VoxEU Column Frontiers of economic research

Volunteering and the State

Why do individuals volunteer? This seemingly personal question is not fully explained by individual characteristics. This column examines the state’s capability to affect individuals' decisions to volunteer. Macroeconomic stability increases volunteering, but higher confidence in government and democratisation reduce participation.

Volunteering is a widespread economic phenomenon with a significant impact on society. Organisations that benefit from volunteer hours offer social and health services, education and youth work, rescue, culture, recreation, religious, and many other services. From society's point of view, volunteering is a highly desirable leisure activity, and volunteers contribute significantly to economic welfare, which would otherwise require paid resources.

Incidence of volunteering

The share of the population offering voluntary labour varies considerably from country to country. Sweden exhibits the highest participation rate among OECD member countries – 56% of all Swedish adults volunteer – followed by the Slovak Republic (54%), the US (50%), Canada (48%), and the Netherlands (44%). The countries with the lowest participation rates (14% to 16%) are Poland, Japan, Spain, and Hungary.

Economic explanations for volunteering

To explain the motivation to volunteer, the economic literature has focused on determinants at the individual level. There are three basic explanations for volunteering. Menchik and Weisbrod (1987) suggest in their seminal paper that volunteering can either be treated as an ordinary consumption good (consumption hypothesis) or as a way of increasing an individual's income on the paid labour market over time (investment hypothesis). Freeman (1997) argues that organisations looking for productive volunteers address people with high human capital. This strategy is successful since volunteering is something that people feel morally obligated to do when asked (good conscience hypothesis).

Heterogeneity across countries

However, as the following figure shows, even after controlling for a comprehensive set of individual characteristics, there is large variation in rates of volunteering across countries.

Figure 1. Variation in volunteering

Figure 1 is based on a logit estimation of the participation in volunteering, based on almost 38,000 respondents from the European and World Values Survey with age, sex, family status, number of children, household income, education, size of place of residence, and labour market status as explanatory variables. The dots represent the distribution of individuals' propensity to volunteer, the diamonds show the country-specific predicted means of these distributions, and the x-symbols indicate the corresponding 25% and 75% confidence intervals. Predicted participation rates can directly be compared with countries' actual participation rates represented by triangles. This comparison shows that individual-level variables can only partly explain the pattern of volunteer behaviour. The unexplained remainder is substantial – the actual participation rate does not lie within the 25%–75% confidence interval of predicted values in more than half of all countries. For instance, the predicted participation rate for the US based on individual characteristics is 32.8% whereas the actual participation is 50%.

Volunteering in a wider context

These results indicate that the analysis of the motivation to volunteer has to be extended to a wider context including the social, economic, and political environments in which individuals live. In our recent paper, we focus on the role of the state and examine its capability to affect the individual's decision to volunteer (Hackl, Halla, and Pruckner 2009). For this purpose, we combine individual-level data from the European and World Values Survey with macroeconomic and political variables for OECD member countries.
We find evidence for crowding out of voluntary labour provision on three dimensions.

  • Fiscal crowding out: An increase in public social expenditure by 1% of GDP decreases an individual's probability of volunteering by about 2%.
  • Political consensus crowding out: A political consensus between the voter and the prime minister reduces the probability of volunteering by 1.8%. A distinct government ideological orientation leads those individuals who are sympathetic to the government to volunteer less.
  • Participatory crowding out: The more a government supports democratisation, the lower is the incentive for individuals to be involved in volunteering activities. An increase of Vanhanen's degree of democratisation by one index point decreases the probability of volunteering by 1%.

Similarly, redistribution policies have also negative consequences for the probability of volunteering. For the first time, this paper documents that – apart from individual-level impacts – both non-monetary and monetary government activities determine the provision of voluntary labour.

Social policy implications

What is the social policy implication of these results? Social policy measures and the promotion of democratisation are beneficial to society per se. However, their negative effects on volunteering behaviour need to be taken into account. Another striking result of this analysis is the necessity of economic stability as a prerequisite for volunteering engagement. This can be seen both on the micro and the macro level. Lower rates of inflation and unemployment, as well as income and employment on the individual level, contribute to greater volunteering. As a consequence, a government can make a contribution to the provision of volunteering activities through macroeconomic stabilisation policy.

Disaggregated measures of public social expenditure show that some components are complementary to volunteering. Governmental family support measures promote volunteering activities. This result is another argument in support of the importance of personal economic stability for volunteering. On the individual political level, we find that politically motivated people (voters) are more likely to volunteer. Right-wing people show a higher propensity to volunteer compared to their left-wing counterparts. These differences disappear, however, if we control for religiosity.

Finally, our results seem to be relevant for the debate on the measurement of social capital. Generally, the measurement of social capital includes four dimensions: (i) interpersonal trust, (ii) institutional trust (e.g. confidence in parliament), (iii) participation in civil society (e.g. volunteering), and (iv) trustworthiness of the respondents themselves. Our results show that increasing confidence in parliament reduces the participation in volunteering. The trade-off between increasing institutional trust and participation in civil society complicates the concept of social capital formation considerably and highlights the need for a more thorough analysis of the interrelation among the four components.


Freeman, Richard B. (1997), Working for Nothing: the Supply of Volunteer Labor, Journal of Labor Economics 15(1), 140-166.
Hackl, Franz, Martin Halla, and Gerald J. Pruckner (2009), Volunteering and the State, Johaness Kepler University of Linz Working Paper 0901, February
Menchik, Paul L. and Burton A. Weisbrod (1987), Volunteer Labour Supply, Journal of Public Economics 32(2), 159-83.

1,050 Reads