We investigate the effect of introducing information about peer portfolios in an experimental Arrow-Debreu economy. Confirming the prediction of a general equilibrium model with inequality averse preferences, we find that peer information leads to reduced variation in payoffs within peer groups. Information also improves risk sharing, as the data suggests that experiencing earnings deviations from peers induces a shift to more balanced portfolios. In a treatment where we highlight the highest earner, we observe a reduction in risk sharing, while highlighting the lowest earner has no effects compared to providing neutral information. Our results indicate that the presence of social information and its framing is an important determinant of equilibrium in financial markets.

Previous research has documented strong peer effects in risk taking, but little is known about how such social influences affect market outcomes. The consequences of social interactions are hard to isolate in financial data, and theoretically it is not clear whether peer effects should increase or decrease risk sharing. We design an experimental asset market with multiple risky assets and study how exogenous variation in real-time information about the portfolios of peer group members affects aggregate and individual risk taking. We find that peer information ameliorates under-diversification that occurs in a market without such information. One reason is that peer information increases risk aversion and induces a concern for relative income position that may reduce or amplify risk taking, depending on whether the context highlights the most or least successful trader. Thus, contrary to conventional wisdom, we show that social interactions may help to reduce earnings volatility in financial markets, and we discuss implications for institutional design. JEL-codes: D53, D83, G11.

The recent financial crisis highlighted the importance of compensation schemes for excessive risk taking in the financial industry, with consequences on asset price bubbles (Rajan, 2006;Bebchuk, 2009). To foster financial stability, a number of reforms have been discussed. Bonus caps have received the most attention and are now being implemented in the European Union. Another proposition suggests giving bankers more skin in the game, by making them liable for losses.In an experimental setting in which investors can entrust their money to traders, we investigate how compensation schemes of traders affect liquidity provision and asset prices. Traders can trade assets that yield dividends and thus potentially high returns. As a consequence, investors face a trade-off between either entrusting their money and enjoying potentially higher returns but with the risk of dealing with an untrustworthy trader, or keeping their money and receiving a safe but low return.We study how subjects solve this trade-off under different compensation schemes. First, we vary the extent to which traders are liable for losses. The introduction of limited liability creates a conflict of interest between the trader and the investor, because they value assets differently. Since traders have no downside risk, they highly value assets and are willing to buy them at a high price. By contrast, investors suffer all the losses while they share the gains. They are thus willing to pay less than traders. Since they anticipate that traders are ready to pay more, they can restrict liquidity provision to bring prices more in line with their valuation. Introducing limited liability should thus result in higher asset prices or lower liquidity provision if investors discipline traders.We also study the effect of capping gains. A cap generates a different type of conflict of interest. It reduces the potential gains of the trader and increases those of the investor. Assets become less valuable to traders which can thus decrease the pressure on prices. At the same time, assets become more valuable to investors, who become willing to provide more liquidity. While caps can have the desired effect of reducing risk taking by traders, the higher liquidity could, at the same time, increase pressure on prices.First, we find that liquidity provision is lowest when traders are liable for losses. This is unexpected given the absence of conflict of interest. Trust should be the highest. Liquidity provision is higher in the presence of a cap and/or limited liability. This implies that investors fail to discipline traders in the presence of limited liability. Also, the cap seems to improve liquidity provision by making the asset more valuable to investors. Second, we find that asset prices are closer to the fundamental value when traders are liable for losses than under limited liability. This suggests that, as expected, traders take more risk when they are not liable for losses. Furthermore, the higher liquidity entrusted to traders increases asset prices. We indeed find...

We investigate the effect of leverage on bubbles in an asset market experiment. We expect higher leverage to produce larger bubbles because (i) it creates moral hazard in a setup with limited liability and (ii) it increases aggregate liquidity. Inconsistent with the moral hazard channel, which we test by holding aggregate liquidity constant, higher leverage does not produce larger bubbles. To understand this unexpected result, we run the same experiment with a different framing: instead of repaying debt, participants can earn a bonus. This bonus treatment produces larger bubbles, suggesting that more leveraged participants trade more cautiously to avoid default. Finally, bubbles are larger and increase over time when we keep leverage constant over time by injecting liquidity in the economy. Overall, these results suggest that higher leverage inflates bubbles not because of moral hazard but because of more abundant liquidity.

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in Non-Technical SummaryThe recent financial crisis highlighted the importance of compensation schemes for excessive risk taking in the financial industry, with consequences on asset price bubbles (Rajan, 2006;Bebchuk, 2009). To foster financial stability, a number of reforms have been discussed. Bonus caps have received the most attention and are now being implemented in the European Union. Another proposition suggests giving bankers more skin in the game, by making them liable for losses.In an experimental setting in which investors can entrust their money to traders, we investigate how compensation schemes of traders affect liquidity provision and asset prices. Traders can trade assets that yield dividends and thus potentially high returns. As a consequence, investors face a trade-off between either entrusting their money and enjoying potentially higher returns but with the risk of dealing with an untrustworthy trader, or keeping their money and receiving a safe but low return.We study how subjects solve this trade-off under different compensation schemes. First, we vary the extent to which traders are liable for losses. The introduction of limited liability creates a conflict of interest between the trader and the investor, because they value assets differently. Since traders have no downside risk, they highly value assets and are willing to buy them at a high price. By contrast, investors suffer all the losses while they share the gains. They are thus willing to pay less than traders. Since they anticipate that traders are ready to pay more, they can restrict liquidity provision to bring prices more in line with their valuation. Introducing limited liability should thus result in higher asset prices or lower liquidity provision if investors discipline traders.We also study the effect of capping gains. A cap generates a different type of conflict of interest. It reduces the potential gains of the trader and increases those of the investor. Assets become less valuable to traders which can thus decrease the pressure on prices. At the same time, assets become more valuable to investors, who become willing to provide more liquidity. While caps can have the desired effect of reducing risk taking by traders, the higher liquidity could, at the same time, increase pressure on prices.First, we find that liquidity provision is lowest when traders are liable for losses. This is unexpected given the absence of conflict of interest. Trust ...