SP500 price to trailing twelve month earnings

Trailing twelve months is too short time frame to measure long term earnings potential. Nevertheless, past twelve months earnings (Ettm) is often used as a short hand for real fundamental analysis. Figure below shows the SP500 vs Ettm. Currently, the P/E=27 which is high by historical standards but not a record. Again, just one year P/E is not very predictive of the near future but it shows that stock market is not cheap.

Figure: SP500 vs. trailing twelve month earnings.

Using Topin's q to estimate future stock market return

Tobin's q is the ratio between a physical asset's market value and its replacement value. We can use Tobin's q gauge to value of stock market relative to its replacement cost:

q = (value of stock market) / (replacement cost)

Or written differently

q = P/B

where P is the price all the stocks of all the companies and B is the book price of what it would cost if we were to build the companies from scratch. In essence, q is a measurement of stock market valuation relative to their assets.

Figure 1 shows historical q for U.S. corporations. It has varied widely over time depending on business cycle, inflation expectations, and most importantly, the investor sentiment. The average q = 0.71 indicating that historically the corporations have traded below their replacement value. Currently q = 1.1 which is historically high. Only during the internet bubble years has Topin's q been higher.

Figure 1: The U.S. stock market Tobin's q.

Valuation is not a good indicator for near term market movement but it can be used to estimate longer term stock market returns. Over several years, the valuation metrics tend to revert to to long term mean. We can use this fact to estimate the future stock market return by writing the price N years into the future as

Here qAVE is the historical average q, qNOW is the current q value, g is the average growth rate for the assets, and PNOW is the current stock valuation (Geek note: derivation at the end of this article). Note that this relationship does not account for dividends.

Figure 2 shows the modeled annualized stock market returns and actual realized returns over the 7 year time span. The plot does not account for dividends and is not adjusted for inflation. Currently, the projected return is negative. With dividends included, the nominal 7 year return is about break even meaning that one could just hold the cash and avoid the stock market risk.

Figure 2: The stock market return estimate based on Tobin's q ratio.

APPENDIX: The derivation between the current stock market price and future price estimate using q ratio:

Sector rotation for market timing: theory and practice

Figure 1 shows the classical sector rotation and its relationship to stock market cycle. The theory is simple:

1. In early bull market, the financials, technology, and consumer discretionary are the best performing assets. These assets have been heavily beaten in the preceding bear market and are ready to bounce.

2. In later bull market, when business cycle has turned, industrials, materials, and energy are the best performing assets.

3. In the bear market, investors seek safety in defensive sectors that are not affected by business cycle: Staples, utilities, and health care do well.

Figure 1: Sector rotation and market cycle (theory).

If the above holds, we can use it to time the market. Figure 2 shows historical data. The correlation is not perfect but usable. The bull market in 2003 with Stage I sectors (green) showing strongest performance. Later, Stage II sectors (yellow) took over. In 2007 we had a brief warning with defensive sectors briefly leading (red) but the bull was not yet done. The bottom of bear market in 2009 was market with Stage III sectors leading (red). 

Figure 2: Historical sector rotation and market cycle.

The cycle since 2009 has been more confusing. The cycle started with Stage I sectors leading as expected but subsequently we have had Stage I, II, III sectors alternating. Currently, Stage III sectors are the relative leaders which is not a bullish sign.

Buy and hold for the next decade

As I note in "Market breath is bad and deteriorating"  on 11/17/2014, now is not a good time to buy stocks:There is a strong risk that stocks will be lower in one year time frame. Long term buy and hold investors have different horizon. What can they expect for the next decade?

In the long run, the investment return is sum of dividend return and price change. In long term, the price tends to mean revert to value. This allows estimation of the long term stock market returns shown in Figure 1.

Figure 1: Modeled and actual historical stock market return (dividend + price change).

The model is not perfect (how could it be?) but correlation is significant. Based on this, the annualized return for the next decade is about 5%. This is well below historical returns as the stocks are overvalued compared to historical norm. As shown next, this may actually be the Goldilocks case. 

The inflation adjusted historical returns are lower. Shown in Figure 2, adjusted for inflation, the stock market returns can be negative for a long period of time. The 70's high inflation was a investment equivalent of hell. This is the real danger of buy and hold in the current environment: Should the central banks succeed in releasing the inflation, the future stock returns could significantly underperform the model return of 5%.  

Figure 2: Modeled and actual inflation adjusted historical stock market return (dividend + price change).