Warning: Wonkiness ahead:

Chris Turner took a look at Yale’s Bob Shiller “cyclically adjusted price to earnings ratio” (CAPE). Shiller uses an inflation adjusted S&P 500 Index (using simple monthly CPI data). The professor then divides that a 10 year average of trailing earnings (similarly CPI adjusted) earnings.

Chris wanted to know what happens if we pull the Cycle out of the CAPE? (Chris’ paper is here).

Short answer: You end up with a long term chart of inflation adjusted SPX valuation that implies the market, by Shiller’s metrics, has been overvalued (i.e, “Not Cheap”) for a long time.

Chart 1: Long term P/E and Interest Rates

More charts, and the paper’s conclusion, can be found here.

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.

27 Responses to “What is the Cyclically Adjusted S&P500 P/E Ratio ?”

  1. torrie-amos says:

    i’m getting the feeling none of this matter, until oil hits 100 with a peak at 120

    it would seem to me, money is running from europ, refuses to go to asia, ie, china, and everyone is happy in american where they have zirp, tbtf policy, and see no evil

    jobs will not matter based on how things are acting, this dollar looks like it is going to rest, and we’ll explode higher

  2. cognos says:

    Uh… arent we skeptical of anyone or any analysis that says… “the market has been way overvalued for a long-time”. Long-time returns have been pretty good. So how can something deliver good returns? Continue to have good balance sheet and cash-flow value? When starting from a “highly overvalued” point 20-yrs ago?

    Its an easy tell… that the analysis sucks.

  3. cognos says:

    And dont tell me… “the current overvaluation supported the overall returns”. That makes no sense.

    Since 1990… SPX is up 400% or 8.4% annually.
    Roughly 1/2 of that is from dividends (real cash divs).
    Dividends continue to be paid, and grow at a nice rate (watch em grow from here!).
    Dividends are <50% of EPS and EPS signficantly understates cashflow.
    Companies continue to get bought in M&A transactions at 50% premium to current prices (about $20B/wk in this recovery phase).

    The next PE and leveraged buyout phase will be even bigger and more profitable. PE firms kill it at almost ALL times except for deals done at the top. This despite mgmt fees, large legal costs, and paying 50% premiums to public stock prices to take companies private. How is this possible… if the original market prices are so over-valued?

    Makes no sense.

  4. bsneath says:

    PEs are a function of expectations on future growth, inflation, risk, interest rates, etc..

    For domestic companies with little foreign exposure, growth in the aggregate will be limited by less consumption due to credit contraction and demographics (aging population and slower immigration growth).

    This should place downward pressure on PEs since profit growth would be expected to be constrained. I think Japan’s experience is an example of this.

  5. RonN says:

    Turner’s paper didn’t convince me that Shiller’s CAPE (or PE10) is flawed in any significant way. Taking CPI out or using, as Turner did, an average of the individual year Price/tmte doesn’t make any sense. In the later, it’s the current not historical price that one needs to use to average over cycles.

    But the idea is good of using as a benchmark the PE10 over various time frames that take out “abnormal” times like the last 15 or so years and perhaps pre WW2 times, and using post 1950 thru 1996 as a “normal” benchmark. I did this and other analysis on Shillers data about 18 months ago (see http://seekingalpha.com/article/101934-how-low-are-pe-ratios-a-comment-on-mark-hulberts-take ). As one example of the results, the Shiller PE10 post 1950, taking out years with inflation over 4.5% had a median value of 15, making the market today overvalued by about 30%.

    Bottom line — one’s favorite fair value calculation is a matter of personal preference, but the difference between most of them is certainly much smaller than the “emotion or animal spirit” driven variations in stock prices and current PE ratios…. And remember, “valuation matters”!

  6. hgordon says:

    What happens to this model if you make some different assumptions about CPI ? Maybe use a basket of commodities or some other proxy for dollar purchasing power ?

  7. RonN says:


    “Since 1990… SPX is up 400% or 8.4% annually.”

    Remember, that the PE10 was 15 back in 1990, and one would say it was “fair valued”, so one would expect “historical (good) returns” over a long period of time, especially if we continue to support an “overvalued” condition, except for short periods of time….

  8. cognos says:

    1) The CAPE does not really seem helpful for making investment decisions (isnt that the point?)

    2) The main mechanism of “average 10-yr” earnings seem to do exactly the opposite of what should be done. That is, you trail a very bad (or very good) year… through your analysis for 10-yrs forward. Why? Isnt the future much more important than the past?

    I think a much more useful measure would be to just run the “Positive EPS” measure and look at PEs. Large losses by single companies (AIG, FRE, FNM) or by an industry being highly spec valued (Internet) dont negatively influence the “value” of the rest of the index.

    This has the potential to smooth the PE reading… making it more valuable for investment decision making. Such a measure might have better caught attractive valuations in 1982-84, 1991-93, 2002-03, and 2009.

  9. cognos says:

    RonN -

    Why is the “fair” return to a pretty tough valuation point (today, 2010) 8.4%? Why did “fair” get you many opportunities for 15-20% returns at other exit points? That seems a little better than “fair”.

    Todays PE is what… 18x? And TTM EPS will grow $10 after Q1 2010… dropping PE to 16x in like 2 months? Focusing on the FUTURE is way better than this CAPE stuff.

  10. dead hobo says:

    The objective for both appears to be the best way to draw the most magical chart.

    I’m not going to over analyze the issue because that is a waste of time. Is a company that loses money or has high non operating charges worth zero or close to zero because of a bad year? If all have a bad year but that situation might not continue worth a fixed multiple or is the multiple concept a non started in some cases? Can an objective P/E multiple ever be computer so that all agree it is the best one to use?

    Sky hooks, liquidity combined with demand for credit, and algos prop the price of stocks. Nothing else. The rest is pleasant contentment, posturing, sales pitches, and occasional shock and awe.

  11. Mr.E. says:

    It’s an interesting analysis that mixes some things that Shiller and others have looked at and published previously. Shiller typically averages the S&P 500 monthly values to derive his annual S&P 500 values – they are not single point in time values, but rather an average of the 12 months for each year (as used in his book “Irrational Exuberance”). Second, the issue of using averaged earnings versus recent earnings has been looked at both by Shiller and by Ed Easterling (see “Unexpected Returns” and Crestmont Research site).

    Personally I prefer Easterling’s treatment of secular cycles and his relationships between CPI and P/E’s.

  12. dead hobo says:

    Or, as the lizard brain in most of you might ignore, What you are looking for is a pattern you can anchor yourself on when reality is nothing but randomness connected by flows and interactions.

    Now, go buy stocks because the patterns look favorable. My special prognosticator says “Go For It”.

  13. Purewater says:

    Thanks for posting this, very interesting.

  14. Maj Tom says:

    Props to BR for posting my research – thanks to all those that read the paper – sorry to those that don’t like “magical charts.” (he he he)

    Don’t get too excited about what the data does and does not say or over-generalizations.

    Just a common guy here that 1) didn’t quite believe in assuming that the cpi adjusted 10 years of earnings were relevant, 2) didn’t think the timeframe involved was long enough (due to rapid growth of earnings from Mortgage Equity Withdrawal, Boomer peak earnings, change in accounting, etc… in the mid-late 90′s), and 3) wanted to find a more relevant “fair value.”

    The goal was to simply see if the cpi adjustment meant anything or if different timeframes changed the over/under value metrics. Additionally, financial TV was quick to tout Shiller’s proclamation last March that based on CAPE 10, the market was “undervalued” – yet say nothing now that his data shows clearly overvalued.

    Not to be too wonky (again) – but Page 14 has the summary – nominal vs Shillers CAPE and all time periods averaged – equal roughly the same numbers. I think the more applicable charts are the last 4 that use the 5, 10, 20, and 30 years of monthly pe’s averaged. Those track the S&P pretty well and catch the periods from over valued to undervalued – especially the 5 and 10 year. Those charts also show how crazy the last 15 years have been – can anyone argue against that? Especially since the S&P is back at 1998 levels…

    I also compared Shiller’s CPI data to Williams shadowstats numbers and have charts with those, but in the end, the “cpi adjustment” is simply not needed and doesn’t affect the computations very much.

    Did it help me? Sure – I bailed out of the 401K in May of 06 with S&P @ 1306. Bonds ever since. When will I get back into S&P index – when it gets below my fair value… Will it take awhile – probably. Will I miss 63% runs – probably. Will I sleep well at night – Most definitely. Will we reach “fair value” again – most definitely…

    Thanks for the discussion – CT

  15. dead hobo says:

    Maj Tom Says:
    February 26th, 2010 at 4:51 pm

    Thanks for the discussion – CT

    Don’t disagree with the over valued opinion. Even using the charts above and eyeballing an average, the S&P should be about 840. Since the S&P is clearly above that with no hint of going to ‘fair value’, I would opine that the entire concept of P/E ratio is a cocktail discussion and a sales tool and a way to sell books for the long run. As I said above, it is the lizard brain’s attempt to create patterns out of randomness. Much like the concept of a ‘very important technical level’. Show me the correct and pure P/E ratio and historicalize it and I will be impressed.

  16. for my own sense, after seeing this Headline:

    “What is the Cyclically Adjusted S&P500 P/E Ratio ?”

    ~4 different times, it keeps registering as: “What is the Cynically Adjusted S&P500 P/E Ratio ?”

    maybe it’s a good thing it’s Friday afternoon, or, maybe, I should be, down South, shaggin’ Flies in the Grapefruit League..
    past that, w/ this: “Warning: Wonkiness ahead:”

    Who are these People–~afraid of ‘Wonkiness’ ??
    or, When does the plumbing/endeavor for the *Truth cross the line into ‘Wonkiness’ ??
    or, differently, When, exactly, was it that We gave up on ‘Thinking for a Living’ ??

    Seriously, if all We can stomach is “All the News, That’s Fit to Print.”, we’re going to f****** starve..
    Maj Tom Says: February 26th, 2010 at 4:51 pm

    yon’ Maj.,

    if you’re up for some unsolicited advice, plainly, “Squawk it, as you See it.”, and, forget all about this: “Not to be too wonky (again)..”-type of self-impugnment..~

    to be clear, I appreciate your offering.

  17. cognos says:

    Except getting out in 2006 was just luck right? Bc on your own measure… you never really should’ve been in?

    And… you may get “fair” again and then buy. And it may be on a good dip. But you might miss out on 10% annual returns from now until that point. As long as the E rises. So I dont see how this is at all a value-added approach from a performance perspective.

  18. Maj Tom says:

    DH – the “correct and pure P/E” ratio is page #9 – hope you are impressed… btw – nothing is for sale, just trying to give out a little education for free… Sent this over to BR since he typically challenges the status quo – and he graciously posted it.

    MH – thx for the advice – after all the time spent digging into Shiller’s data and using his data for other research – gotta say, proud to be wonky now!

    Cognos – the computations with Shiller’s data were recent, the S&P exit was due to how crazy I thought the market had become – but the entrance will be planned, & no worries about the value of the research – just glad you are a fellow reader of BR.

    Best to all – CT

  19. MayorQuimby says:

    This tells me we should go higher. We need to get more ‘irrational exuberance’ on board before punishing them with another collapse. I think tops are usually formed from good news unless there’s a black swan (ie 9/11, Lehman). Eventually – people will realize – long term cd’s on a rolling basis is probably the best investment out there (as long as Bernanke or Greenspan aren’t working at the Fed).

  20. budhak0n says:

    Hey how’s about the people who requested to get out, especially in their 401′s in MAY 2008 according to the new IRS regulations that allowed trustees ( if they so chose) to release 401(k) assets to another form – IRA/Roth IRA/ Individual bank trustee?

    See it doesn’t matter anyway because in order to fight THAT fight you’d have to leave your employer.

    But seriously, if we were really interested in having everybody play from an even playing field, what we would do is we would force Social security to be voluntarily funded, we’d force Trustee’s and mutual fund managers and other smack arses to disclose fully every trade, we’d return funding back to the owners without all the fighting.

    But we’re not going to do any of that because if they have control of your liquid or illiquid assets, they aren’t going to let go unless you pry their dead hands from the booty.

    And now what they are pissed about is you could ACTUALLY own your own property. Actually hold title to it.

    Without their funding involved. The entire Bank hissy fit originates not from the fact that the economy sucks or that there’s no revenue streams etc etc etc.

    It originates from the fact that PEOPLE paid their obligations. The worst thing you could ever do is settle up.

    Because once you settle up, there’s no more revenue, and worse yet, NO more claim of “clients” with which to issue backdoor paper as in stock or other equity instruments which can then be sold on public markets so the corporate officers can benefit without having to pass through any FICO or other hoops.

    Meanwhile John Q Public is stuck with a pissant banker with an attitude, and the local convenience store clerk who imagines themselves to be in the securities business.

    God bless the Internet.

  21. Pat G. says:

    “has been overvalued”

    Absolutely, but investment fund managers are required to be well, invested… Let’s look at their other options. Oil… overvalued based on its bearish supply buildup. USD… overvalued based on issues in Euroland. Treasuries… overvalued based on risk versus yield. I’m told we are a global marketplace which means we will all feel each others pain eventually.

  22. kmckellop says:

    Looks like the constant dollar DOW to me.


    Another 50% haircut in the DOW at least is needed to return to normalcy.

  23. constantnormal says:

    @Maj Tom

    THANK YOU — nice bit of analysis, great thought igniter. (not that my thoughts become flames or anything)

    If only everyone would take the time to grind out data to support/study their ideas.

    And thanks to BR, for promoting good stuff, no matter that it’s from outside the NY financial community or the halls of academia. There are precious few places in this world where you can get a decent discussion/dissection of ideas these days.

  24. constantnormal says:


    I think you’re prolly right … dammitall

  25. Mr.E. says:

    Ground contol to major Tom (anyone else remember David Bowie?) …

    Your analysis and conclusions are good. The significance of P/E relative to a nominal or “fair value” has been beaten to death by many, most notably Shiller and Easterling. After a pause for some liquid refreshment and further consideration I realized a couple of additional elements you may want to consider in your assessment of “fair value”. One is inflation, the other is expected return vs. risk or the risk premium expected by the market. Both play an important role in determining the fair value of an investment asset. While it is quite clear that returns on stocks are optimized over the long-term (i.e., 20 years or more) when purchases are made in periods of low P/E’s, the question remains when are P/E’s low enough that stocks offer attractive returns in light of inflation and the risk that stocks represent vs. other options?

  26. hgordon says:


    Nice chart. That was essentially my question further up. CPI is a manipulated number, but purchasing power of the dollar is real, however you choose to measure it. I think Maj Tom’s logic is that inflation doesn’t matter because it should hit both price and earnings, and therefore it factors out of the ratio.

    Here’s a year-old version of a similar chart with S&P/gold ratio -

    Not a lot different from the Dow version, and we’re still hanging around 1.0.

    BR had this chart back in 2006 when the ratio was at 2.5 -

  27. ivanhoff says:

    P/E is just a measure of risk appetite and investors’ confidence in the stock market, which are derivatives of liquidity in the system, perception of personal wealth growth and access to credit.