The IMF and Economic Recovery: Is Fund Policy Contributing to Downside Risks?


The IMF’s most recent World Economic Outlook (WEO) projects world economic growth will slow, from 4.8 percent in 2010 to 4.2 percent next year.  Throughout the report, there are numerous concerns expressed about the “fragility” of the global economic recovery.  The Acting Chair of the Executive Board states that “[t]he recovery is losing momentum temporarily during the second half of 2010 and will likely remain weak in the first half of 2011, as extraordinary policy stimulus is gradually withdrawn.”1

The WEO expresses particular concern about the high-income countries, where “low consumer confidence and reduced household incomes and wealth are holding consumption down. . . .  Growth in these economies reached only about 3½ percent during the first half of 2010, a low rate considering that they are emerging from the deepest recession since World War II.  Their recoveries will remain fragile for as long as improving business investment does not translate into higher employment growth.”2

Chapter 3 of the WEO provides empirical research on the effect of fiscal consolidation, concluding that it “typically reduces output and raises unemployment in the short term,” with a one percent of GDP fiscal consolidation leading to a one percent fall in private demand.3  Although the report concludes that the decline in GDP is typically smaller, on the order of 0.5 percent, that is due to an increase in net exports; and the authors note that this cannot happen for all countries at once.

Furthermore, the report’s empirical research on trade flows finds that the high-income countries’ demand for imports will remain, for some time, below pre-crisis trends — even more than would be expected on the basis of reduced growth.  Since the U.S., Europe, and Japan are more than half of the world economy, this translates into significantly reduced demand as well as increased downside risks for low-and-middle-income countries, despite the latter’s much more rapid rebound from the world recession.

In view of these findings and considerations, one might expect a strong bias toward continuing fiscal stimulus in weak economies, and a bias against fiscal consolidation.  However, the IMF continues to support pro-cyclical policies in some countries, fiscal consolidation in many others, and clearly does not support central bank financing of fiscal stimulus — even in countries such as the United States — where the threat of high inflation is very remote.

Pro-cyclical Policies among IMF Borrowing Countries

Figure 1 shows the change in government expenditure from 2009 to 2010 for all 55 countries that currently have IMF programs.  For 32 of these countries, government spending has decreased.  Six of these countries had economies that either contracted this year (Antigua and Barbuda, Greece, Iceland, Romania, and Jamaica) or are projected to grow at less than one percent (Grenada).  We can expect that the spending cuts worsened the downturn or weak economy in these countries.

FIGURE 1: Change in Total Government Expenditure, 2009-2010, IMF Program Countries
Figure 1
Source: IMF 2010 and author’s calculations.

TABLE 1: Expansionary and Contractionary Elements of IMF Agreements, 2008-09
Table 1
Source: IMF agreements and authors’ calculations.

However, it is questionable whether spending cuts were appropriate in other countries as well.  Even for those economies that are in economic recovery in 2010, it is not clear that such cuts would be necessary or desirable.

One year ago we reviewed 41 IMF agreements that were in place at that time, which had been initially negotiated at various times during the world recession of 2008-2009.  We found 31 of 41 agreements to have included either pro-cyclical fiscal or monetary policy, with 15 having both.4  These are shown in Table 1.  This was an underestimate, since in most cases cyclically-adjusted budget balances were not available, and in some cases the direction of monetary policy could not be determined.  In some cases, the fiscal and/or monetary conditions of the agreements were relaxed, but since there is a time lag between the agreements and reviews, there was still damage from the mistakes.

A year later, of course, the situation is not directly comparable, since the vast majority of low-and- middle-income countries are in recovery; therefore there are very few countries with IMF agreements that could end up with contractionary, pro-cyclical fiscal or monetary policy.  Nonetheless, it is worth noting that a number of the IMF’s pro-cyclical policies in 2008-2009 were part of agreements that were based on over-optimistic projections.  For example, of 26 countries that had at least one IMF review by October 2009, 11 IMF reports lowered previous forecasts by at least 3 percentage points, and three of these had to correct forecasts that were overestimated by at least 7 percentage points.  These are huge errors, and they were made while the world recession was well under way, and there was considerable evidence that it would be severe.5

This underscores the need for the IMF to err on the side of caution this time, as its own WEO has emphasized the fragility of the current economic recovery and the preponderance of downside risks — so as not to repeat the pro-cyclical policy mistakes of 2008-2009.

Pro-cyclical Policies in Europe

Table 1 lists seven countries where the IMF is involved in the implementation of pro-cyclical policies in Europe.  In six of the seven countries the economy is actually projected to shrink for 2010; Hungary is also included because it is projected to grow by only 0.6 percent this year.

Greece, Latvia, and Romania are operating under IMF agreements.  Spain, Ireland, and Portugal do not have agreements with the IMF, but the Fund, in coordination with the European authorities, is involved in the determination of macroeconomic policy in these countries.

Figure 2 shows the fiscal consolidation underway in Greece, with the IMF’s projected budget deficits.  The biggest step was taken this year, with the budget deficit shrinking from 13.6 percent of GDP in 2009 to 7.9 percent of GDP in 2010, or 5.7 percentage points of GDP.  Not surprisingly, this has seriously worsened the recession in Greece.  For the first half of 2010, the Greek economy shrank at a 5.3 percent annual rate.  As shown in Figure 3, the IMF has downgraded its forecast for Greece’s projected 2010 growth from negative 2 to negative 4 percent, in just the 6 months since April.

FIGURE 2: Greece: General Government Budget Balance
Figure 2
Source: IMF 2010.

FIGURE 3: Greece: Annual GDP Growth
Figure 3
Source: IMF 2010.

Figure 4 shows the IMF projections for Greece’s economic growth through 2015.  As can be seen, Greece does not reach its 2008 level of GDP until after 2015.  However, there are large downside risks.  Figure 5 shows the IMF’s projections for Greece’s public debt.  The baseline projection, which is based on the IMF’s growth projections, shows the debt peaking at 144 percent of GDP, up from 115 percent of GDP in 2009.  It is not clear that the financial markets will see this as a sustainable debt burden — in fact the markets have currently priced in a high probability of a default/restructuring.6  Although the European Financial Stability Facility could potentially see Greece through any turbulence caused by the rising debt burden, since the European authorities have committed to requiring fiscal consolidation as a requirement for accessing these funds, any future crisis is likely to cause further GDP losses.  This is another reason that the IMF growth projections are likely to prove over-optimistic, as has often been the case when pro-cyclical policies are implemented.

FIGURE 4: Greece: Annual GDP (Index)
Figure 4
Source: IMF 2010.

FIGURE 5: Greece: Gross Public Debt
Figure 5
Source: IMF 2010.

Figure 6 shows unemployment in Greece, which rose from 7.5 percent in September 2008 (pre-crisis low) to 12.2 percent today.

FIGURE 6: Greece: Unemployment Rate
Figure 6
Source: Eurostat 2010.

One problem with pro-cyclical policies is that when a downward spiral is initiated, it is often difficult to predict where it will end.  This is especially the case for a country such as Greece, which does not have control over its exchange rate or monetary policy.  The most common exit from such a situation would involve a boost from net exports.

An extreme case of economic collapse in the face of pro-cyclical macroeconomic policy can be seen in the case of Latvia.  As shown in Figure 7, Latvia’s contraction of more than 25 percent for the two years 2008-2009 is a historic record, surpassed only by the four-year decline (1929-1933) of the United States’ Great Depression.  Latvia’s economy is projected to decline another one percent for 2010.  As can be seen in Figure 8, the IMF projects that Latvia is still well below its pre-crisis (2007) level of GDP in 2015.

FIGURE 7: Economic Crises: Lost Output
Figure 7
Sources: IMF International Financial Statistics Online, Author’s calculations, and Reinhart and Rogoff.

FIGURE 8: Latvia: Annual GDP (Index)
Figure 8
Source: IMF 2010.

Like Greece, Latvia has been trapped in a recession in which all of the major macroeconomic policies — the exchange rate, fiscal policy, and monetary policy — are either going in the wrong direction or cannot be utilized to help stimulate the economy.  This is because of the government’s commitment — supported by billions of dollars of loans from the IMF and the European authorities — to maintain its exchange rate peg to the Euro.  The overvalued exchange rate hurts the tradable goods sector, by overpricing exports and under-pricing imports.  It also lowers investment due to the fear of devaluation, which has caused spikes in interest rates as well as capital flight at various points in the past three years.7  Then there is the pro-cyclical fiscal policy, which can be seen in Figure 9.  Although the fiscal consolidation that was planned in the IMF’s first review (IMF 2009b), from 2009 to 2010, did not materialize, it is now being planned for the next two years, with a large deficit reduction of 4.3 percent of GDP from 2010 to 2011, and 5.8 percent of GDP for 2011-2012.  If these targets are achieved, it would not be surprising to see much lower growth than projected in Figure 8.

FIGURE 9: Latvia: General Government Budget Balance
Figure 9
Source: IMF 2010.

The problem for both the weaker Eurozone economies, as well as countries such as Latvia that are pursuing pro-cyclical policies, is that they are attempting to recover by means of an internal devaluation.8  This means that the economy must contract and unemployment must rise sufficiently in order to push down wages and other costs to the point where there is a significant real devaluation, while keeping the nominal exchange rate fixed.  So far, none of the countries that have attempted to do this have succeeded in moving the real effective exchange rate very much.  It is not clear that such a strategy will succeed even after a long time; we are likely to see a prolonged period of slow growth with high unemployment, or even a relapse into recession — as may be happening now in Ireland, Romania, and possibly Spain.  Greece, which has been subject to the most severe fiscal tightening, has had negative growth for seven consecutive quarters.

Spain provides a good example of a case where the public debt itself is not the source of the problem, but rather the pro-cyclical policies in response to recession.  In Spain, from 2000-2007, the gross debt-to-GDP ratio declined sharply, from 59.3 to 36.2 percent of GDP.  Net debt — a more relevant measure of the debt burden — had declined to 26.5 percent of GDP in 2007.9  Even after the crash, in 2009, interest payments on Spain’s debt were just 1.8 percent of GDP, a modest interest burden.  The cause of Spain’s current debt problems, as well as its unemployment and weak recovery, was thus not an over-expansion of government but the collapse of private demand.  The country had built up a large housing bubble that began to collapse in 2007, at the same time that the economy was hit with external shocks from the world recession.  Between 2000 and 2006, construction increased from 7.5 percent of GDP to a peak of 10.8 percent.  After the collapse, housing starts fell by more than 87 percent.  Spain also suffered from the collapse of an enormous stock market bubble: the stock market peaked at 125 percent of GDP in November 2007 and dropped to 54 percent of GDP a year later.  The wealth effect of this huge drop in stock values would be expected to be very large, in the range of a 1.3-1.75 percent fall-off in GDP.10

Figure 9 shows the planned fiscal tightening for Spain: 1.9 percent of GDP for 2010 and 2.4 percent of GDP for 2011.  After shrinking for six consecutive quarters during 2008 and 2009, Spain showed positive growth for the first two quarters of this year — although just barely, at 0.1 and 0.2 percent (quarter-over-quarter, not annualized).  The IMF is projecting -0.3 percent growth for 2010.

Are there feasible alternative policies to the fiscal austerity being imposed on Spain?  Figure 10 shows projections for Spain’s net debt under various scenarios.  The dark blue line shows the government’s baseline projections, assuming that the planned fiscal tightening takes place.  In this scenario, if the country does not lapse into prolonged recession or stagnation, and/or face large increases in its borrowing costs, net debt stabilizes at 64.3 percent of GDP in 2020.  Just above this line, the lightest blue line shows a scenario in which the government continues a modest fiscal stimulus of 3.9 percent of GDP over the next two years, instead of the planned fiscal tightening.  Based on standard estimates of multipliers and revenue elasticities, we project a net public debt of 68.3 percent of GDP in 2020.  Thus it is possible to greatly reduce the risk of prolonged and unnecessary recession with only a modest increase in long-term net public debt.

FIGURE 10: Spain: Net Public Debt
Figure 10
Source: Weisbrot and Montecino (2010).

Figure 10 also shows an alternative that would result in an even lower net public debt.  This would involve the European Central Bank buying the bonds necessary to finance the stimulus, creating reserves in the process and agreeing to refund the interest payments on this debt to the Spanish treasury.  This would add to Spain’s gross debt but not to its net debt.  This alternative, shown in Figure 10 by the line below the baseline scenario, would leave Spain with an even lower net debt burden of just 60.5 percent of GDP in 2020.  In the discussion below, we will explain why central bank financing of the stimulus is a feasible alternative in the economic conditions currently faced by a number of countries.

Spain’s unemployment rate has soared from 7.9 percent in May 2007 to 20.5 percent today.  There is little prospect of it returning to normal levels in the foreseeable future, under the projected pro-cyclical policies.

Ireland’s planned fiscal consolidation is shown in Figure 11.  The budget deficit is projected to shrink by an enormous 6.5 percent of GDP (from 17.1 to 11.2 percent of GDP) over the next year.  Further tightening is planned for 2012 (2.4 percent) and 2013 (1.3 percent), with consolidation continuing through 2015.

FIGURE 11: Ireland: General Government Budget Balance
Figure 11
Source: IMF 2010.

Although Ireland had positive growth for the first half of 2010, the IMF projects negative 0.3 percent growth for the year.  For 2011, the IMF projects growth of 2.3 percent, but this is difficult to believe given the massive fiscal consolidation taking place.  It is worth noting that when Ireland began its budget cuts at the end of 2008, the IMF projected 1 percent growth for 2009; the actual result was negative 10 percent.

Figure 12 shows Ireland’s unemployment rate, which has more than tripled from 4.4 percent in August 2007 to 13.9 percent today.

FIGURE 12: Ireland: Unemployment Rate
Figure 12
Source: Eurostat 2010.

Portugal is also undergoing fiscal consolidation while the economy is weak, as shown in Figure 13.  The budget deficit is projected to shrink by 2 percent of GDP for this year and another 1.9 percent of GDP next year.  Although the IMF projects growth of 1.1 percent for 2010, it is projecting no growth for next year (-0.05 percent).  Unemployment has also risen substantially in Portugal over the last two years, from 7.4 in March 2007 to 10.7 percent.

FIGURE 13: Portugal: General Government Budget Balance
Figure 13
Source: IMF 2010.

Romania had zero GDP growth in the second quarter of this year and the IMF projects that the economy will shrink for the year, by 1.9 percent.  Yet the budget deficit is projected to narrow by 2.4 percent of GDP over the next year, as shown in Figure 14.

FIGURE 14: Romania: General Government Budget Balance
Figure 14
Source: IMF 2010.

It is worth noting, in view of the discussion that follows, that inflation is low or negative in all of the above-mentioned countries, with the exception of Romania, where it is projected to fall from 7.9 percent this year to 3 percent in 2011.  This is shown in Figure 15.

FIGURE 15: Consumer Price Inflation, Annual Percent Change
Figure 15
Source: IMF 2010.

Financing Fiscal Stimulus in the World’s Largest Economies

As noted above in the example of Spain, it would be possible to finance a fiscal stimulus without adding to net debt, if the European Central Bank were to agree to such a program.  This is true for the entire Eurozone, and for the United States — which, together with the Eurozone and Japan, makes up more than half of the global economy.  And it is these largest economies, and the high- income countries as a group, that the IMF’s World Economic Outlook finds most vulnerable to downside risks in the near future.

In the United States, the Federal Reserve has already created reserves of more than $1.4 trillion since August 2007 and used these to purchase debt.  Although most of this debt consists of mortgage-backed securities from Fannie Mae and Freddie Mac, some consisted of U.S. Treasury securities, and there is no reason that the Fed could not create reserves to purchase more of these.  In fact, the Fed is currently considering another round of “quantitative easing” that is expected to involve such purchases.  But such quantitative easing would have little effect by itself, since U.S. 10-year treasury yields are already very low — about 2.5 percent — and a reduction of, e.g., another 0.1 or even 0.2 percent would not have much impact on the economy.  However, if the Fed were to create reserves to purchase bonds, with this borrowing used to finance a new fiscal stimulus, that would have much more effect on economic growth.  Although the financing of a fiscal stimulus in this manner is, for political reasons, not currently under consideration, it makes economic sense.11

Under current conditions, there is little or no risk of unwanted inflation resulting from such policy.  Consumer price inflation in the United States is currently at 1.4 percent, below the Federal Reserve’s implicit target of 2 percent.  This is despite the Fed’s prior addition of $1.4 trillion to its balance sheet over the last three years (inflation was negative in 2009).  The experience of Japan over the last two decades shows that it is possible, under certain conditions, to finance large amounts of deficit spending through the creation of reserves by the Central Bank.  Japan’s gross public debt is 226 percent of GDP, more than 100 percent of GDP higher than its net debt of 121 percent of GDP.  This is the result of this type of central bank financing for trillions of dollars worth of borrowing over the last two decades.  Since 1990, cumulative inflation in Japan has totaled only about 5 percent — that is the total over 20 years, not annually.

Public debt created in this manner does not add to the debt burden of the government, since interest collected by the Federal Reserve from this debt is refunded to the Treasury.  The net debt of the government, which is what matters with regard to the burden of the debt, does not increase.

IMF economists must understand the basic economics and accounting of these policies.  Yet the Fund has so far not given any consideration to central bank financing of additional stimulus, in spite of the WEO’s assessment of the fragility of the current recovery in the high-income economies.  Instead, it has emphasized the need for fiscal consolidation in the high-income countries — if not immediately as in the pro-cyclical policies described above, then beginning very soon: “fiscal adjustment needs to start in earnest in 2011.  Specific plans to cut future budget deficits are urgently needed now to create new room for fiscal policy maneuver.”  The Fund allows that “If global growth threatens to slow appreciably more than expected, countries with fiscal room could postpone some of the planned consolidation.”12  But that may be easier said than done, if the slowdown in the high-income countries gains momentum.  It would be better to err on the side of caution.  And since inflation is below target in the U.S., Eurozone, and Japan, the side of caution is the side of fiscal stimulus — even if it involves central bank financing.  The IMF should be exploring the best possible options, rather than opting for premature and risky fiscal consolidation.


1  IMF (2010), p. 161.

2  Ibid. p. XV.

3  Ibid. p. 93.

4  Weisbrot et al (2010).

5  Ibid, p. 4.

6  Hyde, “Junk Yields in Europe Tumble to Lowest Since Before Crisis.”

7  See Weisbrot and Ray (2010).

8  See Weisbrot and Ray (2010).

9  Net debt excludes debt owed to the government, for example to the central bank.  Since the interest on this debt is returned to the treasury, the net debt measure gives a better measure of the debt burden faced by the government.

10  This assumes a decline in consumption of 3-4 percent for the loss of stock market wealth, and takes into account that stocks in Spain were about 61.5 percent domestically owned during this period.

11  See Baker (2010) for more on this need for this kind of stimulus and its feasibility.

12  IMF (2010), p. XV.



Baker, Dean.  2010.  “The Myth of Expansionary Fiscal Austerity.”  Washington, DC: Center for Economic and Policy Research. Available at

Eurostat.  2010.  Population and Social Conditions Database.

Hyde, Caroline.  “Junk Yields in Europe Tumble to Lowest Since before Crisis.”  Bloomberg, October 11, 2010. Accessed October 20, 2010,

IMF.  2010.  World Economic Outlook Database.

IMF.  2010.  “World Economic Outlook: Recovery, Risk, and Rebalancing.”

IMF.  2010.  “Greece: First Review Under the Stand-By Arrangement.”  Country Report No. 10/286.

Reinhart, Carmen and Kenneth S. Rogoff.  2009.  “The Aftermath of Financial Crises.”  NBER.

Weisbrot, Mark and Juan A. Montecino.  2010.  “Alternatives to Fiscal Austerity in Spain.”  Washington, DC: Center for Economic and Policy Research.

Weisbrot, Mark and Rebecca Ray.  2010.  “Latvia’s Recession: The Cost of Adjustment with an ‘Internal Devaluation.'”  Washington, DC: Center for Economic and Policy Research.

Weisbrot, Mark, Rebecca Ray, Jake Johnston, Jose Antonio Cordero, and Juan Antonio Montecino.  2009.  “IMF-Supported Macroeconomic Policies and the World Recession: A Look at Forty-One Borrowing Countries.”  Washington, DC: Center for Economic and Policy Research.

Mark Weisbrot is an economist and Co-director of the Center for Economic and Policy Research in Washington, D.C.  Juan Montecino is a Research Assistant at CEPR.  This article was first published by CEPR in October 2010 under a Creative Commons license.

| Print