8.3 percent of GNP, turning the 1993 deficit into a surplus of 2.7 percent of GNP by 1995. However, this adjustment was short lived, and the fiscal primary balance fell into deficit again in 1996–97. In 1998, there was another attempt at fiscal adjustment, but this also unraveled soon afterwards. The most recent adjustment came in 2000, and this has so far been successful in improving the primary balance by more than 8 percent of GNP. Given Turkey’s track record, however, it may be instructive to reexamine the earlier failed adjustments in order to identify their strengths and weaknesses, and to use these findings to determine how to improve the chances of success of the current adjustment.
Summary of Primary Surpluses
(In percent of GNP)
Summary of Primary Surpluses
(In percent of GNP)
| 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Consolidated budget | −0.9 | 3.6 | 3.2 | 1.6 | −0.2 | 4.1 | 1.4 | 5.0 | 5.0 | 2.6 | 4.9 | |
| Rest of public sector | −4.7 | −3.8 | −0.5 | −2.8 | −1.9 | −3.6 | −3.4 | −2.0 | 0.5 | 1.5 | 1.3 | |
| Primary surplus | −5.6 | −0.2 | 2.7 | −1.2 | −2.1 | 0.5 | −2.0 | 3.0 | 5.5 | 4.2 | 6.2 | |
| Change in primary surplus | 5.4 | 2.9 | −3.9 | −0.9 | 2.6 | −2.5 | 5.0 | 2.5 | −1.3 | 2.0 | ||
| Change from consolidated budget | 4.5 | −0.4 | −1.7 | −1.7 | 4.3 | −2.7 | 3.6 | 0.0 | −2.3 | 2.3 | ||
| Change from rest of public sector | 0.9 | 3.3 | −2.2 | 0.8 | −1.7 | 0.2 | 1.3 | 2.6 | 1.0 | −0.3 | ||
| Memorandum items | ||||||||||||
| GNP (real growth rate) | 8.4 | −5.0 | 6.9 | 6.9 | 7.6 | 3.1 | −4.7 | 7.4 | −7.5 | 7.9 | 5.8 | |
| Net debt to GNP | 35.1 | 44.7 | 41.3 | 46.5 | 42.9 | 43.7 | 61.0 | 58.3 | 93.9 | 79.2 | 70.9 | |
Summary of Primary Surpluses
(In percent of GNP)
| 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Consolidated budget | −0.9 | 3.6 | 3.2 | 1.6 | −0.2 | 4.1 | 1.4 | 5.0 | 5.0 | 2.6 | 4.9 | |
| Rest of public sector | −4.7 | −3.8 | −0.5 | −2.8 | −1.9 | −3.6 | −3.4 | −2.0 | 0.5 | 1.5 | 1.3 | |
| Primary surplus | −5.6 | −0.2 | 2.7 | −1.2 | −2.1 | 0.5 | −2.0 | 3.0 | 5.5 | 4.2 | 6.2 | |
| Change in primary surplus | 5.4 | 2.9 | −3.9 | −0.9 | 2.6 | −2.5 | 5.0 | 2.5 | −1.3 | 2.0 | ||
| Change from consolidated budget | 4.5 | −0.4 | −1.7 | −1.7 | 4.3 | −2.7 | 3.6 | 0.0 | −2.3 | 2.3 | ||
| Change from rest of public sector | 0.9 | 3.3 | −2.2 | 0.8 | −1.7 | 0.2 | 1.3 | 2.6 | 1.0 | −0.3 | ||
| Memorandum items | ||||||||||||
| GNP (real growth rate) | 8.4 | −5.0 | 6.9 | 6.9 | 7.6 | 3.1 | −4.7 | 7.4 | −7.5 | 7.9 | 5.8 | |
| Net debt to GNP | 35.1 | 44.7 | 41.3 | 46.5 | 42.9 | 43.7 | 61.0 | 58.3 | 93.9 | 79.2 | 70.9 | |
Defining a “Successful” Adjustment Episode
As a preliminary step, what constitutes a “successful” fiscal adjustment needs to be defined. For many countries, fiscal adjustment can best be defined in terms of changes in the overall balance. However, in Turkey’s case, any definition should also take into account potential problems from debt composition. The large share of foreign currency debt and its short maturity makes the overall balance very volatile. Since the interest bill is largely outside the direct control of policymakers, changes in the overall balance may not be representative of policy efforts, and the primary balance should be preferred instead. A key question then is the coverage of the primary surplus. Large fiscal operations outside the central government budget, including numerous state enterprises with quasi-fiscal operations, would call for a definition that covers the general government. However, it is difficult to obtain reliable information on off-budget operations and state-owned enterprises. Thus, while the adjustment is defined here in terms of the general government primary surplus, the composition of the adjustment can only be examined at the consolidated budget level.
Thus, fiscal adjustment episodes are defined here as when the change in the general government primary surplus was 2.5 percent of GNP in a year or 5 percent of GNP in two consecutive years. Such an adjustment is considered successful when the primary surplus does not fall by more than 2 percent of GNP two years after the adjustment.
In identifying episodes of fiscal adjustment, the impact of the business cycle should also be considered. Year-to-year changes in the primary surplus may be driven not only by policy effort, but also by changes in economic conditions that affect the primary balance via the automatic stabilizers. Thus, following Alesina and Perotti (1996), this chapter considers as episodes of fiscal adjustment only those times when there is a negative fiscal impulse (Figure 8.1).
Measurement of the composition of the adjustment should also be corrected for the cycle. This is approached here in two different ways. The first assumes that the elasticities of revenues and expenditures to GNP are unity. This allows for looking at the changes in the primary surplus as a ratio to GNP as fully driven by discretionary policy. The problem with this approach is that revenues may have larger elasticities—as a result, for example, of the progressive personal income tax system. Also, high inflation can have a significant impact on expenditures and revenues from year to year. As a result of these concerns, the second method estimates cyclically-neutral expenditures and revenues for the period. The difference between the estimated revenues and expenditures and the actual budget outcomes is then regarded as the policy effort.2
Using this definition, three significant episodes of fiscal adjustment in Turkey have been identified over the past 10 years. The first was in 1994–95, when the primary surplus increased by 8.3 percent of GNP; the second in 1998, with an adjustment of 2.6 percent of GNP; and the third in 2000–01, when the primary surplus increased by 7.5 percent of GNP. The first and last episodes are large adjustments by any standard. In addition, both were multiyear adjustments, and in both cases the consolidated budget and the rest of the public sector contributed to the effort almost in an equal manner (Table 8.1).
Using another definition explained earlier, only the 2000–01 adjustment can be considered successful. The 1994–95 adjustment failed to be sustained, since in the two years that followed more than half of it unraveled. Likewise, the 1998 adjustment was fully unraveled in 1999. On the other hand, in the case of the 2000–01 adjustment, the primary surplus did not fall by more than 2 percent of GNP two years afterwards—in fact, it has continued to increase.
However, it is paradoxical that the two adjustments that failed apparently had compositions that should have made them sustainable (Table 8.2). The adjustment in 1994–95 was mostly based on expenditure reduction, although revenue mobilization also played a small role. In 1994, 65 percent of the improvement in the primary surplus of the central government was due to expenditure reduction, with composition such that it would have been expected to result in a lasting adjustment. Noninterest current spending was reduced significantly, and cuts in the wage bill represented more than one-third of the total adjustment. In 1998, the composition of the adjustment was more balanced between revenues and expenditures. However, like in 1994–95, cuts in noninterest current spending were a key element of the adjustment.
Central Government Operations and Composition of Adjustment
(In percent of GNP)
Central Government Operations and Composition of Adjustment
(In percent of GNP)
| 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Revenue | 18.3 | 17.0 | 17.3 | 18.6 | 20.7 | 22.1 | 24.6 | 25.7 | 23.9 | 25.6 | |
| Tax revenue | 14.3 | 13.0 | 14.3 | 15.3 | 16.2 | 17.4 | 19.8 | 20.9 | 19.7 | 21.4 | |
| Direct | 5.5 | 5.3 | 5.6 | 6.2 | 7.6 | 7.7 | 8.0 | 8.4 | 6.5 | 6.9 | |
| Indirect | 8.8 | 7.7 | 8.7 | 9.1 | 8.6 | 9.7 | 11.8 | 12.4 | 13.2 | 14.5 | |
| Nontax revenue | 4.0 | 4.0 | 3.0 | 3.3 | 4.5 | 4.6 | 4.8 | 4.9 | 4.2 | 4.1 | |
| Primary expenditure | 14.7 | 13.8 | 15.7 | 18.8 | 16.6 | 20.7 | 19.6 | 20.7 | 21.3 | 20.6 | |
| Noninterest current spending | 12.8 | 12.5 | 14.0 | 16.6 | 14.8 | 18.7 | 17.6 | 18.4 | 18.7 | 18.6 | |
| Personnel | 7.0 | 6.4 | 6.5 | 7.1 | 7.2 | 8.8 | 7.9 | 8.6 | 8.4 | 8.5 | |
| Investment | 2.0 | 1.3 | 1.7 | 2.2 | 1.9 | 2.0 | 2.0 | 2.4 | 2.5 | 2.0 | |
| Primary surplus (–deficit) | 3.6 | 3.2 | 1.6 | –0.2 | 4.1 | 1.4 | 5.0 | 5.0 | 2.6 | 4.9 | |
| Change in surplus | 4.5 | –0.4 | –1.7 | –1.7 | 4.3 | –2.7 | 3.6 | 0.0 | –2.3 | 2.3 | |
| Composition of adjustment | |||||||||||
| Change in percent of GNP | |||||||||||
| Revenue | 1.6 | –1.3 | 0.3 | 1.3 | 2.1 | 1.3 | 2.6 | 1.1 | − 1.8 | 1.7 | |
| Tax revenue | 2.1 | –1.3 | 1.3 | 1.0 | 0.9 | 1.2 | 2.4 | 1.1 | −1.1 | 1.7 | |
| Direct | –0.1 | –0.2 | 0.3 | 0.6 | 1.4 | 0.1 | 0.3 | 0.5 | − 1.9 | 0.4 | |
| Indirect | 2.2 | –1.1 | 1.0 | 0.4 | –0.4 | 1.1 | 2.1 | 0.6 | 0.7 | 1.3 | |
| Nontax revenue | –0.6 | 0.0 | −1.0 | 0.3 | 1.2 | 0.1 | 0.2 | 0.0 | −0.7 | 0.0 | |
| Primary expenditure | –2.9 | –0.9 | 1.9 | 3.0 | –2.2 | 4.1 | –l.l | 1.1 | 0.5 | −0.6 | |
| Noninterest current spending | –2.0 | –0.3 | 1.5 | 2.6 | –1.9 | 4.0 | –l.l | 0.7 | 0.4 | −0.1 | |
| Personnel | –1.5 | –0.6 | 0.1 | 0.5 | 0.2 | 1.6 | –0.9 | 0.7 | −0.2 | 0.1 | |
| Investment | –0.9 | –0.7 | 0.4 | 0.5 | –0.3 | 0.1 | 0.0 | 0.4 | 0.2 | −0.5 | |
| Memorandum item | |||||||||||
| Cumulative improvement from revenues | 34.7 | 6.4 | 49.4 | 70.2 | 100.0 | 72.1 | |||||
| Cumulative improvement from expenditures | 65.3 | 93.6 | 50.6 | 29.8 | 0.0 | 27.9 | |||||
| Cumulative improvement from wage bill | 32.7 | 50.7 | –4.2 | 24.3 | 5.8 | –3.2 | |||||
Central Government Operations and Composition of Adjustment
(In percent of GNP)
| 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Revenue | 18.3 | 17.0 | 17.3 | 18.6 | 20.7 | 22.1 | 24.6 | 25.7 | 23.9 | 25.6 | |
| Tax revenue | 14.3 | 13.0 | 14.3 | 15.3 | 16.2 | 17.4 | 19.8 | 20.9 | 19.7 | 21.4 | |
| Direct | 5.5 | 5.3 | 5.6 | 6.2 | 7.6 | 7.7 | 8.0 | 8.4 | 6.5 | 6.9 | |
| Indirect | 8.8 | 7.7 | 8.7 | 9.1 | 8.6 | 9.7 | 11.8 | 12.4 | 13.2 | 14.5 | |
| Nontax revenue | 4.0 | 4.0 | 3.0 | 3.3 | 4.5 | 4.6 | 4.8 | 4.9 | 4.2 | 4.1 | |
| Primary expenditure | 14.7 | 13.8 | 15.7 | 18.8 | 16.6 | 20.7 | 19.6 | 20.7 | 21.3 | 20.6 | |
| Noninterest current spending | 12.8 | 12.5 | 14.0 | 16.6 | 14.8 | 18.7 | 17.6 | 18.4 | 18.7 | 18.6 | |
| Personnel | 7.0 | 6.4 | 6.5 | 7.1 | 7.2 | 8.8 | 7.9 | 8.6 | 8.4 | 8.5 | |
| Investment | 2.0 | 1.3 | 1.7 | 2.2 | 1.9 | 2.0 | 2.0 | 2.4 | 2.5 | 2.0 | |
| Primary surplus (–deficit) | 3.6 | 3.2 | 1.6 | –0.2 | 4.1 | 1.4 | 5.0 | 5.0 | 2.6 | 4.9 | |
| Change in surplus | 4.5 | –0.4 | –1.7 | –1.7 | 4.3 | –2.7 | 3.6 | 0.0 | –2.3 | 2.3 | |
| Composition of adjustment | |||||||||||
| Change in percent of GNP | |||||||||||
| Revenue | 1.6 | –1.3 | 0.3 | 1.3 | 2.1 | 1.3 | 2.6 | 1.1 | − 1.8 | 1.7 | |
| Tax revenue | 2.1 | –1.3 | 1.3 | 1.0 | 0.9 | 1.2 | 2.4 | 1.1 | −1.1 | 1.7 | |
| Direct | –0.1 | –0.2 | 0.3 | 0.6 | 1.4 | 0.1 | 0.3 | 0.5 | − 1.9 | 0.4 | |
| Indirect | 2.2 | –1.1 | 1.0 | 0.4 | –0.4 | 1.1 | 2.1 | 0.6 | 0.7 | 1.3 | |
| Nontax revenue | –0.6 | 0.0 | −1.0 | 0.3 | 1.2 | 0.1 | 0.2 | 0.0 | −0.7 | 0.0 | |
| Primary expenditure | –2.9 | –0.9 | 1.9 | 3.0 | –2.2 | 4.1 | –l.l | 1.1 | 0.5 | −0.6 | |
| Noninterest current spending | –2.0 | –0.3 | 1.5 | 2.6 | –1.9 | 4.0 | –l.l | 0.7 | 0.4 | −0.1 | |
| Personnel | –1.5 | –0.6 | 0.1 | 0.5 | 0.2 | 1.6 | –0.9 | 0.7 | −0.2 | 0.1 | |
| Investment | –0.9 | –0.7 | 0.4 | 0.5 | –0.3 | 0.1 | 0.0 | 0.4 | 0.2 | −0.5 | |
| Memorandum item | |||||||||||
| Cumulative improvement from revenues | 34.7 | 6.4 | 49.4 | 70.2 | 100.0 | 72.1 | |||||
| Cumulative improvement from expenditures | 65.3 | 93.6 | 50.6 | 29.8 | 0.0 | 27.9 | |||||
| Cumulative improvement from wage bill | 32.7 | 50.7 | –4.2 | 24.3 | 5.8 | –3.2 | |||||
In contrast, the 2000–01 adjustment was driven mainly by revenue increases. In 2000, 70 percent of the improvement in the primary surplus came from revenue increases, especially indirect taxes (value-added tax rates were raised 2 percentage points). For the two year period of 2000–01, the cumulative adjustment was entirely due to revenue increases. Primary expenditures were allowed to grow, particularly the government wage bill and the Social Security deficits. This composition would not seem to bode well for a lasting adjustment, yet it was sustained for at least two years.
As a cross-check, the raw figures for fiscal adjustment were reestimated to include the effects of the business cycle (Table 8.3). To do this, we have estimated cyclically neutral revenues and expenditures, and compared these with the actual budget outcomes. The growth rate of potential real GNP was estimated using a Hodrick-Prescott filter and used to forecast revenues and expenditures. Revenues were assumed to grow with actual nominal GNP, estimating the elasticity by using the average buoyancy of revenues with respect to GNP growth, and expenditures to grow with potential GNP. The difference between these estimated cyclically neutral revenues and expenditures and the actual outcomes can be used as a measure of policy effort.
Estimate of Composition of Adjustment
(In percent of GNP)
Estimate of Composition of Adjustment
(In percent of GNP)
| 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Revenues (actual) | 18 | 17 | 17 | 19 | 21 | 22 | 25 | 26 | 24 | 26 | |
| Primary expenditures (actual) | 15 | 14 | 16 | 19 | 17 | 21 | 20 | 21 | 21 | 21 | |
| Projected performance | |||||||||||
| Revenue | 17.9 | 19.6 | 18.2 | 18.5 | 19.8 | 21.7 | 23.2 | 25.6 | 27.0 | 24.7 | |
| Tax revenue | 13.1 | 15.5 | 14.0 | 15.4 | 16.4 | 17.1 | 18.5 | 20.7 | 22.1 | 20.5 | |
| Direct | 5.8 | 5.7 | 5.5 | 5.8 | 6.4 | 7.8 | 7.9 | 8.1 | 8.6 | 6.6 | |
| Indirect | 7.3 | 9.8 | 8.6 | 9.7 | 10.0 | 9.3 | 10.6 | 12.7 | 13.5 | 13.9 | |
| Nontax revenue | 4.8 | 4.2 | 4.2 | 3.2 | 3.5 | 4.7 | 4.8 | 5.0 | 5.0 | 4.3 | |
| Primary expenditure | 19.0 | 13.7 | 12.5 | 15.2 | 19.0 | 18.9 | 18.9 | 22.6 | 18.8 | 20.6 | |
| Estimated policy effort (actual-projected) | |||||||||||
| Revenue | 0.4 | –2.6 | –0.8 | 0.1 | 0.9 | 0.4 | 1.4 | 0.1 | –3.1 | 0.9 | |
| Tax revenue | 1.2 | −2.5 | 0.3 | −0.1 | −0.2 | 0.4 | 1.3 | 0.1 | −2.3 | 1.0 | |
| Direct | −0.3 | −0.4 | 0.1 | 0.4 | 1.2 | −0.1 | 0.1 | 0.3 | −2.1 | 0.3 | |
| Indirect | 1.5 | −2.1 | 0.1 | −0.6 | −1.4 | 0.5 | 1.2 | −0.2 | −0.3 | 0.6 | |
| Nontax revenue | −0.8 | −0.2 | −1.2 | 0.1 | 1.0 | 0.0 | 0.0 | −0.1 | −0.9 | −0.1 | |
| Primary expenditure | –4.2 | 0.1 | 3.3 | 3.6 | –2.4 | 1.8 | 0.8 | –1.9 | 2.5 | 0.0 | |
| Memorandum | |||||||||||
| Cumulative improvement from revenues | 9.2 | –52.8 | |||||||||
| Cumulative improvement from expenditures | 94.8 | 100.6 | |||||||||
| Fiscal impulse | −4.7 | 2.7 | 4.1 | 3.4 | −3.3 | 1.4 | −0.6 | −2.0 | 5.6 | −0.9 | |
Estimate of Composition of Adjustment
(In percent of GNP)
| 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | 2003 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Revenues (actual) | 18 | 17 | 17 | 19 | 21 | 22 | 25 | 26 | 24 | 26 | |
| Primary expenditures (actual) | 15 | 14 | 16 | 19 | 17 | 21 | 20 | 21 | 21 | 21 | |
| Projected performance | |||||||||||
| Revenue | 17.9 | 19.6 | 18.2 | 18.5 | 19.8 | 21.7 | 23.2 | 25.6 | 27.0 | 24.7 | |
| Tax revenue | 13.1 | 15.5 | 14.0 | 15.4 | 16.4 | 17.1 | 18.5 | 20.7 | 22.1 | 20.5 | |
| Direct | 5.8 | 5.7 | 5.5 | 5.8 | 6.4 | 7.8 | 7.9 | 8.1 | 8.6 | 6.6 | |
| Indirect | 7.3 | 9.8 | 8.6 | 9.7 | 10.0 | 9.3 | 10.6 | 12.7 | 13.5 | 13.9 | |
| Nontax revenue | 4.8 | 4.2 | 4.2 | 3.2 | 3.5 | 4.7 | 4.8 | 5.0 | 5.0 | 4.3 | |
| Primary expenditure | 19.0 | 13.7 | 12.5 | 15.2 | 19.0 | 18.9 | 18.9 | 22.6 | 18.8 | 20.6 | |
| Estimated policy effort (actual-projected) | |||||||||||
| Revenue | 0.4 | –2.6 | –0.8 | 0.1 | 0.9 | 0.4 | 1.4 | 0.1 | –3.1 | 0.9 | |
| Tax revenue | 1.2 | −2.5 | 0.3 | −0.1 | −0.2 | 0.4 | 1.3 | 0.1 | −2.3 | 1.0 | |
| Direct | −0.3 | −0.4 | 0.1 | 0.4 | 1.2 | −0.1 | 0.1 | 0.3 | −2.1 | 0.3 | |
| Indirect | 1.5 | −2.1 | 0.1 | −0.6 | −1.4 | 0.5 | 1.2 | −0.2 | −0.3 | 0.6 | |
| Nontax revenue | −0.8 | −0.2 | −1.2 | 0.1 | 1.0 | 0.0 | 0.0 | −0.1 | −0.9 | −0.1 | |
| Primary expenditure | –4.2 | 0.1 | 3.3 | 3.6 | –2.4 | 1.8 | 0.8 | –1.9 | 2.5 | 0.0 | |
| Memorandum | |||||||||||
| Cumulative improvement from revenues | 9.2 | –52.8 | |||||||||
| Cumulative improvement from expenditures | 94.8 | 100.6 | |||||||||
| Fiscal impulse | −4.7 | 2.7 | 4.1 | 3.4 | −3.3 | 1.4 | −0.6 | −2.0 | 5.6 | −0.9 | |
Correcting for the business cycle changes the composition of adjustment somewhat, but the basic conclusions are unchanged.3 The revised estimates for 1994–95 suggest that the entire adjustment came from expenditure reduction, and that taxes were even allowed to decline. In 1998, the fiscal adjustment was still shared between expenditures and revenues, although the cyclically-adjusted estimates point to an even larger reduction of expenditures. When the cyclical adjustment is made, the results for 2000–01 suggest that the effort is more balanced between revenues and expenditures than the raw data suggest, with almost 40 percent of the adjustment coming from reductions in primary expenditure. Still, the overall picture is the same: the first two episodes were based mainly on expenditure adjustment, while the last one relied much more on revenue measures.
However, other factors aside from the composition of adjustment may have played a key role in the sustainability of these episodes of fiscal consolidation. Perhaps the more important factor behind the lack of success in the first two adjustments was political. Both adjustments took place under weak coalition governments. The unraveling of these governments and the ensuing elections tended to result in an unwinding of the fiscal consolidation. The current, more successful adjustment was also initiated by a coalition government, and it too showed signs of unraveling in the run-up to elections in 2002. The election of a strong single-party government may have been key to ensuring the continuation of the consolidation. Another key factor may have been the growing concern over debt sustainability. In 1994 and 1998, debt sustainability concerns were not yet at a critical stage, so there was seemingly less cost to slippage. But during the 2000–01 crisis, debt sustainability emerged as the main concern. The cost of failure was so great that unsuccessful fiscal consolidation could not be contemplated.
In addition, the analysis presented earlier is very aggregate and does not capture well the quality of individual expenditure measures. Although the first two fiscal adjustments relied on expenditure reduction, the measures may have been of poor quality and therefore unsustainable. Although the reduction in the wage bill in 1994–95 was impressive, it did not deal with the underlying problem of growing public sector employment. The wage bill as a share of GNP in 1995 was still almost 0.5 percent higher than in 1989, and the adjustment of 1994–95 seemed to have been due more to inflation surprises than design, and was therefore difficult to sustain over time. Finally, despite measures to curtail primary expenditures in 1994–95 and 1998, the Social Security deficit was allowed to continue to grow, rising from 0.5 percent of GNP in 1993 to 2.6 percent of GNP in 1998.
Implementation of structural reforms may also help support the latest adjustment. The institutional framework for fiscal policy has been strengthened. Together with a significant reduction in off-budget operations, this might help maintain fiscal control. This was missing in earlier episodes. There also has been a significant effort to curtail quasi-fiscal operations in state enterprises. Measuring the composition of adjustment at the budget level misses these developments in the rest of the public sector. For example, in 2000–01 a substantial effort went into eliminating implicit subsidies and redundant employment in state economic enterprises.
Conclusions
Turkey’s experience does not appear to sit well with the main findings in the literature on successful fiscal adjustment. Earlier episodes of large fiscal adjustment had a composition that should have increased the probability of success, yet they unraveled quickly. Conversely, the current adjustment, with its emphasis on revenue measures, would be expected to be short lived, yet it has been sustained for more than two years.
In Turkey’s case, a better understanding of fiscal sustainability may require extending the traditional analysis beyond fiscal composition considerations. For starters, a stable political backdrop is critical. A supportive institutional setup—aided by wide-ranging structural reforms—also appears central to the sustainability of fiscal adjustment efforts in Turkey. Nevertheless, if the country’s fiscal adjustment efforts are to be sustained over the medium term, greater attention will likely need to be devoted to compositional issues, especially in controlling current expenditures.
They defined a fiscal adjustment in terms of a negative fiscal impulse of 1.5 percent of GDP or more and defined an adjustment as successful when, three years after the adjustment, the debt-to-GDP ratio was 5 percentage points lower.
We have used the OECD formulation for fiscal impulse and for the estimate of cyclical effects on revenues and neutral expenditures: FI = [(Gt–Tt)–(Gt–1 * (1 + ỹt)–Tt–1 * (1 + αyt)]/Yt, where FI is the fiscal impulse; G is primary expenditure; T is government revenues; y is actual nominal growth rate of GNP; Y is actual nominal GNP; ỹ is the estimated potential real GNP growth rate plus the actual GNP deflator; and α is revenue elasticity to GNP.
The results from these estimates are sensitive to changes in the assumptions. For example, increasing the potential growth rate and the revenue elasticity would tilt the estimated composition of adjustment toward expenditure reduction.
