This dissertation examines the existence of heterogeneous motives in markets, particularly how a tension between profit motives and other utility can shape outcomes for organizations and individuals. I explore this tension in the context of biases, organizational identity, and investment behavior. Each of the three empirical chapters employs decision-level data from a different online crowdfunding platform.

Academic researchers and the general public are increasingly interested in the phenomenon of "crowdfunding." The term, however, encompasses an incredibly diverse set of activities---ranging from the facilitation of for-profit start-up investments to the charitable funding of medical procedures. This diversity can make it difficult to generalize research insights from studies of any particular instance of the phenomenon. In the introductory chapter I develop a general framework for understanding the source of observed behavior on crowdfunding platforms given some simple assumptions about platform policies.The goal is to provide context for the subsequent chapters of the dissertation.

The first empirical chapter examines biases against demographic groups, which are typically explained by one of two mechanisms: either decision makers have a taste for one demographic group over another, or demographics are employed as informational proxies for other unobserved but economically important traits. These mechanisms are difficult to empirically untangle despite the theoretical and practical importance of separating them. I attempt to do so in a Chinese peer-to-peer lending market by leveraging a loan guarantee policy that reduces the economic rationale for lenders to discriminate on borrower demographics such as gender and geography. Comparison of pre- and post-policy periods therefore provides a fruitful tool for measuring the degree of taste versus informational bias. I find that female borrowers appear to receive a preferential informational bias but a negative taste bias, while lenders' geographic bias toward borrowers located in the same province appears to be driven predominately by informational processes and not taste. These findings have implications for multiple sets of decision makers and underscore the theoretical importance of accounting for motives.

Chapter two examines the potentially conflicting investment motives found on a non-profit hybrid identity crowdfunding platform, where simultaneous market-like and charity-like motives may lead lenders to respond differently to funding requests from entrepreneurs who appear to have high economic ability and high personal need. I survey actual lenders on the platform to measure their stated preferences for borrowers who fit each of these categories. I find that 1) lenders vary in their preference for these categories and this preference is correlated with their demographics, and 2) past loans made by lenders with an above-average preference for both need and ability were funded faster than loans in other categories. These results highlight how actors' preferences are largely endogenous to the market in which they are observed.

In the final chapter I present the results of a simple online experiment conducted in conjunction with a peer-to-peer lending website. Potential lenders were presented randomized versions of the platform's lender registration web page. The content of the page varied in whether it promoted the potential social benefit of lending versus only the financial benefit. No difference was found between the treatment and control groups. The experiment provides some insight into how lenders self-select into crowdfunding activity and may serve as a model for similar experiments on other platforms.