7:23 a.m. Thursday 7/24/96
Dr. Stein is still reacting to my statement that a 40 percent drop in the real value of the U.S. stock market over the next 10 years is “plausible.”
Did I mean this as a forecast? Well, I wasn’t meaning to say that I was at all sure that this will happen. My forecasting model (which, I should add, I developed jointly with Professor John Campbell of Harvard University) suggests that we have something like a 2/3 probability that the U.S. real (inflation corrected) stock market will be lower than it is now in 10 years, and a 1/3 probability that it will be higher.
The problem is, this forecasting model is not the only one that one might look at. Other models will certainly give different probabilities. Unfortunately, when it comes to the stock market, we have no scientific way to assess probabilities.
There is fundamental uncertainty, that no statistical test can resolve, whether the growth of earnings we have seen recently in the United States can be expected to continue in the long run, or be reversed. We must rely on our personal judgments to decide which is more likely. People are telling engaging stories why earnings ought to continue to grow (downsizing, tighter management techniques, the growth of computers), but these stories sound no better than the ones people were citing in 1929 to forecast a continuation of the rapid earnings growth of the 1920s (lagged effects of the electrification of America, of the vertical integration movement of the 1920s, of the emergence of efficiency experts and of modern management techniques.)
I do not like to give advice to others. It would be constructive, though, to add that this might be as good a time as any for investors to consider whether they are diversified enough around the world and across asset classes; many aren’t.
8:33 a.m. Thursday 7/24/96
Some of us say that earnings in the high-tech sector were grossly overestimated, as were valuations. “But,” the moderator writes, “there was a lot of stock issued by that sector at those valuations.” So, was the allocation of resources a mistake? Should the nation’s savings have gone elsewhere?
Actually, the typical pattern for a high-tech stock over the past year is that the company raises money with a public offering at a pretty reasonable price. Investors then bid it to the stratosphere. Then, the price comes down to close to the public offering. A good example is America Online, which raised $250 million in August 1995 by selling shares at $29. By May 1996, the stock was trading at $70. Now, it’s back down to $26. But the actual allocation of capital to the corporation came at the $29 figure.
While some capital is clearly being “wasted” in young high-tech ventures, on the whole, the bubbling market in initial public offerings has been a boon to the economy.
Think of it this way: Large companies like General Electric have been using their profits to buy back their own stock from outside investors. Those investors, with cash in their pockets, buy shares in new high-tech start-ups. That’s a beneficial transfer–even if the investors overpay.
And who is to say that they have overpaid? The pricing of stocks is a continual process of under-shooting and over-shooting. In May, it seemed that, based on current estimates of future profits, America Online as a company was worth about $6 billion (its market capitalization: the price of a share times the number of shares outstanding). Today, based on new estimates, the market thinks it’s worth $2 billion. Perhaps the truth lies somewhere in between.
The real problem with a market as volatile as this one is that it will discourage investors from putting their money into new ventures like America Online in the future.
8:56 a.m. Thursday 7/24/96
Our moderator asks one of the classic questions posed by stock analysts and theoreticians: Is the market “efficient”? Academics have pondered this question for years and have come up with many different definitions of efficiency–“weak” efficiency, “strong” efficiency, and so forth. I will spare the reader a summary of this debate (and will gladly cede the field to Professor Shiller, who has contributed to it, if he is so inclined). The common sense view of the market–to which I subscribe–is that while in the long run stock prices are likely to accurately reflect market fundamentals (estimates of future earnings discounted by some reasonable “discount” or risk-adjusted interest rate), in the short run prices can be, and often are, moved by the collective emotions of investors, their fears and their (often unreasonable) hopes. As a result, market prices on any given day are likely to deviate from a reasonably objective “fundamental” valuation. This is one reason, among many, that I and many others who are distant from Wall Street (and therefore out of earshot of inside tips) forego any attempt at “timing” the market and instead, like bears in the winter, hibernate with our portfolios (and take the long-run view). Does all this mean that the stock market is a poor vehicle for channeling the nation’s investable funds? No. The stock market is like democracy–it’s messy but there’s no better alternative. Moreover, it is important to remember that virtually all of the money that changes hands in the stock market every day goes for stock that has already been issued. Only a small fraction is invested in “new issues” and thus goes to fund new investments. By the same token, the firms whose shares are traded on the exchanges fund their investments largely out of their own cash flow or by issuing short-term debt (commercial paper) or sometimes long-term debt. Issuing stock is generally a last resort. So even if the market is imperfectly channeling the funds of new equity investment, most investment is funded through other means (the debt markets, banks, and so forth). James Glassman notes that in a highly volatile market such as the current one, investors are scared of initial public offerings (IPOs) and this discourages new ventures such as America Online. This is true in the short run, but in the long run, the possibilities offered by the market for start-up ventures to grow by offering stock are still there. And, incidentally, they will still be there even if the capital gains tax remains where it is. (I know it’s another subject but I can’t help pointing out what a boom in IPOs we’ve had over the last several years, as well as a boom in the stock market, even at current capital gains rates.) Herb Stein
1:47 p.m. Thursday 7/25/96
I thank the panel for their explanatory submissions on Thursday. Friday is the last day of this soiree and I invite you to submit any last words you would like to send into cyberspace. I call your attention to a few subjects on which I would welcome further comments if you have them. I raised at the beginning the question of the market mechanism. After the October 1987 crash there was a lot of talk about defects in the market mechanism, mainly having to do with program trading, that tended to exaggerate market movements. I never understood all that but took it from better-informed people that there was a problem there. What is your view of that after almost nine years? Was there a problem and has it been corrected? I go back to the comments, mainly from Litan, about the effect of a market decline on the general economy. Litan offers a back-of-the envelope estimate that the recent market decline could reduce growth by 0.2 percentage points. I have no opinion about the number but I wonder whether he really means growth in the sense of the long-term trend of output–so that if we previously thought it would be 2.25 percent on the average for a long time in the future we should now think that it will be 2.05 percent. Or does he mean that the rise of output for a brief period, a year or two, would be 0.2 percent less than otherwise? Can we say that the effect of a decline in perceived wealth will be to raise the proportion of their incomes that people will want to save, which we (or many economists) usually think will raise the long-term growth rate? The ratio of profits to national income has risen lately but on the average of recent times is significantly below the level of the earlier postwar years. Does anyone have an opinion about the probability of getting back to the earlier ratio and what that would do to the values of stocks?