Dynamic diversifications and transactions

This thesis is an inquiry into the process of how households' adoption of financial innovation (dynamic diversification) affects the way households carry out transactions. A recent survey by Frame and White (2004) bemoans the lack of research in the area of financial innovation. The goal of this thesis is to make a contribution to monetary economics in this area. The key contribution of this thesis is to utilize microeconometric methods (dynamic panel data) to explain the heterogeneity of households' decisions and actions. Unlike previous studies which focus on cross-sectional data this study employs panel data. To carry out this study, microdata from the Bank of Italy Survey of Household Income and Wealth is analyzed. Chapter 2 contains a description of the data.The first issue addressed is the interplay between financial innovation and the welfare costs of inflation. Endogenous switching regressions techniques are used to investigate the behaviour of households with and without financial innovation. A household money demand function is estimated and the elasticities are used to compute welfare costs of inflation. This study finds that the welfare costs of inflation are quite low, but represent a regressive tax.The next issue reflects the observation that household adoption of financial innovation is a persistent process. To explain this persistence the role of fixed costs is often cited as reason why households do not adopt financial innovation. Dynamic, discrete-choice, panel data methods are used to test for the presence of fixed costs or state dependence. The alternative explanation is the presence of unobserved heterogeneity which must be taken into account so as not to make an erroneous conclusion of spurious state dependence. The econometric evidence shows that state dependence is present. The implication of state dependence is that short term policies aimed at encouraging households to adopt financial innovation could lead to long term usage.