Institutional Heterogeneity Among Stock Exchanges: Implications for Market Efficiency

Abstract

The current financial crisis is a crisis of theory as well. The dominant theory of financial markets, the efficient market hypothesis (EMH), states that in an efficient market the price of a financial asset reflects publicly available information about that asset. Competing theories, such as behavioral finance, argue that other factors, including irrational investor behavior, impact the price of financial assets. We argue, however, that an analysis of market institutions can help explain when and why the EMH works. Although not widely examined, we argue it is significant that until very recently the New York Stock Exchange (NYSE), whose listed companies’ price behavior inspired the EMH, was a nonprofit organization. Thus, we apply an economic theory of nonprofits to the NYSE, and NASDAQ, to identify the incentives of Exchange members and the various governance mechanisms they created in response. We find that the NYSE, which was organized by underwriters, generated mechanisms that made prices on the NYSE relatively well behaved, what we term synthetic inertia. The NASDAQ, on the other hand, was organized by traders who had polar opposite incentives with regard to disclosure. We hypothesize that NYSE demutualization - converting from nonprofit to for-profit - altered the incentives of the NYSE and undermined this synthetic inertia and thus informational efficiency. We test our hypothesis by comparing bid-ask spreads, a measure information quality, at the NYSE and NASDAQ and find that bid-ask spreads on the NYSE were consistently lower than the NASDAQ (suggesting better information quality at the NYSE) but that spreads converged after demutualization. We believe that our approach helps resolve an apparent tension between competing theories of market behavior and contributes an analytical framework from which to consider regulatory changes.

Keywords: stock exchange, institutions, efficient market hypothesis, lemons problem

JEL Classification: K22, A1, B15, B21, D2, D23, G14, L1, L31