We investigate the possible explanations for variations in aggregate levels of participation in large-scale political demonstrations. A simple public choice inspired model is applied to data derived from the annual May Day demonstrations of the Danish labor movement and socialist parties taking place in Copenhagen in the period 1980–2011. The most important explanatory variables are variations in the weather conditions and consumer confidence, while political and socio-economic conditions exhibit no robust effects. As such accidental or non-political factors may be much more important for collective political action than usually acknowledged and possibly make changes in aggregate levels of political support seem erratic and unpredictable.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Price excludes VAT (USA)
Tax calculation will be finalised during checkout.
May 1st was chosen as the “International Workers’ Day” at the first Congress of the Second Socialist International (1889). The specific reason was to mark officially the Haymarket Massacre in Chicago, May 4th 1886, which took place following a number of rallies and strikes across the United States on May 1st 1886, although May Day had since at least the medieval period been used as a celebration for the coming of the spring. The Danish labor unions followed immediately (1890).
Quoted from “1. maj i en tid uden lokum i gården”, in Midtjyllands Avis, May 2nd 2009; author’s translation.
Demonstration data for the years before 1980 could be obtained from newspapers, but would then create problems vis-à-vis what data could be included for various important socio-economic control variables.
A case could be made for using the actual number of unemployed workers rather than the unemployment rate: (1) The dependent variable is the number of participants (rather than, say, a rate of all potential participants) and therefore it makes for a more symmetric modelling of the variables to use numbers where the two are comparable. (2) It makes intuitive sense that the number of unemployed workers may affect the number of workers turning out, whereas the unemployment rate is more sensitive to other factors (e.g., changes in the total size of the workforce). Nonetheless, at the suggestion of a reviewer the unemployment rate has been used, and as it turns out it actually does not matter: All statistical analyses have been run using the ratio of the number of unemployed persons relative to the total number of workers, and all analyses produce, for all practical and analytical purposes, similar results.
According to a 2009 survey 21.1% of Danes say that they will reconsider attending a May Day demonstration if the weather is bad (Wilke 2009).
A case could be made that we might expect a non-linear (inverted U) relationship between temperatures and number of participants, since at some point it might become so warm that other activities (e.g., going to the beach) might become preferable. For that purpose tests have been made by including both daily minimum and daily maximum temperatures, as well as squared versions of these variables. No robust results of this nature appear.
Alternatively it is also conceivable that a larger number of socialist parties might be the expression of a more fragmented, conflict-ridden left-wing and that the correlation therefore might be the reverse.
An alternative to including prosperity could be to include income inequalities (e.g., as measured by the Gini coefficient). However, Denmark is characterized by one of the most equal income distributions in the world and there are only very small changes in it over the sample period.
At least five other Danish left-wing parties have been known to be (or been suspected of) receiving smaller or larger amounts of funding from socialist regimes that fell or cut their collaboration with socialist parties abroad in the years 1989–1991 (e.g., the Soviet Union, China and Albania).
The results of models 16 and 17 are basically unchanged when and if average temperatures and sunshine hours are substituted for the good weather interaction term: In that case the best-fit models will contain the same statistically significant variables, with the same signs and roughly same sized coefficients. The explanatory power also is roughly the same (adj. R 2=0.647 and 0.599 respectively).
As expected average temperatures and number of sunshine hours are positively correlated, but as most people familiar with the Danish tempered coastal climate might guess, the correlation is not necessarily perfect and not at all of an extent so as to produce multicollinearity problems (r=0.37; p=0.04).
All models have been run using the unemployment rate, i.e., the ratio of number of unemployed to total number of workers. The best fit models (models 16–17) and almost all other model specifications produce virtually identical results, i.e., with no significant change in coefficients, statistical significance or overall explanatory power. Results are obtainable from the author.
In order to control for autocorrelation all of the reported models were run including the trend variable, except model 8 which includes the Post-Soviet dummy variable which for obvious reasons is strongly correlated with the trend variable (r=0.84; p=0.00). In no instance did this affect the overall results of the individual models or the importance of the weather variables. Results are obtainable from the author.
DeNardo, J. (1985). Powers in numbers: the political strategy of protest and rebellion. Princeton: Princeton University Press.
Downs, A. (1957). An economic theory of democracy. New York: Harper Collins.
Finkel, S. E., Muller, E. N., & Opp, K.-D. (1989). Personal influence, collective rationality and mass political action. American Political Science Review, 83(3), 885–903.
Goldstone, J. A. (1994). Is revolution individually rational? Groups and individuals in revolutionary collective action. Rationality and Society, 6(1), 139–166.
Gomez, B. T., Hansford, T. G., & Krause, G. A. (2007). The Republicans should pray for rain: weather, turnout, and voting in US Presidential elections. Journal of Politics, 69(3), 649–663.
Granovetter, M. (1978). Threshold models of collective behavior. American Journal of Sociology, 83(6), 1420–1443.
Knack, S. (1994). Does rain help the Republicans? Theory and evidence on turnout and the vote. Public Choice, 79(1–2), 187–209.
Kuran, T. (1989). Sparks and prairie fires: a theory of unanticipated political revolution. Public Choice, 61(1), 41–74.
Kuran, T. (1991). The East European revolution of 1989: is it surprising that we were surprised? American Economic Review (Papers and Proceedings), 81(2), 121–125.
Kuran, T. (1995). Private truths, public lies: the social consequences of preference falsification. Cambridge: Harvard University Press.
Kurrild-Klitgaard, P. (1997). Rational choice, collective action and the paradox of rebellion. Copenhagen: Institute of Political Science, University of Copenhagen & Copenhagen Political Studies Press.
Kurrild-Klitgaard, P. (2004). The paradox of rebellion. In C. K. Rowley & F. Schneider (Eds.), The encyclopedia of public choice (Vol. 2, pp. 403–406). Dordrecht: Kluwer.
Leeson, P. T. (2010). Rational choice, Round Robin, and rebellion: an institutional solution to the problems of revolution. Journal of Economic Behavior and Organization, 73(3), 297–307.
Lichbach, M. I. (1995). The rebel’s dilemma. Ann Arbor: University of Michigan Press.
LO (2006). 1. maj—derfor. København: Landsorganisationen i Danmark (LO). http://www.ugebreveta4.dk/~/media/LO/Fokus/1maj2009/1majderfor2006pdf.ashx.
Matsusaka, J. G., & Palda, F. (1999). Voter turnout: how much can we explain? Public Choice, 98, 431–446.
Merrifield, J. (1993). The institutional and political factors that influence voter turnout. Public Choice, 77(3), 657–667.
Mueller, D. C. (2003). Public choice III. Cambridge: Cambridge University Press.
Oberschall, A. R. (1994). Rational choice in collective protests. Rationality and Society, 6(1), 79–100.
Olson, M. ( 1971). The logic of collective action: Public goods and the theory of groups (2nd ed.). Cambridge: Harvard University Press.
Opp, K.-D. (1991). Processes of collective political action. Rationality and Society, 3(2), 215–251.
Tullock, G. (1971). The paradox of revolution. Public Choice, 11, 89–100.
Tullock, G. (1974). The social dilemma: the economics of war and revolution. Blacksburg: Center for Study of Public Choice.
Wilke (2009). Arbejdernes internationale kampdag lever endnu. Wilke Nyhedsbrev (maj).
An earlier version of the present article was presented at the Annual Meeting of the Public Choice Society, 12th–14th March 2010. I owe thanks to my former research assistants Sebastian Gibson and Henrik Dahl Rasmussen, to Christian Bjørnskov, Catherine Hafer, Kasper M. Hansen, Jacob G. Hariri, Pete Leeson, Peter Nedergaard, Bill Shughart, and two anonymous reviewers for helpful suggestions, to the Copenhagen Police Department for assistance with data on numbers of demonstrators, and to John Cappelen and the Danish Meteorological Institute for access to historical weather data.
About this article
Cite this article
Kurrild-Klitgaard, P. It’s the weather, stupid! Individual participation in collective May Day demonstrations. Public Choice 155, 251–271 (2013). https://doi.org/10.1007/s11127-012-9914-3