In March 2009, I got involved into the raging equity valuation debate by publishing S&P 500 P/E Ratio at Troughs: A Detailed Analysis of the Past 80 Years. I showed that the conventional absolute PE ratio approach widely used by the bears to recommend continued selling of equities was inadequate for the circumstances as it failed to take into account the significant decline in inflation rates (and interest rates). I explained, backed with 80 years of history, that equity valuations were then at a true historical low and that barring deflation, equities were at or near their lows and could advance 20-40% during 2009 with little downside risk. Since then, I have continued to successfully use The Rule of 20 to support my equity valuation work. This post explains why it is the superior valuation tool for the US equity market.

The Rule of 20 simply states that fair PE is 20 minus inflation with the total of PE plus inflation generally fluctuating between 15 and 25.

The chart below plots the S&P 5oo Index actual PE ratio (in red on the right axis) against the Rule of 20 ratio (in blue on the left axis). The median line is shared as 20 for The Rule of 20 and 15 for the Actual PE (PE 15) so that deviation around the median is visually similar. Arrows point to market highs and lows. The S&P 500 Index semi-log chart below is for reference.



During the 1960’s, the PE 15 was in the overvalued area most of the decade while the Rule of 20 gave 2 buy signals (1962 for a 69% gain and 1966 for a 40% gain) that the PE 15 failed to signal. In May 1974, the PE 15 gave a strong  buy signal while the Rule of 20 gave but a feeble one that it feebly reversed in mid-1975. The PE 15 failed to give a sell signal at that time. In fact, it remained in the undervalued area between 1974 and 1985 even though the market was unchanged between December 1976 and July 1982.  The Rule of 20 gave a strong buy signal during 1977 and a sell signal in 1980, after the market had gained some 75% and just before it tanked 24%.

The Rule of 20 gave another strong buy signal in mid-1982 (remember, the PE 15 had been flashing BUY since 1974). The PE 15 reached fair value in April 1986, which the Rule of 20 only reached in March 1987 after another 24% appreciation in the S&P 500 Index. Both ratios signaled SELL during 1992 but only the Rule of 20 gave a BUY at the end of 1994 with the PE 15 then only at the fair value level. Both methods moved to overvaluation and extreme overvaluation between 1997 and 2002. The PE 15 remained in overvalued territory until August 2010 when it reached fair value.

Meanwhile, the Rule of 20 ratio reached fair value in September 2002, gave a buy signal in September 2006, started flashing overvaluation in October 2007 and gave a strong sell signal in May 2008. It went to undervaluation in November 2008 and reached what could prove to be a generation low in February 2009. The previous such low for the Rule of 20 was recorded in June 1955.

During the early 1960’s, US inflation hovered around 1.0% for nearly 7 years following violent inflation and disinflation periods after WWII. During the 1970’s through 1982, inflation fluctuated between 3% and 15%. After that and right up to 2008-09, inflation was relatively benign between 3% and 5%. It weakened rapidly in 2008-09 and remains in the 1% range since.


It is not a coincidence that when inflation is either very low or very high that the PE 15 fails to provide equity investors with the better signals provided by The Rule of 20. The former, being merely a number (15) with “a trading range” around it (from 10 to 20) takes no account of meaningful changes in the inflationary environment, even though inflation has a direct impact on interest rates which directly (but inversely) impacts the discount rate that is the PE ratio. When inflation rises, interest rates also normally rise to maintain real rates within an appropriate range. PE ratios need to decline to reflect the increase in the earnings discount rate. Another way to look at it is that equities then face more competition for money from fixed income instruments. The cost of equities must therefore decline to keep or attract investors.

Furthermore, high inflation rates tend to reduce the quality of earnings through inventory profits for FIFO-accounting companies. If a larger percentage of earnings come from illusory and temporary inventory profits, these earnings should sell at reduced valuations.

Finally and importantly, investors know that rising inflation is generally not tolerated by central banks. An eventual rise in short term interest rates, often followed by an economic slowdown and lower profits, therefore gets factored into equity values when inflation rises.

At the other end of the spectrum, very low inflation rates (but not deflation) generally boost equity valuations. Interest rates then being generally low, investors tend to favor equities, especially when earnings are buoyant. Monetary authorities are often very accommodative in such periods, providing equity investors with an extended positive investment horizon as well as ample liquidity.

The PE 15 approach makes no adjustment for these meaningfully changing economic and monetary conditions. It leaves investors guessing what should be an appropriate multiple between 10 and 20 in spite of what could be very significant and fundamental shifts in the investing environment. The Rule of 20, while not perfect, helps investors adjust to the changing conditions as they happen. Eighty years of experience of equity markets valuation through thick and thin back the legitimacy of the Rule of 20 valuation method.



  1. Hey Dennis can you post/send me a template of how you did this chart – very interested in following this more often.


  2. Mark, the chart is simply a plot of trailing operating PE + CPI (YoY% ch). The result generally fluctuates between 15 and 25. I hope that answers. I update the chart fairly regularly on the blog.

    • Hi, very interesting work.

      However did you mean to say you plotted PE MINUS inflation rather than PLUS inflation given that the blue line is the rule of 20 ie “20 minus inflation”?

      This comment confused me.

      If not, why are you plotting pe PLUS inflation? What purpose does that serve? Am I missing something – I am no great statistical brain so am not sure of its significance!!
      thanks in advance for your response.

      • Hi sixpack

        Sorry for the confusion.

        The Rule of 20 states that PE on trailing EPS should be 20 minus inflation. If (Theoretical PE) = 20 – CPI, it follows that 20 = (Theoretical PE) + CPI.

        The plotted line measures the variation of the actual (PE + CPI) from “20″ illustrating that (PE + CPI) in reality fluctuates around 20, generally within a range of 15 and 25.

        Measuring the gap between actual and theoretical “Rule of 20″ PE is the risk/reward exercise that all investors should perform before committing big $ to equities.

        I hope that helps.

        Thanks for reading me


  3. Great post. While a useful back of the envelope gauge,I think there are a couple limitations to the rule of 20. The two I’d point to are profit margins and earnings cycles. The rule of 20 ignores potential for profit margins to change. There’s a robust debate about whether current year margins are sustainable. Implicitly you’re saying they are. One could possibly normalizing margins I guess? Second, earnings are cyclical and the rule of 20 ignores where we are in the market cycle. Notably, the S&P appears cheap/reasonable on a current year basis, but considerably more expensive using, say, a Shiller PE.

    Just some thoughts, I don’t claim to have any answers!

    • The Rule of 20 makes no forecast which eliminates the limitations you mention. Of course, one can look forward and adjust his own risk/reward ratio according to one’s own risk profile and cyclical bias. The Rule of 20 simply helps us understand where we stand on the valuation band, by itself an extremely valuable tool. The Schiller PE never got to undervalued level even before the S&P 500 Index more than doubled. Its problem is with the earnings base which got totally disrupted in 2008-2009. Companies with HUGE losses then are no longer in the Index but their impact is carried for 10 years!

      Thanks for reading me and for your comments

  4. Pingback: What If The Fed Has It All Wrong? « Exploring the World

  5. Pingback: THE “RULE OF 20” EQUITY VALUATION METHOD « Exploring the World

  6. Should the Rule of 20 take into account the changes in how inflation is now calculated compared to earlier methods used pre-1990? As I see on the web site the CPI as measured by pre-1990 methods has been running about 3% higher than current methods. So should the Rule of 20 maybe now be the Rule of 17?

    • Tks John, good point. This is a contentious and debatable issue. In any event, to the extent financial markets use the official CPI or PCE deflator in their assessment of inflation, the CPI remains the best benchmark. I am more concerned of the impact that all the QEs are having on interest rates. The CPI in the Rule of 20 is a convenient proxy for a `real discount rate`to apply to equity valuation. If the Fed keeps managing interest rates down, below their natural, free market, level, it has to have an impact on the discount rate. I should try to discuss that in the not to distant future.

      • That is a very good point to bring up– the Fed essentially owns a good chunk of Treasury curve for the most part, and is distorting the pricing of risk across maturities.

        Of course, when the Fed is faced with unwinding all this self imposed liquidity… one would think that (a) inflation would already be running rampant; and/or (b) interest rates would pop dramatically higher. I think when the market arbitrates what the interest rate will be– the Rule of 20 works very well.

        Factoring in the Fed suppression of rates across the curve, I would suspect that the Rule of 20 valuations are a little distrorted. How much they are distorted is obviously the $64 trillion question.

  7. Pingback: What If the Fed Has It All Wrong? | The Big Picture

  8. Pingback: What If the Fed Has It All Wrong? | CapitalInvestments

  9. I understand the calculation and use of rule of 20 to guide in estimating fair value on the index, however I’m a little confused on the use and interpretation of what you called the total p/e in the 2nd paragraph (p/e + inflation usually between 15 and 25).
    Is the total p/e measure just an additional metric that implies where a ceiling and floor on the p/e on the index should be?
    Thanks for your insight,

    • Tks Andrew and sorry for the confusion.

      Over the past 60 years, the sum of actual pe + inflation has fluctuated between 15 and 25 with a median of 20 (hence the name). The range enables us to measure risk vs reward which is what a good valuation tool should do. The closer to 25 we get, the higher the risk and the lower the reward and vice versa. Many people use absolute pe to assess risk/reward but absolute pe is generally meaningless without considering inflation. At the extreme, an absolute pe of 5 might look appealing but not so much if inflation is 20.

      I hope that helps

  10. Dennis I love the KISS aspect of this. I get to similar outcomes thru the more cumbersome steps of DDM assumptions on appropriate risk free rates given inflation, current yield curve geometric ERP means etc. but this goes straight to the heart. I’m curious if you also extrapolate is to individual securities?

  11. Okay, I’m a bit confused on the ‘buy signal’.

    Can I infer that, if the PE is below 20 – inflation, that constitutes a buy signal? For instance, if 20 – inflation = 17 but current PE of the S&P 500 is 15, that is a buy signal?

    Thanks for the great info.

    • Sorry for the confusion. I will try to make it clearer shortly. In the meantime, the Rule of 20 is a Risk/Reward analytical tool. Two ways to use it:
      1- If (Fair PE) = 20 – CPI, it follows that 20 = (Fair PE) + CPI. Since the Rule of 20 historical range is 15-25, actual PE + CPI is cheap nearer to 15 and expensive closer to 25.

      2- Fair PE is 20 minus inflation. If that = 17 and that the actual PE is 15, it means that the upside to fair PE is 13%.

  12. Pingback: Watched the debates? Now read this. | Casino Capitalism and Crapshoot Politics

  13. Very intriguing – I’ve been wondering how to use the Schiller PE index but haven’t been really comfortable with it.

    Question: Are you double-dipping on the impact of inflation? Increased money supply would already be built into the stock prices. It would also be built somewhat into previous earnings, although it wouldn’t be built into earnings taken prior to the inflation. Why deduct inflation from 20 to derive the “fair PE” when it’s already built into the “P” and partially into the “E” of “PE”?

    • Thanks Matt,
      1- You can’t assume everything is built into P. Too easy and too generous to the crowd.
      2- The P/E is a discount mechanism that needs a discount factor which is generally 5 or 10yr interest rates. Using inflation is another way that makes no assumption on the “real” interest rates. Sort of a real earnings yield…

  14. What objective or mechanical parameters did you use to determine the historical buy and sell “signals” you reference above in evaluating the historical performance of the “Rule of 20?”

    • Randy,
      the Rule of 20 is not a “buy-sell” indicator. It is best used to objectively measure the upside/downside relationship vs “fair value” to enable investors to determine the level of risk they are each willing to take at any point in time. Over the last 80 some years, it has been a very useful tool to objectively gauge valuation and therefore measure risk. You can judge by yourself its usefulness on the charts. For my part, I tend to be more aggressive when the R20 is below 17.5 and I gradually reduce my risk as the R20 gets closer to 20. Beyond 20, you know that valuation risk is getting historically high and increasingly dangerous as it keeps rising.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>