Valuation indicators tell you whether a stock, market, or asset class is “undervalued” or “overvalued”.
- A market is “cheap” when it is “undervalued”. Conventional wisdom states that investors and traders want to buy assets that are “undervalued”.
- A market is “expensive” when it is “overvalued”. Conventional wisdom states that investors and traders want to sell assets that are “overvalued”.
This concept isn’t only used by value investors. A lot of mean-reversion (contrarian) traders use this as well. The standard logic states that valuations should “mean revert”.
- Assets whose prices are “overvalued” should fall in the future, thereby bringing their valuations down.
- Assets whose prices are “undervalued” should rise in the future, thereby bringing their valuations up.
But this is the point that many traders and investors don’t realize:
Valuation can only be used as an EXTREMELY LONG TERM indicator. It cannot be used as a medium term indicator for market timing.
- When valuations are “high” (expensive), the expected future return for 5-10 years later is low.
- When valuations are “low” (cheap), the expected future return for 5-10 years is high.
HOWEVER, the market’s 1-2 year future returns have almost no correlation with valuations. Hence you should not be bullish right just because valuations are low, and you should not be bearish right now just because valuations are high. Valuations can’t be used for timing, and timing is everything in trading and investing.
Here’s an example, using the “Stock market’s market cap / U.S. GDP” valuation indicator.
You can clearly see the flaw in this valuation indicator: it isn’t useful for picking tops in bull markets. Based on this indicator, valuations have been “insanely expensive” from 1994-present. The S&P 500’s nominal value in 1995 was 450. Anyone who stayed out of stocks because this indicator says that the “stock market is overvalued” would have missed out on massive gains. (The S&P is up multiples from 1994-present).
Here are some commonly used valuation indicators. Tobin’s Q is the best.
The Tobin’s Q Ratio is the “total price of the market” (nominal value) divided by the “replacement cost of all its companies”. This replacement cost is essentially akin to “book value”.
“Replacement cost” is a simple concept in theory but insanely difficult to calculate. That’s why the Federal Reserve calculates this number once a quarter in the Z.1 Financial Accounts of the United States. This number is usually released 2.5 months AFTER the quarter is over.
Tobin’s Q looks like this:
Tobin’s Q is arguably the best and most comprehensive valuation indicator for the U.S. stock market. It includes the pre-existing assets of U.S. companies, something that a valuation indicator like the S&P 500’s P/E ratio doesn’t.
Some traders and investors dislike Tobin’s Q because they say that the Q Ratio doesn’t take into account U.S. corporate earnings. Well it does. All earnings become assets for these companies.
The Tobin’s Q Ratio also fluctuates within a nicely defined band from 1900-1994. Since then this band has shifted upwards, which is what’s happened to every single valuation indicator.
S&P 500 P/E Ratio
There are 2 S&P 500 P/E Ratios.
- The 12 month forward S&P 500 P/E ratio. This divides the value of the S&P 500 today by the S&P’s projected earnings in 1 year.
- The 12 month trailing S&P 500 P/E ratio. This divides the value of the S&P 500 today by the S&P’s earnings over the past 1 year.
Both of these valuation indicators essentially treat the S&P 500 as a bond. “Valuation” is treated as how much ROI you’re getting on an investment.
The problem with the S&P 500 P/E ratio is that unlike Tobin’s Q, this valuation indicator doesn’t have clearly defined peaks and bottoms. As you can see in the following chart, the stock market can be “overvalued” for a long time.
This valuation indicator has another problem.
- Valuations should be going down in a bear market because the market is falling.
- Valuations should be going up in a bull market because the market is rising.
Sometimes the S&P 500 P/E ratio will go UP in a bear market! This is because corporate earnings are falling even faster than the S&P 500.
These declines in corporate earnings tend to be one-time events as companies write-off losses. Based on this valuation indicator, the stock market would be getting more “expensive” in a bear market, and anyone who avoids equities because of this indicator will miss out on the bottom of the bear market.
The S&P 500’s forward P/E ratio tends to have a similar problem.
Hence, this valuation indicator is inferior to Tobin’s Q.
Market capitalization as a % of Nominal GDP
We’ve already mentioned the basic flaws for this valuation indicator. This is arguably one of the least useful valuation indicators. I don’t know why this is Warren Buffett’s favorite stock market valuation indicator.
For starters, “valuation” is supposed to divide the market’s price by the market’s “value”. Why would you bring in a third variable (e.g. nominal GDP)? It doesn’t make any sense!
The stock market and GDP move in the same direction over the long run, but they don’t move in a 1-to-1 ratio. U.S. companies have become much more profitable over the past 30 years, hence this valuation indicator’s mean has gone up over the long run.
In addition, it’s worth nothing that there’s nothing “special” about a Market Capitalization to GDP % of 75%. For example, Hong Kong’s Market Cap as a % of Nominal GDP is almost 300%!
S&P 500 relative to its regression trendline
This is arguably the worst valuation indicator because it doesn’t actually measure valuation at all.
The S&P 500 returns an average of 7-8% a year over the long run.
This indicator basically just plots the S&P 500 (inflation adjusted) against a 7-8% year-over-year increase in the regression trendline.
The idea behind this indicator is that when the market is above its long term trendline for too long, it will eventually mean-revert to the trendline. When the market is below its long term trendline for too long, it will eventually mean-revert to the trendline.
But as you can see in the above chart, the market can be above or below this regression trendline for a long, long time (i.e. 1-2 decades). Hence it can’t be used to pick the top of bull markets or the bottom of bear markets at all!
This valuation indicator is useless because it just assumes that the S&P 500 will forever yield an average of 7-8%. I don’t expect it to in the future. There is nothing “magical” about the 7-8% number. The S&P 500’s long term performance might be higher or lower in the future depending on a number of factors:
- Long term interest rates. Interest rates have been much lower in the past 30 years than the 30 years before that.
- U.S. corporate profitability. U.S. corporate profit margins have been consistently elevated over the past 30 years.
- Investors’ appetite for stocks. The concept of investing in stocks wasn’t really prevalent until the 1980s. Before that, mom and pop investors tended to put most of their savings into bonds. A large part of this mentality change was thanks to the 401(k).