Doubts about the accuracy with which outside investors can assess a banking firm’s value motivate many government interventions in the banking market. Although the available empirical evidence is somewhat mixed, the recent financial crisis has reinforced a common assessment that banks are unusually opaque. This paper examines bank equity’s trading characteristics during “normal” periods and two “crisis” periods between 1993 and 2009. We find only limited (mixed) evidence that banks are unusually opaque during normal periods. However, consistent with theory, crises raise the adverse selection costs of trading bank shares relative to those of nonbank control firms. A bank’s balance sheet composition significantly affects its equity opacity, but we cannot detect specific balance sheet categories that have robust effects.

The rapid increase in U.S. house prices during the 2001–2006 period was accompanied by a historically rapid expansion of bank assets. We exploit cross-regional variation in local housing booms to study how housing demand shocks affected the growth of the banking sector. We estimate the effect of housing demand shocks that are orthogonal to observed non-housing demand shocks and credit supply shocks in each bank’s market area. We employ several instrumental variables that plausibly identify variation in local housing demand that is exogenous to local banks. We find that the housing boom had a large effect on bank asset growth—the cross-regional elasticity of bank growth with respect to housing demand shocks is around 0.6. The regional elasticity estimate suggests that housing demand shocks can potentially account for a large fraction of the growth of the banking sector during this period.