Crash Redux

May 17, 2013

With the market maintaining a perpetual upward trajectory in 2013, it could be time to revisit our theories about crashes and market peaks.

In general, stocks – and the economy — go up, and trying to predict tops and drops can leave you out of the market for too long. And in today's market, with essentially zero returns in cash, the punishment for staying out is high. 

That being said, the Powerfund Portfolios are somewhere between buy and hold and active trading. Buying solely because stocks are going up can cause problems. Although we can’t predict a top or a bottom, we like doing what other investors are not doing, which usually means owning out-of-favor investments as well as decreasing stock allocations when euphoria builds. We primarily use fund flows to decide where to invest or not to invest.

Let’s start with crashes, which we define as a greatly accelerated bear market – a 20% drop or more within a few days.

There's no way to predict a crash, although a few lucky predictors gain fame when a crash happens somewhere in the ballpark of their prediction – fame they soon squander away, making numerous bad predictions in ensuing years.

Too Much of a Good Thing

One key warning sign of a crash is rapid price increases far above historical growth rates. This applies to real estate as much as stocks. For example, in 1987, there was a 40% one-year gain and a roughly 100% gain in the three years leading up to October just before the crash. Before the 1929 October crash the Dow had gained almost 50% in the previous year and 140% over the previous three years. For more on stock market crashes, read our piece from 2006.. 

While today’s stock market is up over 130% from its 2009 low, the gains are not quite irrational as they occurred off a big drop. More importantly, we're not far above previous highs, as we were in 1929 and 1987. Today’s bull market may presage a bear market, but probably not a crash.

The Price is too Darn (D!@*) High

Valuations (how much you pay for a dollar of earnings [or even future earnings]) are a good prediction of longer-term future returns, but less of a quick crash. Stocks were really not that expensive in 1987 with a 20 P/E and a 2.7% yield. Heck, by 2000 standards, 1929 stock prices weren’t even that lofty – 18 times earnings and a 3.4% yield right before the crash.

Valuations, unfortunately, are only one piece of the puzzle. It's how those valuations compare to future earnings growth (an unknown) that determines future returns. In the early 1990s – before a great period for stocks – P/Es were high historically in the low 20s, but massive profit growth followed.

As we’ve noted, the 2007-2009 slide did more damage across the board than the 2000-2002 drop, and did it from a lower stock valuation, since stocks were cheaper in 2007 than they were in 2000 (P/Es of 20 vs. 30 and a yield of just under 2% vs. just over 1%). Too much money was looking for yield and value (the flipside of too much money looking for growth in 2000). That's one reason banks underperformed biotechs. 

The problem was that the bubble was in earnings. Whole swaths of the market, from homebuilders to banks, had booming earnings from the housing market bubble. So when the underlying bubble popped, earnings fell with it, making formerly cheap stocks look expensive. 

We’ve seen the same thing happen in precious metal mining stocks over the last year or so. The underlying gold bubble, which is popping, is making the mining stocks seem very expensive at the old prices, much like housing stocks.

Today, valuations are lower than the previous two peaks in stocks (2000 and 2007,) but high relative to historical values. We don’t think you can pay too much attention to the pre-1980 markets for valuation cues, because throughout much of this early period, valuations were too low. At that time, investors feared markets and the economy more than they do today – which is why stock returns were so good throughout much of history. They were cheap. Before modern central banking (no laughing, please,) deep and frequent recessions were commonplace and recent enough in memory to put a huge risk premium on owning stocks.

The trouble with today’s cheap record stock market is that soon, corporate earnings growth rates are going to hit a GDP growth rate ceiling, and that sluggish rate may not warrant P/E ratios over 15. Just look at Apple, which may not grow much in the future and trades around 10x earnings due to that growth disillusionment. That could carry over to stocks across the board in coming years.

Theoretically (as in Dow 36,000 theoretically,) stocks should be more expensive and reach some sort of permanently-high plateau, because today’s paltry 2% dividend yield will grow to be much better than bond’s cash flow over time, even with a, say, 5% fixed bond yield. 

Consider that only high-risk bonds yield over 5% today, and stocks should be to the moon, Alice. In reality, the 50% drops you get in stocks every so often (too often for most retired investors) should keep the risk premium alive on stocks and total market P/Es under 20.

Skyscraper Indicator

These pages have made the case before that the sheer hubris in the world at any given time is a better indicator as to when the stock market's due for a slide than any cadre of economists or financial experts. Bottom line? Don’t be heavily invested in the stock market while the world’s tallest building is under construction. Currently, that indicator is no blaring sell, although the Freedom Tower just became the tallest building in New York City, now that the cheater spire's been placed on top.

Stock Market Cap to GDP

This is our favorite overall market value gut-check. The economy as a whole is a glass ceiling that publicly-traded corporations will just keep on hitting. Fortunately, the economy grows over time, especially when not adjusting for inflation, which lifts all boats. 

The worst episode of excess was early 2000, when the value of all stocks hit just about 200% of our (then nominal) GDP – worse than 1929 — but then we have far more publicly-traded companies covering most of the earnings of corporate America. Things got a little irrationally exuberant in 2007, when the market value of all stocks got to about 150% of GDP – right before the GDP fell apart with housing.

With the market’s recent rise, today we're over $20 trillion in total market capitalization and around 140% of GDP – roughly the same percentages we saw at the end of 2006. 

This is a concern, especially if we have another recession or even just very sluggish earnings growth. While it's quite rare to be able to invest today at less than 100% of GDP (with opportunities occurring only briefly at the bottom of the 2000 and 2007 slides,) closer to 100% is better than pushing 200%. 

Another way to look at it is this: our stock market has only been above 150% of GDP twice — once in 1998-2000 and again in 2007. Both were poor times to invest in stocks. Our last two major bear markets didn’t even take stocks back to pre-1994 valuations, relative to GDP.

The glass ceiling math works like this: A $15 trillion dollar economy probably can’t pump out more than 10% in profits (best-case) to publicly-traded corporate America. That’s $1.5T a year (even with record earnings, we're not at this level yet.) That’s already nearly a 14 P/E ratio to the current market cap of all stocks. 

How much can nominal (no inflation adjustment) GDP and therefore corporate earnings, grow from here? 5% average a year? That’s the best case, assuming no recessions and ongoing record profit margins. Without inflation, stocks could sink to a 10 P/E ratio, because growth would be a lackluster 2%.


Margin loans to buy stocks are near record levels. This is a sign of optimism (and low interest rates,) and in general, means future returns could be lackluster. Considering there are so many ways to leverage an investment, this figure is probably understated, anyway. As a long-term investor, you don’t want heavy exposure to a market investors are comfortable borrowing to invest in.

Fund Flows

Ah, our favorite indicator. If individual investors aren’t piling in, there's usually more upside left. During the first quarter, we saw more going into stock funds than we’ve seen since the first quarter of 2007 – a bad year in which to start investing. But investors pulled so much out of ordinary stock funds during and since the 2007-2009 slide that this relatively recent move in is hardly 2000-esque.

Bond fund inflows have continued to be the real boom, and should foreshadow low future returns, although probably not the outright crash the rising-rate crowd fears. This feature is the main reason we aren’t cutting way back on stocks and go into more bonds. Others are still favoring bonds.

ETF flows have been another thing entirely, with much of the new asset growth going to ETFs, not mutual funds. Even so, investors have pulled out almost as much money from U.S. stock mutual funds since 2007 as the total assets in U.S. stock ETFs. 

While you can point to overvaluation in foreign stock funds a few years ago and specific categories today by looking at the flows of new money, probably the best thing going for the U.S. stock market as a whole is that the individual investor hasn’t gone in whole hog. 

Bottom Line

If this recent upward trajectory continues or accelerates, or we see investors piling in like the old days, this will end badly once again. For now, it's more a question of low returns from these relatively high levels (without faster GDP growth) and avoiding popular areas like higher-risk bonds, real estate funds (again!), commodities, and more recently, "safer" stock funds owning less cyclical lower-volatility stocks.