Emerging markets are characterized by volatile, but substantial returns that can easily exceed 75% per annum. Balancing these lofty returns are liquidity costs that, using the bid–ask spread as a basis, range from 1% for the Taiwanese market to over 47% for the Russian market. However, the paucity of bid–ask spread information across countries and time requires the use of liquidity estimates in emerging markets even though little is known about the efficacy of these estimates in measuring bid–ask spread costs. Using firm-level quoted bid–ask spreads as a basis, I find that price-based liquidity measures of Lesmond et al. [Review of Financial Studies 12 (1999) 1113] and Roll [Journal of Finance 39 (1984) 1127] perform better at representing cross-country liquidity effects than do volume based liquidity measures. Within-country liquidity is best measured with the liquidity estimates of either Lesmond, Ogden, and Trzcinka or, to a lesser extent, Amihud (2002). Examining the impact of legal origin and political institutions on liquidity levels shows that countries with weak political and legal institutions have significantly higher liquidity costs than do countries with strong political and legal systems, even to the exclusion of legal origin or insider trading enforcement. Higher incremental political risk is associated with a 10 basis point increase in transaction costs, using the Lesmond, Ogden, and Trzcinka estimate, or a 1.9% increase in price impact costs, using the Amihud estimate.

This paper proposes a new framework for understanding financial intermediation. In contrast to previous research, we consider a setting in which intermediaries possess no inherent information processing or monitoring advantages. Instead, in an economy with overly optimistic entrepreneurs who require funding from pessimistic investors, we show that intermediaries can arise endogenously. In such a setting, only a rational intermediary will be sufficiently optimistic to find it worthwhile to invest in a technology for screening entrepreneurs' projects, and yet be pessimistic enough to use this technology. Our framework produces implications consistent with heretofore unexplained stylized facts, and conjectures which are thus far untested.

We examine interactions between flexible financing and investment decisions in a model with stockholder–bondholder conflicts over investment policy. We find that financial flexibility encourages the choice of short-term debt thereby dramatically reducing the agency costs of under- and overinvestment. However, the reduction in agency costs may not encourage the firm to increase leverage, since the firm's initial debt level choice depends on the type of growth options in its investment opportunity set. The model has a number of testable predictions for the joint choice of leverage and maturity, and how these choices interact with a firm's growth opportunities.

We consider how fund managers respond to the conflicting preferences of their investors. We focus on the conflict between the taxable and retirement accounts of international funds, which face different tradeoffs between dividends and capital gains. In principle, managers could resolve this conflict through dividend arbitrage, but a proprietary database of dividend-arbitrage transactions shows that in practice they cannot. Thus, managers must resolve it through their investment policies. We find robust evidence that managers with more retirement money favor the preferences of retirement investors and further evidence for this view in the difference between U.S. and Canadian funds’ portfolio weights.