Abstract
The Selected Issues paper and Statistical Appendix analyzes output developments in the Russian Federation since the 1998 crisis. It outlines near-term growth prospects for the economy. The paper highlights that output growth accelerated in 1999 and the first half of 2000, but has slowed since then. The initial output recovery was led by import substitution as a result of the large exchange rate depreciation in 1998. One finding in the context of an overall policy package is that the real exchange rate and oil prices were the main determinants of growth after the 1998 crisis.
V. Fiscal Sustainability1
1. Despite an impressive improvement in Russia’s fiscal position since the crisis in 1998, questions remain about the sustainability of public finances over the medium to long term. The enlarged government primary balance strengthened by 11 percent of GDP during 1997–2000, reaching a surplus of almost 8 percent of GDP in 2000 and is estimated to have declined only moderately in 2001. At the same time, the level of federal government debt was roughly halved to 64 percent of GDP at end-2000, that is estimated to have declined further to about 50 percent of GDP by end-2001, reflecting also strong GDP growth and the real appreciation of the ruble during 1999–2001. The strengthening of the fiscal position reflects a sharp compression of noninterest expenditure and, more recently, an increase in revenues owing to rapid income growth and high global energy prices since 1999. The question is to what extent this improvement can be sustained in the face of weaker growth, lower energy prices, a lowering of the tax burden, and potentially significant budget pressures arising from compressed social spending and reform costs. The latter issues are particularly important at the subnational government level.
2. Our analysis suggests that Russia’s fiscal position should be sustainable, but that it remains vulnerable over the medium term to a sharp downturn in global energy prices. While there should be considerable room for the fiscal surplus to decline from its current high level, this should be implemented cautiously to reduce vulnerability. In this context, pressures to spend current windfall gains should be contained and the fiscal cost of reforms carefully assessed so that the implementation of these reforms is appropriately paced and sequenced. Further, intergovernmental fiscal relations should be re-examined to ensure that policies are sustainable at all levels of government.
3. We examine the issue of fiscal sustainability in a normative framework. We assume that economic developments in Russia will gradually resemble those in advanced transition countries and emerging markets. We also assume that global energy prices return to their “normal” level. Fiscal sustainability is examined both from a fiscal debt sustainability and from a balance of payments sustainability perspective (or macroeconomic balance approach). In addition, we look at fiscal policy over the medium term where financing constraints also need to be addressed.
A. Conceptual Framework
4. Long-term fiscal sustainability is typically analyzed in terms of public debt sustainability.2 The standard analysis of debt sustainability looks at the primary fiscal balance consistent with maintaining a certain “sustainable” ratio of public debt to GDP given output growth, real interest and exchange rates:
where PB is the primary fiscal balance, Y is nominal GDP, r is the real interest rate on government debt, g is the real GDP growth rate, and D is the stock of public debt. This formula assumes a constant nominal exchange rate and ignores potential access to nondebt financing.3
Thus, if the government is in a net debtor position, and the real interest rate exceeds the rate of real GDP growth, a primary surplus would be required to maintain a constant debt ratio.
5. For a transition economy, the macroeconomic balance approach may be more relevant. This approach looks at the fiscal balance consistent with a viable or target external current account position and the outlook for private savings and investment:
where (S−I)gov denotes the overall fiscal balance, CA* the sustainable current account, and (S−I)priv the private sector savings-investment balance.4
This approach may be more relevant for a country like Russia, where long-term debt sustainability considerations are likely to be dominated by concerns about medium-term balance of payments viability and where prudential thresholds on public debt ratios for industrial countries (as established notably for members of the European Monetary Union by the Maastricht criteria) are probably of limited value in determining sustainable fiscal policy. Of course, one would need to ensure that the fiscal balance determined in this way does not raise concern about debt sustainability.
6. We employ a medium- to long-term perspective and a “normative” framework. We assume that the economy will reach during 2010–15 a “steady-state” economic development path similar to that observed in advanced transition and emerging economies, and—given the dependence of both public finances and the balance of payments on energy prices—where global energy prices have stabilized at a “normal” level. It is thus necessary to determine “normal” energy prices and a viable external current account balance as well as to analyze the prospects for real GDP growth, private investment and savings, and real interest rates and exchange rates.
7. We draw on both theoretical considerations and evidence from comparator countries, and we distinguish three groups of countries:
non-transition emerging markets (Argentina, Brazil, China, Korea, Malaysia, Mexico, South Africa, Thailand, and Turkey);
advanced transition economies (Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovak Republic, and Slovenia); and
resource rich economies (Chile, Indonesia, Kazakhstan, Mexico, Nigeria, Norway, Saudi Arabia, South Africa, United Kingdom, and Venezuela).
The latter group is included as these otherwise diverse economies share certain characteristics owing to their dependence on income from natural resources. In addition to historical data, we include the latest World Economic Outlook (WEO) projections.
B. “Normal” Energy Prices
8. Although it has proved difficult to model oil prices, various methods suggest remarkably similar values for their long-term level. Most studies have been unable to reject the hypothesis that—over the long run—oil prices follow a random walk either in nominal or real terms (see e.g., Engel and Valdes, 2000). One statistical approach to determining long- run oil prices is to examine the 10-year moving average. This has fluctuated in the range of $17.6–18.9 a barrel since 1996.5 Alternatively, it could be assumed that prices revert to the level prevailing in a stable period (say 1987–97, where the average price stripping out the Gulf war spike was $17.8). A third approach is to run an ARIMA regression for this stable period, which suggests that oil prices are stationary, with a long-run level of $19.3.6 However, these statistical constructs based on historical prices ignore any information regarding structural shifts in oil markets and thus in future oil prices. The WEO forecasts oil prices five years ahead based on market forecasts—futures prices and/or forward contracts—and trade publications. For 2007, the December 2001 WEO forecast is $19.5.
9. These methods thus suggest that a plausible “normal” level of oil prices is in the range $18–20. In line with the medium-term WEO forecast, we thus assume a long-term average oil price of $19 a barrel. We use this oil price both for extrapolating backward and for projecting forward the constant-oil-price fiscal and external balances, including to assess fiscal sustainability.7
C. Growth Prospects and Private Sector Investment/Savings
Potential output growth
10. Output growth in comparator countries has broadly been in the range of 4–6 percent a year. In emerging markets, output growth has generally been in the range of 4–6 percent a year over the last 20 years, except for the period of the Asian and Russian crisis. Similar growth rates have been experienced by advanced transition economies in the second half of the 1990s, with improvements in total factor productivity accounting for about one-half of the growth.8 Growth rates appear to have been somewhat lower in resource rich economies, both in a longer time perspective and in recent years, when it has hovered around 4 percent.
11. Empirical evidence on the determinants of growth confirms that Russia should be able to achieve growth rates similar to comparator countries. In analyzing growth in transition economies, several authors have used variations of the following general specification:
fy<0; fgov<0; fpop<0;fps>0;fss>0; finv>0
where g is per capita growth, y is per capita income, PS and SS are the primary and secondary school enrollment rates (proxies for human capital), respectively, INV is gross capital formation in percent of GDP, GOV is government consumption in percent of GDP, POP is the population growth rate, and f denotes the first derivative.
Real GDP growth
(In percent to previous year)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Real GDP growth
(In percent to previous year)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Real GDP growth
(In percent to previous year)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
12. Table 2 below shows the values of these key variables for Russia and for a sample of fast-growing and of slow-growing developing economies. The high level of education in Russia suggests that growth prospects are relatively favorable. Based on the empirical equation first developed by Levine & Renelt (1992), the forecasted per capita growth rate for Russia would be around 4.7 percent with the relatively high value, in large part, reflecting the relatively high level of human capital.9
Factors Affecting Long-Term Growth: Russia in a Global Perspective
Factors Affecting Long-Term Growth: Russia in a Global Perspective
| Variable | Russia | Fast-Growers | Slow-Growers |
|---|---|---|---|
| Primary-school enrollment rate (pet) | 91 | 90 | 54 |
| Secondary-school enrollment rate (pet) | 92 | 30 | 10 |
| Gross investment/GDP (pet) | 0.23 | 0.23 | 0.17 |
| Government consumption/GDP (pet) | 0.21 | 0.16 | 0.12 |
| Population growth rate (pet) | −0.28 | … | … |
Factors Affecting Long-Term Growth: Russia in a Global Perspective
| Variable | Russia | Fast-Growers | Slow-Growers |
|---|---|---|---|
| Primary-school enrollment rate (pet) | 91 | 90 | 54 |
| Secondary-school enrollment rate (pet) | 92 | 30 | 10 |
| Gross investment/GDP (pet) | 0.23 | 0.23 | 0.17 |
| Government consumption/GDP (pet) | 0.21 | 0.16 | 0.12 |
| Population growth rate (pet) | −0.28 | … | … |
Real Deposit and Lending Rates
(1990–2000)
Real Deposit and Lending Rates
(1990–2000)
| Industrial Countries | Advanced Countries | Developing Countries | Transition Countries | |
|---|---|---|---|---|
| (In percent) | ||||
| Real deposit rate: mean | 1.5 | 2.5 | 3.5 | 1.7 |
| Real deposit rate: median | 1.7 | 2.0 | 4.0 | 1.0 |
| Real lending rate: mean | 5.0 | 6.6 | 12.2 | 7.8 |
| Real lending rate: median | 5.0 | 6.0 | 9.9 | 8.3 |
Real Deposit and Lending Rates
(1990–2000)
| Industrial Countries | Advanced Countries | Developing Countries | Transition Countries | |
|---|---|---|---|---|
| (In percent) | ||||
| Real deposit rate: mean | 1.5 | 2.5 | 3.5 | 1.7 |
| Real deposit rate: median | 1.7 | 2.0 | 4.0 | 1.0 |
| Real lending rate: mean | 5.0 | 6.6 | 12.2 | 7.8 |
| Real lending rate: median | 5.0 | 6.0 | 9.9 | 8.3 |
13. Fischer, Sahay, & Vegh (1998), using a similar approach, forecasted per capita growth rates in the range 4.8–5.3 percent. Gomulka (2000) studied real GDP growth in transition economies using as explanatory variables the ratio of gross investment to GDP, per capita income, and inflation. He found that for investment to GDP ratios in the range of 20 to 30 percent, Russia’s forecasted growth rate over the next decade would be between 4.5 percent and 6 percent. The EBRD in its 1997 Transition Report similarly noted that Russia should be able to grow at rates of around 5 percent per year. In sum, evidence suggests that a potential growth rate of 5–6 percent in Russia is a reasonable working assumption. This, of course, assumes that structural reforms are implemented in a comprehensive and sustained manner.
14. Russia’s growth potential could also be examined in a growth accounting framework (see e.g., IMF 2000a and Laursen 2000). This would be based on the specification of a production function and assumptions about developments in inputs, output elasticities, and total factor productivity. We use such a framework below to examine the level of investment required to sustain output growth at its potential.
Gross investment
15. Private investment in transition economies has generally been higher than in advanced emerging markets. Over the last 20 years, gross investment in emerging markets has fluctuated in the range of 15–20 percent of GDP (it appears to be stabilizing in the upper end of this interval). In contrast, investment in transition economies has been 24–25 percent of GDP in the second half of the 1990s implying a higher incremental capital-output ratio. While there is no evidence of higher investment ratios in resource-rich countries, rather the opposite, this may not be all that relevant for Russia given its under-developed extraction and transportation infrastructure. Turning to more disaggregated data, private investment ratios in advanced transition economies typically lie in the range 19–24 percent of GDP, with public investment accounting for another 3–6 percent of GDP; in other advanced emerging markets, private investment generally hovers around 17–20 percent of GDP.
16. Investment needs may be examined in a simple growth accounting framework. Assuming a constant-returns-to-scale Cobb-Douglas production function, the gross investment ratio required to support the long run potential growth rate equals:
where I is gross investment, Y is output, TFP is total factor productivity, K is capital, L is labor, gx denotes the net growth rate of any variable x, α and (1-α) are the marginal elasticities of output with respect to labor and to capital, respectively, and δ is the rate of capital depreciation.
Private Investment
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Private Investment
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Private Investment
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
17. As discussed above, real GDP should potentially be growing at 5–6 percent per year in the medium to long run, with around one-half being accounted for by total factor productivity growth. According to the national accounts, the prevailing capital-output ratio is around seven. However, this implies unrealistically high values for the contribution of total factor productivity growth, and various studies suggest that the capital-output ratio is considerably overestimated owing to a low value of pre-transition capital (see e.g., McKinsey, 1999). A more plausible number would be around three, implying that pre-transition capital is worth only about 10 percent of post-transition capital. The depreciation rate has been around 4 percent a year in recent years according to the national accounts. The labor force should remain broadly stable based on demographic projections. Finally, international evidence suggests that the marginal elasticity of output with respect to labor—equal to the labor share assuming competitive labor markets—is fairly stable at around 65 percent.10 With these parameters, the required total investment-to-GDP ratio would be about 26 percent.
18. In sum, it seem reasonable to assume a gross investment ratio of 26 percent, of which the private investment ratio would be about 21 percent (reflecting both the evidence from advanced transition economies and the prediction from the growth accounting framework).
Private sector saving
19. Private saving ratios in emerging markets and advanced transition economies have been fairly similar and stable. In these groups of countries, private saving has fluctuated in the range of 18–22 percent of GDP over an extended period. It has been somewhat lower in resource-rich economies, fluctuating around 15 percent of GDP.
Private Savings
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Private Savings
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Private Savings
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
20. The theoretical and empirical analysis of private saving has focused little on the unique institutional structures of transition economies. While the literature generally finds that a number of factors are important determinants of saving levels in countries independent of their level of development (e.g., the dependency ratio, public saving, the rate of output growth, and changes in terms of trade), it also suggests that private saving in developing countries is influenced importantly by other factors than in industrial countries (Appendix I). In particular, the degree of financial liberalization and foreign saving. Given the different findings for industrial and developing countries, it is conceivable that emerging markets and the more advanced transition economies may also exhibit a unique behavior that may be more relevant for Russia. This could, for example, relate to the significant role of enterprises in providing social safety nets in transition economies or the deeper integration with international financial markets in emerging markets, as well as macroeconomic instability in both groups of countries.
21. We have therefore estimated a private savings equation for a sample consisting of our two groups of advanced emerging markets and transition economies (Annex I). The results show a significant positive effect of real GDP growth and financial deepening (proxied by the ratio of M2 to GDP), and a negative effect of the budget deficit and the dependency ratio. The impact of inflation volatility is very small. Per capita income was not found to be significant.11 Using this equation and projected long-run values for these variables for Russia suggests a private savings rate of about 20 percent of GDP.
Real interest rate
22. Evidence on real interest rates in emerging and transition economies is difficult to interpret. This owes to imperfect financial markets, capital controls, unreliable data on interest rates on government debt, typically large spreads between lending and deposit rates, volatile inflation, and uncertain inflation expectations. Estimates suggest that real lending rates in transition economies have been around 8 percent, somewhat less than in developing countries but higher than in advanced and industrial countries (Table 2).
23. While real interest rates are currently very low in Russia, they can be expected to gradually mirror those in transition and advanced economies. The current low rates reflect the absence of government borrowing requirements. Yield spreads on Russian Eurobonds have fallen substantially since the crisis but remain relatively high compared to other transition and emerging markets given the current strength of macroeconomic fundamentals. This reflects both the lingering memories of the default on government debt in 1998 and the large debt repayments falling due in coming years. Over time one would expect spreads on new Russian long-term borrowing to be broadly similar to those faced by advanced transition economies and emerging markets, which have typically been around 200–300 basis points (abstracting from crisis periods). Thus, with real interest rates in industrial countries around 4 percent (average of the 1990s), it seems reasonable to assume that Russia would be facing real interest rates the order of 6–7 percent (the choice of ruble debt versus foreign currency debt is essentially a debt management issue).12
D. Sustainable External and Fiscal Balances
24. Russia should be targeting an external current account deficit of 2–3 percent of GDP. This is well within the danger zone experienced by a number of emerging markets and transition economies where deficits of 4–6 percent have generally proved to be sustainable. However, most of these countries, especially the advanced transition economies, have benefited from very high levels of foreign direct investment, and are generally converging toward lower external deficits in the 2–4 percent of GDP range. A cautious current account target is also warranted by Russia’s experience with large capital flight and potentially volatile capital flows in the future.
Current Account Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Current Account Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Current Account Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
25. This implies that the general government deficit should be limited to 1–2 percent of GDP. With private investment expected to expand rapidly, our analysis above suggests that the private sector savings-investment balance is likely to turn slightly negative, by about 1 percent of GDP, requiring that the fiscal deficit is contained to 1–2 percent of GDP to be consistent with the current account target (this compares to projected fiscal deficits of ½–2 percent of GDP in advanced transition economies and emerging markets). Given projected interest payments, this in turn requires a primary surplus of 1–2 percent of GDP.
General Government Primary Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
General Government Primary Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
General Government Primary Balance
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
26. These external current account and fiscal balances are consistent with debt sustainability. External and public sector debt ratios would be declining fairly rapidly over the medium term, reflecting both the large current surpluses and the relatively low interest rate on the current debt stock, as well as real appreciation of the ruble (see below). Thus, staff projections suggest that by 2005 total federal government debt would have declined to about 30 percent of GDP and it would continue declining thereafter.13 This compares to debt ratios in emerging market and transition economies that increased to about 45 percent of GDP during the 1990s but are expected to gradually decline to 30–40 percent over the medium term.14
Total Public Debt
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Total Public Debt
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
Total Public Debt
(In percent of GDP)
Citation: IMF Staff Country Reports 2002, 075; 10.5089/9781451833034.002.A005
27. Russia should aim for conservative fiscal and external balances and debt levels. Fiscal revenues and exports remain highly dependent on volatile global commodity prices, and budget financing is subject to both swings in international financial markets and limited development of domestic capital markets. Also, both reform costs and contingent liabilities may prove to be substantial although difficult to quantify ex ante. Further, the balance of payments remains characterized by large private capital outflows and a relatively low level of FDI and it is uncertain how fast these flows will improve. Under these circumstances, it would be prudent to target debt levels at the lower end of those aimed for by advanced emerging markets and transition economies.
28. The adjustment of external and fiscal balances over the medium term to the “steady-state” discussed above is of paramount importance to policymakers. While there is considerable room for the current account balance to deteriorate and the real exchange rate to appreciate over the medium term, this process should proceed cautiously—both to maintain a prudent reserve cover in the face of only a slow reduction in capital flight and to avoid jeopardizing growth while the investment climate and domestic demand are strengthened. Similarly, while there is room to ease the underlying fiscal position by about 3 percentage points of GDP, caution is warranted as financing remains limited, the fiscal balance remains vulnerable to a sharp downturn in global energy prices, and additional budget pressures could arise from the related expenditures.
Key Parameters
(Share of GDP unless otherwise noted)
Cash basis.
“Steady Stale” refers to the average over 2010–15.
Share of federal government revenues.
Key Parameters
(Share of GDP unless otherwise noted)
| 1999 | 2000 | 2001 | Steady-State | |||
|---|---|---|---|---|---|---|
| Real GDP growth | 5.4 | 8.3 | 5.2 | 5 to 6 | ||
| Gross investment | 15.1 | 17.2 | 19.5 | 26 | ||
| o/w Private | 6.4 | 9.3 | 12.2 | 21 | ||
| National savings | 26.9 | 35.2 | 30.6 | 23 to 24 | ||
| o/w Private | 21.4 | 23.7 | 20.1 | 20 | ||
| External current account | 11.8 | 18.0 | 11.0 | −2 to −3 | ||
| Federal government 1/ | ||||||
| Primary balance | 4.5 | 5.3 | 1 to 2 | |||
| Overall balance | 1.9 | 2.7 | −1 to −2 | |||
| Total debt 2/ | 96.5 | 63.8 | 48.4 | 24 | ||
| External debt service 2/3/ | 35.6 | 25.3 | 24.7 | 16 | ||
| Enlarged government | ||||||
| Primary balance | 6.4 | 5.5 | 1 to 2 | |||
| Overall balance | 3.8 | 2.9 | −1 to −2 | |||
| Memo items: | ||||||
| World oil price | 18.0 | 28.2 | 24.3 | 19 | ||
| REER (change; e.o.p) | 12.3 | 22.0 | 12.8 | 2 | ||
| Underlying fiscal | ||||||
| primary balance | 3.5 | 4.7 | 1 to 2 | |||
Cash basis.
“Steady Stale” refers to the average over 2010–15.
Share of federal government revenues.
Key Parameters
(Share of GDP unless otherwise noted)
| 1999 | 2000 | 2001 | Steady-State | |||
|---|---|---|---|---|---|---|
| Real GDP growth | 5.4 | 8.3 | 5.2 | 5 to 6 | ||
| Gross investment | 15.1 | 17.2 | 19.5 | 26 | ||
| o/w Private | 6.4 | 9.3 | 12.2 | 21 | ||
| National savings | 26.9 | 35.2 | 30.6 | 23 to 24 | ||
| o/w Private | 21.4 | 23.7 | 20.1 | 20 | ||
| External current account | 11.8 | 18.0 | 11.0 | −2 to −3 | ||
| Federal government 1/ | ||||||
| Primary balance | 4.5 | 5.3 | 1 to 2 | |||
| Overall balance | 1.9 | 2.7 | −1 to −2 | |||
| Total debt 2/ | 96.5 | 63.8 | 48.4 | 24 | ||
| External debt service 2/3/ | 35.6 | 25.3 | 24.7 | 16 | ||
| Enlarged government | ||||||
| Primary balance | 6.4 | 5.5 | 1 to 2 | |||
| Overall balance | 3.8 | 2.9 | −1 to −2 | |||
| Memo items: | ||||||
| World oil price | 18.0 | 28.2 | 24.3 | 19 | ||
| REER (change; e.o.p) | 12.3 | 22.0 | 12.8 | 2 | ||
| Underlying fiscal | ||||||
| primary balance | 3.5 | 4.7 | 1 to 2 | |||
Cash basis.
“Steady Stale” refers to the average over 2010–15.
Share of federal government revenues.
29. Fiscal policy in the next couple of years is likely to be constrained on the financing side. With limited access to international capital markets and large debt service payments falling due, especially in 2003, foreign financing will be negative over the medium term. While the government should be able to gradually substitute foreign for domestic financing, the ability of the financial system to absorb new government debt will be constrained by the slow process of remonetization and increasing demand for credit to support a broad based expansion of investment. Under these circumstances, there is little scope for the budget to go into deficit, at least through 2003–04, and thus for the primary surplus to decline significantly from its envisaged level in 2002.15
30. This will place the federal budget under considerable pressures. With revenues projected to decline by about 2½ percent of GDP over the medium term (2002–04) as a result of lower oil prices and macroeconomic developments, considerable expenditure restraint will be required in the coming years. While it should be possible to maintain noninterest expenditures close to the level envisaged for 2002 of 13 percent of GDP, expected reform costs and increased transfers to regions (see below) would mean that expenditures on existing needs would have to be scaled back.
31. Further, subnational governments would need to undertake significant adjustment to cope with losses from tax reform and increased federal expenditure mandates. Agreed tax reforms will cost local governments about 1 percent of GDP in 2002 (see Chapter IV—Tax Reform), while planned tax reforms (phasing out of the road fund tax and abolition of the sales tax) could cost another l½ percent of GDP in 2003–04. 16 At the same time, the federally mandated increase in government wages will cost about l½ percent of GDP in 2002. Thus, subnational government would need to find measures on the order of 2½ percent of GDP in 2002 and further savings over the medium term to preserve overall balance. While part of the adjustment could come from increased cost recovery in housing and communal services, it is not clear whether net savings can be achieved in the short term given the plan for introducing targeted subsidies (see Chapter VI). There may also be scope for reducing employment at the subnational level, but again the short-term savings are uncertain.
32. The medium-term outlook and the sustainability of fiscal policy are subject to a number or risks. Over the medium term, the key risk is that rapid output growth is not sustained. Sustaining output growth in the range of 5–6 percent over the medium term while the external current account is deteriorating rapidly in conjunction with the real appreciation of the ruble, implies that domestic demand continues to recover strongly. Although some rise in private consumption from its still relatively depressed level can be expected, the driving force should be private investment. This in turn requires comprehensive structural reform and an improvement in the investment climate. In particular, financial sector reform is needed to ensure an efficient intermediation of resources and the provision of adequate liquidity from the banking system. Over the shorter term, the main risk is a major deterioration in the external environment, notably a sharp decline in global energy prices.
APPENDIX I: Estimating Private Sector Savings
1. A number of studies have discussed the variables determining private savings in developing countries and have examined these relationships empirically (see e.g., Dayal-Gulati and Thimann 1997, Masson and others 1995, Savastano 1995, and Ogaki and others 1996).
2. However, these studies have only to a limited extent included transition economies and emerging markets, which may be more relevant for Russia, and are somewhat outdated.17 We have therefore performed an empirical analysis covering the period 1985–99 for two groups of countries: (i) Argentina, Brazil, China, Korea, Malaysia, Mexico, South Africa, Thailand, and Turkey; and (ii) Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia. The sample period for the first group is 1985–98 and for the second 1993–99.
Descriptive Statistics for Group 1
Descriptive Statistics for Group 1
| Private Savings | Government Debt (in percent of GDP) | Money Supply (in percent of GDP) | Inflation | Real GDP Growth | Income Per Capita | Dependency Ratio | |
|---|---|---|---|---|---|---|---|
| Mean | 20.0 | −2.8 | 37.3 | 203.2 | 4.5 | 4614 | 61.1 |
| Median | 20.6 | −2.0 | 30.2 | 13.4 | 5.5 | 4069 | 63.6 |
| Maximum | 32.2 | 4.7 | 98.9 | 3079.8 | 13.3 | 11864 | 82.9 |
| Minimum | 8.4 | −18.6 | 6.2 | 0.2 | −16.2 | 1357 | 40.1 |
| Std. Dev. | 4.3 | 4.7 | 21.7 | 589.8 | 5.4 | 2318 | 9.3 |
Descriptive Statistics for Group 1
| Private Savings | Government Debt (in percent of GDP) | Money Supply (in percent of GDP) | Inflation | Real GDP Growth | Income Per Capita | Dependency Ratio | |
|---|---|---|---|---|---|---|---|
| Mean | 20.0 | −2.8 | 37.3 | 203.2 | 4.5 | 4614 | 61.1 |
| Median | 20.6 | −2.0 | 30.2 | 13.4 | 5.5 | 4069 | 63.6 |
| Maximum | 32.2 | 4.7 | 98.9 | 3079.8 | 13.3 | 11864 | 82.9 |
| Minimum | 8.4 | −18.6 | 6.2 | 0.2 | −16.2 | 1357 | 40.1 |
| Std. Dev. | 4.3 | 4.7 | 21.7 | 589.8 | 5.4 | 2318 | 9.3 |
Descriptive Statistics for Group 2
Descriptive Statistics for Group 2
| Private Savings | Government Debt (in percent of GDP) | Money Supply (in percent of GDP) | Inflation | Real GDP Growth | Income Per Capita | Dependency Ratio | |
|---|---|---|---|---|---|---|---|
| Mean | 18.9 | −1.8 | 36.8 | 22.0 | 2.8 | 4085 | 48.6 |
| Median | 19.8 | −1.4 | 31.8 | 17.6 | 4.0 | 3460 | 49.1 |
| Maximum | 26.9 | 2.6 | 71.3 | 89.8 | 10.6 | 10716 | 54.0 |
| Minimum | 11.6 | −6.7 | 16.6 | 4.7 | −11.9 | 1031 | 42.8 |
| Std. Dev. | 4.0 | 2.1 | 16.4 | 18.1 | 4.5 | 2522 | 2.9 |
Descriptive Statistics for Group 2
| Private Savings | Government Debt (in percent of GDP) | Money Supply (in percent of GDP) | Inflation | Real GDP Growth | Income Per Capita | Dependency Ratio | |
|---|---|---|---|---|---|---|---|
| Mean | 18.9 | −1.8 | 36.8 | 22.0 | 2.8 | 4085 | 48.6 |
| Median | 19.8 | −1.4 | 31.8 | 17.6 | 4.0 | 3460 | 49.1 |
| Maximum | 26.9 | 2.6 | 71.3 | 89.8 | 10.6 | 10716 | 54.0 |
| Minimum | 11.6 | −6.7 | 16.6 | 4.7 | −11.9 | 1031 | 42.8 |
| Std. Dev. | 4.0 | 2.1 | 16.4 | 18.1 | 4.5 | 2522 | 2.9 |
3. The OLS estimation shows a significant effect of the budget deficit, financial deepening, real GDP growth, inflation volatility (measured as the difference between inflation and a two year moving average), and the dependency ratio on private savings (Appendix Table 3). The estimated coefficients, however, are biased since significant correlation pattern in residuals is present. This could be as a result of missing explanatory variables or endogeneity problems inherent in this type of single equation regressions.
4. We use these estimation results to forecast the long run savings ratio for Russia. Applying the “steady-state” values for real GDP growth and fiscal balance discussed in the main text, inflation volatility and financial deepening ratios corresponding to the average for Groups 1 and 2 in recent years (70 percent and 34 percent, respectively), and a dependency ratio of 42 percent (projected average for the period 2010–15), yields a private savings rate of 20 percent of GDP.
OLS Estimation Results
OLS Estimation Results
| Dependent Variable: PRS? | ||||
|---|---|---|---|---|
| Method: Pooled Least Squares | ||||
| Sampte (adjusted): 1986 1998 | ||||
| Included observations: 13 after adjusting endpoints | ||||
| Number of cross-sections used: 17 | ||||
| Total panel (unbalanced) observations: 136 | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. |
| C | 23.75 | 2.57 | 9.24 | 0.00 |
| GVBGDP? | −0.41 | 0.11 | −3.81 | 0.00 |
| GDPDOT? | 0.31 | 0.07 | 4.11 | 0.00 |
| M2GDP? | 0.16 | 0.02 | 9.87 | 0.00 |
| INFLVOL? | 0.00 | 0.00 | 2.29 | 0.02 |
| DEP? | −0.21 | 0.04 | −4.99 | 0.00 |
| R-squared | 0.56 | Mean dependent var | 21.13 | |
| Adjusted R-squared | 0.54 | S.D. dependent var | 6.52 | |
| S.E. of regression | 4.41 | Sum squared resid | 2523.24 | |
| F-statistic | 33.11 | Durbin-Watson stat | 0.65 | |
| Prob (F-statistic) | 0.00 | |||
OLS Estimation Results
| Dependent Variable: PRS? | ||||
|---|---|---|---|---|
| Method: Pooled Least Squares | ||||
| Sampte (adjusted): 1986 1998 | ||||
| Included observations: 13 after adjusting endpoints | ||||
| Number of cross-sections used: 17 | ||||
| Total panel (unbalanced) observations: 136 | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. |
| C | 23.75 | 2.57 | 9.24 | 0.00 |
| GVBGDP? | −0.41 | 0.11 | −3.81 | 0.00 |
| GDPDOT? | 0.31 | 0.07 | 4.11 | 0.00 |
| M2GDP? | 0.16 | 0.02 | 9.87 | 0.00 |
| INFLVOL? | 0.00 | 0.00 | 2.29 | 0.02 |
| DEP? | −0.21 | 0.04 | −4.99 | 0.00 |
| R-squared | 0.56 | Mean dependent var | 21.13 | |
| Adjusted R-squared | 0.54 | S.D. dependent var | 6.52 | |
| S.E. of regression | 4.41 | Sum squared resid | 2523.24 | |
| F-statistic | 33.11 | Durbin-Watson stat | 0.65 | |
| Prob (F-statistic) | 0.00 | |||
| Where | |
| PRS? | − is Private savings for the pool of countries |
| C | − Constant |
| CVBGDP? | − Government deficit as a percent of GDP |
| GDPDOT? | − Real GDP growth |
| M2GDP? | − Broad money supply to GDP ratio |
| INFLVOL? | − Inflation volatility |
| DEP? | − Dependency ratio |
| Where | |
| PRS? | − is Private savings for the pool of countries |
| C | − Constant |
| CVBGDP? | − Government deficit as a percent of GDP |
| GDPDOT? | − Real GDP growth |
| M2GDP? | − Broad money supply to GDP ratio |
| INFLVOL? | − Inflation volatility |
| DEP? | − Dependency ratio |
References
Bernanke, B. S., 2001, “Is Growth Exogenous? Taking Mankiew, Romer, and Weil Seriously,” Paper presented at NBER conference, April 20 –21.
Chalk, N., 1998, “Fiscal Sustainability with Non-Renewable Resources,” IMF Working Paper 98/26 (Washington: International Monetary Fund).
Chinn, M.D., and E.S. Prasad, 2000, “Medium-Term Determinants of Current Accounts in Industrial and Developing Countries: An Empirical Exploration,” IMF Working Paper 00/46 (Washington: International Monetary Fund).
Daseking, C., 2001, “How Large Should Poland’s Current Account Deficit Be?,” IMF Poland—Selected Issues 2001 (February).
Dayal-Gulati, A., and C. Thimann, 1997, “Saving in Southeast Asia and Latin America Compared: Searching for Policy Lessons,” IMF Working Paper 97/110 (Washington: International Monetary Fund).
EBRD, 1997, “Transition Report.”
Engel, E., and RO. Valdes, 2000, “Optimal Fiscal Strategy for Oil Exporting Countries,” IMF Working Paper 00/118 (Washington: International Monetary Fund).
Fischer, S. (2001), The Robbins Lectures delivered at the London School of Economics.
Fischer, S., R. Sahay, and C. Vegh, 1998, “From Transition to Market: Evidence and Growth Prospects,” IMF Working Paper 98/52.
Gomulka, S., 2000, “Macroeconomic Policies and Achievements in Transition Economies, 1989–1999,” Economic Survey of Europe 2000, 2/3, pp. 69 –83, Geneva.
IMF, 2000, “Russian Federation: Selected Issues,” IMF Staff Country Report No. 00/150.
IMF, 2001a, “International Financial Statistics Yearbook.”
IMF, 2001b, “World Economic Oudook,” December 2001.
Laursen, T. (coord.), 2000, “Russia: Medium-Term Macroeconomic Outlook,” paper presented to conference on Investment Climate, Moscow, April.
Levine, R., and D. Renelt, 1992, “Sensitivity Analysis of Cross-country Growth Regressions,” American Economic Review, 82, September, pp. 942 –63.
Liuksila, C., A. Garcia, and S. Bassett, 1994, “Fiscal Policy Sustainability in Oil-Producing Countries,” IMF Working Paper 94/137 (Washington: International Monetary Fund).
Masson, P. R., T.A. Bayoumi, and H. Samiei, 1995, “International Evidence on the Determinants of Private Saving,” IMF Working Paper 95/51 (Washington: International Monetary Fund).
Ogaki, M., J.D. Ostry, and CM. Reinhart, 1996, “Saving Behavior in Low- and Middle-Income Developing Countries,” Staff Papers, International Monetary Fund, Vol. 43 (March).
Savastano, M., 1995, “Private Saving in IMF Arrangements,” in IMF Conditionality: Experience Under Stand-By and Extended Arrangements, Part II: Background Papers, ed. by Susan Schadler and others, IMF Occasional Paper No. 129 (Washington: International Monetary Fund).
United Nations, 2001, “Economic Survey of Europe,” Economic Commission for Europe.
Prepared by Thomas Laursen (EU2).
For a resource-rich economy, fiscal and external sustainability should ideally be examined in the context of intertemporal welfare maximization, which takes into account the depletion of non-renewable resources. Such an approach has been the object of extensive research and attempts have been made to apply it in some countries (see e.g. Liuksila and others 1994 and Chalk 1998). However, it requires considerable information and/or assumptions for which there is substantial uncertainty (e.g., existing reserves of natural resources and future extraction rates, costs, and prices, as well as the social discount factor), and it generally also ignores the substitution of natural resource wealth with other and potentially higher yielding forms of wealth, including the stock of physical and human capital which affect future income growth in the economy or the stock of financial assets. In Russia, future generations are likely to be better off than current generations as transition takes hold and a dynamic market economy is established.
Ceteris paribus, the use of nondebt financing such as privatization proceeds does not change the government’s net worth
The macroeconomic balance approach abstracts from the issue of private sector external assets owing to the difficulties of estimating both their size and yield. This is potentially an important caveat as the cumulative external current account balances during the 1990s suggests that Russia should be in a net creditor position vis-à-vis abroad.
Oil prices in this chapter refer to a simple average of spot prices of U.K. Brent, Dubai, and West Texas Intermediate.
The baseline specification, selected by both the Akaike and Schwartz Information Criteria, is an ARMA (5,3) regression:Yt = 18.3 − .26 Yt−1 + .03 Yt−2 + .43 Yt−3 + .21 Yt−4 − .36 Yt−5 + 1.8 MAt + 1.77 MAt−1 + .95 MAt−2 where Y denotes the world oil price.
This is a useful rough indicator of the underlying or structural fiscal (or external) balance and has been applied by Fund staff in assessing changes in the fiscal stance (see Staff Report for the 2001 Article IV Consultation). While one should, in principle, also adjust for deviations in output from its potential level and for deviations (or unrealized effects of past changes) in the real exchange rate from its equilibrium level, such adjustments pose well-known problems in a case like Russia. It is also assumed that neither imports nor public expenditure are affected by changes in energy prices.
On average, estimates suggest that annual TFP growth in advanced transition economies during the period 1991–97 amounted to about 2 percent, but with large variations (Poland had the highest growth rates of close to 4 percent per year).
The baseline specification for the Levine & Renelt equation is as follows:
gt = −0.83 − 0.0035* Y0 −0.38* POP + 3.17* SS0 + 17.5* INV,
where gt the growth rate of per capita income, is expressed in percentage points; and Y0 the initial level of real per capita income, is divided by 1000. This equation was estimated over the period 1960–89 using data for 119 countries.
The strengths of this equation are that it tries to relate growth to its underlying, exogenous determinants, that it explains a surprisingly large fraction of the sample variance in growth rates (R2= 0.46), and that its specification is relatively robust The main weakness for our purposes is that the estimation sample does not include any transition economies or resource exporters. Hence, the relevance of this (or any other) cross-country growth regression to the long-term growth prospects of transition economies in general, and Russia in particular, rests on the assumption that in the long run such economies will not fundamentally differ from other developing economies.
This elasticity in Russia (and many other transition economies) is currently closer to one-half, significantly lower than in advanced economies and many other developing countries. There are two main explanations: (i) the labor share may increase as an economy develops; and (ii) developing countries in general, and transition economies in particular, have a large unincorporated business sector whose operating surplus should be included in the labor share. Also, part of the remuneration of labor is not presently recorded as wages, e.g., subsidized rents and utilities. Adjusting for this, the true labor share may be quite constant around the world at about 0.65 (see Bernanke 2001).
Private savings and its determinants in transition economies are discussed in UN (2001)—see Appendix I for further details. The analysis emphasizes the importance of the level of per capita income and the depth and strength of the financial system. The latter underlines the crucial importance of financial reform and the creation of sound institutions for encouraging higher levels of saving and investment.
Over the medium term, and even beyond, interest rates on aggregate public debt will be lower reflecting the large stock of outstanding, restructured commercial debt as well as the lower rates on bilateral debt.
With output growth of 5½ percent and a real interest rate of 6½ percent, the primary surplus consistent with maintaining a stable debt-to-GDP ratio of 30 percent would be ½ percent of GDP (the analysis here abstracts from subnational government debt, which is relatively small).
The Maastricht criteria for membership of the European Economic and Monetary Union establishes a public debt limit of 60 percent of GDP, but such a threshold is clearly too high for developing countries. Fischer (2001) notes that since the variability of the real interest rate in emerging market countries is significantly higher (up to a factor of 5) than in industrialized countries, emerging market countries are likely to face a very large fiscal shock if their debt to GDP ratio is around 60 percent. Thus, ratios nearer 30 percent are much safer.
The budget plans for 2002 envisage a decline in the enlarged government primary balance at constant oil prices to less than 3 percent of GDP (and to 2 percent of GDP at actual oil prices based on the staffs output and revenue projections).
These costs may be smaller to the extent that tax compliance improves.
An exception is the UN Economic Survey of Europe 2001, which includes a chapter on domestic savings in transition economies. The study identifies the following relevant variables: current account balance, government savings, social security expenditure, the level and growth of per capita income, the level of monetization, the real interest rate, CPI inflation, change in the terms of trade, and the age dependency ratio. The estimated equation for Eastern Europe and the Baltic States (1995–98) is:
Sp = 0.09*CA − 0.6*Sg + 0.3*SS+0.7*y + 0.09*dy + 0.1*M−0.2*r−0.01 *dCPI+0.16*dTOT + 0.13*DEP (R-square=Q. 79),
where Sp is private saving, CA is the current account balance, Sg is government savings, SS is social security expenditure, y is per capita income, M is broad money, r is the real (ex-post) interest rate, CPI is the consumer price index, TOT is the terms of trade, DEP is the dependency ratio, and d denotes rate of change. The fitted value for Russia using our medium-term/steady-state projections of the explanatory variables is 19 percent of GDP.
