Economic performance does not necessarily improve when under-performing governments are ejected by the voters

Elections are the “instruments of democracy” so long as voters hold incumbents accountable for their performance in office. Periodic elections in which the process of accountability works well should generate electoral incentives for politicians to pursue good economic policy during their term in office. Dani Marinova uncovers little systematic support for this logic. Unemployment was no lower when voters correctly perceive and assign blame for the state of the economy than in elections where voters misattribute economic performance.

Credit: Andrew_WrIter, CC BY 2.0

The question of electoral accountability – to what extent and under what conditions voters punish incumbents for past economic performance – has generated a host of empirical studies in the social sciences. And rightly so; the presence of accountability in elections is an essential ingredient for a functioning democracy. When voters correctly perceive and assign blame for the state of the economy, they hold incumbents accountable for their performance in office. What is more, as voters punish underperforming policymakers and reward the stewards of a growing economy, they create electoral incentives for incoming officeholders to pursue good economic policy.

Thus far, scholars have largely turned their back on what difference, if any, the presence of electoral accountability makes for policy outcomes. It is often unappreciated that while voters decide based on their evaluations of past performance, the vote is ultimately about who will govern in the future. Electoral accountability is desirable because it keeps politicians in line after the election. Under the threat of losing office, elected officials ought to deliver desirable policy outputs, such as low inflation and jobs. Electoral accountability should therefore not be a mere exercise of punishment for past deeds. A strong mass-elite linkage ought to bring about tangible economic goods after the election.

Even though a strong link between accountability and the economy is normatively desirable, much stands in the way. For starters, it is not clear what level of punishment supplies sufficient incentives to politicians to pursue sound economic policy. Is it enough for the incumbent’s vote or seat share to shrink? Or does only a strong signal, e.g., loss of governing power, provide sufficient incentives to elected officials? And second, even when elected officials have all the right incentives to perform at their best, the role of globalization should not be underestimated. To varying extent, the forces of global markets undermine national governments’ ability to exercise unilateral control over the economy, thus casting doubt on any post-election economic outcomes accountability can guarantee.

I took a first stab at these questions in a paper presented at the Elections, Public Opinion and Parties meeting in Edinburgh this fall. I traced how accountability in elections shaped subsequent economic performance in eighteen developed democracies over thirty years. I argued that if we are to find evidence of elections making a difference in policy outputs, it should be precisely in these established democracies. I focused on the narrow case of unemployment as it has been shown to have a decisive impact on voters’ decision and incumbents’ electoral fortunes. I also examined how ties to the global economy and the strength of electoral signals mitigate the effects of accountability.

Take the most recent UK elections as an example. The incumbent Labour Party at the time oversaw a sharp increase in the levels of unemployment. Between 2005 and 2010 the number of unemployed increased from 1.44m to 2.47m, or a staggering 70%. In May of 2010, voters punished the Brown government for the poor state of the economy. Labour lost ninety seats in the House of Commons and yielded power to the coalition government of David Cameron and Nick Clegg. Having successfully held incumbent accountable in the most recent election, did voters see economic improvements in the years after? In three out of the four years since 2010, the number of unemployed has stayed at or above 2010 levels when the coalition government took office. Only in the past year has a drop in unemployment registered, from 2.49m in August of 2013 to 1.97m in August of 2014, or roughly 20%.

By examining cross-national differences and within-country variation over time, I was able to test for a systematic effect of accountability in elections on changes in unemployment. The results lent little empirical support to the idea that electoral accountability brings about improvements in the national economy. Even after an election where voters effectively sanctioned incumbents, unemployment was no lower than in elections where voters were not able to assign blame for macroeconomic performance. This finding held after accounting for the strength of electoral signals voters send to incumbents (e.g., a decline in vote or seat share or loss of governing power) and the ties of the national economy to global economic markets. In short, the empirical findings suggest that electoral accountability does not pay off.

The analyses of this paper are a first cut at understanding how elections shape economic policy. The data address the research question only narrowly, where performance is studied as the national rate of unemployment. While accountability in elections does not seem to generate a drop in unemployment, it may well affect other aspects of economic performance. In future research I will examine the effects of accountability on economic, social and political performance. More importantly, I will search for the missing link between accountability and policy outcomes – the policy programs implemented by incumbents. It may well be that electoral accountability provides incentives to politicians to pursue policies they deem beneficial to the economy but which do not bring about tangible improvements in macroeconomic indicators. Studying the causal chain, from accountability to policy programs to policy outcomes, should shed light on the causal mechanism through which the economic vote shapes, or fails to shape, policy outputs.

If future research lends credence to these preliminary findings, then the implications for electoral democracy are grim. Accountability in elections has been heralded as the saving grace of democracy as it requires only minimal levels of knowledge about the state of the economy. If it is citizens’ best hope to shape policy outcomes, then the findings of this paper spell bad news for democratic theory and practice. Given that accountability has only a negligible effect on the quality of governance, then one must ask if periodic elections serve merely as an instrument of political legitimacy.

—

Note: this post represents the views of the author and not those of Democratic Audit or the LSE. Please read our comments policy before posting.

—

Dani M. Marinova is a postdoctoral researcher at the Hertie School of Governance in Berlin.