The electoral consequences of large fiscal adjustments

Alberto Alesina, Giampaolo Lecce, Dorian Carloni, 29 May 2010



The conventional wisdom about the political economy of fiscal adjustments is that deficit reduction policies cause recessions. Recessions, combined with the direct political costs of tax increases and spending cuts create serious problems for incumbent governments at the polls. The governments which raise taxes or cut spending – the latter especially in Europe – are voted out of office. Therefore Europe is doomed to failure. Fiscally responsible governments which try to cut spending will lose elections and fiscally irresponsible ones will survive and fiscal problems will persist or worsen.

Conventional but wrong

This view is conventional but vastly incorrect. First of all a large literature starting with Giavazzi and Pagano (1990) has shown that large fiscal adjustments can be expansionary. Recently Alesina and Ardagna (2010) have provided the latest piece of evidence on this issue showing that, indeed, fiscal adjustments based upon spending cuts are more successful (i.e. lead to more stable consolidations of the budget) and cause less contraction of the economy than tax increases. In fact more often than not spending cuts boost growth even in the very short run.

But what about the electoral consequences? Alesina et al. (1998) found no clear evidence suggesting that governments which reduce budget deficits decisively are routinely voted out of office. Some are. Many are not. But that paper stopped with data at the early nineties. What happened next? This is the topic of our work in progress (Alesina et al. 2010).

Here we look at the largest cases of fiscal adjustments in the last 25 years in Western Europe as compiled by Gros and Alcidi (2010) in a recent Vox column and we look at their political consequences1. The point of these authors was that large adjustments of the size needed today in Europe are economically feasible, but politically? We list as “government changes” the cases when there was a change in the political orientation of the government with or without an election during the adjustment period. We list as “no change” cases in which an election was followed by the reappointment of the same party or coalition or there was a new government of the same party without an election.

In the Gros and Alcidi sample we find 13 government changes and 26 no changes, thus the changes were 33% of the total. Is this high or low? In the sample of these countries from 1975 to today there were about 39% of government changes over the total. So there were slightly fewer changes during fiscal adjustments than in the entire sample. This difference is probably not statistically significant but certainly there is no evidence that fiscal adjustments lead to a higher likelihood of defeat. Also we do not find any pattern of more electoral losses for government which cut spending more than taxes. In fact the third column of Table 1 reports the share of change in deficit obtained with spending cuts; when it greater than 100 it means that taxes were actually cut during the adjustments. Of the three countries with a share of spending cuts about 100 two had no government change during the period (Ireland and Belgium), the third country (the Netherland) had two but in a decade. The three countries with the lowest cut in spending (Italy, France, and Greece) had 5 government changes and 7 no changes – a frequency of changes well higher than the average.

The bottom line is that it is possible for fiscally responsible governments to engage in large fiscal adjustments and survive politically. Moreover, acting on the spending side is no more costly that doing on the tax side. A sense of urgency because of impending crisis, a bit of time between the adjustment and the next election, good communication with the public, are ingredients that help.

Table 1. Fiscal adjustment and changes of government

Primary balance adjustment per year (% GDP)
% of total primary balance adjustment through cut in expenditure
Number of Government changes
Number of times no Government changes happen
Denmark 82-86
Greece 89-94
Sweden 93-98
Ireland 87-89
Portugal 80-86
Italy 92-97
UK 93-99
Finland 92-2000
France 93-97
Belgium 81-90
Austria 95-2001
Netherlands 90-2000
Government changes
No changes



Fiscal problems in Europe can be solved. Large adjustments based on the spending side work, they are not recessionary and are not the kiss of death for the governments implementing them. It can be done. If the fiscal crisis in Europe worsens, threatening the world recovery, we will know whom to blame: the current leaders of Europe.


Alcidi, Cinzia and Daniel Gros (2010), “The European experience with large fiscal adjustments”,, 28 April.

Alesina A and Ardagna S (2010), “Large Changes in Fiscal Policy: Taxes Versus Spending”, forthcoming in Tax Policy and The Economy.

Alesina A, D Carloni and G Lecce (2010), “The electoral consequences of large fiscal adjustments”, work in progress.

Alesina A, R Perotti, and J Tavares (1998), “The Political Economy of Fiscal Adjustments”, Brookings Papers on Economic Activity, Spring.

Giavazzi F, and M Pagano (1990), "Can Severe Fiscal Contractions Be Expansionary? Tales of Two Small European Countries”, NBER Macroeconomics Annual, MIT Press:95-122.

1 We have made two minor changes relative to Gros and Alcidi. The timing of the Irish and Italian adjustment which we report is slightly different than theirs and it is more consistent with the literature. In both cases the really decisive part of the adjustment started a year later than what Gros and Alcidi report. Only our data sources are different and the data do not perfectly correspond.

Topics: Europe's nations and regions, Politics and economics
Tags: Eurozone crisis, fiscal adjustment, Fiscal crisis


Authors' reply

Thank you for the constructive comments. The idea of this article is to present the main points of our work on the electoral consequences of fiscal adjustemnts. Our analysis is going beyond the evidence presented in the table above. In particular the complete work is going to deal with all the aspects you suggested: in our regression analysis we use both macroeconomic and political variables and try to control for the "stability" of governments. Our study considers the possibility of "reverse causality", and is aimed at addressing this problem using variables that control for the type of cabinet: mainly if a coalition cabinet is in power or not, and if the cabinet has a majority support in parliament or not.

Control Variables

Agree with the previous post. Controlling for the type of electoral system (proportionate vs majority, maybe through the share of seats in parliament of the major party) would shed more light on the significance and causality of this relationship.

Currency Devaluations

It would be useful to add data about currency devaluations in the adjustment periods. An adjustment including a nominal component through a currency devaluation could be seen by voters as more "fair" than an adjustment implemented exclusively through the real side of the economy. The former affects everybody, rich and poor, while the latter affects the poor much more. Controlling for devaluations, would the interpretation of the 13/26 result change, given that this tool is not available in the Eurozone?

Reverse Causality?

But are more stable governments not more likely to push for fiscal adjustments in the first place? Then your analysis does not show, as you claim, that fiscal adjustments do not increase the likelihood of electoral defeat. Your data could rather highlight the difficulties that unstable governments (that fear electoral defeat) have in implementing fiscal adjustment programmes. Solving Europe’s fiscal crisis will then not be as easy as you suggest.

Nathaniel Ropes Professor of Political Economy, Harvard University; and Research Fellow, CEPR
Graduate Student at University of California at Berkeley
Graduate student at New York University