Agénor, Pierre-Richard, and Devrim Yilmaz. 2011. “The Simple Dynamics of Public Debt with Productive Public Goods.” Working Paper.
Alesina, Alberto, and Roberto Perotti, 1996, “Fiscal Adjustments in OECD Countries: Composition and Macroeconomic Effects”, NBER working paper series.
Almunia, Miguel, Agustin Benetrix, Barry Eichengreen, Kevin O’Rourke and Gisela Rua (2010), “From Great Depression to Great Credit Crisis: Similarities, Differences and Lessons,” Economic Policy 25.
Auerbach, Alan, and Yuriy Gorodnichenko, 2012a, “Measuring the Output Responses to Fiscal Policy,” American Economic Journal – Economic Policy 4(2012): pp. 1–27.
Auerbach, Alan, and Yuriy Gorodnichenko, 2012b, “Fiscal Multipliers in Recession and Expansion,” in Fiscal Policy after the Financial Crisis, Alberto Alesina and Francesco Giavazzi, eds., University of Chicago Press, 2012.
Batini, Nicoletta, Giovanni Callegari, and Giovanni Melina, 2012, “Successful Asuterity in the United States, Europe and Japan,” IMF Working Paper, No. 12/190.
Baum, Anja, Marcos Poplawski-Ribeiro, and Anke Weber, 2012, “Fiscal Multipliers and the State of the Economy,” IMF Working Paper, No. 12/286.
Cotteralli, Carlo, and Laura Jaramillo, 2012, “Walking Hand in hand: Fiscal Policy and Growth in Advanced Economies”, IMF Working Paper No. WP/12/137.
Hall, Robert E., 2009, “By How Much Does GDP Rise If the Government Buys More Output?” Brookings Papers on Economic Activity, Fall 2009, pp. 183–249.
Ilzetzki, Ethan, Enrique G. Mendoza and Carlos A. Végh, 2011, “How Big (Small?) are Fiscal Multipliers?”, IMF Working Paper, No. WP/11/52.
International Monetary Fund, 2010, “Will It Hurt? Macroeconomic Effects of Fiscal Consolidation”, World Economic Outlook 2010 (Washington, DC: IMF).
Michael Woodford, 2011. ”Simple Analytics of the Government Expenditure Multiplier,” American Economic Journal: Macroeconomics, American Economic Association, vol. 3(1), pages 1–35, January.
Padoan, Pier Carlo, Urban Sila, and Paul van den Noord, 2012, “Avoiding Debt Traps: Fiscal consolidation, financial backstops and structural reforms”, OECD Journal: Economic Studies.
Yakita, Akira. 2008. “Sustainability of Public Debt, Public Capital Formation, and Endogenous Growth in an Overlapping Generations Setting.” Journal of Public Economics 92:897914.
This paper builds on a background note by the authors and Ferhan Salman prepared for the 2012 IMF Crisis Programs Review. We are grateful to Abebe Selassie, Ali Abbas, Anke Weber, Daniel Leigh, Donal McGettigan, Paolo Mauro, Helge Berger, Rakesh Mohan, seminar participants of the Fund Fiscal Macro Working Group and the Fiscal Policy Group in the Strategy, Policy and Review department, and colleagues in the Fiscal Affairs Department for useful comments and discussions. We thank Malika Pant and Trung Thanh Bui for outstanding research assistance.
Recent exceptions include Batini, Callegari, and Melina (2012), Cherif and Hasanov (2012), and Eyraud and Weber (2013).
Although the framework may indirectly provide evidence on size of multipliers, by pointing to inconsistencies arising from particular multiplier assumptions, as discussed in Section IV.
We also tried to include the growth impact of debt overhang in the framework, using the findings from Cecchetti, Mohanty, and Zampolli (2011): once public debt reaches 85 percent of GDP, an additional 10 percentage point increase in the ratio of public debt to GDP is associated with a 17–18 basis point reduction in subsequent average annual growth. However, this has only a marginal impact on our results, and hence is not reported.
See also IMF Staff Position Note, SPN/09/11.
Standard aggregate cyclical adjustment methodology effectively assumes that, absent structural changes, nominal revenues would move in line with nominal actual GDP, while nominal expenditures would move in line with nominal potential GDP. Thus if revenues fall faster than GDP and measures are taken to compensate, they will show as fiscal tightening from the measures perspective but as no change in the structural balance. Similarly, if nominal potential GDP falls, either from the real component or the deflator, then measures would be needed to reduce nominal expenditures correspondingly, to remain at zero structural change. It is also possible for these factors to reverse (especially in an expansion) so the change in structural balance would exceed identified measures.
Thus, assuming a multiplier of 1, a 1 percent permanent structural tightening would reduce the level of output by 0.8 percent in the current year, reaching 1.0 percent in the second year, and then 0.8 percent, 0.6 percent, 0.4 percent, 0.2 percent, and finally reverting to zero level effect in subsequent years.
CBO (2004) presents a summary of methodologies for estimating potential GDP.
To calculate debt levels, one needs not only the structural primary balance and interest payment, but also the automatic stabilizer. For this purpose, we assume, as a common practice in the literature, that revenues change in proportion to real output and hence the revenue-to-output ratio remains constant. This implies that all fiscal consolidation is from spending cuts. This is not a critical assumption—it can be adjusted depending on the country-specific composition of fiscal adjustment measures.
Eyraud and Weber (2013) provide an extensive discussion on alternative measures of cyclically-adjusted debt ratios (CADRs), and express reservation about using debt-to-potential GDP as an appropriate CADR because this measure still has a cyclical component in the numerator (from past cyclical components of fiscal deficits). Therefore, they argue against using it to design short-term fiscal policy. This concern, though valid, is not very relevant for our purposes, as we do not propose to use debt-to-potential GDP as a fiscal anchor, but instead as a simple improvement over the headline debt-to-GDP ratio to monitor a country’s debt sustainability risks.
Technically, the output loss is slightly larger under the gradual adjustment scenario, because potential GDP is assumed to start growing from 2013—as a result, a cumulative adjustment of 10 percent (of potential GDP) over 2010-14 is slightly larger in level terms than a 10 percent adjustment over 2010-12, but the difference is negligible.
Auerbach and Gorodnichenko (2012a, 2012b), and IMF (2012), Delong and Summers (2012), Baum, Poplawski-Ribeiro, and Weber (2012), Batini, Callegari, and Melina (2012).
IMF (2010, 2012), Auerbach and Gorodnichenko (2012b), Blanchard and Leigh (2013).
We also considered a different variant where the multiplier is 2 when the negative output gap is wider than 4 percent, and 0.5 otherwise. Results from this alternative assumption are qualitatively the same, and hence are not reported here.
The result on total output loss is sensitive to the effective size of the multipliers used. With smaller effective multipliers (e.g., with a multiplier varying linearly from 0.5 to 2 under an output gap from 0 to -10 percent), the total output loss would be smaller under the even adjustment scenario.
The choice of the threshold does not affect the hysteresis effects under the front-loaded scenario much as the economy is driven immediately into deep recessions, or the “hysteresis zone”, anyway. However, the hysteresis effects under the gradual adjustment is more sensitive to the assumed threshold—a more negative threshold would reduce the likelihood that the economy is driven into the “hysteresis zone” and hence result in smaller hysteresis effects.
In practice, some countries may have already lost market access before interest rate reaches 12 percent. In this framework, for simplicity, we do not model the loss of market access explicitly—instead a potential debt crisis is illustrated by showing how the debt path could be explosive when the debt-interest rate feedback loop is in play.
Assuming fixed fiscal multipliers generates similar results, and hence not reported here.
Fiscal Monitor, April 2012, Chapter 3 “Easy Does It: The Appropriate Pace of Fiscal Consolidation”.
Please note that the potential output estimated using the HP100 filter is only as good as the team’s real GDP projections (on which the HP filter is applied). If, for instance, the real GDP projections are overly optimistic, the HP100 filter would not be able to correct the problem—all it does is to avoid double counting of the fiscal effect and provides a smoother trend.
In theory, there are various reasons why the multipliers could significantly exceed unity. For example, as the global economy remains weak and trading partners undertake fiscal consolidation simultaneously, external demand could not help offset the negative impact of fiscal austerity. In addition, the macro-financial feedback loop could play an important role: fiscal consolidation reduces growth with adverse effect on the financial sector, which, in turn, would reduce credit extension, hurting economic growth. Finally, some of the economies in need of large multi-year fiscal adjustment are in the euro zone and hence cannot rely on the exchange rate to partially absorb the negative growth impact of fiscal consolidation.
The peak-to-trough output change is the product of the change in fiscal stance and the fiscal multipliers. As the fiscal actions are the same under both the actual GDP (the blue line) and the implied output level (the red line), the huge discrepancy in the output collapse under the two paths can only be attributed to the wrong calibration of the fiscal multiplier.