PKSG Working Paper Series

Policy note PKWP1404

Government spending multipliers in contraction and expansion

Walid Qazizada, Engelbert Stockhammer

This policy note is based on Working Paper PKWP1404

The financial crisis that began in 2007 has brought an old question of macroeconomics to the forefront again: What should fiscal policy do in a recession? Should it counteract swings in private demand, accepting higher budget deficits, or should it aim at sustainable public balances? The answer to these questions depends among other things on how large the fiscal multiplier is.

There has been a tension between economic policy, economic theory and empirical economic research. While fiscal policy had a prominent place in the economic policy repertoire, with renaissnae of New Classical economics in the 1970s and 1980s the role of fiscal policy was downgraded. The Ricardian equivalence theorem argued that anticipated fiscal expansions would be ineffective because people expect future tax increases (Barro, 1989) and the ‘expansionary fiscal contractions’ argument even held that austerity could result in growth (Giavazzi and Pagano 1990). The New Keynesian response to this was that a part of the population would be credit constraint and will react to current income. Thus fiscal policy would work, but theoretically the New Keynesians conceded grounds to the Classicals. Prior to the crisis the dominant view, the so-called New Consensus Model, then assigned only a passive role to fiscal policy. Governments should let automatic stabilizers, such as unemployment benefits, work, but not use fiscal policy actively. However, this was never well founded in empirical research. The majority of studies had long shown that multipliers are substantially larger than zero; in fact they were thought to be reasonably close to one.

However, with the deep recession following the global financial crisis, this view has been put into question. At the peak of the crisis all countries resorted to expansionary fiscal policy, but within a few years, in most countries, notably the UK, fiscal orthodoxy is back and governments aim at running a balanced budget. The Keynesian idea of countercyclical policy is passé again. However, is this backed by economic research? The IMF, formerly a bastion of economic orthodoxy, has highlighted that existing macro models consistently underestimated fiscal multipliers (Blanchard and Leigh 2013) and rejected the validity of expansionary fiscal contractions (Guardjardo et al 2011). Moreover, the Keynesian argument has been developed further. Several studies have argued that multipliers will be different during a crisis from those of normal times. For example, de Long and Summers (2012) argue that in a depressed economy “the Keynesian multiplier is likely to be substantially greater than the relatively small value it is thought to have in normal times” (de Long and Summers 2012, 233f).

There is a large empirical literature estimating fiscal multipliers econometrically (Hemming et al 2002; Bouthevillain et al. 2009). Only very recently have there been attempts to provide estimates of the multiplier that differ during boom and downswing. There are various econometric challenges in identifying the multiplier Afonso et al. (2010) and Turini et al. (2012) use a panel two stage least square approach for large panels of advanced and developing economies. In Qazizada and Stockhammer (2014) we follow their approach and apply it to a panel for 21 OECD countries for the period 1979-2011. We control for standard growth theory variables such as a catching up term, population growth, export shocks, and inflation as well as for the degree of private sector debt and the short-term real interest rate. We find that the size of the multiplier does indeed differ substantially in the different phases of the business cycles. While expenditure multipliers are close to one in the upswing, they are substantially higher, around 3, in the downswing. Overall our results suggest that fiscal policy is a potent tool for countercyclical economic policies. These results are qualitatively consistent with the findings of Gechert and Rannenberg (2014), who conduct a meta analysis of 98 econometric studies and find expenditure multipliers close to one in times of growth, and multipliers substantially higher in times of recession.

There is thus a big gap between empirical economic research and the dominant economic policy view on the effectiveness of fiscal policy. While economic policy explicitly or implicitly assumes that fiscal multipliers are small or zero, economic research suggests they are much larger. An increase of government expenditures by £100 million is likely to increase national income by the same amount and by a lot more in times of underutilized resources. In other words, fiscal policy works well in times of crisis.



Afonso, A, Gruner, H, & Kolerus, C 2010, 'Fiscal policy and growth: do financial crises make a difference?', ECB Working Paper No. 1217

Barro, R (1989a) ‘The Ricardian approach to budget deficit’, The Journal of Economic

Perspectives, 3 (2), pp. 37-54

Blanchard, Olivier, Leigh, Daniel, 2013. Growth Forecast Errors and Fiscal Multipliers.

International Monetary Fund Working Paper WP13/1

Blyth, M,. 2013. Austerity: The History of a Dangerous Idea. Oxford University Press

Bouthevillain. C. et al (2009): Pros and cons of various fiscal measures to stimulate the economy, Banque Central de Luxembourg Working Paper 40

DeLong, B, Summers, L, 2012. Fiscal policy in a depressed economy. Brookings Papers on Economic Activity, 44, 1:33-297

Gechert, S., Rannenbert, R. (2014). Are Fiscal Multipliers Regime-Dependent? A Meta Regression Analysis, IMK Working Paper 139-2014,

Giavazzi, Fransesco; Pagano, Marco (1990). "Can  Severe Fiscal Contractions Be Expansionary?

Tales of Two Small European Countries". NBER Macroeconomics Annual 5: 75–111

Guajardo, Jaime; Leigh, Daniel; Pescatori, Andrea (2011). "Expansionary Austerity: New International Evidence". International Monetary Fund Working Paper WP11/158.

Hemming, Richard, Kell, Michael and Mahfouz, Selma (2002): The Effectiveness of Fiscal Policy in Stimulating Economic Activity – A Review of the Literature. IMF Working Paper 02/208

Qazizada, W, Stockhammer, E, 2015.  Government spending multipliers in contraction and expansion, International Review of Applied Economics 29, 2: 238-258

Turrini, A., Roeger, W. and Szekely, I. P. (2012) 'Banking crisis, output loss, and fiscal policy', CESifo Economic Studies, 58 (1), pp. 181-219.

Download: Policy Note PKWP1404

Download: Working Paper PKWP1404