Market’s ‘summer doldrums’ depress trading activity but not necessarily stock prices “Don’t sell a dull market short” is good advice anytime, but especially now as summer nears.

That’s because many investors go on vacation during June, July and August and stock-market trading volume declines. Some worry that this means the market will struggle, on the theory that price follows volume. Indeed, some analysts blame vacationing traders for the market adage, “Sell in May and Go Away.”

Read more: The stock market is defying ‘sell in May and go away.’ Why the U.S. presidential election could aid summer rally.

I searched for evidence for this theory and came up empty, however. To conduct that search, for each month since 1972 I calculated the ratio of NYSE average trading volume to its trailing 12-month average. (This is the proper way to measure relevant changes in trading volume, given volume’s strong uptrend over the past five decades. The chart below plots each month’s average ratio.)

I first measured the correlation of each month’s ratio to the S&P 500’s SPX total return in that same month. This test would detect whether volume is a coincident indicator: does the market tend to decline at the same time as volume? It doesn’t. 

I next tested whether volume is a leading indicator. I did this by measuring the correlation of the monthly ratios to the S&P 500’s return over the subsequent 1-, 3-, 6- and 12-month periods. In all instances I came up empty at accepted standards of statistical significance.

As far as my tests are concerned, therefore, volume is neither a coincident nor a leading indicator of price.

It’s possible that I incorrectly designed my tests, of course. Some technicians, for example, insist that it’s most crucial to focus on the interaction between volume and price rather than volume alone. But it’s not immediately obvious that focusing on those two together does a better job of forecasting the stock market’s direction.

Consider what I found when focusing on just those months since 1972 in which both trading volume and the S&P 500 were lower than in the immediately prior month. The stock market performed no worse in the wake of those months than after other months in which both volume and price had risen.

It’s also possible there’s some other way of slicing and dicing the volume data that is a reliable market timing indicator. But if it’s this difficult to detect anything meaningful in the volume data, it would be better to focus on other indicators with more obvious connections to the market’s direction.