Long term look at Composite Earnings and P/E, using 10 year average:


click for ginormous version

All charts courtesy of Bianco Research

Category: Earnings, Valuation

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

12 Responses to “S&P Composite Earnings, Long Term P/E”

  1. wj says:

    One caveat about the long term PE ratio is interest rates…
    Long term rates have been in a downtrend since about 1980, and you can see the corresponding increase in the S&P PE ratio during that time.
    The question going forward is, what happens now that interest rates are at rock bottom and eventually will start back up?

  2. rd says:

    The current generation of investors believe that the dips we have had over the past decade are worse than average despite their CAPE levels that are still at levels that would have signalled a market top over the previous hundred years (with the exception of 2009 which did simply drop to average levels).

    Every major market-wide fundmantal value indicator has indicated that the market has been at least one standard deviation above average levels for over a decade now with one brief exception of spring 2009 where it dropped to average levels.

    I don’t think Bernanke can abandon ZIRP until this graph shows a sustained PE ratio of less than 16 or else the raising of interest rates by itself will cause the stock market to crash. He has stated that one of the primary reasons he has maintained ZIRP is to keep asset values elevated, which is what this graph is proving is occurring in equities. He has to wait until something else implodes stocks (the Europeans are working hard to accomplish this) since it is unlikely that earnings will continue to streak upward.

  3. Sechel says:

    Superior returns tend to be in periods of p.e expansion, and one reason that p.e. ratios supposedly expand is lower interest rates. Since we are already at zero interst it’s hard to imagine lower interest rates being a catalyist, that leaves expectations that earnings growht will be sustainably higher than expecations which is fairly hard to imagine given world events and where we are on using CAPE values.

  4. Futuredome says:

    Bernanke will abandon “zirp” when the market tells him to. The market leads, they follow.

    If you want to get the McCoy take: real corporate profits peaked in 1979 and have been declining since then overall.

  5. littleboyblue says:

    Is it only coincidental that the PE chart heads higher in the late 80s just as 401(k)s become widespread. I am curious as to how much retirement money is tied up in the stock market just waiting to sell out on the next spike and move into fixed income as retiring boomers look to follow the formula and move into “safer” assets.
    Is there a bearish bias to everything in the stock market due to the boomer’s money looking to get out?
    How much 401(k) money is actually out there?

  6. CANDollar says:

    Lower Price Earnings Ratios may be here to stay… the 10 year average graph indicates an average lower than todays – I would not be surprised if the bottom in ratios is in the neighbourhood of where it was in 1982.

    Why? Usually with low interest rates the p/es would be higher but rates are kept low artificially with great externalities such as penalizing savers and retirees.

    Higher taxes on the horizon imply lower p/es. New economics research indicates this expectation can be powerful. Higher taxes are coming and it is dangerous to expect not.

    People are moving out in duration in order to get yield. Stocks are longer duration because they discount relatively far into the future but the low growth prospects of the economy imply higher discount rates and therefore lower p/es. In a market where p/e’s are declining the market often does poorly. However when the bottom of p/e’s is reached its a good buying opportunity over the long term. These could be p/es at 10 and below, or even the 1982 lows.

    The real interest rate is very low and this is often a leading indicator of p/es for the market… implying p/es have potential to drop even more.

    Lower p/es of course mean the market may not have the potential to grow rapidly in the next year. The average level that it goes to over the past 140 years when interest rates are negative is 11.6.

    All this adds up to tremendous vulnerability to shocks and a lower growth trajectory.

  7. nofoulsontheplayground says:

    Three patterns on that P/E chart suggest a probable bottom around 8.

  8. Iamthe50percent says:

    FutureDome: “… real corporate profits peaked in 1979 and have been declining since then overall.”

    Notice how this coincides with the ascendancy of Reaganism, the Chicago school, and Alan Greenspan?

    Corporations are like children, you can’t give them anything the want. They will try to eat only candy and ice cream, skipping the milk and vegetables. Instead of growing healthily, they get a gut ache.

  9. rd says:


    I think the PE expansion/contraction is more like to track demographics.

    The labor force saw a significant rise in the pecentage of the labor force that are in the most productive age group of 35-54 year olds from the 1970s to the late 1990s. It plateaued in the late 90s and early 2000s, and is now in a steady decline. These are the people working and paying money into savings and pension plans. The demographic slump is likely to go on for a couple of decades. I suspect the PE multiple will grind lower over the next decade as the demographic slump takes a firm grip on the economy and the market.

  10. rootless says:

    I prefer the ratio of the current price to the inflation adjusted 10-year average of the earnings over the P-E ratio averaged over 10 years as valuation metric. Latter could paint a distorted picture of current stock valuations due to large deviations from the average of the P-E ratio in the past.

  11. blackjaquekerouac says:

    i must say i’m struck with how the collapse of 2009 was far worse than the Great Depression. It only goes to show how important Ben Bernanke and what he did really was to this nation. That the people who still stand idly by and watch with pure hate those who eviscerate those who obliterated the entitrety of the American economy…AS A WAR WAS GOING ON…(don’t worry, “we’ll kill those people too!” i imagine them thinking) speaks volumes about how the “age of the internet” truly has crashed head long into the need for “a proper viewing of the historical record.” Simply put “you don’t see Ben Bernanke granting interviews” and “that’s why he get’s it.” Only now is the government realizing the need for “dummying up” now that everything (and nothing) is an open secret. Of course as anyone who has commanded anything having to do with an infantry unit will tell “holding back on the info is not a good way of endearing you to the troops.” God forbid if you did it to the Commander as well. I mean “there’s more oil in North Dakota than all of the Middle East”? Who knew? Anywho…”it’s quite a war”…that “war between the ears.” I think i’ll leave that to the government people myself.

  12. blork says:

    Try running the same analysis using average earnings EXCLUDING the 2009 and 2008 bust.

    Not trying to game the numbers just being true to the intent of the analysis. The whole point of the ten year average is to smooth out statistical anomalies and capture a full cycle. It makes sense to strip out at least one of those crisis years as they are clearly outliers in the data set (unless you are forecasting a continuing series of disasters in the next ten years). Stop and consider how it would actually MAKE sense to include the outliers. At least add in a “dummy” guess for what EPS would have been in a deep recession excluding the financial crisis if you want to preserve a 10 year sequence and a more reasonable business cycle. Result is a 17x-18x trailing PE if my (off the web S&P earnings stats) math is correct. Not screamingly cheap, but not as expensive as it looks. At least we will be grandfathering out the 2001 30% earnings decline that made this ten-year average-approach even less useful before.


    BR: Within a decade, you have two major market crashes — are those really “statistical anomalies” to be smoothed out? If you want to capture a full cycle, don’t you need to see how they end also?

    Or, do we simply pull out the data that we don’t like . . . ?