Welfare Studies Blog

A Vote at the Opera: Why Politicians Like to Manipulate Highbrow Culture

Written by Professor Pieter Vanhuysse, Department of Political Science and Public Management

Are Politicians Really Done with Election-Time Tactics?
For a long time it has been a commonplace wisdom within political science that politicians in government possess the weapon (power), the motive (re-election) and the opportunity (election times) to try and attract votes by strategically manipulating public policies. There has been plenty of evidence since the 1960s and 1970s that politicians boost the economy through expansionary monetary and fiscal policies during election periods, when voters most need to be won over. This is the key argument of the important theory of political business cycles.

But in recent years, evidence of such tactics has been much less conclusive. Monetary and fiscal cycles have tended to be smaller and less robust in advanced Western democracies. Some have argued that this is because stronger domestic and supra-national (e.g. EU) institutional constraints on budget deficits and budgetary decisions today prevent politicians from manipulating the economy. Moreover, better-informed voters using more critical and accessible media may be less easily fooled.

But in an article in the European Journal of Political Economy, Markus Tepe (University of Oldenburg) and I argue that conclusions of the death of electoral manipulation tactics are premature. Even today politicians still need to be reelected. So they might merely shift the use of political business cycle mechanisms and other manipulation tactics away from monetary and fiscal policy as in the past, and towards other policy domains that are more easily manipulable, targetable, and timeable today. Tepe and I (2014) extend the political business cycle theory and apply it even to a seemingly implausible policy domain – highbrow culture. 

Taking the theory to the opera
At first sight, highbrow culture seems like a rather farfetched claim, as it is elitist, rather than populist or even popular. After all, parties do not really compete on culture. It is even unlikely that manipulating artist hiring might provide significant additional support from artists as voters (electoral clients): the number of actors, singers, dancers, and choir and orchestra members is very small relative to the size of the electorate.

And yet, we argue that politicians aiming to be reelected will often strategically manipulate subsidies for highbrow culture and the number of artists employed. The reason is that politicians can use artists as conduits of indirect competence signals aimed, through them, at their audiences. These audiences - highbrow culture-consuming voters – are obviously more numerous than the artists themselves. But they are politically even more important than their numbers. That is because they are, sociologically speaking, ‘high-multiplier’ voters.

We claim that highbrow culture consumers are electorally key voters because of two direct and two indirect reasons. First and second, highbrow culture consumers are more likely to be consummate, or at least involved, ‘political animals’ themselves: they are especially likely to go voting and to otherwise actively participate in politics. Third and fourth, highbrow culture consumers may be unusually able to influence other voters’ political behavior because they may be more interested in politics, and they may have more social network ties with other voters.

Highbrow cultural politics in Germany
We test our theoretical propositions using data on the German states between 1993 and 2010. Together with education and domestic security, culture policy is one of three policy domains that are still decided largely at the state, not federal, level in the Federal Republic of Germany. In the notorious model of German ‘cultural federalism,’ state-level and local-level politicians have joint legal and funding authority for theatres and operas. This gives them the ‘opportunity’ and the ‘means’ for manipulating cultural policies to gain votes.

And we do find evidence of such manipulation tactics. Controlling for a whole array of socio-economic variables, the number of actors, singers, dancers, and choir and orchestra members employed in German public theatres and orchestras tends to increase during state-level election years, and even more so during municipal-level election years. This positive association between state-level elections and artist numbers is in turn particularly strong in states in contemporary Western Germany, whereas the positive association with municipal-level elections is equally strong in both parts of Germany. In parallel papers, we have found similar evidence also of strong political manipulation of the hiring of teachers (Tepe and Vanhuysse 2009) and police officers (Tepe and Vanhuysse 2013 ) in Germany.

Looking for high-multiplier voters in the audience
But why would German politicians manipulate spending on operas and theatres? The number of votes they might buy directly from the artists seems electorally so negligible as to disqualify the ‘artists-as-clients’ thesis. In the period 1993-2010, there were on average only around twelve thousand persons working as dancers, singers, actors or musicians in public theatres and orchestras in all of Germany; scarcely a key voting constituency. We therefore checked our alternative ‘artists-as-conduits’ for signals to high-multiplier voters thesis with the German General Social Survey dataset ALLBUS. Even after controlling for a whole array of socio-economic variables, we find that Germans who consume more highbrow culture in their leisure time tend to go voting more frequently. We call this a direct electoral effect. They also more frequently tend to actively participate in politics in their leisure time. We dub this a direct political activity effect.

In substantive terms, increasing the frequency of cultural leisure time activities from its minimum to its maximum corresponds with an average increase in personal political activism by 23 percentage points, and in voting likelihood by 20 points. Compared to the impact of core socio-economic control measures such as age (35 points) and income (38 points), this indicates a considerable direct effect of cultural consumption on active political engagement. Moreover, as posited, these highbrow culture consumers tend to be more interested in politics (a political multiplier effect) and to have more social network ties with whom they might discuss politics (a sociological multiplier effect).

Increasing the frequency of cultural activities from its minimum to its maximum corresponds with an average increase in interest in politics by 19 percentage points, and in spending one’s leisure time with friends, neighbors and acquaintances (predominantly weak social ties) by 8 points and with family (strong social ties) by 5 points.

It’s the politics, stupid!
Even a seemingly elitist domain such as operas and theatres can be an attractive tool for political manipulation by vote-seeking policy-makers. Stronger institutional constraints and better-informed voters in advanced democracies today may lead incumbents to shift the use of political manipulation mechanisms to those policy domains that remain more easily targetable, and timeable.

Under these constraints, manipulating the number of artists is a politically efficient way to signal cultural competence to voters. In sum, politics remains politics. And political manipulation tactics continue to matter even in developed and accountable democracies such as Germany.

Tepe, Markus and Vanhuysse, Pieter (2014) ‘A Vote at the Opera? The Political Economy of Public Theatres and Orchestras in the German States,’ European Journal of Political Economy, Vol. 36, pp. 254-273.

Tepe, Markus and Vanhuysse, Pieter (2013), ‘Cops for Hire: The Political Economy of Police Employment across German States,’ Journal of Public Policy, 33 (2): 165-199.

Tepe, Markus and Vanhuysse, Pieter (2009), 'Educational Business Cycles: The Political Economy of Teacher Hiring across German States, 1992-2004,' Public Choice, 139 (1-2): 61-82.

To give you the best possible experience, this site uses cookies  Read more about cookies

Accept cookies