Today, we present a guest post written by Jeffrey Frankel, Harpel Professor at Harvard’s Kennedy School of Government, and formerly a member of the White House Council of Economic Advisers. This is an extended version of a column that appeared at Project Syndicate on February 22nd.
Much has been written about the stock market correction in the short time since the US market peaked on January 26 and abruptly fell 10 % (as of February 8, followed by a partial bounce-back). Some of what is said is useful, some is not. Three pronouncements concern advice to investors, the role of machines, and the connection to the real economy.
Is it time to sell?
First, how should an intelligent investor react? “Don’t be scared into selling stocks in reaction to a short-term plunge,” they say. “Think longer term.” They are right. That stocks fell in early February is not a reason to sell.
Rather, the reason to sell stocks is that they are too high in a longer term perspective. Prices are very elevated relative to fundamentals such as earnings. The Cyclically Adjusted Price Earnings, for example, is still at a level that has only been surpassed twice in the last century: the peaks just before the stock market crashes of 1929 and 2000-02, respectively. The implication is that the rate of return to stocks is likely to be substantially lower over the next 15 years than over the last 15 years.
When I suggest that investors should sell stocks, I refer to those who are fully invested in the stock market or, worse yet, have leveraged themselves to a position that is more than 100% into stocks. But of course an appropriately diversified portfolio will still have a large allocation to equities.
The uprising of the machines
A second declaration that one suddenly hears everywhere runs along the lines, “The market has been made more volatile because machines are taking over the trading.”
They say that algorithmic trading means that when stocks start to fall, computers kick in, selling more and driving the price down further. This can happen. And I am not one of those who believe that automated or high-speed trading accomplishes a useful social purpose. But neither do I think that it is necessarily destabilizing. Human beings are as likely to be stampeded into unwise decisions as are machines.
It all depends on how the algorithm is designed (which is ultimately done by humans, needless to say). A computer that has been instructed — whether directly or indirectly — to instantly “buy on the dip,” will generate demand for falling stocks and thus tend to stabilize prices, not destabilize them.
We have always had “stop-loss orders,” whereby an investor leaves instructions with his or her stock broker to sell if the price falls to a pre-specified level. They are de-stabilizing, in that they generate sell orders in response to an incipient price fall, thereby working to exacerbate the price fall. Or the investor can leave instructions to buy when the price falls to a certain pre-specified level, which is stabilizing. It doesn’t matter whether the order is executed by a human broker or a machine. What matters is whether the instruction is stabilizing or destabilizing.
If anything, perhaps when a human being programs a computer he or she is more likely to think about it calmly than when a human watches a plunge on his or her monitor in real time, susceptible to a sense of panic and to jumping on the bandwagon.
I make an exception for intra-day volatility. A flash crash such as occurred on May 6, 2010 – when the Dow Jones fell and rose over 1,000 points within 15 minutes – almost certainly would not have been possible without high-speed algorithmic trading. This sort of volatility matters to those who make their living by intra-day trading; but it is not clear why it should matter to the rest of us.
Wall Street vs. Main Street
One also encounters a third wisdom which goes, “The stock market is not the economy.” Yes, this one is very true. The market can crash while the economy is doing well, and vice versa. Three reasons.
For one thing, stock market busts (and booms) can be driven by rises (and falls) in interest rates, (respectively), which in turn are often the result of economic expansions (and recessions), respectively, rather than the other way around. The bits of news that seem to have precipitated the February market correction were reports of a strong job market in the US (hourly earnings up 2.9% in the year to January, announced February 2); inflation in the US, UK and some other countries; and correspondingly stronger anticipations of future increases in interest rates on the part of the Federal Reserve and the Bank of England. On February 7, the Bank of England announced that rates “could be tightened somewhat earlier and by a somewhat greater extent” than had previously been forecast. Inflation news in the US made it near-certain that the Fed will raise interest rates in March.
In the second place, there is a lot of randomness in the markets, including cases where market prices depart from economic fundamentals, as in true speculative bubbles. People buy because everyone else is buying, and then all sell at once when the bubble bursts. The “risk on” mood in financial markets last year comprised unnaturally low perceptions of future volatility. When the VIX hit all-time lows in 2017, it was not based on fundamentals. It was not hard to think of substantial risks that were lying in wait, such as the inflation-plus-interest-rate shock. But it was predictable that the VIX would not adjust until the shock materialized and securities prices fell from their heights. As of the third week of February, the VIX has indeed adjusted to more normal levels – the stock market correction having served as a useful “wake-up call” for complacent investors. But the securities prices themselves probably still have a substantial distance to fall. After all, the S&P 500 is still higher than it was in 2017.
In the third place, even in pure textbook theory, stock prices represent the profits accruing to corporations — current and expected future profits – not the income of all of us. In the past, there was a high correlation between profits and the economy because a relatively stable share of national income went to workers versus owners of capital. But that stability has broken down in the last decade or so. The share of GDP going to capital has increased remarkably, probably as a result of what economists call rents – increased monopoly power and decreased competition in many sectors. It is noteworthy that the markets are down from where they were after December’s Republican bill to slash corporate taxes. Some part of last year’s stock market boom may have been attributable to anticipation of the possible tax cut. If so, this reflected a policy that will almost certainly redistribute the pie away from labor and toward capital, more than it expands the size of the pie.
Still, having reminded ourselves that Wall Street is not Main Street, there are of course important connections between the stock market and the real economy. If the market goes down, consumption and investment spending fall: a household that holds stocks becomes less wealthy and so may cut back, while a corporation that had been considering building a factory may be less inclined to do so if it becomes harder to raise new capital.
One can’t predict when the next market plunge will occur or whether it will be coincide with the next recession. But one can predict that the unusually low financial and economic volatility of last year is over.
This post written by Jeffrey Frankel.