Public Goods and Minimum Provision Levels: Does the Institutional Formation Affect Cooperation?*

• Author: Emil Persson,Peter Martinsson
• DOI: http://doi.org/10.1111/sjoe.12321
• Date: 01 October 2019
• Published date: 01 October 2019

Scand. J. of Economics 121(4), 1473–1499, 2019

DOI: 10.1111/sjoe.12321

Public Goods and Minimum Provision

Levels: Does the Institutional Formation

Affect Cooperation?*

Peter Martinsson†

University of Gothenburg, SE-40530 Gothenburg, Sweden

peter.martinsson@economics.gu.se

Emil Persson

Link¨oping University,SE-58183 Link ¨oping,Sweden

emil.persson@liu.se

Abstract

We investigate experimentally the role of institutional formation on the implementation of a

binding minimum contribution level to a public good. Groups either face the minimum level

exogenously imposed bya central authority, or are allowed to decide for themselves by means of

a group vote whether a minimum level should be implemented.We find that a binding minimum

contribution level has a positiveand substantially significant effect on cooperation. Interestingly,

we do not find an additional positive effect of democracy in the context of our experiment;the

minimum-levelintervention is as effective when exogenouslyimplemented as when endogenously

chosen.

Keywords: Experiment; minimum level; public goods; voting

JEL classification:C91; D72; H41

I. Introduction

Free riding is a frequently discussed reason for creating institutions to

increase contributions to public goods. Many of these institutions focus

specifically on the free riders or those who contribute a small amount to the

public good, for example, by allowing punishment.1In real life, however, a

frequently used institution to combat free riders is to require some minimum

*Weare grateful to Martin Kocher,H ˚akanHolm, Kjell Arne Brekke, Åsa L ¨ofgren,Martin Sj ¨ostedt,

two anonymous reviewers, and seminar participants for valuable comments and suggestions.

We also thank Lisa Bj¨ork and Andrea Martinangeli for excellent research assistant. Financial

support from Formas (COMMONs) and Vetenskapsr˚adet (ref. 348 2013-6348) is gratefully

acknowledged.

†Also affiliated with Link¨oping University.

1There are also experiments focusing on incentivizing people to give a high amount byallowing

rewards (e.g., Sefton et al., 2007).

C

The editors of The Scandinavian Journal of Economics 2018.

1474 Public goods and minimum provision levels

contribution level. A centralized institution with exogenous imposition of

rules and regulations of minimum levels is, in many cases, the only feasible

option for policy-makers. This is because letting individuals themselves

be responsible for punishment, rewards, and potentially ostracizing others

is not possible in many situations, including areas such as taxation and

environment (e.g., see an overview in Chaudhuri, 2011). For example, when

it comes to the environment, there are certain standards for emissions from

cars and for recycling at household level. The most common example is

probably the financing of public goods, such as health care and schools,

through taxation. Entrance fees to museums or national parks, and workload

in teams, are other examples. The focus of this paper is whether a minimum

level per se helps to increase contributions to a public good. Alternatively,

these public goods could be provided in a regime without any minimum

standards for cars or entrance fees, and could rely solely on voluntary

contributions. From an institutional design point of view, the two key

questions are whether a binding minimum level results in crowding out

and whether the way the institution is implemented affects contribution

levels. The objective of this paper is to investigate, by using a public goods

experiment, the effect of imposing a minimum contribution level for a

public good, and the effect of the mode of implementation (i.e., whether

the way the minimum level is introduced, either endogenously by majority

voting or exogenously, affects contributions).

Binding minimum contribution levels have been studied experimentally

with mixed results in terms of how levels of contributions to public goods

have been affected. For instance, Andreoni (1993) and Gronberg et al.

(2012) find that minimum levels increase contributions in public goods

games with concave payoff functions, and Chan et al. (2002) find the

same effect but also that it wears off as the minimum level is increased.2

A related body of literature focuses on the effects of letting subjects in

public goods experiments both suggest and subsequently agree on the level

for contribution obligations (e.g., Kroll et al., 2007; Dannenberg, 2012;

Dannenberg et al., 2014; Kesternich et al., 2014, 2018; Kube et al., 2015;

Gallier et al., 2017).3The two papers closest to our paper are Kocher

et al. (2016) and Keser et al. (2017). Kocher et al. (2016) investigate

the effect of exogenous minimum levels and also the effect of having

2For publicgoods experiments using non-binding minimum levels, see, for example, Bochet and

Putterman (2009), Galbiati and Vertova (2008), and Riedel and Schildberg-H¨orisch (2013).

3A related paper is Gerber et al. (2013), who build on Kosfeld et al. (2009), and examine

minimum requirements with respect to coalition formation in the context of cooperation. See

also Putterman et al. (2011) who let subjects vote on the parameters of a centralized punishment

scheme for public good provision, and Ertan et al. (2009) and Noussair and Tan (2011) who let

subjects vote on the conditions for when peer-punishment should be allowed.

C

The editors of The Scandinavian Journal of Economics 2018.