ASSET HOLDING AND CONSUMPTION VOLATILITY: Introduction 3

Posted by Kathryn Schwartz on May 18, 2014
ASSET HOLDING AND CONSUMPTION VOLATILITY

The data are drawn from the UK Family Expenditure Survey, which has been collected continuously and consistendy since 1968.2 This gives us a long time-series of data on consumption, a crucial factor in estimating Euler equations and asset pricing models. Changing patterns of direct share ownership over the last twenty years also make the UK an interesting case for analysis. The Conservative government in the early 1980s introduced a number of measures designed to create a ‘share-owning democracy’, including the heavily-advertised flotation of public utilities and tax-breaks for employee share schemes. Largely as a result of these measures, there was a near trebling of the level of share ownership over a very concentrated period, 1985-88. These changes induced a change in the composition of shareholders and may also have changed the nature of the process generating prices and returns. this

The plan of the paper is as follows. In the next section we review briefly the consumption CAPM model and its testable impHcations on the time series properties of asset prices. Section 3 presents evidence on recent changes in share-ownership in the UK and also shows asset returns over this period. Using the techniques developed in Hansen and Jagannathan (1991), we present mean-variance bounds on the IMRS computed from data on the returns to shares and Treasury Bills between 1978 and 1995. It can be seen dearly that the IMRS computed from aggregate consumption data over this period does not satisfy these bounds, generating an equity premium puzzle over this period. Section 4 discusses in detail the econometric technique we develop to characterise the time series properties of a variable for a group whose composition changes over time in a manner endogenous to the variable of interest. This technique has a number of potential applications. Section 5 compares the time series properties of consumption growth of ‘likely’ shareholders and non-shareholders as defined by the estimated probabilities and interprets the results. Section 6 concludes the paper with some remarks about potential extensions.

Tags: , ,