Christoph Siemroth

A Field Experiment in Motivating Employee Ideas

We study the effects of a field experiment designed to motivate employee ideas, at a large technology company. Employees were encouraged to submit ideas on process and product improvements via an online system. In the experiment, the company randomized 19 account teams into treatment and control groups. Employees in treatment teams received rewards if their ideas were approved. Nothing changed for employees in control teams. Our main finding is that rewards substantially increased the quality of ideas submitted. Further, rewards increased participation in the suggestion system, but decreased the number of ideas per participating employee, with zero net effect on the total quantity of ideas. The broader participation base persisted even after the reward was discontinued, suggesting habituation. We find no evidence for motivational crowding out. Our findings suggest that rewards can improve innovation and creativity, and that there may be a tradeoff between the quantity and quality of ideas.

Crowdfunding, Efficiency, and Inequality

We show how decentralized individual investments can efficiently allocate capital to innovating firms via equity crowdfunding. We develop a model where consumers have privately known consumption preferences and may act as investors. Consumers identify worthwhile investments based on their own preferences and invest in firms whose product they like. In the presence of aggregate demand uncertainty, an efficient capital allocation is achieved if all groups of consumers have enough liquidity to invest. If some groups of consumers cannot invest, capital flows reflect preferences of liquid investors but not future demand. Comparing with traditional financing forms, crowdfunding in the absence of liquidity constraints can be superior unless traditional financiers are fully competitive and perfectly informed.

An experimental analysis of information acquisition in prediction markets

We study which factors in terms of trading environment and trader characteristics determine individual information acquisition in experimental asset markets. Traders with larger endowments, existing inconclusive information, lower risk aversion, and less experience in financial markets tend to acquire more information. Overall, we find that traders overacquire information, so that informed traders on average obtain negative profits net of information costs. Information acquisition and the associated losses do not diminish over time. This overacquisition phenomenon is inconsistent with predictions of rational expectations equilibrium, and we argue it resembles the overdissipation results from the contest literature. We find that more acquired information in the market leads to smaller differences between fundamental asset values and prices. Thus, the overacquisition phenomenon is a novel explanation for the high forecasting accuracy of prediction markets.

The Informational Content of Prices When Policy Makers React to Financial Markets

When can policy makers use policy-relevant information from financial market prices and how does policy affect price informativeness? I analyze a novel setting with noise where a policy maker tries to infer information about a state variable from prices to improve policy decisions, and policy in turn affects asset values. I derive a necessary and sufficient condition for the possibility of information revelation in equilibrium, which might not be possible if the policy reaction to prices punishes traders for revealing their information. If the policy maker is uninformed, then policy objectives do not change price informativeness, but they do if the policy maker has independent information about the state. I also analyze policy maker transparency, and find that policy makers with objectives having a large impact on asset values should publish their information before trading to make prices more informative. In other cases, intransparency can be optimal.

When Can Decision Makers Learn from Financial Market Prices

I analyze a general setting where a policy maker needs information that financial market traders have in order to implement her optimal policy, and market prices can potentially reveal this information. Policy decisions, in turn, affect asset values, hence forward looking traders may have incentives to withhold information. I derive a necessary and sufficient condition for the existence of fully revealing equilibria in competitive financial markets, which identifies all situations where learning from prices for policy purposes works, and where it does not. Full revelation may be impossible because the pricing problem is a self-defeating prophecy. Using the result, I demonstrate that some corporate prediction markets are ill-designed and punish traders for revealing their information, and show how to fix it. I also discuss the possibility of using market information for banking supervision and central banking, and the general problem of asset design.

An Experimental Analysis of Information Acquisition in Prediction Markets

We study which factors in terms of trading environment and trader characteristics determine individual information acquisition in experimental asset markets. Traders with larger endowments, existing inconclusive information, lower risk aversion, and less experience in financial markets tend to acquire more information. Overall, we find that traders overacquire information, so that informed traders on average obtain negative profits net of information costs. Information acquisition and the associated losses do not diminish over time. This overacquisition phenomenon is inconsistent with predictions of rational expectations equilibrium, and we argue it resembles the overdissipation results from the contest literature. We find that more acquired information in the market leads to smaller differences between fundamental asset values and prices. Thus, the overacquisition phenomenon is a novel explanation for the high forecasting accuracy of prediction markets.