Benchmarking Liquidity Proxies: Accounting for Dynamics and Frequency Issues

We revisit a central task of the extant liquidity literature, which is to identify effective measures of liquidity. We critically assess the influential practice of identifying the best liquidity measures based on monthly correlations by comparing and contrasting correlations between monthly and daily averages of high-frequency benchmarks and low-frequency proxies of liquidity, as well as by examining the coherences between such measures. Furthermore, we propose MIDAS regressions as a way of investigating the bilateral relationships between benchmarks and proxies without averaging out potentially valuable high-frequency information, as is common practice. We conclude that the empirical correlations between benchmarks and proxies in general become weaker as the frequency over which these relationships are examined becomes higher, and that standard practices may lead to misleading conclusions. One implication of our results is that any liquidity measure needs to be assessed against the relevant timeframe for conversion into cash.