“The time to sell is when you are getting stock tips from the shoeshine boy”. This idea has been variously attributed to Barton Biggs, Bernard Baruch, Joe Kennedy and JK Galbraith. Its parentage deserves to be disputed, because – despite its snobbery – it is a sensible idea.
If shares are widely owned there might not be a marginal buyer, and there’s a heightened danger that a marginal seller will appear. Also, wide ownership means that lots of good news and optimism is already embedded in share prices – which means the market is vulnerable to either bad news or to sentiment mean-reverting downwards.
All of which poses the question: just how big is share ownership among ordinary Americans?
The Federal Reserve has data on this going back to 1952. The latest numbers tell us that households now own $18 trillion of equities and another $9.1 trillion of mutual funds. This is equivalent to almost 30 per cent of their financial wealth and 20.8 per cent of their total wealth, which includes housing. Although this is above average, it is less than we saw at the peak of the tech bubble, and around the same proportion as in 1968. (A high point for popular capitalism closely followed the summer of love: draw your own conclusions.)
The fact that the share of equities in wealth isn’t higher might surprise you. After all, the market has had an astounding rise: the S&P 500 has quadrupled since its 2009 low point. What’s more, equities should be, in economic jargon, a superior good. Demand for them should increase more than one-for-one as incomes rise simply because higher incomes enable people to cope better with equity risk. This predicts that the share of equities in wealth should gradually rise over time. Taking the past 60 years altogether, however, this has not been the case: the equity share of wealth is much the same today as it was 50 years ago.
In fact, there’s a good reason why households’ ownership of equities is not higher. It’s that the stock market has in one important sense been shrinking. Since the start of 2010 non-financial companies have bought $4.1 trillion of shares in the form of buybacks and takeovers. This has tended to reduce households’ shareholdings thereby partly offsetting the fact that rising prices have tended to increase them.
Which brings us to the question: are households’ shareholdings now so high as to point to prices falling?
Not quite. True, there is a negative correlation between the share of equities and mutual funds in total wealth and subsequent changes in the S&P 500, adjusted for inflation. High equity weightings tend to lead to lower returns. As is usually the case, these correlations are stronger for longer-period returns. For annual real changes in the S&P 500 since 1952, the correlation is only minus 0.15, but for five-yearly changes it has been minus 0.32. (The correlations are much the same if we take the share of equities in financial rather than total wealth.)
Even the latter, however, is not very strong. Equity weightings have sometimes been a misleading guide to subsequent medium-term returns. For example, high weightings in the early 1960s led to good returns; low weightings in the 1970s sometimes led to the market falling; and equity weightings were not unusually high in the run-up to the 2008 financial crisis.
This shouldn’t be surprising. Although investors’ sentiment does help us predict returns, households are only a subset of those investors: the sentiment of fund managers and overseas investors also matters. (In fact, given that foreign buying of US equities has been a great predictor of equity returns, the latter is especially important.)
What’s more, share prices are ultimately determined by the state of the real economy. If GDP continues to grow, and inflation stays low, equities should do okay. And if they don’t, they won’t. And (for what it’s worth) economists expect steady growth and low inflation to continue.
Granted, the post-1952 relationship between households’ equity holdings and subsequent returns now points to low but positive real returns over the next five years, with an above-average chance of the market falling over this period. But this is not (yet?) a screaming sell signal.
