Testing the Equity Mutual Fund Liquidity Ratio

A reader citing Mark Hulbert’s recent column on the Fosback Index and its Ned Davis variant requested an evaluation of the indicators. The reasoning for these indicators is that a high (low) ratio of cash equivalents to assets among equity mutual funds indicates strong (weak) potential for near-term purchasing of stocks. Norman Fosback adjusts the raw liquidity ratio based on current interest rates, reasoning that mutual fund managers have more (less) incentive to hold cash when interest rates are high (low). The Investment Company Institute (ICI) surveys mutual fund managers monthly to measure the aggregate mutual fund liquidity ratio. However, only the most recent survey results and year-end values of the liquidity ratio since 1984 are publicly available. Note that current-month values are available with a lag of about one month. Using year-end and February closes of the S&P 500 index, a year-end short-term Interest Rate Composite and year-end values of the equity mutual fund liquidity ratio for 1984 through 2009 (26 observations), we find that:

The first step to approximating the Fosback Index is to correct the raw equity mutual fund liquidity ratio for current interest rates. The following scatter plot relates the raw year-end equity mutual fund liquidity ratio to the year-end short-term interest rate over the entire sample period. The R-squared statistic for the relationship is 0.49, indicating that variation in short-term interest rate explains about half the variation in the aggregate liquidity ratio.

We calculate an “excess” liquidity ratio as the vertical difference between the raw liquidity ratio and the best-fit line. When an observation lies above (below) the best-fit line, excess liquidity is positive (negative). For example, the December 2009 observation of a 3.6% raw liquidity ratio and 0.1% short-term interest rate lies almost exactly on the best-fit line (at the far left), so the excess liquidity ratio is approximately zero.

A non-linear (quadratic) best-fit curve improves the R-squared statistic modestly to 0.52.

The next chart summarizes raw and excess year-end equity mutual fund liquidity ratios over the entire sample period. It shows that the raw liquidity ratio has been relatively low in recent years, and the excess liquidity ratio has often been negative since the mid-1990s.

To measure the power of these series to predict stock returns, we relate them to subsequent stock returns. Since ICI measures the ratio monthly, we consider monthly returns. Since there is a one-month lag in reporting, we investigate the power of the year-end ratios to predict the S&P 500 Index return for the ensuing February.

The following scatter plots relate the S&P 500 Index return for February to the immediately preceding year-end raw and excess equity mutual fund liquidity ratios over the period 1985 through 2010 (the 2010 February return is partial only through 2/9/10). Results from this small sample indicate that the raw liquidity ratio is a better predictor of future stock market returns (explaining 22% of the variation for February) than is the excess liquidity ratio (explaining 9% of the variation).

The sample is too small to employ ranking of average returns by ranges of liquidity ratios.

In summary, evidence from a simple test on a small sample supports belief that the aggregate equity mutual fund liquidity ratio has some power to predict future stock market returns.

A larger sample might generate materially different results.

Note that the equity mutual fund liquidity ratio might be interpreted as a measure of sentiment among a sophisticated group of investors (mutual fund managers), as well as a measure of availability of funds to buy stocks. However, unlike many sentiment measures, the liquidity ratios do not relate strongly to past (prior-year) stock market returns.