Chapter

2. Fiscal Policy in Advanced Countries: Effectiveness, Coordination, and Solvency Issues

Author(s):
International Monetary Fund. European Dept.
Published Date:
May 2009
Share
  • ShareShare
Show Summary Details

On the heels of the global financial crisis, active fiscal policy is back on the agenda of the advanced European economies. Indeed, a fiscal expansion could be particularly effective in the near-term economic environment: the recent tightening of credit constraints could make spending more sensitive to current income and, thus, taxes and subsidies. Given the increased integration of European economies, policy coordination is nonetheless key to magnifying the effects of national fiscal expansions. While it is important for countries to support their economies in the face of this unprecedented slowdown, a clear and credible commitment to long-run fiscal discipline is now more essential than ever: any loss of market confidence may raise long-term real interest rates and debtservice costs, partly offsetting the stimulus effects of measures taken to deal with the crisis and further adding to financing pressures. Hence, it is particularly crucial that any short-term fiscal action be cast within a credible medium-term fiscal framework and envisage a fiscal correction as the crisis abates.

Overview

Policy actions taken to address the global crisis have been increasingly broad in scope as financial problems have spread and activity has deteriorated. Overall, policies have aimed at restoring confidence in global financial markets and institutions, supporting aggregate demand, and, ultimately, breaking the corrosive feedback loop between the financial and real sectors of the global economy. In particular, fiscal policy is providing important support through direct stimulus, automatic stabilizers, and the use of public balance sheets to shore up the financial system.

While such support is critical to bolster aggregate demand and to limit the impact of the financial crisis on the real economy, it implies a significant deterioration in the fiscal positions of advanced European economies.11 In fact, public deficits and debt are projected to rise dramatically in the coming year, exacerbating existing long-run fiscal challenges and raising questions about sustainability. Financial markets seem to have responded to these developments by requiring higher sovereign default risk premiums for most countries, and differentiating across sovereign issuers much more than before.

Against this background, it is important to assess the fiscal implications of the crisis so far, including for the sustainability of public finance, and to explore how governments can maximize the effectiveness of fiscal support while minimizing the impact of this support on their solvency. This entails gauging the fiscal costs of the measures taken to address the financial crisis and sustain aggregate demand, and assessing the impact of automatic stabilizers on government balances. Other questions of relevance include the fiscal policy mix that is likely to enhance effectiveness, the benefits of policy coordination, and the impact of the crisis on fiscal sustainability. It is also important to review the role of budgetary frameworks, and the Stability and Growth Pact in particular, as an essential device of commitment to long-term fiscal sustainability.

Fiscal Costs of the Crisis

Financial Sector Support

Government support to the financial sector has taken various forms, with different implications for debt and fiscal balances. Almost all advanced European economies have provided capital injections and guarantees for financial sector liabilities. Some have purchased illiquid assets from financial institutions or extended direct loans (Table 4).12 Altogether, the immediate impact of these measures to support government financing has reached 6.3 percent of 2008 GDP on average, ranging widely from 1.1 percent of GDP in Switzerland to 20.2 percent of GDP in the United Kingdom.

Table 4.Headline Support for the Financial Sector and Upfront Financing Need(As of April 15, 2009; percent of 2008 GDP)
Capital Injection(A)Purchase of Assets and Lending by Treasury(B)Central Bank Support Provided with Treasury Backing(C)Guarantees1/ (D)Up-front Government Financing2/(E)
Austria5.30.00.030.05.3
Belgium4.70.00.026.24.7
France1.21.30.016.41.53/
Germany3.80.40.018.03.7
Greece2.13.30.06.25.4
Ireland5.30.00.0257.05.3
Italy1.30.00.00.01.3
Netherlands3.42.80.033.76.2
Norway2.015.80.00.015.8
Portugal2.40.00.012.02.4
Spain0.04.60.018.34.6
Sweden2.15.30.047.35.84/
Switzerland1.10.00.00.01.1
United Kingdom3.913.812.951.220.25/
Average6/2.53.72.125.06.3
Source: IMF, Update on Fiscal Stimulus and Financial Sector Measures (published April 26, 2009), available via the Internet: www.imf.org/external/np/fad/2009/042609.htm.

Part of the up-front costs, though, is expected to be recovered. In previous episodes of banking crises around the world, recovery rates varied considerably (Laeven and Valencia, 2008). In the Nordic countries in the early 1990s—the most recent episodes of banking crises in advanced European economies—recovery rates ranged from around 15 percent in Finland to more than 90 percent in Sweden. Empirical analysis based on a large sample of financial crises indicates that recovery rates are positively correlated with per capita income and the fiscal balance at the beginning of the crisis—probably an indicator of sounder public financial management (IMF, 2009b and 2009c). Countryspecific projected recovery rates based on this analysis suggest that the medium-term impact on public debt could be substantially smaller than the up-front cost in most cases.

Explicit guarantees provided so far are quite large, especially in Ireland (Table 4, column D). Nevertheless, the ultimate costs are likely to be lower. Indicative estimates based on financial derivative pricing models suggest that outlays from contingent liabilities could be on average around 1–3 percent of GDP, cumulative for 2009–13 for the advanced European economies (IMF, 2009b and 2009c).13 However, governments may need to extend additional support to the financial sector if the crisis persists. De facto, governments are also providing implicit guarantees to financial institutions that could potentially entail significant additional fiscal costs.14

Downturns, Asset Price Reversals, and Stimulus Plans

Fiscal deficits will increase, owing to the operation of the automatic stabilizers during the economic downturn. In most countries, the estimated effect of automatic stabilizers is expected to increase significantly in 2009 (Figure 15).

Figure 15.Advanced European Economies: Estimated Impact of Automatic Stabilizers on Fiscal Balances, 2008–091/

(Percent of GDP)

Source: IMF, World Economic Outlook.

1/ The impact on fiscal balance from automatic stabilizers is computed as the change in the cyclical balance between two consecutive years.

On top of that, there might be a further negative impact on revenues because of the ongoing reversal in house and equity prices, which is not fully reflected in the conventional estimates of cyclical balances. Fluctuations in asset prices affect the taxation of capital, financial transactions and capital gains, and corporate and personal income tax proceeds—as well as indirect tax revenues—through their impact on wealth and consumption. Even though asset values and business cycles are correlated, dramatic declines in asset prices may reduce revenues even more than currently anticipated. Indeed, during previous episodes of house and equity price busts in advanced European economies, even cyclically adjusted revenues fell in the aftermath of asset price reversals and, in the case of house price reversals, they bottomed out with a long lag (Figure 16).

Figure 16.Advanced European Economies: Fiscal Revenues During Episodes of House and Equity Price Busts

(Percentage change from a year earlier in cyclically adjusted revenues as a share of potential GDP; X-axis in quarters)

Sources: Organization for Economic Cooperation and Development; Bank for International Settlements; Claessens, Kose, and Terrones (2008); and IMF staff calculations.

Note: The solid blue line denotes the median of all observations. Zero is the quarter when a bust begins.

Most countries have also adopted fiscal stimulus plans, amounting to, on average, 1 and 0.8 percent of GDP in 2009 and 2010, respectively (Table 5).

Table 5.Advanced European Economies: Estimated Cost of Discretionary Measures, 2008–101/(As of April 15, 2009; percent of 2008 GDP)
200820092010
Austria0.31.51.7
Belgium0.00.80.4
Cyprus0.31.70.0
Denmark0.00.00.0
Finland0.01.70.5
France0.00.70.8
Germany0.01.62.0
Greece0.00.10.0
Ireland0.00.00.0
Italy0.00.20.1
Luxemburg0.03.73.6
Malta0.00.60.4
Netherlands0.00.80.7
Norway0.01.81.8
Portugal0.31.00.0
Spain1.92.30.3
Switzerland0.00.70.0
United Kingdom0.21.4−0.1
Average2/0.41.00.8
Discretionary impulse3/0.40.6−0.3
Source: IMF staff estimates.

Most of the announced measures are temporary, but some are permanent, implying a lasting effect on deficits and a cumulative impact on public debt (IMF, 2009b and 2009c).15 Overall, fiscal accounts will be severely affected by the crisis, with potential implications for fiscal sustainability (see section on fiscal solvency).

Making the Most Out of Fiscal Interventions

Targeting the Response in Times of Financial Distress

The current crisis has acquired the hallmarks of a full-fledged recession driven by a strong contraction in aggregate demand. The fall in aggregate demand is due to a large decrease in financial wealth, an increase in precautionary saving on the part of households, increasing difficulties in obtaining credit, and a wait-and-see attitude on the part of consumers and firms in the face of uncertainty. A further fall in demand will increase the risk of setting in motion a perverse dynamic of deflation, rising debt, and associated feedback loops to the financial sector.

These factors—together with the unavailability of any export-led recovery strategy due to the global nature of the recession—call for an expansionary fiscal response and, at the same time, shape its design. In practice, the crisis may affect the impact of fiscal measures, including the speed of their transmission through the economy.16 The ongoing financial turmoil is part of an adjustment toward more sustainable macroeconomic conditions around the globe, and the behavior of economic agents may change as a result of this process and the policy actions it entails. As a result, fiscal policies that contribute to lowering existing uncertainties, bolster confidence, and promote the expectation of sustainable public finances over the medium and long term will be most effective in fighting the current downturn.

In the current circumstances, assessing the effectiveness of fiscal policy—namely, the likely impact of the discretionary fiscal stimulus on activity—is even more difficult than normal. Generally, researchers have used a variety of econometric techniques and model specifications to get reliable estimates of fiscal multipliers, reaching a wide range of outcomes. This broad range of results mostly reflects country conditions, the persistence of the stimulus, and policy mix—including the type of instruments used, the response of monetary policy, constraints on borrowing, trade openness, and longterm sustainability.17

With regard to the type of instruments featured in the policy mix, both the fiscal multipliers used by IMF country teams (Table 6) and model-based empirical research on multipliers for the various components of countries’ fiscal packages (revenue measures, infrastructure spending, and other spending) seem to suggest that infrastructure spending is likely to have the largest impact on growth, although it also has the longest implementation lags.18 In contrast, tax cuts are likely to have a more modest growth impact—particularly if they are not targeted to credit-constrained consumers—as they rely entirely on propensities to spend out of wealth or income, and have no direct effect on the demand for goods and services.

In general, econometric analysis tends to back skepticism regarding the effectiveness of discretionary fiscal actions by documenting a trend of small and declining consumption multipliers since the 1980s, as greater financial deregulation, larger wealth accumulation, and better policies that might have helped lower uncertainty about future income, relax credit constraints on households and firms, and lengthen private sectors planning horizons.19

Because of the enhanced opportunities to smooth consumption vis-à-vis temporary fluctuations in income and diversify income risks, household demand is likely to have become—over time—less and less dependent on current income, thereby shaking the foundations of fiscal multipliers.20

Table 6.Ranges of Fiscal Multipliers Used by IMF Country Teams
Lower BoundUpper Bound
Tax cuts0.30.6
Infrastructure investment0.51.8
Other 1/0.31.0
Source: IMF staff estimates.Note: Includes additional spending on safety nets, transfers to state and local governments, assistance to small and medium-sized enterprises, and support for housing markets.

In the current slowdown, however, it seems that this argument can be run in reverse. As the financial turmoil is spreading across borders and across asset classes, the share of credit-constrained households is likely to rise and the planning horizon of creditunconstrained consumers to shorten. Similarly, the correction of housing prices is bound to lower the value of the collateral that households in many European economies can count on to borrow. For all these reasons, spending patterns would become much more dependent on current income, thus boosting the effectiveness of any tax cut and/or transfer increase. Thus, while in normal circumstances a discretionary fiscal stimulus might have limited impact, in the current economic conditions the case for fiscal action seems to have become stronger, giving reasons to believe that the multipliers would be closer to the upper bound of the range of estimates provided above.

The same argument also speaks in favor of targeted fiscal measures. Specifically, as the problem of financial distress is not uniform across groups in the economy, the effect of income support may differ vastly across groups of households, depending on their initial debt level and their equity losses in the crisis. Fiscal support should be targeted consistently to specific groups of households and firms that are most vulnerable to the economic downturn. Concentrating income support may maximize its insurance value for the population, while guaranteeing a relative strong stimulus to the economy—thereby creating more output for a given deterioration of the budget.

The anticipated depth and length of the downturn should also shape the fiscal reaction. With a significant risk of a prolonged downturn, the stimulus should be designed to support demand over a long period and rely, more than in the past, on spending measures that directly boost final demand. One advantage of direct spending—for instance, on government investment—over tax cuts or increase in transfers, which operate by raising the purchasing power of households and firms, is that the impact of direct spending will be less mitigated by a possible increase in the saving rate of the private sector.

The potential effectiveness gain of carrying out fiscal stimuli in circumstances where uncertainty is higher and financial markets are disrupted—hence causing households and firms to be more “myopic” and risk averse—can be illustrated via simulations of the IMF’s Global Integrated Monetary and Fiscal (GIMF) Model.21 The version of the model used here features two (asymmetric) economic areas within a monetary union—a large euro area advanced economy (which has been calibrated using German data) and the rest of the euro area. The model has a number of features that make it especially suitable for assessing the effectiveness of fiscal measures: as a new-Keynesian, intertemporal model based on household- and firm-optimizing behavior, GIMF allows for non-Ricardian responses to fiscal actions by allowing for overlapping generations with finite economic lifetimes, life cycle incomes, and liquidity-constrained households. In the model, taxes on labor income have distortionary effects and changes in these taxes cause agents in the model to adjust their behavior. GIMF also features a number of nominal and real rigidities that capture well the characteristics of labor and product markets in Europe, and allow for monetary policy to have real effects in the short to medium run.22

Under selected scenarios, simulation outcomes focus on the effects on GDP, private consumption, public debt, and the real interest rate of a temporary fiscal stimulus. The fiscal measures considered here involve taking expansive actions equal to 1.0 percent of GDP in the first year and 0.5 percent of GDP in the second year, either by cutting labor income taxes or by increasing public investment. In addition, lump-sum transfers are used to offset the resulting endogenous changes in the budget deficit, so that the working of automatic stabilizers is suppressed.23 There is no monetary accommodation, so the Taylor-type rule in the GIMF model continues to operate in both years of the fiscal expansion.

It is important to stress that the purpose here is to present a “likely range” regarding the effectiveness of a given fiscal stimulus under different—and necessarily stylized—scenarios. In other words, by focusing on a temporary tax cut, each simulation attempts to give an idea of the “smallest possible” multiplier, as tax cuts that are perceived to be temporary have only a limited effect on the behavior of households that are not liquidity constrained. In contrast, at the other end of the spectrum, the same targeted stimulus engineered through government investment expenditures would provide “the upper bound” of the multiplier effect, given the existence of direct effects on aggregate demand and secondary effects on household spending, as incomes and wealth would also increase due to the higher productivity of the economy.

With respect to the large euro area country, Table 7 shows the cumulative fiscal multipliers—defined as the cumulative impact on real GDP and consumption over the two years divided by the cumulative deficit over the same period—for a simultaneous euro area stimulus. Two cases are distinguished: (1) “normal” circumstances, defined by a planning horizon for forward-looking households of 20 years, a share of credit-constrained households of 25 percent, and a risk aversion coefficient equal to 4; and (2) distressed financial markets, characterized by a shorter planning horizon for the optimizing consumers (10 years), a larger share of liquidity-constrained consumers (50 percent),24 and greater uncertainty, thereby raising the degree of risk aversion coefficient to 6.

Table 7.The Case for Fiscal Stimulus: Effects of Fiscal Stimulus Under Distressed Financial Markets1/(Percentage point deviation from control, unless otherwise stated)
First YearSecond YearThird Year
GDP2/
”Normal” circumstances0.2-1.50.3-1.60.3-1.8
Distressed financial markets0.7-1.70.8-1.70.8-1.8
Consumption 2/
”Normal” circumstances0.5-0.90.5-1.00.6-1.2
Distressed financial markets0.7-1.20.7-1.20.7-1.4
Public debt-to-GDP ratio
”Normal” circumstances−0.2-0.90.3-1.40.3-1.4
Distressed financial markets−0.4-0.80.2-1.30.2-1.3
Real interest rate
”Normal” circumstances0.1-0.40.0-0.30.0-0.1
Distressed financial markets−0.1-0.4−0.1-0.3−0.1-0.1
Source: IMF staff calculations.

The cumulative multipliers on GDP and consumption appear to be larger in times of financial distress—despite the increase in risk aversion—and accompanied by lower fiscal costs and slower debt accumulation over the short term. This is particularly true for the tax stimulus because—with distressed financial markets—all agents in this model become more myopic with respect to future tax liabilities. Even the least constrained group of households is assumed to halve its planning horizon and its remaining working life. This means that such households perceive the temporary tax cut as an increase in wealth, which, in turn, leads them to spend more. This effect is even stronger for the group of liquidity-constrained agents, who are assumed to consume their after-tax income in every period, without any possibility for borrowing or saving. Changes in taxes directly affect the disposable income and, therefore, spending of these agents. Doubling the size of this group—which is assumed to be the same in the two economic regions—is therefore critical to increase the multipliers of tax stimulus measures.

Coordinating the Response in Integrated Economies

Another reason for fiscal skepticism is the belief that—in highly integrated economies—an increasing share of a given fiscal stimulus benefits employment and output abroad, rather than in the country sustaining the cost of the fiscal expansion. In the traditional jargon, the problem consists of the “leakages” that reduce the additional output one can stimulate with a given amount of government spending or tax cuts, as a large share of domestic consumption and investment falls on imported goods or on imported inputs.25

However, even in the presence of trade and financial linkages, fiscal policy may still be powerful at the aggregate level, if all governments expand at the same time. More specifically, a fiscal expansion in a large country affects net exports in trading partners through different channels. Part of the public spending rise in the expanding country falls directly on imports. Also, the fiscal stimulus boosts domestic demand in the expanding country, leading to more imports. Finally, the prices of the expanding country’s products rise faster than those of its trading partners, inducing consumers in those countries to substitute imports for locally produced goods. Overall, trading partners benefit from the boost in net exports and GDP, although in the expanding country part of the stimulus “leaks” abroad. But, if a fiscal stimulus is also adopted simultaneously by the trading partners, all countries mutually benefit from each other’s expansion. Hence, the broader the multicountry participation in the fiscal expansion, the larger the multiplier of a given stimulus.

The gains of a coordinated fiscal expansion can also be evaluated by simulation of the IMF’s GIMF model. Table 8 summarizes the effects of a temporary fiscal expansion through either lower taxes or higher investment in GDP and private consumption in the large euro area economy under two scenarios: (1) the fiscal expansion is undertaken by the large euro area country alone; and (2) the fiscal expansion is undertaken jointly by all euro area countries.

Table 8.Coordination Gains: Cumulated Effects of Fiscal Stimulus on a Large Euro Area Country1/(Percentage point deviation from control, unless otherwise stated)
First YearSecond YearThird Year
GDP2/
Own stimulus effect0.1–1.10.2–1.20.2–1.4
Coordinated stimulus effect0.2–1.50.3–1.60.3–1.8
Consumption 2/
Own stimulus effect0.3–0.70.3–0.90.4–1.1
Coordinated stimulus effect0.5–0.90.5–1.00.6–1.2
Source: IMF staff calculations.

Coordination gains in the stylized large economy appear to be nonnegligible, amounting to 0.1–0.4 percentage point of the cumulated effects of the simultaneous fiscal stimulus on output—depending on the composition of the fiscal package. This is due simply to existence of very large trade effects within the euro area, which have been reproduced in the calibration of the two-region model. The lesson here is that the potential for intra-EU externalities in fiscal policy, makes policy coordination more essential than ever.26

Treasuring Fiscal Solvency

The Crisis Is Putting Fiscal Solvency at Risk …

The benefits of fiscal expansions do not come for free. Current gains should be assessed against the future costs of a larger stock of public liabilities. Indeed, in most countries, government debt is expected to rise sharply as a share of GDP over the next year, reflecting support for the financial sector, fiscal stimulus packages, and revenue losses caused by the economic downturn and declining asset prices (Figure 17). The debt-to-GDP ratio is expected to increase, on average, by 9.4 percentage points—the largest jump since the early 1980s. This deterioration exacerbates existing long-run fiscal challenges related to population aging, thereby raising concerns about fiscal solvency.

Figure 17.Projected Changes in Public Debt

(Percent of GDP; change end-2009–end-2008)

Sources: IMF, World Economic Outlook; and IMF staff calculations.

1/ Weighted by PPP GDP.

As debt dynamics depend on the initial debt stock, the real growth rate of the economy, the real interest rate, and the size of future primary balances, the current crisis creates risks to fiscal sustainability in different ways. Growth rates will be negative in 2009, while the time and extent of the recovery remain highly uncertain. Even beyond the short term, potential growth is expected to be lower than in the precrisis period. While average nominal yields (weighted by maturity) have fallen in most countries, sovereign spreads have risen sharply in some of them (see next section), signaling that governments’ marginal funding costs may increase. At the same time, inflation is expected to ease significantly in the near term, thus boosting real interest rates. Primary balances are projected to remain weak in the immediate future and, in many cases, are expected to achieve levels insufficient to ensure debt stabilization (IMF, 2009b). Another reason for concern is that rollover risks are likely to increase in the short run, given both the limited availability of credit and the dramatic and simultaneous increase in debt issuance across the world (Box 5).

… Markets Are Concerned27

In the last few years, credit spreads have been extremely narrow. Countries with vastly different ratios of debt and deficits to their GDP have been able to borrow at essentially the same interest rate. But it seems increasingly unlikely that countries undertaking substantial fiscal expansion in the current circumstances will be granted the same treatment going forward. To be sure, in recent months financial markets have been requiring higher default risk premiums across a variety of bond instruments and across most issuers. Sovereign spreads and credit default swap premiums have widened since September 2008 and are now very large by historical standards.

In the euro area, in particular, spread variance across common currency members has risen dramatically, suggesting that markets are differentiating more and more among government issuers. From a range of 0–25 basis points in 2007, 10-year spreads versus the German bund by the end of January 2009 had reached close to 300 basis points in the case of Greece and over 100 basis points for Ireland, Italy, Portugal, and Spain (Figure 18).28 Increasing sovereign spreads could have a major impact on governments’ marginal funding costs in the euro area, possibly undoing the benefits of declining risk-free interest rates.

Figure 18.Selected Euro Area Countries: Sovereign Spreads and Financial Institutions’ Expected Default Probabilities, 2008–January 20091/

Sources: Datastream; Moody’s Creditedge; and IMF staff calculations.

1/ Solid blue line denotes the country’s 10-year sovereign bond spread vis-à-vis Germany (basis points, left scale). Dotted red line denotes expected default frequencies of the country’s median financial institution (percent, right scale).

Behind these developments are, among other things, growing concerns about the fiscal solvency of some euro area member states in the face of the expected increase in public debt associated with the fiscal costs of the financial crisis. The medium-term net budgetary cost of financial support operations will also depend on the extent to which the assets acquired by government or the central bank will hold their value, on future losses from explicit guarantees, and on additional costs from implicit guarantees to the banking sector. The high correlation between the expected default frequency (EDF) of financial institutions and market concerns about the fiscal implications of the government support, as indicated by wider sovereign spreads, points in this direction (Figure 18).

Considerations about the relative liquidity of different government bond markets may be yet another factor. The financial turmoil could lead to a flight to safety and liquidity, resulting in a decline in the yields of the most liquid sovereign bond markets.

But global discrimination among different classes of default risk may have also contributed to the widening of the sovereign risk premium differentials. Assuming that risk premiums embedded in country-specific sovereign yields are determined jointly in the market and influenced by the riskiness of the specific asset as well as common risk factors—such as the willingness and the ability of investors to bear that risk—a common component in sovereign spreads has been estimated (Figure 19).29 This common component seems able to capture the general downward trend in sovereign spreads due to the Economic and Monetary Union convergence over 2001–02. It also reflects high liquidity and falling market volatility over 2003–07, and the jarring risk repricing since 2008—proxied in Figure 19 by the implied volatility of the German stock market index.

Figure 19.Estimated Common Component in Sovereign Spreads, 2001–January 2009

Sources: Datastream; Bloomberg L.P.; and IMF staff calculations.

1/ The fan chart plots, at each point in time, the 5th, the 50th, and the 95th percentile of the estimated probability distribution for the expected common component across euro area sovereign spreads. Hence, there is a 90 percent chance that the common spread will be inside the blue-shaded range. The central thick black line denotes the estimated median common spread.

2/ Implied volatility of German stock market.

Shedding additional light on the role played by common and country-specific factors, a panel estimate shows that the sensitivity of sovereign spreads to projected debt changes has significantly increased after September 2008 (Table 9). This model suggests that the markets are concerned about fiscal sustainability more than in the early years of the common currency. In a few countries, markets also appear to be progressively more concerned about the solvency of national banking systems. The liquidity of sovereign bond markets—proxied by traded volume—seems also to be a relevant factor in explaining spread behavior.

Table 9.Sovereign Spreads: Estimated Panel Regression1/
Dependent Variable: D(Spreads)
Coefficientt-Statistic
Constant_preOct081.37[2.64]
Constant_postOct097.68[0.91]
D(Common component)4.49[3.40]
D(Projected debt)_preOct080.05[0.64]
D(Projected debt)_postOct080.29[2.09]
D(Financial EDF)_preOct080.37[0.90]
D(Financial EDF)_postOct088.38[3.24]
Traded volume_preOct08−0.07[−2.70]
Traded volume_postOct08−0.59[−1.96]
R-squared0.43
Adjusted R-squared0.42
Durbin-Watson statistic2.29
Mean dependent variance1.35
S.D. dependent variance6.65
S.E. of regression5.06
Sources: Datastream; Moody’s Creditedge; and IMF staff calculations.

Similarly, seemingly unrelated regression estimates highlight the importance of countryspecific issues alongside the common component (Figure 20). Decomposing the contributions to the actual change in the country-specific sovereign spread between end-January 2009 and end-September 2008 indicates that concerns about fiscal sustainability are significant for countries such as Greece and Ireland and—to a lesser extent—Austria, Italy, and Portugal. For Ireland, the main contributing factor to the widening in sovereign spreads is the deterioration in the solvency of the country’s financial sector, mirroring market concerns about the potential fiscal implications of financial sector fragility. The extent to which rising EDFs in the financial sector translate into increases in government spreads is found to be large and significant also in Austria. Finally, ceteris paribus, the liquidity of the sovereign bond market appears to lessen the Italian government’s financing costs.

Figure 20.Contributions to the Change in Spreads1/2/

(Basis points, change January 2009 over September 2008)

Sources: Economist Intelligence Unit; Datastream; Moody’s Creditedge; and IMF staff calculations.

1/ For each country, the explanatory variables included in the model are the changes in the common factor, the expected default frequency of the median financial institution (except for Greece, for which the 25th percentile of the distribution is used), the projected debt-to-GDP ratio over the next year, and the traded volume of government debt.

2/ For each country, the actual change in spread over the period is reported above the corresponding histogram.

Box 5. Sovereign Financing Needs of Advanced Economies and the Rollover Risk

In 2009, advanced economies around the world are planning to issue a large amount of new public debt to finance fiscal deficits and financial sector support. This issuance, together with the need to refinance a large amount of maturing debt, raises concerns about rollover risk.1

The currently growing rollover risk looks different from the past, even though that experience is limited in advanced economies. The largest increases in public debt occurred during the First and Second World Wars, when extensive government control over the economy and moral suasion facilitated sovereign borrowing (IMF, 2009b). During the 1970s and 1980s, many advanced economies accumulated sizable government debt, but problems with placing debt were few. The only notable exceptions were Italy and the United Kingdom in the 1980s. Both countries responded to the difficulties by adjusting their debt-management strategy to attract a wider class of investors. However, today’s circumstances differ in several respects: a number of countries are planning to issue large amounts of sovereign debt simultaneously, thus compounding rollover risk; new debt is projected to be issued within a relatively short period of time; and a large accumulation of contingent liabilities related to guarantees of financial sector obligations could require additional debt financing.

What are the global funding needs?

  • The gross financing need—calculated as the sum of fiscal deficits plus financial sector support measures, maturing medium- and long-term debt, and the inherited stock of short-term debt—is particularly sizable in the United States and Japan. In Europe, Ireland and Belgium are the countries with the largest financing needs (in percent of GDP), followed by the Netherlands, Italy, and Portugal. Among European countries, the average annual debt rollover—defined as the stock of outstanding debt (as percent of GDP) divided by its average maturity—is highest in Belgium, Italy, Greece, and Portugal (figure). In total, gross fiscal funding needs of the 20 advanced countries in 2009 (including new borrowing and refinancing) could reach almost $10 trillion, representing about 23 percent of the aggregate GDP—an increase of over 50 percent relative to 2008, and even more compared with previous years.

  • Emerging economies’ new sovereign borrowing needs in 2009 are likely to be much less, given the limited intervention support of governments in their financial sector, less aggressive use of fiscal stimuli and weaker automatic stabilizers.2 However, they will still add to the global supply of new debt. Moreover, their refinancing needs will be large.

  • Corporate borrowing needs also will be large. By some estimates, corporate bond issuance (mainly high grade) reached a record $373 billion in January 2009, and there could be an additional $450 billion in corporate debt issuance during the year, as companies are forced to look for alternative funding while banks cut back lending. In addition, in a number of countries, government-guaranteed bonds will be issued by private corporations or banks.3 These bonds typically offer somewhat higher yields than the corresponding sovereign bonds, but they are also highly rated, and, therefore, enter into direct competition with other sovereign bonds.

While the demand for funds is escalating quickly, it is unclear whether the supply of funds will grow correspondingly. On the one hand, the private sector’s increasing savings could provide additional funds to sovereign borrowers. On the other hand, slower reserve accumulation in China and oil-exporting countries could reduce the amount invested in sovereign bonds. Also, elevated risk aversion could curtail appetite for less creditworthy sovereign bonds.

What could governments do to limit rollover risks and ensure that sovereign bonds will be absorbed without pushing up financing costs? First, governments could adjust borrowing strategies to adapt to market needs and better match investors’ preferences, even if this would imply a temporary departure from their longstanding debt-management strategy. They could also take measures to improve the functioning of the treasury debt market, for example, by increasing the number of primary dealers or allowing direct access of final investors to auctions, and seek to diversify the investor base. Third, and very important, governments need to assure markets that measures taken to mitigate the current downturn are consistent with medium- to long-term fiscal sustainability (IMF, 2009b).

Advanced Countries: Average Annual Government Debt Rollover1/

Percent of GDP)

Sources: Organization for Economic Cooperation and Development; and IMF staff estimates.

1/ Central government debt in 2007 divided by its average maturity.

Note: The main authors of this box are Jiri Jonáš and Philippe Karam.1 Rollover risk is defined as “the risk that debt will have to be rolled over at an unusually high cost or, in extreme cases, cannot be rolled over at all.” See IMF and World Bank (2001).2 IMF (2009b) estimates for the G-20 emerging economies a fiscal deficit of 3.2 percent of GDP, and very low up-front financial sector intervention financing.3 According to Fitch Ratings (2009), European governments have guaranteed about €1.5 trillion new issues of banks over the next three years.

… Curbing the Effectiveness of Fiscal Stimulus

While it is important for countries to support their economies in the face of the current unprecedented slowdown, it is key that the entailed short-term fiscal costs not be seen by markets as undermining longrun fiscal sustainability. In fact, the empirical link between country spreads and sustainability concerns suggests that the impact of higher current expenditure is strengthened when complemented with a credible plan that ensures it is financed at least in part by future spending cuts. This is because future spending cuts tend to raise current private consumption and investment via their effects on the long-term interest rate.

Simulations with the GIMF model illustrate this point: in countries where fiscal sustainability is perceived to be in jeopardy, expansionary fiscal measures lead to an increase in risk premiums and long-term real interest rates, which would tend to offset part of the fiscal stimulus effects on spending. To highlight the impact of government borrowing levels on domestic interest rates, the GIMF model has been modified to make the domestic risk premium dependent on the government debt-toGDP ratio in an asymmetric way, hence allowing for a steeply increasing risk premium at large debt-to GDP ratios.30

Table 10 provides quantitative ranges to highlight the effects of the temporary fiscal stimulus described above on GDP, private consumption, public debt, and the real interest rate in an euro area economy in two cases: (1) the fiscal expansion is undertaken by a member state featuring a low government debt-toGDP ratio (60 percent); and (2) the fiscal expansion is undertaken by an euro area member state featuring a high government debt-to-GDP ratio (120 percent).

Table 10.Solvency Concerns Increase Risk Premiums Thereby Reducing the Effectiveness of Fiscal Stimulus1/(Percentage point deviation from control, unless otherwise stated)
First YearSecond YearThird Year
GDP2/
Low-debt scenario0.2–1.40.2–1.40.2–1.5
High-debt scenario0.1–1.30.1–1.30.1–1.4
Consumption 2/
Low-debt scenario0.2–0.60.3–0.80.4–1.0
High-debt scenario0.1–0.50.1–0.70.1–0.8
Debt
Low-debt scenario−0.2–0.90.2–1.40.2–1.4
High-debt scenario0.3–1.00.3–1.60.3–1.7
Real interest rate
Low-debt scenario0.1–0.30.0–0.20.0–0.1
High-debt scenario0.3–0.40.2–0.20.1–0.1
Source: IMF staff calculations.

Consistent with the findings in the previous section, the simulation results show that the fiscal stimulus is more effective in countries with low debt—and hence with more fiscal room—than it is in high-debt countries. Although the fiscal expansion is assumed to be temporary, the low-debt economy is still found to enjoy real interest rates that are approximately 20 basis points below the high-debt economy. This finding could have implications in the context of international policy coordination. It seems to suggest that the effectiveness of a joint fiscal expansion would be maximized if countries where fiscal sustainability is less at risk—and hence that are enjoying more fiscal space and lower risk premiums—were able to help finance the fiscal packages announced by countries with less fiscal room.

Ensuring Fiscal Sustainability: Policy Options

For the fiscal response to the current crisis to be the best possible, it will have to be coordinated regionally, focus on the most effective measures, and take into account the sustainability of public finances. This requires both a credible medium-term strategy and the fiscal framework to support it.

Credible Medium-Term Strategy31

To improve the trade-off between the needed fiscal expansion and the risk of a loss of market confidence, governments should supplement their support packages with a clear and credible strategy to ensure fiscal solvency and improve confidence. Such a strategy should comprise the following:

  • A clear plan for fiscal consolidation in the aftermath of the crisis, including fiscal stimulus packages that rely as much as possible on temporary or selfreversing measures; announcement at an early stage of how the deficit will be reduced in the medium term; and a firm commitment and a clear strategy to contain the trend increase in aging-related spending.

  • Growth-enhancing structural reforms. As growth is a key factor in restoring debt sustainability, directing expenditures toward productive areas—such as government investment in transportation, infrastructure, and education—would be beneficial. This also holds for tax reforms that reduce distortions.

  • Transparency. The cost of government interventions to safeguard the financial sector and cushion the downturn should be recorded transparently to avoid adding to current uncertainties. Where reliable market information is not available to estimate asset prices, contingent liabilities, and other information pertaining to fiscal costs, alternative scenarios should be considered. In general, countries should improve their capacity to identify, disclose, and manage fiscal risks.

  • A clear communications strategy. Government plans should be communicated to the public and markets in a clear and consistent manner.

Strengthening Fiscal Frameworks: Role of the Stability and Growth Pact32

Given the demands on fiscal policy, fiscal frameworks can make an important contribution to implementing such a medium-term strategy, in particular by adding to its credibility. In advanced European economies, a wide range of fiscal rules is in force at both the national and subnational levels. Their role could be usefully strengthened to foster fiscal discipline once growth has resumed.33 Arguably, though, the most binding and visible fiscal constraint in the European Union (EU) is the Stability and Growth Pact (SGP). So, a crucial question is, what role can the SGP play in anchoring expectations of sustainability while allowing an adequate crisis response?

Although the SGP was not set up to facilitate an EU-wide fiscal stimulus in response to a large recession, a closer look at it suggests that the fiscal framework’s provisions are likely to be flexible enough to accommodate the currently envisaged fiscal responses to the crisis (Box 6). At the same time, the SGP’s excessive deficit procedure (EDP) is likely to provide a useful anchor for fiscal adjustment, at least in the short run, by mandating reductions in the structural deficit once the recovery has set in.

Looking beyond the short term, however, there is room to strengthen the SGP’s role in ensuring the longer-term health of public finances. Under the current framework, countries are required to steer fiscal policy in line with medium-term objectives (MTOs) (the so-called preventive arm).34

Nevertheless, the SGP so far has failed to provide sufficiently strong reasons for EU members to adhere to their targets. Observance of MTOs is subject only to regular budgetary surveillance by the European Commission (EC), which is lacking the political “teeth” the EDP provides, and analyses of the SGP have consistently emphasized this as a major flaw in the anchoring role of the framework.35 As a result, many countries have fallen short of their MTOs even in good economic times.36

There are various options to enhance the anchoring role of the preventive arm:

  • Encourage reforms putting national institutions and rules in tune with the SGP. In particular, MTOs could be better integrated into medium-term fiscal frameworks at the national level, providing the SGP with a suitable institutional interface in each country. The EC has emphasized such governance reforms in its surveillance exercises, but pressure on member states remains minimal.

  • Strengthen the preventive arm. One possibility would be linking the EDP and the MTOs, for instance by abrogating the EDP only after the MTO has been reached. Alternatively, MTOs could be treated as the reference value under the EDP for those countries deemed to have unsatisfactory debt dynamics. Both options would require an amendment to the EU Treaty, however.

  • Enhance the commitment value of MTOs. The process leading to the definition of MTOs is rather opaque, reducing their public visibility and, correspondingly, their signaling value and political relevance. Establishing a simpler link between MTOs and public debt could help put them at the core of EU budgetary surveillance and, more generally, public scrutiny.

Box 6.Is the Stability and Growth Pact an Obstacle to Adequate Fiscal Stabilization?

In principle, the Stability and Growth Pact (SGP) allows for considerable leeway in using fiscal policy for macroeconomic stabilization:

  • Escape clause (definition of excessive deficit). A deficit above the reference value of 3 percent of GDP may not be considered “excessive” if (1) the deficit is exceptional (i.e., due to unusual events outside the control of the member state); (2) temporary; and (3) close to the reference value. Negative GDP growth qualifies as an exceptional event.

  • Excessive deficit and enforcement procedure (EDP). If the deficit of year t has been deemed excessive, the initial deadline for correction (normally t + 2) can be extended by one year (to t + 3) in case of “special circumstances,” including negative GDP growth, and “all other relevant factors.” Moreover, the EDP can be put in abeyance, effectively leading to one-year extensions of the initial deadline. There are two conditions for abeyance: (1) the member state complied with the previous Council’s recommendations or notices; and (2) “unexpected adverse economic events” with major budgetary effects occurred after the recommendation or notice was issued. Again, “all other relevant factors” can be taken into account by the European Commission (EC) and the Council in extending the deadline for correcting the excessive deficit.

However, once a country operates under the EDP, there are limits to this flexibility, as an improvement in the structural balance is required. As a rule, any country under the EDP will face recommendations (or notices—which are more specific and legally binding) requiring an improvement in the country’s structural balance (cyclically adjusted and net of one-offs) by at least 0.5 percent of GDP. In practice, this rule applies from t + 2 onward (the first year after the excessive deficit has been identified), and it precludes the possibility of discretionary fiscal stimuli in countries that are under the EDP in the preceding year (automatic stabilizers would still be allowed). In practice, it is reasonable to expect that a continuation of financial sector stress and anemic or negative growth into 2010 would lead to the extension of deadlines under the EDP. In this case, a country experiencing an excessive deficit in 2009 would not be expected to correct it before 2013 at the earliest; however structural adjustment would be expected to begin in 2011 at the latest.

The EC has clearly indicated that the EDP should not be an obstacle to an adequate fiscal response to the downturn. In December 2008, the EC proposed the European Economic Recovery Plan, subsequently endorsed by the European Council, calling for a discretionary stimulus of about 1.5 percent of European Union GDP. In presenting the plan, the EC stated that “for Member States considered to be in excessive deficit, corrective action will have to be taken in time frames consistent with the recovery of the economy.”1

Note: The main authors of this box are Xavier Debrun and Jean-Jacques Hallaert.

1 In a similar vein, in considering the continued breach of the 3 percent target by the United Kingdom already operating under an EDP, the EC invited the British authorities to “proceed in financial year 2009/10 with the stimulus measures consistent with the European Recovery Plan while avoiding any further deterioration of public finances.”

Conclusions and Policy Implications

In the face of a crisis of historic proportions, countries need to continue to support financial systems and the economy. Failure to do so would result in a prolonged recession, additional financial losses, and a further worsening of fiscal accounts. With limited scope for further stimulus through monetary policy, fiscal policy remains the main option available for policymakers to preempt such a downward spiral. It is crucial, though, for governments to try to maximize the effectiveness of fiscal support, while limiting the impact of such support on sustainability.

In this respect, the analysis presented in the chapter suggests the following:

  • With regard to the policy instruments mix, spending on infrastructure is likely to exhibit a larger growth impact than tax cuts and transfers increases, despite having the longest implementation lags.

  • That said, under the present tight credit conditions and low collateral values, tax cuts and transfer increases can be more effective than in the precrisis period in fueling aggregate demand, as the share of credit-constrained agents—whose consumption pattern is highly sensitive to current disposable income—is rising.

  • By the same argument, targeting fiscal support to specific groups of consumers and firms that are mostly credit constrained would enhance the effectiveness of the stimulus.

  • Given the importance of intraregional trade in Europe, coordination of expansionary measures would greatly amplify the size of fiscal multipliers.

  • If fiscal sustainability is perceived to be in jeopardy, market interest rate increases could partly offset the expansionary effects of stimulus packages. Therefore, support measures should be accompanied by a clear and credible strategy to ensure fiscal sustainability, including a plan to withdraw the stimulus as the crisis abates.

  • Fiscal frameworks, including national and subnational fiscal rule as well as the SGP, can and should be strengthened to anchor expectations of fiscal sustainability and foster the implementation of medium-term strategies.

Note: The main authors of this chapter are Silvia Sgherri and Edda Zoli.

It has to be stressed, though, that—given the severity of the crisis and the potential risks of an economic meltdown—public finances would worsen significantly even in the absence of government intervention.

In several countries, central banks have extended assistance to financial institutions through credit lines, purchase of assets, asset swap, and liquidity provisions without direct treasury funding. While such operations do not require up-front treasury funding, they could eventually generate fiscal costs.

These estimates are obtained by applying the expected default frequency implied credit default swap spreads—which are indicators of the “insurance” premium for providing the guarantee—to the guaranteed amounts. For details, see IMF (2009b and 2009c).

Some estimates, based on the assumption that governments will provide implicit guarantees to all systemic institutions, suggest that the potential fiscal costs could be on average around 3–10 percent of GDP, cumulative for 2009–13 for the advanced European economies (IMF, 2009b and 2009c).

For example, France, Germany, and Italy introduced measures that entail permanent reductions in personal income tax or in indirect taxation (IMF, 2009c, Table 13).

Monetary policy effectiveness is also somewhat impaired (see Box 1 in Chapter 1) and policy rates are very low almost everywhere, leaving limited scope for further easing.

For recent studies on fiscal policy effectiveness as a countercyclical tool, see Blanchard and Perotti (2002), Christiansen (2008), IMF (2008b), and Freedman and others (forthcoming).

For an analysis of the stimulating impact of different fiscal instruments, see Freedman and others (2009).

On this point, see, for example, Perotti (2002).

In a recent empirical study, Bayoumi and Sgherri (2009) explore the relationship between changes over time in policy effectiveness and time variation in both households’ planning horizons and the degree of persistence of policy shocks.

See Kumhof and Laxton (2007) for a more detailed discussion of GIMF’s structure and properties.

At the same time, though, the fact that GIMF is a rational expectations dynamic stochastic general equilibrium model rules out—a priori—any meltdown scenario over the medium term: expectations are, indeed, well anchored, and policy actions are assumed to be fully credible and anticipated by (rational) economic agents.

In the period following the temporary fiscal action, lump-sum transfers adjust to return government debt back to its baseline value over time. They are assumed to do so very gradually to minimize the effects of fiscal consolidation on postcrisis GDP.

Following the seminal work of Campbell and Mankiw (1989), several papers have estimated of the share of liquidity-constrained consumers. Available estimates for euro area countries over the 1960–early 1980s period range around 50 percent (Jappelli and Pagano, 1989).

For an analysis and empirical investigation of fiscal policy spillovers in Europe, see Beetsma, Giuliodori, and Klaassen (2006), and references therein.

On this point, see also Krugman (2008).

The underpinning analytical work for this section is presented in Sgherri and Zoli (forthcoming).

Sovereign debt ratings for Greece, Ireland, Portugal, and Spain have been downgraded by Standard & Poor’s in recent months.

The common risk factor—which is not directly observable—is assumed to be represented by a dynamic stochastic process. It is also assumed that sovereign spreads are simultaneously determined within a multivariate generalized autoregressive conditional heteroscedasticity framework, capturing fat-tail noise. The model is estimated using a nonlinear Bayesian estimation technique based on the model’s likelihood function.

The empirical literature finds that larger government deficits tend to increase long-term interest rates. Typical estimates show that a persistent increase in debt equal to 1 percent of GDP increases long-term real interest rates by between 1 and 6 basis points. The effect of a persistent increase in deficits of the same magnitude is associated with a 10- to 60-basis point increase in long-term real interest rates (Engen and Hubbard, 2004; Gale and Orszag, 2004; and Ardagna, Caselli, and Lane, 2004).

Prepared with the help of Manal Fouad and Edouard Martin drawing on IMF (2009b).

This section was prepared by Xavier Debrun, Jean-Jacques Hallaert, and Helge Berger.

For an analysis and empirical evidence on national fiscal rules in EU members, see Debrun and others (2008).

MTOs are country specific and reflect the requirement to (1) keep a sufficient safety margin with respect to the 3 percent limit (depending on the size of automatic stabilizers); (2) ensure stable debt dynamics around prudent levels (or rapidly declining debt toward these levels); and (3) also take into account public investment needs.

See, for instance, the discussion in Beetsma and Debrun (2007).

    Other Resources Citing This Publication