Essays on public good provision: experimental and empirical evidence

Abstract

This dissertation contains three different essays that study the provision of public goods. These three studies share the common overarching theme of how the nature of the public good can be expected to have a negative effect on its provision (development of climate mitigation technologies in Chapter 2, and local public goods in Chapter 3 and 4). These three studies also use empirical approaches that aim to find causal relationships, although the methodologies differ – Chapter 2 uses a quasi-experimental design, while Chapters 3 and 4 use laboratory experimental settings. Lastly, two of the chapters also study how this provision can be improved, by setting new environmental regulations (Chapter 2) or new transfer mechanisms (Chapter 3). Chapter 2 examines how environmental regulation can spur green innovation. By studying a tax reform introduced in Germany in 1999, this study analysis whether firms responded by developing new green technologies. By matching German firms with similar, foreign companies and using a difference-in-differences design, this chapter estimates causal effects of this reform. This study finds no evidence of an effect of the reform on green transport innovation, while small firms decrease the number of green energy technologies developed after the introduction of the reform, although this result is not robust to different specifications. Finally, some potential reasons for this null effect are discussed, such as exemptions and reductions in the tax rate. Chapter 3 and Chapter 4 investigate the role of networks in public good provision. Chapter 3 studies the provision of a weakest link public good in a circular network, where the provision of the good is calculated at a local level – formed by a participant’s direct connections in the network. The treatments vary (i) the endowment of participants, whether it is homogeneous or heterogeneous, (ii) in the case of heterogeneous endowment, how it is spatially distributed – alternating or clustered together – and (iii) whether participants can transfer funds freely to each other or through an intermediary common account. This chapter finds that participants tend to transfer a sizable part of their endowment to their direct neighbors. Endowment heterogeneity is detrimental to poorer participants, but only when they are clustered together, showing that spatial distribution matters for final payoff inequality. Using an intermediary account does not improve average provision, but it decreases inequality when participants are clustered together. Chapter 4 analyses the provision of local public goods in networks. In this setting, participants benefit directly from the investments in the public good done in their own location, as well as in the location of their direct neighbors. The treatments vary in terms of the (i) participants’ endowment – homogeneous or heterogeneous –, (ii) and in the case of heterogeneous endowment, whether participants with the same endowment are alternating or clustered together. This chapter finds that participants tend to invest a large share of their endowments on their direct neighbors’ locations. When participants are clustered together regarding their endowment, there is a net transfer from rich to poor participants, facilitated by the rich participants at the border of the cluster. Participants also show a reward (punishment) behavior, where they increase (reduce) the investments to their (un)cooperative neighbor.