My Sunday Washington Post Business Section column is out. This morning, we look at a famous aphorism from Sir John Templeton. The print version has the hedder When is ‘this time’ really different? while the online version used the fuller Investors must recognize what ‘this time it’s different’ really means.

The column tries to draw the difference between the psychology of when investors are rationalizing paying anything in a speculative bubble versus the times when there are actual differences in fundamental economic metrics. Despite the differences between these two situations, people apply — and mis-apply — Templeton’s famous aphorism.

Here’s an excerpt from the column:

“So what is the formula for recognizing what “this time it is different” means? (And why is it that it’s never different? The answer, I believe, is two parts math, two parts psychology.

The math part begins with the traditional formula for valuation. In its most simple expression, it is price relative to earnings, or the P/E ratio. That is the oversimplified version. There are many other factors that also impact valuation: economic growth, inflation, interest rates, cost of capital, investment options, etc. And there are countless variations of how to measure it. Last year, Merrill Lynch’s quant team looked at 15 metrics that measured equity valuation — but the basic formula is typically a version of price relative to a profit related metric.”

I really like what the Post did in the dead tree version of the paper — the chart below, and the headline on the jump – Think this time is different? Take your temperature. Then check the data.

click for ginormous version of print edition

different data



Think this time is different? Take your temperature. Then check the data.
Barry Ritholtz
Washington Post, May 18 2014

Category: Apprenticed Investor, Psychology, Quantitative

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.

18 Responses to “When is ‘this time’ really different?”

  1. ch says:

    Barry – CPI calculated as we did in 1990 and 1974 would yield an inflation rate running somewhere between 7-10%. We know this b/c John Williams calculates it.

    That changes the calculus above quite a bit.


    ADMIN: Those numbers are highly suspect, and numerous people have challenged the underlying assumptions and math behind SGS

  2. jib10 says:

    The links are broken. Looks like WP mundged them.


    ADMIN: That was my error; I omitted the “http://” — I just fixed it

  3. Typos: “Sir Josh Templeton”, “hedder”. There is in fact a Josh Templeton, but he’s not a Sir, and he’s a cartoonist.


    1974? Seriously? WHEN in 1974? I seem to recall there was sort of a wide range that year, what with the huge market crash and all…

    CPI? Seriously? Is there any historical correlation between CPI and market valuations? Have there, umm, been any changes in how CPI is computed over the decades, that might make this series an apples-to-oranges comparison? /snark. (P.S. Hussman looked at valuations vs. CPI in 2007 here: Future total returns are BETTER with high inflation, and worse with LOW inflation, so why should valuations be higher with low inflation?)

    What about profit margins, and corporate profits as a share of GDP? (Both mean-reverting series, both at or near all-time unprecedented highs.)

    And about those “all time low” interest rates. How’d the late 1960s work out for investors (last time we had high valuations, low inflation and low rates)?

    BR writes: “There is a lack of competition from fixed-income products for your investment dollar, thereby making equity dividend stocks that much more appealing. This is a significant fundamental difference in the markets and valuation and prices.”

    There is a lack of guarantee of “return of capital” from “equity dividend stocks”, whereas most fixed income products return your original cash when they mature (especially those not issued by Wall Street, but say by the U.S. Treasury). That is a significant fundamental difference between the markets… one which the above discussion overlooks. Dividend stocks should have substantially higher yields to provide a proper margin of safety.

    Well, it’s nice to know today looks just like 2007… Oh wait. Why do investors get the willies when we see 2007 in a comparison, again? Oh yeah, that was the LAST market peak.


    Admin: Sir Josh is fixed. “hedder” or “hed” are all slang for headline

  4. 4whatitsworth says:

    One thing to consider is that the S&P 500 now has a very larger percentage of earnings coming from international developed and emerging markets and these markets may have changed the nature of fair value for the S&P 500. It would be interesting to compare developed and emerging market historical valuations to the S&P.

  5. Frilton Miedman says:

    Another way to look at rates & inflation vs P/E & valuation is to step aside & look at home prices vs income.

    The historic ratio of median home price to median household income has been about 2.2 (2.2 X annual median income = median home prices)

    As of 2007, that ratio was about 3.5x, now it’s 3x, with low rates keeping the cost to service mortgage debt low.

    For home prices vs income,
    This implies that either; 1 – incomes have to rapidly increase to 45% of median home prices (now roughly 30%),
    2 – home prices must fall in line with 2.2x median income from the current 3x
    3 – Households continue increase their debt to income ratio under the guise that these rates are normal. (not good, with household debt already @ ~ 105%)

    I can’t imagine the same concept doesn’t apply for consumption in terms of stock valuations, after all, earnings are based on a 70% consumer driven economy, right?

    For now, the Fed has interest rates low enough to enable mortgage servicing cost as a % of income to match early 1980′s levels, but with the taper already in progress, how can higher rates effect disposable incomes without wage growth and not induce deflation & reduced earnings?

    The % change in the home price to income ratio needed to normalize consumption is massive, that difference has to be either in wage increases, deflation, or both.

    Did I miss anything here?

    • DeDude says:

      Actually I don’t think the most relevant metrics is “price to income”. Affordability is much better judged by “price to monthly payment”. At current interest levels it is much easier to afford a big loan.

      • Frilton Miedman says:

        Nope, reread what I said about the cost to service debt being at the same % of income as 1980. (despite household debt to income ratio being up 35% in the same time)

        My point revolves around the massive size of excess reserves, 2500% more than the historic norm, and what might happen if banks start putting that quantity of excess reserves to use.

        Low rates are the only thing saving our asses right now, household income is back to where it was in 1988, lower mortgage payments are the only reason households have money to spare.

        Makes for an interesting conundrum if excess reserves, now at 2500% the norm, start flowing & inflation gets out of hand.

        Is the Fed to raise rates without income growth? …while household debt to income is still hovering @ 105%?

        A sick game of political favoritism for big campaign contributors who expect an ROI for their bribery (circa; T. Roosevelt), leaving the Fed as last resort, has us looking at a potential secular see-saw until we’re finally forced to confront wage stagnation instead of using cheap money to supplant wages.

      • DeDude says:

        I do agree that there is ground for concern for what happens if the Fed decide to slam the breaks to early and to fast. Another housing bust will not cause a financial crisis but it could give us a nasty recession with potential for serious deflation.

      • Frilton Miedman says:

        Hence my above speculation comparing stock valuations to home price to income ratio.

        In theory, for the Fed to exit, median income would more closely match the 2.2 mark for home prices, whether this means prices fall, incomes rise, or both, households cannot afford higher mortgages right now without hurting stock’s earnings.

        I imagine either stocks (and homes) are over-valued, wages are going to increase – OR – the Fed is going to have to facilitate easy money as a band-aid at a time when household debt is still stretched to an extreme. (arousing the question of how much more can we deleverage without wage increases?)

        Brings me full circle to that elephant in the room, excess reserves, seems the herd has forgotten about the subject.

  6. howardoark says:

    Your concluding sentence was rates are at an historic low so this time things are fundamentally different. But the 10 year yield minus inflation was 1.22% in 2007 and now it’s 1.1%. Everything else is the same except price to sales is 10% higher. Though, perhaps fracking and robotics will change things up in the next decade and things will be different this time.

  7. cobaltbluedolphins says:

    Corporate taxes and financing costs . . . how much lower can they go?

  8. Petey Wheatstraw says:

    MEH! Brother, how have you been?

    This time it’s different, because the conquistadors have discovered the infinite fiat vein at Potosi on the Potomac.

    Hundreds of years from now, marine archeologists will be discovering shipwrecks full of valuable paper booty within the sunken hulls of hedge funds ripped apart by the derivatives Kraken.

  9. Futuredome says:

    The market is saying economic growth has been higher in the 2010-14 period than the government says. The government says 2-2.5%. The market says 3-3.5%.

    I also think long term interest rates are ready to take off. There is a lot of money there that once freed, will have to find a home or have inflation eat away at it.

    • rd says:

      A lot of people have been betting that long-term interest rates will take off. The only place that doesn’t seem to agree is the Treasury bond market.

  10. rd says:

    I think there have been two periods of “This time is different” where it really is differnt over the past coupleo f centuries.

    1. Mid to late 1800s – the Industrial Revolution created real economic growth where the standard fo living for the individual began to rise dramatically as people could now produce much more than before, virtually without constraint compared to the previous several millenia. The agricutlural revolution and information revolutions of the mid and late 20th century are just extensions of the original Industrial Revolution.

    2. The switch from gold to fiat-based currencies in the mid-20th century. Roosevelt started the disconnect between gold and the dollar in the mid-30s and Nixon completely severed it in the early 1970s. Continuous inflation was now a major force in the economy and financial markets which made it important to own a stake in the industrial revolution to keep up with it as opposed to just owning bonds to collect interest without concern for long-term inflation. In the 1950s, the dividend yield of stocks dropped below Treasury-bond yields essentially eliminating the “equity-risk premium”. Before the 1950s it was unusual for dividend yields to be lower than long-term interest rates – now it is unusual for them to be higher than long-term interest rates. I don’t think we will see long-term (50 year+) sustained 10% returns of on equities again in the absence of high inflation simply because the dividend yields are not high enough to provide a major portion of the return and it is unlikely that economic growth will proceed at a pace that will allow for that return to be replaced by capital gains.