Remember the famous book by Jeremy Siegel called “Stocks For The Long Run“?
It was the bible of the 1990s.
In the 1990s, every dip in the stock market was hailed as a “buying opportunity,” because the prevailing wisdom was that stocks always do well over the long haul.
And stocks usually do do well over the long haul, especially relative to bonds and cash. But there’s one major exception to this: Stocks don’t do well over the long haul when they’re bought at extremely high prices.
When have we had extremely high prices for stocks?
Well, in the late 1920s, for example, just before the Great Crash and Great Depression. Stocks crashed nearly 80% and then moved sideways for more than two decades. Here’s a chart of 130 years of inflation adjusted stock performance. Check out how much time it took to recover from the Great Crash. Note that it also took 18 years to recover from the malaise market of 1966-1982:
Photo: Professor Robert Shiller
And we also saw extremely high prices in Japan in the late 1980s. Stocks crashed there and are still falling nearly three decades later. (See chart at top).
And, we had extremely high prices in the US in the late 1990s.
During all of those peak periods, stocks hit extreme prices relative to earnings. And in all of those periods so far, the “long run” required to do well if you bought stocks near the peak has been shockingly long. (Below is a chart of the P/E ratio in the US for the past 130 years, again from professor Robert Shiller of Yale. Look how high the PE ratio got at the bubble peaks. Look how much higher it got in the 1990s peak than it had ever gotten before.)
Photo: Robert Shiller
Today, Japanese stocks hit a 28-year low.
A 28-year low!
The Nikkei closed at 8,296.
That’s down from a high of about 39,000 in 1989.
That’s nearly an 80% drop over three decades.
Many, many times during that decade, it looked like Japan and Japanese stocks were on the road to recovery… only to have the market falter again and head for new lows.
Japan’s market would have to quintuple just to get back to the level of nearly 30 years ago. And it’s hard to believe that’s going to happen before at least another decade or two has passed, if not more. So that will be half a century to get back to even (excluding dividends).
So that’s a whole new definition of “stocks for the long run!”
Meanwhile, US stocks are in the middle of a multi-decade workout period after our own series of bubbles–the stock bubble of the 1990s, the housing bubble of the mid-2000s, and the current bond bubble, which has sent interest rates on US bonds to record lows.
US stocks did break their 2000 highs briefly a few years ago, but they’ve since fallen back below them, especially after adjusting for inflation.
Photo: Robert Shiller
US stocks aren’t terribly expensive anymore, on a cyclically adjusted PE basis, but they’re still overvalued. (Here’s that PE chart again–click for larger).So that suggests we have many years left in our post-bubble workout period.
US stocks weren’t as overvalued in the late 1990s as Japan’s were in the late 1980s, so hopefully we won’t have to wait three decades to hit bottom.
But to put the devastation of Japan’s stock market into context for you, here are some numbers.
The Nikkei peaked at about 39,000 in 1989. Now, 28 years later, it’s trading at 8,296, down 80%.
The Dow peaked at 12,000 in 2000.
If the Dow performs from that level the way Japan has performed from its 1989 peak, the Dow will close at 2,400 in 2027. That’s down about 80% from today’s level.
Meanwhile, our broader market measure, the S&P 500, peaked at about 1,500. If the S&P 500 falls 80% from its peak, it will trade at about 300 in 2027.
Here’s an excellent chart from Doug Short showing the US market (blue) mapped against Japan’s market (red), and the US DOW from the Great Crash (grey), with the respective dates of each market peak lined up. As you can see, our market has tracked Japan’s pretty closely so far. Let’s hope it doesn’t continue to track it for the next 15 years! (Click for larger)
How could it possibly ever happen that US stocks could trade down another 50%-80% over the next couple of decades? Would that be inconceivable?
It’s actually not inconceivable.
Four things would have to change:
- Stocks would have to trade down to the usual post-bubble PE-ratio low of high-single digit PEs. Let’s call it 7X-8X, about the level at the bottom of our three prior secular bear-market bottoms in the past century. (See the PE chart above. Look how far our current market is above the usual post-bubble low. We’re at 20X. The usual post-bubble low is ~7X).
- St. Louis FedCorporate profit margins just hit an all-time high. History suggests that that is completely unsustainable. Click for larger.Profit margins would have to return to the long-term mean and then fall below it. Given how high profit margins are today, this seems inconceivable, but look at how fast profit margins have fallen in the past and how long they have stayed below average. (See chart).
- Interest rates would have to rise sharply, back to long-term means or beyond them. This is entirely conceivable. At some point, the clamor for the US to generate some inflation to get out of its debt and unemployment problems will grow deafening. When that happens, the Fed will print even more money. And, at some point, inflation will start running hot. When that happens, cash returns will look much more attractive relative to stocks, and PEs will fall further (they shouldn’t, because stocks are actually a hedge against long-term inflation whereas cash isn’t, but that’s how investors have behaved in the past).
- Investor frustration with the stock market would have to hit a generational peak–to the point where it seems inconceivable that anyone would ever voluntarily choose to buy stocks. This is also quite conceivable. Most people have lost money in stocks over the last decade, especially when costs and taxes are factored in. Another decade of performance like that, and stocks will seem like a crazy place to put your money.
So, is that likely to happen?
No, I don’t think it’s likely. (And I certainly hope it doesn’t–I have about 50% of my savings in stocks).
It’s just relatively easy to see how it could happen.
And it’s even easier to see how stocks might still be trading at the current level in another 10 years, with many ups and downs in the interim.
If stocks do that, and profit margins regress back to means, then stocks will just be modestly undervalued in a decade. And after three decades in which stocks have been overvalued, it seems entirely conceivable that they’ll now move to spending a couple of decades undervalued.
All of which is a long-winded way of saying, when it comes to stock-market performance, the “long run” is, occasionally, very long.