Valuations Seem High... Market Looks Dangerous

If you are a value investor or a contrarian of any sort, it's a tough time to invest right now. I don't have much experience and was away from investing for about 5 years but I did start serious investing a few years before the financial crisis but it seems worse right now. It is really difficult right now to find anything attractive. Everything seems overvalued and the market is rallying unexpectedly (at least in my view).

I was going to write a post on market valuation but ran across an excellent article by John Mauldin that captures my feelings quite accurately so I'm going to quote his work below. I don't always agree with Mauldin and definitely don't share his right-leaning political views but this was a great piece. If you are macro-oriented, I would highly recommend that you read his article, "The Trump Rally Will Morph" (Dec 11 2016), which contains more graphs and thoughts than my post.

"The P/E ratio spent most of the last century between 10 and 25. The times it went below 10 correlate with market bottoms, i.e. “undervaluation,” while the times it spiked over 25 were “overvalued” manic tops.

You might think that the P/E would be a pretty good timing indicator. That’s true for very long periods. The problem is that P/E ratios can stay undervalued or overvalued for years. They can also go to extremes well below 10 and well above 25 and stay there for uncomfortably long spans. Keynes had it right when he said, “The market can stay irrational longer than you can stay solvent.”

Presently, all three P/E versions are near or above 25, indicating overvaluation. This doesn’t mean the end is near – though it could be. But it does suggest that we are not at the beginning of another long-term bull market. The next chart illustrates the past and present trend in a different way."

Mauldin is referring to the S&P 500 P/E ratio above, which has rarely been above 25. Right now it is around 25. The thing is, the P/E ratio has been high for most of the last 15 years so one can't be sure it won't go higher for many years.

Some people argue that the P/E ratio is high because bond yields are low. That certainly is playing a role but the relationship isn't very clear. For instance, bond yields were very low in the 1940's--somewhat akin to now, the FedRes was carrying out unusual activities; basically monetizing debt on a grand scale--but P/E ratios were nowhere near what we have now.

The Wall Street Journal provides a nice summary of the P/Es and dividend yields of the major US indexes and I look at that once in a while. This page also provides the forward-looking P/E which can be useful sometimes. Right now the forward P/E on S&P 500 is 18, which is a forecast of almost 30% higher earnings next year and seems unlikely. Typically the forward P/Es are unreliable.

The S&P 500 and NASDAQ P/Es are about 10% higher than last year, while the DJIA is about 20% higher and the DJTA is a little under 20% higher.


The Russell 2000, which represents small-cap stocks has negative P/E right now. Usually this is a bad sign (because it means small businesses are struggling--you want broad participation during bull markets) but I'm not sure in this case. I wasn't following the markets closely the last few years and I'm not sure if it is negative due to a few companies posting large losses (as opposed to a large number of them posting losses) or if it is due to non-cash charges resulting in losses or something. Based on analyst forecasts, the forward P/E is 20 so the index is expected to be profitable next year.

An interesting thing to note is that the dividend yield on the S&P 500 and DJIA has declined less than 5% (whereas the P/E ratio has gone up 10%). Historically, this is a bullish sign because it means companies are able to raise dividends and support them with cash flow (if this didn't happen then you would see dividend yields fall as stock prices rise, which is common during bubbles). However, in this case, I'm not too sure how sustainable the dividends are given that quite a number of companies are paying out more in dividends than they are earning (basically issuing debt to pay dividends--many commodity businesses like ExxonMobil and Chevron have been doing that this year).

Overall the P/E picture is mixed. On the one hand, the P/Es are really high. But, if earnings grow about 30% as analysts expect then the P/E will drop to around 18 next year, which is only slightly higher than the long-term average of around 15. I think a lot of things have to go right for the earnings to rise so much--it can happen but I wouldn't bet on it.

Mauldin next looks at the so-called Buffett indicator, which is the ratio of value of stocks to GDP:

The interesting thing here is that right now the Buffett Indicator, while down from its late 2014 peak, is still higher than it was before the 2008 financial crisis. That should not be encouraging if you’re a bull. Buffett himself said the 70%–80% area of this indicator was the zone where buying stocks was likely to work out best. Presently we are way above that.

I will add one word of caution regarding this indicator: It assumes that GDP remains a meaningful statistic, even though the economy has changed considerably in recent decades. I’ve written about the quirks and vagaries of GDP before. We still use GDP because we have nothing better, but take it with a grain of salt if you’re using it as part of a timing decision.

I don't generally use this metric because I don't understand it very well. The economy and the companies that operate in them have changed quite a bit--for instance, US companies earn way more profits overseas nowadays--and I haven't thought hard about how that impacts this measure. In any case, this measure is also implying high valuation.

The article finishes off with some references to research from Ned Davis Research.

Steve’s next NDR chart [not included here] takes some explaining. It shows the percentage of household financial assets invested in stocks (blue line) versus the S&P 500 total return for the following 10 years (dotted line). Notice that the right axis is inverted and the dotted line tracks pretty close to the solid blue one. The correlation is 0.91, which is extraordinarily high. What the chart shows us is that a higher percentage of household assets in equities points to a lower annualized return over the next 10 years.

...That’s the good news. The bad news is that the red arrow at 6-30-2016 means that the 10 years ending 6-30-2026 should produce a 3.25% annual gain. That’s not awful, of course, but it is nowhere near the 7% or more that many pensions and insurance companies think they can earn on their portfolios – and since we all know that bonds earn less than equities do, they are really hoping to get closer to 10% on their equity portfolios. It’s therefore a problem for everyone, stock investor or not. Your taxes may have to make up the difference.

Not sure how many realize how long we have gone without any meaningful correction:

Since I’m blatantly borrowing from my friend Steve Blumenthal, let me show yet another piece of good research that he included in yesterday’s letter. Bottom line: We are way overdue for a correction. Again, our situation is not the worst it’s ever been, but we are beginning to bump up against historical lines in the sand. Here’s the chart, which shows the number of days before the start of 5%, 10%, and 20% corrections.

It has been 1955 days since we suffered a 20% correction. Since 1928, the average number of days before a 20% correction occurred was 635. In secular bull periods the average number of days was 1105. In secular bear periods the average number of days was 486.

The current case of 1955 days without a 20% correction is more than three times the average of 635 days (for the whole period from 1-3-1928 to 12-8-2016).

Even if you believe we are in a major bull market, the strongest case for one to be really cautious is the data above. Namely, we have gone 1955 market days without a 20% correction, which is pretty common from a long term point of view and typically occurs every 635 days (look at the black triangles in the bottom chart). Looking at the chart, it seems we are in the second longest bull market without a 20% correction since the 1987-2000 period (which was about 3000 market days).

Having said that, in terms of 10% corrections, the market is not that out of whack since there was a correction in 2014-2015. Does several 10% corrections over many years make up for a 20% correction? I have no idea and that's up to you to decide.


To sum up, I think one should be cautious. Nearly all popular valuation metrics are suggesting the market is overvalued or at least is due for a correction. I am just a newbie but I'm having a hard time finding investment opportunities. If you find good opportunities go for it but don't try to trade down the quality spectrum to earn higher returns. Now is probably not the time to do it. If you are already invested and have a low cost base, it won't be that painful if things get marked down but if you are putting new money, be careful.

I see some amateur value investors putting money into heavily leveraged or small cap commodity stocks or into cyclicals trading at low P/Es, and I think they are playing with fire. I would also be careful with stocks that have outperformed strongly over the last 6 or 7 years. This applies to certain growth stocks, particularly in technology and consumer discretionary sectors, but also to something that some value investors seem to be overloaded on: financials. It would not surprise me if financials underperform most other sectors if we enter a correction (financials always underperform during corrections but I'm talking about them doing much worse than that).

Interesting times we live in...