Here are some charts from a Credit Suisse research report that you should keep in mind when reading articles about high valuations and future returns. I first saw the highlighting done by WSJ Daily Shot, but I can’t find the post anymore.

Here is a chart plotting the starting Forward P/E ratio and the subsequent 10-year annualized returns for US Stocks (S&P 500). As of September 2018, the Forward P/E ratio is approximately 17. As you can see, this is higher than the historical average, with the trendline suggesting that the average expected future return is about 5% annually. Chart like this explain why many articles warn of low returns ahead.

However, step back and you see that the historical range has varied from -1% to 8%. Over 10 years, that’s a big spread. It’s the difference between $100,000 turning into $90,000 or $215,000. Focusing on the line gives you a false sense of accuracy.

Here is a chart plotting the starting Forward P/E ratio and the subsequent ONE-year annual return for US Stocks (S&P 500). In the next year, just about anything could happen! You should shoot up 30%+ or drop 30%+. (There appears to be a typo with the vertical axis missing some zeros. I believe the range should be from -60% to +60%.)

P/E ratios are not reliable for market timing. Yes, P/E ratios are something to watch and consider, especially to keep reasonable expectations for long-term returns. But when people sell all their stocks, they feel a drop is coming soon, not 10 years out. That crash might happen. But it might not. As the John Maynard Keynes quote goes, “The market can stay irrational longer than you can stay solvent.”

In my opinion, it is a much more reliable bet to maintain broad exposure to stocks for decades. You can do some trimming around the edges if that helps you minimize regret.

Warren Buffett had his annual charity lunch today and was on CNBC for a short interview. As usual, he was asked about the current stock market situations and, as usual, he managed to sum everything up in a few folksy sentences. Here’s a direct quote from the full CNBC interview:

If you had your choice between buying and holding a 30-year bond for 30 years or holding a basket of American stocks, there’s just no question, you’re going to do better owning stocks. It’s more attractive than, considerably more attractive than fixed income securities. That doesn’t meant they’re going to go up or down tomorrow, next week, or next year, but over time, a bunch of businesses that are earning high returns on capital are going to beat a bond that’s fixed at roughly 3% or 30 years. And it’s not my field of specialty, but actually they look, stock generally (American businesses), they look cheaper than, generally, real estate.

[…] That’s what you have to do in investing. I mean, you’re sitting with some cash in your pocket. You have savings, and the question is what do you do with it? You can buy a duplex next door and rent it out to people and do fine over time or buy a small piece of farmland or something of the sort or you can put it into something fixed income, bonds, or bank deposits, or whatever it may be.

My interpretation is that you have invest your money somewhere, and if you have a 30-year time horizon and you don’t plan on timing the market in and out, then stocks are still your best bet. Over the long haul, stocks will still outperform bonds and cash (at current interest rates), and he thinks real estate as well right now. Timing the market is too hard to do. Predicting returns over the next 5 years is too hard to do. If you don’t have a long enough time horizon or can’t handle the swings, you shouldn’t be in stocks.

Long-term stock investors just have to take some lumps if prices drop for a while. Keep enough money in bonds and cash so you don’t panic and have money to spend in the meantime.

[I actually have an issue with the CNBC caption “Buffett: Stocks always more attractive than bonds”. He never said that. He specifically noted that the 30-year bond was paying 3%. In the past (1970s?), Buffett has invested in Treasury bonds when the rates were really high and the stock market was overvalued. If today’s rates were 8% instead of 3%, Mr. Buffett would be rational enough to adjust his opinion.]

I’ve read Jeremy Grantham’s quarterly letters and the GMO 7-Year Asset Class Return Forecasts for over 8 years. (Anyone can sign up at GMO.com for free.) Grantham gained successively more fame after avoiding the Japanese bubble, the Dot-com bubble, and the Financial Crisis. Given the recent talk about record bull markets, let’s take a look back and see how accurate the most recently applicable GMO predictions turned out to be.

Let’s compare the GMO 7-year return forecast with reality, in this case actual returns from January 1st, 2011 to January 1st, 2018. Below is the forecast chart as of December 30, 2010. I looked up the closest ETF or mutual fund that I would have invested in for each asset class, and then found the total return via Morningstar or ETF Replay for 1/1/2011 to 1/1/2018. According to BLS.gov and SmartAsset, the total inflation during that same period was 1.6% to 1.7 annualized. I added the actual 7-year real returns onto the old prediction chart below.

(The blurred-out asset classes are basically special investments that GMO was selling to their institutional clients that the Average Jane could not have easily bought via ETF or mutual funds. Since they didn’t make their position clear in the beginning, it’s hard to judge the subsequent results.)

Some quick and simple observations:

• US Large-Cap and US Small-Cap stocks did a lot better than forecasted. Compounded over the full 7 years, the difference was on the order of doubling your money vs. making nearly nothing.
• International Developed Large-Cap and International Small-Cap stocks also did significantly better than forecasted.
• Emerging Market stocks did significantly worse than expected.
• Bonds did about as expected across the board.

Reading the commentary is always illuminating and helps me better understand their forecasts. Grantham has addressed these results here and there, but basically he still thinks a reversion to the mean will happen eventually. Here’s a quote from one of his recent quarterly letters:

“Relative to what we were thinking [in 2010], emerging equities have done surprisingly badly, and the U.S. equity market has done surprisingly well,” said Grantham. “Was that the luck of the draw, which has no bearing on future returns? Was it a temporary phenomenon that will soon reverse? Or does it tell us something important about emerging being a value trap and/or the U.S. being extraordinary that we need to take into account in our forecasting of the future?”

The short answer to these questions is that while emerging markets “deserved” some of their bad luck over the last several years and the outperformance of the U.S. has made some sense, we do not believe that emerging is a value trap, nor do we believe that the U.S. has proved itself particularly extraordinary.

Here’s GMO’s most recent forecast as of July 31st, 2018:

I understand the fundamentals behind these numbers, but I can’t help but think that if you keep calling for a drop long enough, it’ll happen eventually. There’s a reason why market timing is hard and why it’s called a “risk premium”. You never know exactly when the drops will come. All you might really be able to say is that a drop is more likely now than when people were worried in 2014, 2015, 2016, 2017… The rubber band is stretched, but it’s been stretching for a while and it could still stretch even longer.

Bottom line. Forward-looking stock return forecasts can be off. By a lot. Most of them rely on a reversion to historical average valuations, which doesn’t always happen in a timely fashion (who knows, maybe not ever all the way back?). Forward-looking bond return forecasts are made differently, and more likely to be kept within a tighter range. I sitll believe in simple diversification between stocks and high-quality bonds.

After my last quarterly portfolio update, I rebalanced back to my target asset allocation this week. I sold some US stocks and went from 70% stocks/30% bonds back to 67% stocks/33% bonds. (I bought a little Emerging Market stocks as well.) In preparation for living off my portfolio, I also recently increased my separate “emergency fund” to two years of household expenses.

This is all part of my normal investment plan, but current prices do show a market valuation that is pretty darn optimistic. If we get Vanguard’s expected asset class returns along with their projected inflation of a bit under 2% annualized, I’d be actually be fine with that.

Of course, the real fear is something much worse – a severe recession and extended bear market. I like to imagine this ahead of time and give myself a mental “stress test”. This is more about imagining your future behavioral responses than running fancy computer simulations.

A simple version of such a crash scenario is to imagine the following:

1. Your stock holdings drop by half over a short period.
2. Your bond holdings have zero total return for the same period.

Below again is a chart including multiple 50% drops.

In addition, here is a JP Morgan AM slide that shows that sharp intra-year drops are much more common than you might think from just looking back at annual returns:

My personal portfolio is basically 33% US Stocks, 33% International Stocks, and 33% Bonds (33/33/33). In the crash scenario above, my portfolio balance would drop by 1/3rd and my new asset allocation would be 50% stocks and 50% bonds.

Based on past experience, I will probably find it difficult to keep buying stocks as they keep dropping. However, I will probably find it tolerable to hold on without selling. This might be a hoarder thing, although I actually don’t do that much with physical stuff.

How long could I hold out in a crash scenario? I think in term of “years of expenses”. First, I have my two years of expenses in cash. My withdrawal rate is 3%, so if I didn’t want to touch my stock holdings, I could withdraw funds out of my bond holdings alone and still have another 11 years of expenses (33% divided by 3). This would be my form of “rebalancing”. Instead of selling bonds and buying stocks, I’d just gradually sell bonds and buy food. 😉

In addition, my stock holdings would still distribute dividends. Right now the dividend yield is about 2%, but maybe in a severe recession the total dividend also drops to half of the original amount. Taking the cash, bonds, and depressed stock dividends together, I could go about 17 years without selling a single share of stock.

Hopefully, stocks will rebound well before 17 years pass. People like to point out that to get back to even after a 50% drop, you’d have to have a 100% rise. True. But look at the chart above again… 100%+ rises happen more frequently than 50% drops. It’s easy to forget how crazy the swings can be in both ways. Staying out of the market at the wrong time hurts too.

There are other options that are less fun and harder to count on. I could spend less money. Cutting back might come more easily when everyone else is cutting back as well. I could get more work. It might be harder to find a job in a severe recession, but even a lower-paying job would help.

Bottom line. This is my rambling stress test as someone planning to live off their portfolio for another 40+ years. Hopefully, you’ve gone through something similar that fits your situation. If you’ve got a good steady job, maybe it’s most important to ignore the noise and ABC (Always Be Contributing). Some retirees put 5 years of expenses into cash or bank CDs so they “know” that they can last 5 years without having to take money out of a depleted portfolio. If that sounds like a good idea, I’d do it sooner rather than later. There are many bank CDs earning around 3% APY right now.

When it comes to constructing a portfolio, I used to think it was all about numbers and optimization. When you pick an asset class based on historical data, that assumes you hold through both the good times and the really bad times. It has helped me to keep gathering nuggets of knowledge over time to maintain my faith during those really bad times.

I’d like to start a series of posts to document why I own each specific asset class. Somebody asked me about TIPS the other day, so I’ll start with them. I’m not an investing professional, just a semi-retired DIY investor who wants to keep on learning and would like to share with other like-minded folks. I won’t get into the tiny details, mostly a lot of charts, links, and higher-level ramblings.

Treasury Inflation-Protected Securities (TIPS) are bonds issued by the US government that pay interest which is linked to inflation. Inflation is measured by the Consumer Price Index (CPI). In terms of a useful all-around primer, this Morningstar (M*) article 20 Years In, Have TIPS Delivered? covers a lot of the bases. For more nuts-and-bolts mechanics, see this older Vanguard paper Investing in Treasury Inflation Protected Securities. According to M*, TIPS currently make up about 9% of the overall Treasury market.

Here are my reasons for owning Treasury Inflation-Protected Securities (TIPS):

TIPS are backed by the US Government, just like the more common “vanilla” US Treasury bonds. With bonds, I prefer to stay on the safer end of the spectrum. Bonds are debt, and I don’t want to worry about if I get paid back. Buying US Treasury bonds is the lowest amount of credit risk possible.

TIPS provide a “real” inflation rate at purchase, which means it is guaranteed to provide a set return above inflation (before taxes) until maturity. Very few bonds are structured in this manner. In simplified terms, if the real interest rate is 2% and inflation is 3%, then the total interest paid will be 5%. Working backwards from that, you get the concept of breakeven inflation rate, or the expected inflation by the market (via M*):

The introduction of TIPS brought with it a market-determined observable real rate of interest, which is what the yield on a TIPS is. If you subtract the TIPS yield from the comparable-maturity nominal Treasury yield, you get the market’s inflation expectation over the period until maturity of these two bonds. This is called “breakeven inflation,” because it is the level of inflation at which returns for the nominal and inflation-indexed bonds should break even.

Here’s how the expected inflation and actual inflation compared for 5-year periods since 2003. For the most part, they have been pretty close:

TIPS thus provides insurance against *unexpected* inflation. TIPS are often described as an inflation hedge, but it’s more of a hedge against unexpected inflation. All bonds are already priced with inflation in mind. If everyone thinks inflation will be high, then bonds across the board will be priced to pay out more interest to counter that.

The reason why you don’t hear much about TIPS in the media is that over the last several years, there hasn’t been any unexpected inflation. If you bought fire insurance on your house, and your house hasn’t burned down yet, are you going to stop buying the fire insurance?

TIPS also provides a certain amount of protection in case of severe deflation. TIPS are guaranteed to return par at maturity, meaning they have floor value even in a case of severe deflation. This asymmetry helps make TIPS attractive relative to Treasuries, as best explained by this EconompicData post:

Thus, assuming a view that an inflationary and deflationary scenario are equally likely, the unlimited potential outperformance of TIPS vs. Treasuries in an inflationary environment and limited upside of Treasuries vs. TIPS in a deflation environment would sway an investor towards TIPS.

If inflation meets the market expectations, then TIPS and Treasuries will have the same return. If actual inflation is higher than the (expected) breakeven inflation rate, then TIPS will pay more than the regular Treasury bond. If actual inflation is less than the (expected) breakeven inflation rate, then TIPS will pay less than the regular Treasury bond. Here’s a simple graphic from AAII:

Here’s a 2008-2018 Morningstar chart comparing the growth of $10,000 between the Vanguard Intermediate Treasury Fund (VFITX) and the Vanguard Inflation-Protected Securities Fund (VIPSX). You can see while there is definitely a difference – sometimes one leads, sometimes the other – but over the last 10 years the net return has been very similar. Again, inflation has not been much higher (or a lot lower) than expected.

Do you think future inflation will be higher than the current expected number? Here’s the 5-year breakeven inflation rate for the last couple of years. Via WSJ Daily Shot.

If I had to bet, I would bet that the future inflation number will be higher than 2% then less than 2%. However, most likely they will return around the same amount. So this is not a huge risky bet. In terms of the big picture, it’s a relatively wimpy bet. I currently hold about 1/3rd of my portfolio asset allocation in bonds, and about 1/3rd of those bonds are invested in TIPS. That means about 11% of my total portfolio is in TIPS. If the real yields on TIPS were to go back higher to historical levels, I would go back up to 50% of bonds in TIPS.

Here are some reasons for NOT owning Treasury Inflation-Protected Securities (TIPS).

• TIPS are not part of the efficient frontier. If you run an mean-variance blah-blah-blah optimizer, you won’t find TIPS on the ideal risk/return curve.
• TIPS have a low historical correlation with stocks, but not as low as regular Treasuries – regular Treasuries are a better bet to go up when the stock markets crash.
• TIPS have only been around for 20 years. You might argue that they have not been tested in severe high-inflation environment.
• As with nominal Treasuries, the interest is taxable as ordinary income rates, not the lower dividend or long-term capital gains rates as with stock dividends. You’ll have to pay taxes on this interest every year – it can’t be deferred like if you buy a stock and hold it for a long time. If you buy individual TIPS, you’ll also have to pay income taxes on the inflation adjustment without actually getting the interest until maturity. This is called “phantom income” but can be avoided if you buy TIPS via an ETF or mutual fund. TIPS are thus generally recommended to be kept in tax-sheltered accounts. (TIPS interest is exempt from state and local taxes, however.)
• Some people worry that the government will fudge the CPI numbers if high unexpected inflation really becomes a problem.