- International Monetary Fund. Fiscal Affairs Dept.
- Published Date:
- May 2010
Change in the cyclical balance over time.CDS spreads
The spread on Credit Default Swap (CDS) refers to the annual amount (in bps of the notional amount) that the protection buyer must pay the seller over the length of the contract to protect the underlying asset against a credit event.Cyclical balance
Cyclical component of the overall fiscal balance. Typically computed as the difference between cyclical revenues and cyclical expenditure. The latter are typically computed using country specific elasticities of aggregate revenue and expenditure series with respect to the output gap. Where unavailable, standard elasticities (0,1) are assumed for expenditures and revenues, respectively.Cyclically adjusted (CA) expenditure and revenue
Revenues and expenditure adjusted for the effect of the economic cycle (i.e., net of cyclical revenues and expenditure).CA primary balance (CAPB)
Primary balance adjusted for the effects of the economic cycle, usually expressed in percent of potential GDP. This is typically computed as the difference between CA primary revenue and CA primary expenditure.Fiscal stimulus
Discretionary fiscal policy actions adopted in response to the financial crisis that affect the overall fiscal balance.General government
The general government sector consists of all government units and all nonmarket nonprofit institutions that are controlled and mainly financed by government units, comprising the central, state, and local governments. The general government sector does not include public corporations or quasi-corporations.Gross debt
All liabilities that require future payment of interest and/or principal by the debtor to the creditor. This includes debt liabilities in the form of SDRs, currency and deposits, debt securities, loans, insurance, pensions and standardized guarantee schemes, and other accounts payable. The term “public debt” is used in this Monitor, for simplicity, as synonymous of gross debt of the general government, unless otherwise specified (strictly speaking, the term public debt refers to the debt of the public sector as a whole, which includes financial and nonfinancial public enterprises and the central bank).Gross financing needs
Overall new borrowing requirement plus debt maturing during the year.Net debt
Gross debt minus financial assets, including those held by within the broader public sector, for example in some cases, social security funds. These financial assets are monetary gold and SDRs, currency and deposits, debt securities, loans, shares, equities, insurance, pension, standardized guarantee schemes, and other accounts receivable.Output gap
Deviation of actual from potential GDP, in percent of potential GDP.Overall fiscal balance (also “headline” fiscal balance)
Net lending/borrowing, defined as the difference between revenue and total expenditure (using the IMF’s GFSM 2001). Does not include policy lending. During this transitional period to GFSM 2001, not all countries have adopted the new presentation; for some, the overall balance continues to be based on GFSM 1986, defined as total revenue and grants minus total expenditure and net lending.Policy lending
Transactions in financial assets that are deemed to be for public policy purposes but are not part of the overall balance.Primary balance
Overall balance minus interest revenue plus interest expenditure.Public debt
See gross debt.Public sector
The public sector consists of the general government sector plus government controlled entities, known as public corporations, whose primary activity is to engage in commercial activities.Relative asset swap (RAS) spreads
RAS spreads measure the difference between benchmark government bond yields and the interest rate on the fixed-rate arm of an interest rate swap in the same currency and of the same maturity (usually 10 years) as the bond.
Since the November Monitor, a handful of G-20 economies have adjusted their stimulus, adding new measures to address worsening unemployment, as well as to further support key sectors and private consumption. With new measures in Germany, Japan, Russia, and the United States, stimulus will be higher (by 0.3 percent) than previously expected in 2010 (Table 1). In some countries, stimulus has been reoriented, for example from investment to transfers (Japan). Elsewhere, changes reflect a front-loading of 2009–10 stimulus (France), or an earlier withdrawal of part of the crisis related support due to a stronger-than-expected recovery and the objective of containing the rise in the public debt ratio (Korea). On balance, implementation in 2009 was broadly in line with expectations (2 percent of GDP for the G-20 economies as a whole).
|Previously-Reported Crisis-Related Stimulus 1/||Update||Comment|
|Australia||2.9||2.0||2.8||1.8||Estimated stimulus implementation in 2009-10 is slightly lower, due largely to higher nominal GDP.|
|Brazil||0.6||0.6||0.7||0.6||2009 estimates higher, due to greater revenue impacts. 2010 is higher in nominal terms, but the same in percent of (higher) GDP. Public lending via the state-owned bank BNDES were also stepped up in 2009 and is expected to continue at high levels in 2010.|
|Canada||1.9||1.7||1.8||1.7||Implementation in line with expectations and higher GDP.|
|China||3.1||2.7||3.1||2.7||No update; from overall data, stimulus appears to have been implemented, particularly at the central government level.|
|France||0.7||0.8||1.0||0.5||Some stimulus brought forward to 2009.|
|Germany||1.6||2.0||1.5||2.1||2009 estimates lower, partly due to higher GDP and lags in capital spending implementation. 2010 estimates reflect new business tax rebates and higher family benefits. Additional discretionary measures in 2010 are, however, much smaller than the estimated deterioration of the CAPB (2.8 percentage points). This reflects mainly a drop in revenues, which exceeds the estimates obtained by the standard approach (using fixed revenue elasticities) to calculate the CAPB.|
|India||0.6||0.6||0.6||0.4||2009 stimulus implemented in full. Revised estimate for 2010 reflects increased excise tax rates in the 2010/11 budget.|
|Indonesia||1.4||0.6||1.1||0.6||Stimulus allocated in 2009 was lower, 1.3 percent of GDP, due to lower fuel subsidies; outturn reflected less-than-full implementation of tax incentives.|
|Italy||0.2||0.1||0.0||0.1||Updated estimate for 2009 stimulus is near zero on a net basis. New measures introduced for 2010 in a deficit-neutral manner, with spending increase of 0.2 percent of GDP expected to be fully offset by expenditure savings and collecting the postponed income tax installment from 2009.|
|Japan||2.4||1.8||2.8||2.2||Update reflects new measures and a shift from investment to transfers and expanded energy efficient product incentives.|
|Korea||3.6||4.7||3.6||1.1||No update for 2009. With faster-than-expected recovery, stimulus is being withdrawn in the 2010 budget, which implies a withdrawal relative to 2009 of 2.5 percent of GDP and more than 3½ percent of GDP relative to earlier plans.|
|Mexico||1.5||1.0||1.5||1.0||Outturns suggest that the 2009 stimulus (freezing some prices, spending increases, development bank lending) was largely implemented. No new stimulus plans for 2010. Social spending, subsidies, and investment will decline in 2010; sizeable tax package introduced (1 percent of GDP).|
|Russia||4.1||1.3||4.5||2.8||2009 stimulus fully implemented, with more support to strategic sectors. Higher 2010 stimulus reflects new social and labor market measures and further support to strategic sectors.|
|Saudi Arabia||3.3||3.5||3.3||3.5||2009 stimulus implementation in line with expectations. Sizeable stimulus will continue in 2010 as expected.|
|South Africa||3.0||2.1||3.0||2.1||2009 and plans for 2010 in line with expectations.|
|Turkey||1.2||0.5||1.2||0.5||2009 implemented as planned, with offsetting differences in impacts of specific measures. 2010 estimates are unchanged.|
|United Kingdom||1.6||0.0||1.6||0.2||2009 implemented in full. Revised 2010 reflects new transfers and income tax allowances, partly offset by efficiency savings.|
|United States||2.0||1.8||1.8||2.9||Additional corporate tax breaks, extended unemployment benefits, and homebuyer tax credits of 0.3 percent of GDP adopted in November 2009. 2010 also includes March 2010 jobs bill (0.1 percent of GDP), as well as additional possible sizeable mitigation measures to support the unemployed, families with children, and disadvantaged groups under consideration in the Congress.|
|G-20 Average 2/||2.0||1.6||2.0||1.9|
The policy goal driving the illustrative fiscal adjustment scenarios is to reduce public debt to prudent levels over the next two decades. The average adjustment needed for the scenario targeting a gross debt-to-GDP ratio of 60 percent for advanced economies is estimated at around 8¾ percent of GDP (Table 1). For emerging economies, using a gross debt target of 40 percent, the average adjustment need is considerably smaller, at only 2¾ percent of GDP (Table 2). These calculations are based on the following debt dynamics equation:
where dt denotes the general government gross debt-to-GDP ratio; pbt is the primary balance in percent of GDP (defined as the overall balance minus interest revenue plus interest expenditure); at denotes the asset-to-GDP ratio; rD and rA represent the nominal interest rates on gross debt and assets, respectively; and g denotes the nominal GDP growth rate. In using this equation to compute the fiscal adjustment need, it is assumed that asset accumulation is equal to interest revenue. With this assumption, the above equation simplifies to
For the asset ratio, this implies that . The evolution of the asset-to-GDP ratio thus depends on the interest rate-growth differential for assets (rA − g): if it is zero, the asset-to-GDP ratio remains constant over time, but it increases (falls) for a positive (negative) differential.
This framework could also be applied to a net debt instead of a gross debt target, assuming rA = rd and letting d equal net debt. The choice between targeting gross or net debt depends in part on the types of risk one wishes to monitor. In addition, there are data availability considerations that guide the choice of net or gross debt targets (see Box 2 in the main text).
Rollover risk: gross debt tends to be the natural choice to assess this type of risk, as it is the gross debt stock that countries need to roll over. Assets can matter as well if they are sufficiently liquid, as they can be sold and the proceeds used to retire maturing debt. Given that reliable cross-country data on asset composition is scarce, however, basing the analysis on gross debt is often preferable.
Impact on interest rates and growth: whether net debt or gross debt is more relevant for assessing the impact of debt on interest rates and economic growth is mostly an empirical question, as good theoretical arguments could be made to support the choice of either indicator. The analysis in this issue of the Monitor uses gross debt to investigate these impacts, in part because data scarcity makes a similar analysis for net debt difficult.
Cross-country comparability: accounting and reporting methodologies for net debt lack an internationally-agreed standard, which limits cross-country comparability. This is much less of a problem for gross debt (with the notable exception of Japan, where general government gross debt is reported on an unconsolidated basis).
In practice, the choice of a net or gross debt target has limited impact on the amount of adjustment called for under the illustrative scenario considered in the Monitor. Conducting the analysis using a net debt target of 45 percent of GDP (equal to the precrisis median for advanced G-20 economies) results in an illustrative adjustment need that is similar to that which emerged from the gross debt exercise conducted in Section III). Only three countries—Canada, Iceland, and Ireland—have a cumulative adjustment need that differs by more than one percentage point between the net and gross debt exercises. (In each case, the adjustment need associated with the net debt target is smaller than with the gross debt target.) As Canada’s postcrisis net debt is projected to be below the target level, its adjustment need in the net debt exercise is limited to what is needed to eliminate its primary deficit. Iceland and Ireland would benefit from their large asset position, which leave them exiting the crisis with net debt levels that are closer than are their gross debt levels to the respective targets for these variables. Other countries with sizable assets tend to be low-debt countries. The adjustment need under the net debt target scenario is somewhat larger than with the gross debt target for a number of high-debt advanced economies with limited assets, e.g., Germany, Italy, and the United Kingdom.
|Current WEO Projections, 2010||Illustrative Fiscal Adjustment Strategy to Achieve Debt Target in 2030|
|Gross Debt||Primary Balance||Cyclically Adjusted Primary Balance||Cyclically Adjusted Primary Balance in 2020–30||Required Adjustment between 2010 and 2020|
|Hong Kong SAR||0.6||−1.4||−3.5||0.3||3.8|
|Current WED Projections, 2010||Illustrative Fiscal Adjustment Strategy to Achieve Debt Target in 2030|
|Gross Debt||Primary Balance||Cyclically Adjusted Primary Balance||Cyclically Adjusted Primary Balance in 2020–30||Required Adjustment between 2010 and 2020|
|2007||2010||Required Adjustment Between 2010 and 2020||Difference|
|Net Debt||Gross Debt||Net Debt||Gross Debt||Cyclically Adjusted Primary Balance||/W Net debt||/w Gross debt||2010 Assets-to-GDP ratio (Gross minus net debt)||Req adj (Gross debt) minus Req adj (Net debt)|
|Hong Kong SAR||…||…||…||…||…||…||…||…||…|
This appendix presents empirical analysis regarding the potential impact of government debt on investment and growth.28 There is a large literature on the potential adverse effects of high government debt via higher long-term interest rates, and expectations of higher future distortionary taxation (Elmendorf and Mankiw, 1999). Also, high debt may limit space for countercyclical fiscal policies, which can result in higher volatility and lower growth, and increase vulnerability to crises.29 To date, there are few studies that assess the magnitude and significance of potential adverse effects of high public debt (Box 1).
The empirical analysis examines the relationship between initial government debt and subsequent economic growth in a panel of advanced and emerging economies for the period 1970–2007.30 Building on the empirical growth literature (Aghion and Durlauf, 2005), a standard set of explanatory variables is used to explore the impact of initial government debt on growth. The growth regressions are complemented by a growth accounting exercise which allows an exploration of the channels (factor accumulation and factor productivity) through which government debt may influence growth. Nonlinearities and threshold effects—whether there is a certain level beyond which debt begins to have an adverse effect on growth—are also examined. The analysis pays particular attention to a variety of estimation issues that can have an important bearing on the estimation. In particular, by using the initial level of debt, it avoids the “reverse causality” between debt and growth—low growth may raise debt, rather than high debt lowering growth. In addition, the possible endogeneity problem—that is, debt and growth might be jointly determined by a third variable—is taken into account in the analysis.31 Various robustness checks are also conducted.
Box 1.Government Debt and Growth: Existing Empirical Studies
There is little systematic analysis of the impact on GDP growth of high public debt in advanced economies. A notable exception is Reinhart and Rogoff (2010), who find that the median GDP growth rate differential between low debt (below 30 percent of GDP) and high debt (above 90 percent of GDP) countries is 2.6 percent, based on a comparison of annual growth rates of GDP for the four debt level categories over the period 1946 to 2009.
However, a number of studies have looked at the impact of external debt on economic growth in developing economies. Most of the studies are motivated by the “debt overhang” hypothesis—a situation where a country’s debt service burden is so heavy that a large portion of output accrues to foreign lenders and consequently creates disincentives to invest (Krugman, 1988). Imbs and Ranciere (2009) and Pattillo, Poirson, and Ricci (2002, 2004) find a nonlinear effect of external debt on growth, that is, a negative and significant impact on growth of high debt levels (typically, over 60 percent of GDP), but an insignificant impact at low debt levels. Similarly, Reinhart and Rogoff (2010) report that when external debt reaches 60 percent of GDP or above, annual GDP growth declines by about 2 percent or more in emerging economies. Cordella, Ricci, and Arranz (2005) find evidence of debt overhang for intermediate debt levels, but insignificant debt-growth relationship at very low and very high levels of debt.
Some stylized facts
There is a negative relationship between initial government debt and subsequent per capita GDP growth. The fitted line (OLS) of a scatter plot of initial debt against subsequent growth over five-year periods shows a coefficient of initial debt of -0.025 (Figure 1). Taken at face value, this suggests that a 10 percentage point of GDP increase in initial debt is associated with a slowdown in per capita GDP growth of 0.25 percentage points. This magnitude is consistent with that obtained using econometric estimation (see below). Second, the average growth rate during periods of rising debt is lower than that during the periods of falling debt (Figure 2). Third, the adverse impact of debt on growth appears to be larger in emerging economies.
Figure 1.Government Debt and Per Capita GDP Growth
Source: IMF staff estimates.
Figure 2.Growth of Real per Capita GDP during Periods of Rising and Falling Debt
Source: IMF staff estimates.
Econometric analysis 32
Higher initial government debt is generally found to be associated with lower subsequent growth. Econometric results suggest that, on average, a 10 percent of GDP increase in initial debt is associated with a decline in real per capita GDP growth of around 0.2 percent per year (Box 2 provides an illustrative example based on a simple production function framework). Table 1 presents the main results. Columns 1 and 2 show that the coefficients of initial debt are all negative and significant at the 5 percent level in the regressions of per capita GDP growth, ranging from -0.018 to -0.02. The negative impact of initial debt on growth in advanced economies tends to be smaller than that in emerging economies. This may reflect limited borrowing capacity of emerging economies due to less developed domestic financial markets or uncertain access to international capital markets. There is some evidence of nonlinear effects: medium to high levels of debt have significant negative effects on subsequent growth, whereas low levels of debt have insignificant effects on growth.33
The adverse effects of high debt on growth appear to occur through a variety of channels. From a growth accounting perspective, the adverse effects on growth largely reflect a slowdown in labor productivity growth mainly due to slower growth of the capital stock per worker.34 (Figure 3 shows a scatter plot of initial debt against subsequent growth of capital per worker.) There is also a mild negative relationship between higher initial debt and total factor productivity (TFP) growth. The estimated coefficients of initial debt suggest a 10 percent of GDP increase in initial debt is associated with a decline in growth of output per worker of about 0.2 percent per year (Column 3); a decline in TFP growth of around 0.1 percent per year (Column 4); and a decline in growth of capital per worker of around 0.4 percent per year.35 The result also suggests that high debt is associated with lower investment: a 10 percent of GDP increase in initial debt is associated with a decline in investment of about 0.4 percentage points of GDP, with a larger impact (0.9) in emerging economies (Column 5). To some extent, higher initial debt is also associated with greater macroeconomic volatility (Figure 4).36
Figure 3.Government Debt and Capital per Worker Growth
Source: IMF staff estimates.
Figure 4.Government Debt and Macroeconomic Volatility
Source: IMF staff estimates.
Box 2.Debt and Growth: An Analytical Perspective
The above empirical results can be supplemented by an analytical assessment based on a simple Cobb-Douglas production framework. For illustrative purposes, the starting point is the premise, which has support in the literature, that in the United States each dollar of debt crowds out one dollar of capital in the long run (Elmendorf and Mankiw, 1999). If factors of production earn their marginal product, then the marginal product of capital equals the capital share of income divided by the capital-output ratio. Historically, the capital income share has been about one third, and the capital-output ratio in the United States is around 3.7 in 2010. The implied marginal product of capital is about 9 percent.1 An increase in the ratio of net debt to GDP of 40 percent over the next five years amounts to an increase in net debt of around US$6,450 billion (in real terms based on the projection of 2 percent average growth of real GDP). Other things equal, a full crowding out, that is, capital stock declining by the same amount as the increase in net debt, implies that output would decline by about 4.4 percent in total. (This is obtained as a product of the marginal productivity of capital and the decrease in capital stock, as a percent of initial GDP). This is approximately equivalent to growth slowdown of around 0.8 percent; or 0.2 percent per year on average for a 10 percent of GDP increase in government debt (assuming that the output decline mostly occurs in the following five-year period).
The above result will depend on a number of other factors. If private savings rise in response to an increase in public debt, or there are capital inflows from abroad, crowding out would be reduced as would be the effect of debt on growth. Thus, the full crowding out assumption may exaggerate the magnitude of effects of debt on growth. However, there may be externalities that are not captured in standard economic models. The endogenous growth models suggest that the accumulation of capital stimulates technological change and increase economy-wide productivity (Romer, 1987). The opposite will occur in the case of (partial) crowding out.
Elmendorf and Mankiw (1999) estimate the marginal product of capital to be 9.5 percent in the United States at the time of writing.
|Dependent variable: Growth of real per capita GDP||Growth of output per worker||Growth of TFP||Domestic Investment|
|Lagged dependent variable||−2.810**||−4.459*||0.828***|
|Initial per capita real GDP||−2.187***||−2.823***|
|Initial years of schooling||2.863***||4.161**||3.493||0.29||6.985**|
|Initial inflation rate||−2.234***||−2.296||−6.987*||−6.839*||−3.924**|
|Initial government size||0.087**||0.168||0.080||0.077||0.332|
|Initial trade openness||−0.001||−0.004||−0.009||0.002||−0.044**|
|Initial financial depth||0.019***||0.026***||0.025**||0.018*||0.029|
|Terms of trade growth||−0.019||−0.025||−0.063||−0.053**||0.173*|
|Government debt, initial||−0.018**||−0.020**||−0.022**||−0.011||−0.038*|
|Arellano-Bond AR(2) test p-value||0.12||0.16||0.94||0.99|
|No. of observations||166||166||159||159||159|
Subnational government (SNG) spending represents a large share of total government spending in some countries, and fiscal policies at the local level can have important macroeconomic implications. This appendix provides a brief summary of how the crisis affected SNG budgets, how countries responded to revenue losses and spending pressures at the local level, and the implications of these developments for the future.
The crisis had a significant adverse effect on SNG finances. For the OECD countries, SNG revenues fell on average by 3.5 percent in 2009, although in some countries, where reliance on more cyclically sensitive income taxes is greater, the decline was almost 10 percent (Johnson, Collins, and Singham, 2010). In addition, SNG budgets were hit by increased spending, particularly in cases where SNGs were tasked by central governments with providing part of the crisis-related stimulus response without a fully-matching increase in funding.
Country responses to crisis-related fiscal pressures at the SNG level have varied (Table 1). In a number of countries (Canada, Germany, Japan, and Spain), SNGs have pursued countercyclical discretionary easing. In Germany, state and local government balances moved from a surplus of 0.3 percent of GDP in 2008 to a projected deficit of 2 percent of GDP in 2010 and 2011. In other countries such as France, the United Kingdom, and the United States, SNGs responded by procyclical fiscal tightening. The different policy response mainly reflects legal constraints on local deficits and borrowing.
|Procyclical response||Passive policy||Countercyclical response 1/|
To avoid undermining fiscal stimulus, some central governments sought to mitigate the impact of the crisis on SNGs, increasing transfers, particularly grants. In Canada, Japan, and Spain, the increase in grants represented over one-half of national stimulus spending (Blöchliger and others, 2010).37 Some countries eased SNG financing constraints by providing loans (Canada, Switzerland), or guarantees (Australia, Korea, Spain), or temporarily easing balanced budget rules (Spain). Finally, other countries (Finland) have temporarily increased the share of tax revenue going to SNGs.
In countries where borrowing rules were not eased and where central governments covered only part of the budget gap, SNGs had to resort to tightening. For example, in the United States, all states except Vermont are legally required to balance their budgets. The crisis has hit their budgets hard: the total state budget shortfall in FY 2010 is projected to peak at US$196 billion or 1.4 percent of GDP (Table 2).38 State governments were forced to take substantive measures to balance their books, drawing down accumulated reserves from a high of US$66 billion in 2007 to US$32 billion in 2009;39 cutting spending by 4 percent in FY 2009 and 4.8 percent in FY 2010; and (for two-thirds of the states) eliminating tax exemptions, broadening tax bases, or increasing rates. These measures boosted tax revenues by US$32 billion.
Comparing SNG budgetary projections and outcomes in the United States and Germany illustrates the effect of enforcing borrowing constraints. Table 2 shows projected SNG budget balances in Germany and projected total U.S. state budget shortfalls (a proxy for state deficits in the absence of borrowing constraint). The U.S. budget shortfalls should be brought to zero. During 2009–13, balances would need to be cumulatively tighter by about US$600 billion (4 percent of annual GDP, or almost 80 percent of the 2009 stimulus package).
|Germany SLG balance||+0.3||−1.0||−2.0||−2.0||−1.5||−1.5|
|U.S. state budget shortfall 1/||…||−0.8||−1.4||−1.2||−0.75||…|
The state of SNGs finances will need to be considered when designing exit policies. When the crisis erupted and private demand collapsed, strictly enforcing borrowing constraints would have risked weakening the beneficial effects of fiscal stimulus. But as consolidation becomes the priority, SNGs, too, will need to start strengthening their fiscal positions. The consolidation at the local level could be even more difficult than at the central level, as exit from central government stimulus will entail, in part, phasing out discretionary transfers to local governments. Moreover, in some cases, the recovery in tax revenue may be slow.40
Subsidies on petroleum products (gasoline, kerosene, etc.) are again on the rise with the rebound in international oil prices. These consumer subsidies are inefficient, inequitable, and environmentally unfriendly. Eliminating them can make an important contribution to reducing the large fiscal deficits that have accumulated in many countries during the recent financial crisis.
Measuring petroleum product subsidies
A recent paper by Coady and others (2010) estimates the global magnitude of fuel subsidies by comparing retail prices with “optimal” benchmark prices. Reflecting the fact that the optimal level of taxation can vary across countries for many legitimate reasons, the paper estimates subsidies based on pretax benchmark prices as well as on tax-inclusive subsidies for optimal tax rates of $0.30 and $0.40 per liter of gasoline, kerosene, and diesel.
The magnitude of subsidies
Although subsidies fell sharply in the second half of 2008 due to the steep decline in international prices, they have again started to increase with the rebound in international oil prices. Pretax subsidies are projected to reach almost $250 billion, or 0.3 percent of global GDP by end-2010. Projected tax-inclusive subsidies are substantially higher at between 1.0 and 1.3 percent of GDP, depending on the benchmark optimal tax. Whereas advanced economies have zero pretax subsidies, they account for about a quarter of tax-inclusive subsidies.
This increase in subsidies is taking place in the context of a more challenging fiscal backdrop. Of the 83 countries with tax-inclusive subsidies in 2010 (based on a benchmark tax of US$0.40 per liter), 69 have projected fiscal deficits for 2010: in 43, the deficit will exceed 3 percent of GDP; and in 22, it will exceed 5 percent of GDP (Figure 1). Reducing tax-inclusive subsidies by one-half in these countries would result in their average deficit falling from 4.2 percent of GDP to about 2.5 percent of GDP.
Figure 1.Projected 2010 Fiscal Deficit and Fuel Subsidies
Source: Coady and others (2010).
A number of measures can be implemented to support the reform of petroleum subsidies. Replacing subsidies with improved social safety nets can reduce the fiscal cost of protecting the poor from price increases. Transparently recording subsidies in government accounts ensures that they more explicitly compete with alternative uses of public funds. Although a liberalized approach to petroleum pricing is best, countries can adopt an automatic pricing mechanism in the interim while they develop a competitive supply system and an effective regulation capacity.
Methodological and Statistical Annex
This annex comprises four sections: (i) assumptions; (ii) data and conventions; (iii) economy groupings; and (iv) statistical tables. The assumptions underlying the estimates and projections for 2010–15 are summarized in the first section. The second section provides a general description of the data and of the conventions used for calculating country group composites. The classification of countries in the various groups presented in the Fiscal Monitor is summarized in the third section. The last section comprises the statistical tables on key fiscal variables. Data in these tables have been compiled on the basis of information available through mid-April 2010.
I. Fiscal Policy Assumptions
The historical data and projections of key fiscal aggregates are in line with those of the April 2010 WEO (IMF, 2010f), unless highlighted. For the underlying assumptions, other than on fiscal policy, see April 2010 WEO.
The short-term fiscal policy assumptions used in the Fiscal Monitor and WEO are based on officially announced budgets, adjusted for differences between the national authorities and the IMF staff regarding macroeconomic assumptions and projected fiscal outturns. The medium-term fiscal projections incorporate policy measures that are judged likely to be implemented by IMF staff. For countries supported by an IMF program, the medium-term projections are those under the program. In cases where the IMF staff has insufficient information to assess the authorities’ budget intentions and prospects for policy implementation, an unchanged structural primary balance is assumed, unless indicated otherwise. Below are the specific assumptions relating to selected economies.
Argentina. The 2010 forecasts are based on the 2009 outturn and IMF staff projections. For the medium term, the IMF staff assumes unchanged policies.
Australia. The fiscal projections are based on the Mid-Year Economic and Fiscal Outlook (2009–10) and IMF staff projections. Due to the proximity of the release of the Australian government’s budget (May 11) and the issuance of the Monitor, the Monitor does not reflect the budget figures.
Brazil. The 2010 forecasts are based on the budget law and IMF staff projections. For the medium term, the IMF staff assumes unchanged policies, with an increase in public investment in line with the authorities’ intentions.
Canada. Projections use the baseline forecasts in the latest Budget 2010—Leading the Way on Jobs and Growth. The IMF staff makes some adjustments to this forecast for differences in macroeconomic projections. The IMF staff forecast also incorporates the most recent data releases from Statistics Canada, including provincial and territorial budgetary outturns through the end of 2009.
China. For 2010–11, the government is assumed to continue and complete the stimulus program it announced in late 2008. In view of available information, IMF staff assumes the stimulus continues through 2010 and is withdrawn in 2011.
France. Projections for 2010 are based on the 2010 budget and the latest Stability Program, and are adjusted for differences in macroeconomic projections. Projections for the medium-term years incorporate the IMF staff’s assessment of current policies and implementation of announced adjustment measures.
Germany. Projections for 2010 are based on the 2010 budget, adjusted for differences in the IMF staff’s macroeconomic framework. The IMF staff’s projections for the medium-term outlook incorporate the withdrawal of fiscal stimulus, planned income tax cuts envisaged for 2011, and IMF staff’s assessment of feasible adjustment policies already announced.
India. Projections are based on available information on the authorities’ fiscal plans, with some adjustments for the IMF staff’s projections. Projections are based on the budget as well as the semiannual budget review. IMF presentation differs from Indian national accounts data, particularly regarding subsidies and certain loans.
Indonesia. The 2010 projections are based on the revised budget draft but adjusted by the IMF staff to reflect the fact that due to built-in cushions and a track record of under-execution, the deficit could be slightly smaller. Medium-term projections are based on the authorities’ exit strategy (broad-based revenue reforms to support gradual fiscal consolidation), combined with IMF staff’s projections.
Italy. The fiscal projections incorporate the impact of the 2010 Budget Law, and the authorities’ latest revisions to the unchanged legislation scenario, which was presented in the January 2010 “Nota di aggiornamento 2010–2012.” In the absence of specific measures and details underlying their policy scenario, the authorities’ estimates for an unchanged legislation scenario are used as a basis for the projections, adjusted mainly for differences in the macroeconomic assumptions. From 2013 onward, a constant structural primary balance (net of one-time items) is assumed.
Japan. The 2010 projections assume that fiscal stimulus will be implemented as announced by the government. The medium-term projections assume that expenditure and revenue of the general government are adjusted in line with current underlying trends (excluding fiscal stimulus).
Korea. The fiscal projections are based on the 2010 budget. Expenditure numbers for 2010 correspond to the expenditure numbers presented in the government’s budget. Revenue projections reflect the IMF staff’s macroeconomic assumptions, adjusted for the estimated costs of tax measures included in the multiyear stimulus package introduced last year and discretionary revenue-raising measures included in the 2010 budget. The medium-term projections assume that the government will resume its consolidation plans and balance the budget (excluding social security funds) in 2014.
Mexico. Fiscal projections are based on: (i) the IMF staff’s macroeconomic projections; (ii) the modified balanced budget rule under the Fiscal Responsibility Legislation; and (iii) the authorities’ projections of spending on pensions and health care and of wage-bill restraint. For 2010–11, projections take into account the departure from the balanced budget target under the exceptional clause of the fiscal framework, which allows for a small deficit reflecting cyclical deterioration in revenues.
Portugal. Projections do not reflect additional deficit reduction plans announced May 10.
Russia. Projections for 2010 are based on the nominal expenditures in the 2010 Budget and the IMF staff’s revenue projections. Projections for 2011–12 are based on the non-oil deficit in percent of GDP implied by the medium-term budget and on the IMF staff’s revenue projections. The IMF staff assumes an unchanged non-oil federal government balance in percent of GDP during 2012–15.
Saudi Arabia. Projections are based on the 2010 budget but modified to reflect IMF staff projections of oil-related revenue and some expenditure adjustments. The pace of spending is projected to slow over the medium term, leading to a tightening of the fiscal stance.
South Africa. Fiscal projections are based on the authorities’ 2010 budget and policy intentions stated in the budget review published on February 17, 2010, and are adjusted for differences in the IMF staff’s macroeconomic framework. The IMF staff’s projections for the medium-term outlook incorporate the gradual completion of the government’s infrastructure investment plans.
Spain. Projections do not reflect additional deficit reduction plans announced May 10.
Turkey. Fiscal projections assume that the authorities maintain the medium-term program target for 2010 and implement the fiscal rule from 2011 onward.
United Kingdom. The fiscal projections for 2010 and onward are based on Budget 2010, adjusted for differences in IMF staff projections of macroeconomic and financial variables.
United States. The fiscal projections are based on the administration’s draft budget for fiscal year 2011 and the U.S. Congressional Budget Office’s outlook for 2010–19. The projections include the American Recovery and Reinvestment Act and other recent support measures. The projections are adjusted for differences in forecasts of macroeconomic and financial variables, and costs to support financial institutions and government-sponsored enterprises.
II. Data and Conventions
Data and projections for key fiscal variables are based on the April 2010 WEO, unless indicated otherwise. Where the Fiscal Monitor includes additional fiscal data and projections not covered by the WEO, data sources are listed in the respective tables and figures. All fiscal data refer to the general government where available, and to calendar years, with the exceptions of Pakistan and Singapore where data refer to the fiscal year.
Composite data for country groups are weighted averages of individual country data unless otherwise specified. Data are weighted by GDP valued at PPP as a share of the group GDP in 2009. Fixed weights are assumed for all years, except in figures where annual weights are used.
For most countries, fiscal data follow the IMF’s Government Finance Statistics Manual (GFSM) 2001. The concept of overall fiscal balance refers to net lending(+)/borrowing(−) of the general government. In some cases, however, the overall balance refers to total revenue and grants minus total expenditure and net lending.
Data on the financial sector support measures are based on the IMF’s Fiscal Affairs and Monetary and Capital Markets Departments’ database on public interventions in the financial system, revised following a survey of the G-20 economies. Survey questionnaires were sent to all G-20 members in early December 2009 to review and update IMF staff estimates of financial sector support, consisting of recapitalization, asset purchases, liquidity support comprising asset swaps and treasury purchases, and guarantees. For each type of support, data were compiled for the amounts that had been initially announced or pledged, actually utilized, and recovered to-date. The period covered is June 2007–December 2009.
Table 3 of this annex presents IMF staff estimates of the general government cyclically adjusted primary balance. For some countries, the series reflect additional adjustments, including for natural resource-related revenues or commodity-price developments (Chile, Norway, Peru), for land revenue and investment income (Hong Kong SAR), for royalties from a large hydroelectric power station (Paraguay), for tax policy changes and the effects of asset prices on revenues (Sweden), and for extraordinary operations related to the banking sector (Switzerland).
Additional country information, including for cases where reported fiscal aggregates in the Monitor differ from those reported in the WEO:
Argentina. Following the national definition the general government balance, primary balance, cyclically adjusted primary balance, and expenditure include accrued interest payments (while it is excluded in the WEO).
Bulgaria. The general government balance projections for 2010 reflect the data presented in the April 2010 WEO (on a cash basis). They do not yet account for the recently announced new government cash deficit target of 2.5 percent of GDP in 2010, or the government’s new policy measures which are being discussed.
Colombia. Historical figures for the overall fiscal balance as reported in the Monitor and WEO differ from those published by the Ministry of Finance as they do not include the statistical discrepancy.
Estonia. Gross and net debts have been revised with respect to the WEO to reflect full consistency with Eurostat methodology.
Greece. The projections for the overall balance, primary balance, cyclically adjusted primary balance, and gross debt are those under the IMF-supported program and reflect measures in the program affecting both the revenue and expenditure side. However, the government expenditure and revenue ratios in Tables 4 and 5 of this annex do not include these measures, consistent with the presentation of the selected economic indicators provided on May 9, 2010 (Press Release No. 10/187).
Latvia. In accordance with WEO conventions, the fiscal deficit shown in the Monitor includes bank restructuring costs and is thus higher than the deficit in official statistics. Exclusive of bank restructuring costs, which are incurred between 2008 and 2011 only, the overall general government deficit is: 3.3 percent of GDP (2008), 7.0 percent of GDP (2009), 8.5 percent of GDP (2010), and 6.5 percent of GDP (2011).
Philippines. Fiscal data are for central government.
Singapore. Data are on a fiscal year rather than calendar year basis.
Turkey. Information on general government balance, primary balance and cyclically adjusted primary balance as reported in this Monitor and the WEO differ from those published in the authorities’ official statistics or country reports, which still include net lending.
III. Economy Groupings
The following groupings of economies are used in the Fiscal Monitor.
|Advanced Economies||Emerging Economies||G-7||G-20||Advanced G-20||Emerging G-20||Euro Area||Emerging Asia||Emerging Europe||Emerging Latin America||Low-income Economies||Oil Producers|
|Denmark||Colombia||United Kingdom||France||Japan||Mexico||Germany||Central African||Niger||Congo,|
|Finland||Estonia||United States||Germany||Korea||Russia||Greece||Philippines||Poland||Peru||Rep.||Republic of|
|France||Hungary||India||United Kingdom||Saudi Arabia||Ireland||Chad||Papua New|
|Germany||India||Indonesia||United States||South Africa||Italy||Thailand||Romania||Guinea||Ecuador|
|Hong Kong SAR||Kenya||Japan||Malta||Russia||Guinea|
|Iceland||Latvia||Korea||Netherlands||Congo, Dem.||Sao Tome &|
|Israel||Malaysia||Russia||Slovak Republic||Ukraine||Cote d’lvoire||Senegal||Indonesia|
|Italy||Mexico||Saudi Arabia||Slovenia||Eritrea||Sierra Leone||Iran|
|New Zealand||Philippines||United States||Ghana||Tanzania||Nigeria|
|Slovak Republic||Saudi Arabia||Kyrgyz Republic||Vietnam||Timor-Leste|
|Slovenia||South Africa||Laos||Yemen||Trinidad and|
|Hong Kong SAR||−0.6||1.0||4.1||7.7||0.2||0.8||−1.4||−1.2||−0.1||−0.1|
|Hong Kong SAR||…||1.0||4.2||7.7||0.3||0.9||−1.4||−1.2||0.0||0.0|
|Hong Kong SAR 2/||…||−1.3||0.3||1.6||−0.1||−1.5||−3.4||−3.9||−2.0||−2.0|
|united States 4/||…||−1.1||−0.5||−0.6||−3.3||−6.2||−7.5||−5.2||−2.3||−2.3|
|Hong Kong SAR||17.7||16.9||15.4||14.5||18.6||16.7||17.5||13.2||17.4||17.4|
|Hong Kong SAR||17.1||17.9||19.5||22.2||18.9||19.5||16.1||17.0||17.4||17.4|
|Hong Kong SAR||…||1.9||1.7||1.3||1.2||1.0||0.6||0.6||0.5||0.5|
|Hong Kong SAR||0.0||0.0||0.0||0.0||0.0||0.0||0.0||0.0||0.0||0.0|
2009, “How Expensive is the Welfare State? Gross and Net Indicators in the OECD Social Expenditure Database (SOCX),”OECD Social, Employment, and Migration Working PapersNo. 92 (Paris: Organization for Economic Cooperation and Development).
2007, “Cyclical Macro Policy and Industry Growth: The Effect of Countercyclical Fiscal Policy,”Working Paper (Cambridge, Massachusetts: Harvard University).
2005, Handbook of Economic Growth, Vols. 1A and 1B (Amsterdam: North-Holland).
1995, “Another Look at the Instrumental Variables Estimation of Error-Component Models,”Journal of Econometrics, Vol. 68, pp. 29–51.
2010, “Fiscal Deficits, Debt and Interest Rates,”IMF Working Paper (Washington: International Monetary Fund), forthcoming.
2010, “Public Expenditures on Social Programs and Household Consumption in China,”IMF Working Paper 10/69 (Washington: International Monetary Fund).
2010, “Sub-Central Governments and the Economic Crisis: Impact and Policy Responses,”OECD Economic Department Working Paper 2010/8 (Paris: Organization for Economic Cooperation and Development).
2010, “Recession, Recovery, and State-Local Finances.”Presentation to the Forecasters Club of New York, January28.
2010, “Sovereign Spreads: Global Risk Aversion, Contagion, or Fundamentals?”IMF Working Paper (Washington: International Monetary Fund), forthcoming.
2010, “Petroleum Product Subsidies: Costly, Inequitable, and Rising,”IMF Staff Position Note 10/05 (Washington: International Monetary Fund).
2005, “Debt Overhang or Debt Irrelevance? Revisiting the Debt-Growth Link,”IMF Working Paper 05/223 (Washington: International Monetary Fund).
1994, “Some Unpleasant Supply Side Arithmetic,”Journal of Monetary Economics, pp. 507–524.
1999, “Government Debt,” in Handbook of Macroeconomics, Vol. 1C, ed. by, J.B.Taylor andM.Woodford (Amsterdam: North-Holland).
2010, “A Practical Guide to Public Debt Dynamics, Fiscal Sustainability, and Cyclical Adjustment of Budgetary Aggregates,”IMF Technical Note and Manual (Washington: International Monetary Fund).
2006, “Fiscal Policy and Interest Rates in Europe,”Economic Policy, July, pp. 443–89.
2003, Fiscal Vulnerability and Financial Crises in Emerging Market Economies, (Washington: International Monetary Fund).
International Labor Organization, 2010, “World Social Security Report 2010: Providing Coverage in the Time of Crisis and Beyond” (Geneva: International Labor Office, Social Security Department).
International Monetary Fund, 2009, “The State of Public Finances; Cross-Country Fiscal Monitor: November 2009,”IMF Staff Position Note 09/25 (Washington).
International Monetary Fund,, 2010a, “Exiting from Crisis Intervention Policies,”IMF Policy Paper (Washington).
International Monetary Fund, 2010b, “Strategies for Fiscal Consolidation in the Postcrisis World,”IMF Policy Paper (Washington).
International Monetary Fund, 2010c, “From Stimulus to Consolidation: Revenue and Expenditure Policies in Advanced and Emerging Economies,”Staff Paper (Washington), forthcoming.
International Monetary Fund, 2010d, “Fiscal Rules—Anchoring Expectations for Sustainable Public Finances,”Staff Paper (Washington).
International Monetary Fund, 2010e, “Preserving Debt Sustainability in Low-Income Countries in the Wake of the Global Crisis,”Staff Paper (Washington).
International Monetary Fund, 2010f, World Economic Outlook: Rebalancing Growth, World Economic and Financial Surveys (Washington).
International Monetary Fund, 2010g, Global Financial Stability Report: Meeting New Challenges to Stability and Building a Safer System, World Economic and Financial Surveys (Washington).
2010, “State Tax Changes in Response to the Recession,” (Washington: Center on Budget and Policy Priorities).
1988, “Financing versus Forgiving a Debt Overhang,”Journal of Development Economics, Vol. 29, pp. 253–68.
2010, “Government Debt and Economic Growth,”IMF Working Paper (Washington: International Monetary Fund), forthcoming.
2009, “New Evidence on the Interest Rate Effects of Budget Deficits and Debt,”Journal of European Economic Association, Vol. 7, pp. 858–85.
2002, “External Debt and Growth,”IMF Working Paper 02/69, (Washington: International Monetary Fund).
2004, “What Are the Channels Through which External Debt Affects Growth?”IMF Working Paper 04/15 (Washington: International Monetary Fund).
2010, “Growth in a Time of Debt,”Working Paper. Presented at American Economic Association meeting, Atlanta, January4.
1987, “Crazy Explanations for the Productivity Slowdown,”NBER Macroeconomic Annual 1989, pp. 1002–1037 (Cambridge, Massachusetts: National Bureau of Economic Research).
2009, “Banking Stability Measures,”IMF Working Paper 09/4 (Washington: International Monetary Fund).
For detailed results, see Kumar and Woo (2010).
For example, see Baldacci and Kumar (2010) on debt and interest rates; Dotsey (1994) on debt and future distortionary taxation; Hemming, Kell, and Schimmelpfennig (2003) on debt and crises; Aghion and Kharroubi (2007) on countercyclical fiscal policy and industrial growth; and Woo (2009) on procyclicality and volatility of fiscal policy and growth.
The focus is on the medium- or long-run relationship between initial government debt and subsequent economic growth. Panel data comprise eight non-overlapping five-year periods. The reported econometric results are for advanced and emerging economies with over 5 million of population. However, the results using the full sample of countries including developing economies are qualitatively similar. As a robustness check, single cross-country regression was also tried for longer time periods. The results are consistent with those from panel regressions.
Specifically, several estimation methods are employed: pooled OLS, robust regression, between estimator (BE), fixed-effects (FE) panel regression, and dynamic system GMM (SGMM) estimation. While this analysis uses the initial level of government debt and avoids the reverse causality, using pre-determined variables such as initial debt does not necessarily get around the endogeneity problem. Given the difficulty of finding appropriate external instruments, it addresses the endogeneity of all the regressors and incorporates fixed effects by using SGMM regression (Arellano and Bover, 1995).
The baseline panel regression specification is:
where i and t denote country and time period; y is the logarithm of real per capita GDP; vi is the country-specific fixed effect; ηt is the time-fixed effect; εit is an unobservable error term; Xit is a vector of economic and financial variables; Zit is the initial government debt (in percent of GDP). Xi includes a set of explanatory variables (other than lagged per capita GDP): (i) log of average years of secondary schooling, as a proxy for human capital; (ii) initial government size as measured by government consumption share of GDP; (iii) initial trade openness (sum of export and import as a share of GDP); (iv) initial financial market depth (quasi-liquid liabilities as a share of GDP); (v) initial inflation; (vi) terms of trade growth rates; (vii) a measure of banking crisis; (viii) fiscal deficit. For robustness checks, parsimonious specifications were tried and additional variables were included (such as aged-dependency ratio, population growth, urbanization, investment, private saving, or constraints on executive decision making). However, the results do not change appreciably.
At a lower level of debt, it is plausible that the positive effect of growth-enhancing government expenditures financed by debt may outweigh the adverse effect of debt.
This is based on a detailed growth accounting exercise on components of growth of output per worker (capital stock per worker, human capital, TFP). The relation between labor force participation and debt is also examined.
This effect appears to partly reflect the impact of debt on interest rates; for an analysis of the latter, see November 2009 Fiscal Monitor, and Baldacci and Kumar (2010).
Measured by standard deviation of annual GDP growth rates over each five-year period.
The allocation of part of stimulus funds to SNGs can increase its effectiveness. Without intergovernmental transfers, central government would be hard-pressed to spend effectively a significantly higher amount of funds on new investment projects, while the SNGs would be unable to start or complete projects under way.
The Fiscal Survey of States, December 2009. National Governors Association and the National Association of State Budget Officers.
For example, California’s state revenues fell from US$103 billion in FY 2008 to US$88.1 billion in FY 2010, and are projected to increase only slightly to US$89.3 billion in FY 2011. In a slow-growth recovery, tax revenue can take even longer than the three to five years for “normal” recovery to reach the precrisis peak (Donald J. Boyd, Recession, Recovery and State-Local Finances. Presentation at the Forecasters Club of New York, January 28, 2010).