Relationships, Risk and Rents: Evidence from a Market for Ice
Tarek Ghani (WUSTL)
Tristan Reed (World Bank)

Abstract : Firms frequently engage in repeated trade despite the availability of alternative business partners. A large literature shows how such relationships affect market outcomes, but less is known about how changes in market structure affect these relationships. We study a market for an intermediate input--ice--in which customers--fishermen--are regularly loyal to retailers, who prioritize loyal customers for deliveries when supply is scarce from the monopolistic manufacturer. When entry of additional manufacturers reduces supply risk, retailers can no longer extract loyalty as a rent to abate the risk and switching between customers and retailers becomes more common. Retailers respond by expanding trade credit to customers, particularly previously loyal ones. We interpret this as evidence that supply risk and demand volatility can contribute to loyalty in relationships, particularly in developing countries. Further, we show that entry into ice manufacturing leads to substantial improvements in fishermen's productivity and reductions in the consumer price of fish, indicating that increased competition in low income economies can lead to improvements in productivity and welfare.

Paying Smallholders Not to Cut Down the Amazon Forest: Impact Evaluation of a Redd+ Pilot Project
Gabriela Simonet (INRA)
Julie Subervie (INRA)
Driss Ezzine-de-Blas (CIRAD)
Amy Duchelle (CIFOR)
Marina Cronberg (CIFOR)

Abstract : We estimate the effects of a REDD+ pilot project offering a mix of incentives including Payments for Environmental Services to reduce deforestation by smallholders in the Brazilian Amazon. We collected original data from 181 individual farmers. We use DID-matching and find evidence that supports the parallel trend assumption. We estimate that an average of 4 ha of forest have been saved on each participating farm in 2014, and that this conservation came at the expense of pastures rather than croplands. This amounts to a decrease in the deforestation rate of about 50 percent. We find no evidence of within-community spillovers. Finally, we use this estimate and the Social Cost of Carbon (SCC) to perform a cost-benefit analysis of the project.

Analyzing How a Growth in Energy Prices Impact on the Size of the Shadow Economies Around the World
Nikita Suslov (Novosibirsk State University)
Ekaterina Meltenisova (Novosibirsk State University)

Abstract : We assume that a growth of energy prices could create additional incentives for firms to conceal their incomes that results in the higher size of the shadow economy in a country. To verify this hypothesis we apply a formal analysis of the model where a representative firm attempts to ‘op-timize’ its concealed income in the context of a non-rigid outside control. We show that institu-tional improvements would allow the lower SE share in GDP. To test this hypothesis empirically, we construct regressions of the shares of the shadow economy in GDP over 2003-2008 using the estimations both available in the publications and ours calculated by Currency Demand Approach which we modified and applied, unlike the previous modifications, to cross-sectional data. A specific interaction variable used in our regression equations, being a combination of both tax burden variable and institutional quality indicator, allows considering an important fact that the high level of taxes assumes the shadow economy bigger in size in the counties with poor in-stitutional environment and vice versa – in those with sound environment due to a high supply of public goods and services. To calibrate the model, we applied the algorithm which allows con-sidering the fact that foreign currency (US dollars in our case) along with domestic one is in-cluded in the shadow economy’s transactions. Finally, the hypothesis suggested by us concerning the impact of relative energy prices on the size of the shadow economy in a country was verified on the basis of large samples of both cross-sectional and panel data. The estimations of the regression parameters which we obtained can demonstrate a stable character of their values de-spite the methods we applied to in our analysis.