[Miller and Modigliani (1961) show that in perfect and complete financial markets a firm's value is unaffected by its dividend policy. Much of the more recent research has demonstrated that dividend policy becomes important in the presence of taxation, asymmetric information, incomplete contracts, institutional constraints, and transaction costs. By examining the effects of dividend policies on 475 British firms existing between 1895 and 1905, and consequently operating in an environment of very low taxation with an absence of institutional constraints, we find strong support for asymmetric information theories of dividend policy, and little support for agency models.]

Miller and Modigliani (1961) show that in perfect and complete financial markets a firm's value is unaffected by its dividend policy. Much of the more recent research has demonstrated that dividend policy becomes important in the presence of taxation, asymmetric information, incomplete contracts, institutional constraints, and transaction costs. By examining the effects of dividend policies on 475 British firms existing between 1895 and 1905, and consequently operating in an environment of very low taxation with an absence of institutional constraints, we find strong support for asymmetric information theories of dividend policy, and little support for agency models.

We investigate the source of information advantage in inter-dealer FX trading using data on trades and counter-party identities. In liquid dollar exchange rates, information is concentrated among dealers that trade most frequently and specialize their activity in a particular rate. In cross-rates, traders that engage in triangular arbitrage are best informed. Better-informed traders are also located on larger trading floors. In cross-rates, the ability to forecast flows explains all of the advantage of the triangular arbitrageurs. In liquid dollar rates, specialist traders can forecast both order-flow and the component of exchange rate changes that is uncorrelated with flow.