Long-term buy-and-hold investing has plenty going for it. Nonetheless, the U.S. stock market is seems expensive in a number of ways which we'll discuss. As such, now could be a good time to make sure your portfolio is well diversified. Given that the U.S. market has had such a great run it may now be a greater portion of your portfolio than a few years ago. In essence, if you have a big chunk of the U.S. in your portfolio, it may make sense to think about rebalancing to keep your U.S. stock exposure under control.
U.S. Stocks Are Expensive
I think most would agree that U.S. equities are expensive today. Yes, some would argue that the U.S. deserves a premium. Others would argue that valuation matters less than in the past. Still, the fact remains that investors are paying a lot for a dollar of earnings in the U.S. today. Over history, buying a dollar of an American company's earnings has typically cost about $15. Today it costs around $22 or $31 depending on how you calculate those earnings, the former is the current earnings the latter uses the 10-year average of earnings.
To come at it another way, the average yield on U.S. stocks is historically over 4% and now it's closer to 2%, so about half the long-term average. Of course, the increasing use of share buybacks instead of dividends complicates that picture somewhat. There are plenty of valuation metrics out there, Warren Buffett's preferred yardstick is to look at the value of the U.S. stock market to the value of U.S. GDP the U.S. market is at 145% of GDP today. That's about the same as the peak of the tech bubble and virtually all other points going back to 1970 show the U.S. market at lower valuations. I'll stop there, but hopefully you get the point that the U.S. stock market is very richly valued today relative to its own history on a number of commonly-used metrics. That's simply what the data says.
Does It Matter?
So, we've established that the U.S. market is pricey in historical terms. Should you care? The answer is if you're a longer term investor then almost certainly yes. Very few indicators have much luck predicting the markets over the short-term. However, valuation has a degree of success in the pointing the general direction of the market over the next decade or so. This makes sense because valuation becoming more expensive and less expensive accounts for a lot of the market's valuation swings over history. Valuations seldom stay still. Though valuations could rise from here, it's also pretty probable that they could fall. That will be painful for U.S. stocks. Coming back to historically average valuations means potentially a halving in the U.S. stock market. So yes, valuation could be extremely important over the next few years.
What Are The Counter-Arguments?
Obviously not everyone is that concerned about U.S. valuations. Some argue that high valuations are rational now because bond yields are so low. There's some truth to that. Bond yields are extremely low today, but it doesn't quite work for the U.S. market. The U.S. 10-year government bond today carries a roughly 2% yield which is quite high by global standards. For comparison, Germany has negative government bond yields and yet it's stock market is trading far closer to average valuation levels at around 16x earnings. So there may well be a relationship between bond yields and valuation, but it can't account for all the run-up in U.S. stocks as it doesn't appear to be holding internationally to the same degree.
Secondly, the U.S. market contains a high proportion of tech, about a quarter of the S&P 500 is tech stocks. That's higher than other countries, and thus, the argument goes, the U.S. should trade at a higher valuation. First off that argument is a somewhat circular. Today the tech sector trades at around 20x forward earnings according to Ed Yardeni's data. If tech traded at 10x earnings it would represent roughly 14% of the U.S. market (all else equal), that's 10% less share of the market in absolute terms. Therefore, one of the reasons tech is a large proportion of the U.S. market today is perhaps because tech is relatively expensive and trading at a premium to other sectors. It's unclear that's a reason to pay more for the U.S. market. In fact, it could be a source of risk. Tech hasn't historically always traded at a premium. Remember airplanes were once considered cutting edge technology and before that railroads. So we need to be careful about attributing the valuation disparity to tech, because maybe one of the reason tech is a large part of the index today is because it's potentially overvalued. That may not last.
There's also the argument that the U.S. is just in a sense, better at capitalism than other countries. Therefore, you should pay a premium for that market. There may be some truth to this, but again there are some concerns too. First off, if the U.S. is a more profitable market, then a lot for the result of that should appear as higher earnings. Not as a higher multiple for those earnings.
Secondly the U.S. is increasingly the site of operations for global companies. For example, when Apple sells a phone in Brazil it still has to deal with all of the rules and processes that Brazilian firms face. So again, the argument may be directionally right, but can't account for all the run up. Also, when exactly did the U.S. develop this global advantage? Because it was likely true in the 1990s and 2000s as well, yet U.S. stocks weren't trading at such a large premium back then.
Finally, there is momentum. There's quite a lot of data to suggest that stocks that perform well, as the U.S. has, continue to perform well over the short-term. Thus we should continue to hold U.S. stocks. We can't ignore that, but again we should be careful. The same research that shows gains can continue for the short-term shows that this only works for about a year, sometimes less. So though this is perhaps helpful for a few months it's hardly a long-term source of support for the U.S. market.
What To Do Next
As such, as a buy-and-hold investor it's hardly time to exit the U.S. and move to cash. That's almost never a smart move. However, there are some academically backed approaches to consider. International diversification is something to consider. Harvard and Yale are already investing their money this way, perhaps you should follow. The U.S. represents about a quarter of total global GDP. Maybe your equity exposure to the U.S. should be closer to that level, or at least not all of your portfolio. You can make up the difference with other global stock markets, such as Europe, Japan and other international markets. A low-cost ETF that could get you there would be the Vanguard FTSE All-World ex-US ETF. This holds US companies entirely outside the U.S. at an expense ratio of 0.09% annually. If you want to up your non-US stock exposure, that fund is one potentially simple way to do it. It's price to earnings ratio is around 15x, so lower than the U.S. and closer to the long-term averages. Of course, that's not a complete portfolio, your bond exposure matters too for example. This is just a fix for potential elevated U.S. exposure. You can see some suggestions for complete portfolios here. Nonetheless, adding an international ETF could help you get your U.S. exposure down a bit, which may prove useful over the coming years if valuation models are any guide.
