Cooperatives and happiness. Cross-country evidence on the role of relational capital

ABSTRACT

Why is the share of happy people higher in some countries than in their equally developed neighbours? We conjecture that the apparent contradiction might depend on a country’s endowment of relational capital, which we proxy empirically with the extent of cooperativeness. In particular, within the black box of social capital, we consider relational capital as the outcome of the civil economy paradigm and use cooperativeness as the macro and objective proxy of long term face-to-face interaction. Compiling an index of the importance of the cooperative sector, we test whether more cooperativeness associates with more happiness controlling for countries’ HDI and other control variables. Checking for endogeneity, using various country samples, and through different regression methods we find support for our hypothesis. This suggests that, indeed, an institutionalized cooperative culture can promote happiness.

KEYWORDS:

• cooperative sector
• happiness
• cross-country evidence
• relational capital
• social capital

JEL CLASSIFICATION:

• I31
• P13

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ For Australia, see also the specific debate emerged about this country (e.g., Blanchflower and Oswald 2005; Leigh and Wolfers 2006).

² The classic texts are from Adam Smith: ‘By directing that industry in such a manner as may be of the greatest value, [the merchant] intends only his own gain, and he is in this, as in many other cases, led by an invisible hand to promote an end which was no part of his intention. Nor is it always the worse for society that is was no part of it’ (Adam Smith, Wealth of Nations, 1776, 456).

³ When not otherwise specified by a year in parenthesis next to the country name, data refer to WVS (2009): Australia, Austria (EVS 2008), Belarus (EVS 2008), Belgium (EVS 2008), Brazil, Bulgaria, Canada, China, Croatia (EVS 2008), Cyprus, Czech Republic (EVS 2008), Denmark (EVS 2008), Estonia (EVS 2008), Finland, France, Germany (EVS 2008), Greece, Hungary (EVS 2008), India, Ireland (EVS 2008), Italy, Japan, South Korea, Latvia (EVS 2008), Lithuania (EVS 2008), Luxembourg (EVS 2008), Malaysia, Malta (EVS 2008), Moldova, Netherlands, New Zealand, Norway, Poland, Portugal (EVS 2008), Romania, Russia, Serbia, Singapore (WVS, 2010), Slovakia (EVS 2008), Slovenia (EVS 2008), Spain, Sweden, Switzerland, Taiwan, Turkey, Ukraine, United Kingdom, USA.

⁴ Austria, Belarus, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Moldova, Netherlands, Norway, Poland, Portugal, Romania, Russia, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, Ukraine, United Kingdom.

⁵ Australia, Austria, Belgium, Brazil, Canada, China, Denmark, Finland, France, Germany, India, Ireland, Italy, Japan, South Korea, Malaysia, Netherlands, New Zealand, Norway, Singapore, Spain, Sweden, Switzerland, Taiwan, UK, USA.

⁶ To make the ICA rank – based on 26 countries – comparable to the Coopseurope rank – based on 36 countries – we need to adjust it. We rescale the ICA rank multiplying it by 0.7222 = 26/36. Then the mean adjusted rank (MAR) is the mean of the adjusted-ICA rank for the 14 European countries which are observed both from ICA and Coopseurope. Alternatively, the MAR is either simply the adjusted-ICA rank for the 12 non-European countries observed in ICA or simply the Coopseurope rank for the 22 European countries for which we have no observations from ICA.

⁷ Following the influential paper by Djankov et al. (2002), the World Bank compiles data on the various factors of the ease of doing business. Unfortunately, the composite index is only available from 2013 onwards and we had to rely on its sub-components. We chose the two most prominent indicators: ‘days to start up a business’ and ‘number of procedures to start up a business’. As expected, the two are strongly correlated (their pairwise correlation is 0.626, significant at the 1%) and, noticing that the latter turned out insignificant in our regressions, we selected the former.

⁸ See La Ferrara (2002, 2003) for the negative effect of ethnic fragmentation on cooperation. See Clark and Bonggeun (2012), and Cox (2010) for the negative effects of language fractionalization on trust and cooperation.

⁹ A recent paper using population as an instrument is Rose (2013).

¹⁰ Since Coopr is a rank, we couldn’t calculate the same instrument for it. For it we use the first three instruments only.