Portfolio return autocorrelation and non-synchronous trading in UK equities

Although infrequent trading in equity stocks is more prevalent in the United Kingdom (and other non-United States countries), we find that it is proportionally more important in explaining the degree of serial correlation in stock returns in the US than in the UK, in contrast to much of the existing literature.We show that infrequent trading cannot explain more than a small proportion of the serial correlation observed in monthly UK stock returns and hence, other explanations for return predictability must be sought. Many studies have shown that stock market returns in the UK and other international markets are substantially and significantly serially correlated. The success of an investment strategy that is based on the apparent predictability of returns depends on whether the serial correlation is truly random and period specific or due to time varying risk premia or to market microstructure effects. A frequently noted explanation for this serial correlation is market thinness or, more precisely, the infrequency with which a substantial number of UK stocks are traded. Non-synchronous trading results in a measurement error in the observed data for returns on individual stocks, portfolios and market indices. This measurement error generates serial correlation in the observed returns. Here, we assess the extent to which the observed serial correlation in returns can be explained by equity non-trading behaviour. This will reveal whether there is any residual serial correlation left to be explained by alternative sources. We find that, whilst a proportion of the serial correlation in the returns of portfolios of low value stocks can be explained by non-trading, much of it still remains unexplained.