Profits Always Mean Revert

Here’s an interesting tidbit from the mean reversion school of investing.  Like most things in financial markets, mean reversion is a pretty standard occurrence.  Market prices mean revert, valuations mean revert, profit growth mean reverts, you get the picture.  Of course, the problem with mean reversion is not knowing whether it will happen, but know WHEN it will happen.

That said, I found this chart and commentary pretty interesting (via Meb Faber):

 “At the peak of the cycle, when profits are far above average and the economy is doing well, it is hard to imagine earnings collapsing back below the average, as it is to imagine a depressed region recovering. Mean-reversion in earnings, though sometimes delayed, is as undeniable as the economic cycle itself. Cyclically adjusted (or trend) PE calculations will always give a conservative valuation estimate. But that is exactly the point of valuation – to offer a degree of safety (a margin of error) and to smooth the dangers of the economic cycle. That peak profits typically accompany peak valuations only reinforces the point.

One can always discuss the idiosyncrasies of any particular valuation metric, although we reach similar conclusions to Robert Shiller’s CAPE analysis – but using a more modern time frame and a different (and more generous) earnings series. Our conclusions are that the US equity market is currently expensive. We can also reach a similar conclusion using alternative valuation metrics such as dividend yield, trend PE, and Tobin’s Q.

Most significantly, the downside risk of investing when earnings and valuations are far above historical averages should not be underestimated.  from our work, peak earnings go hand-inhand with peak valuations. When earnings revert back to mean (and below), the valuation will also collapse. That many continue to argue against this, and so soon after the collapse of 2008/09, is something we find quite remarkable.”



Source: SocGen


Got a comment or question about this post? Feel free to use the Ask Cullen section, leave a comment in the forum or send me a message on Twitter.
Cullen Roche

Cullen Roche

Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services. He is also the author of Pragmatic Capitalism: What Every Investor Needs to Understand About Money and Finance and Understanding the Modern Monetary System.

More Posts - Website

Follow Me:

  • Andrea Malagoli

    “peak earnings go hand-inhand with peak valuations.”

    This effect can be explained with the tendency of analysts to extrapolate recent trends in a linear fashion to infinity. The use of PE ratios based on trailing 12month earnings is a typical example.

    This is irrational behavior by the supposedly rational analysts.

  • SS

    As you mentioned, timing is everything. That chart is visually appealing, but it serves as a completely useless timing tool. You could have gone 5 years in many cases betting on mean reversion that never came.

  • Anonymous

    Samuel Beckett’s Waiting for Godot………

  • dctodd27

    “At the peak of the cycle, when profits are far above average and the economy is doing well, it is hard to imagine earnings collapsing back below the average, as it is to imagine a depressed region recovering.” – SocGen, quoted above

    “The Shiller P/E, which is based on inflation-adjusted earnings over the past 10 years, currently suggests that stocks are overvalued. However, this metric assumes that the normalized (cyclically-adjusted) EPS for the S&P 500 is today less than $70—well below even our recessionary scenario for EPS.” – S. Subramanian, BofA Equity Analyst 8/9/2013

  • Old Dog

    Of course profits are mean reverting, if they weren’t then Capitalism would be dead!

    Other factors here include the fact that the Fed will continue, probably for years, to buy duration. Investors requirement for duration will continue to drive equities.

    The digital revolution still has a very long ways to go in increasing the profitability of most businesses – and this more than any other single factor is driving profitability.

  • Skateman

    This is not true. If every business has access to the new technology, none will have an advantage and extra profits from lower costs will not be sustained.

  • CharlesD

    Skateman is correct.
    Productivity has no impact on profits at the overall level. Believing differently is called the Fallacy of Composition in economics – that is, what is true at the micro level does not necessarily hold at the macro level. For example, if corporate America suddenly
    doubled its productivity and could produce twice as many widgets with the same labor
    and costs, the income available to buy these widgets would not change. Hence, prices would have to fall by 50% to sell all the widgets and profits would be unchanged. This is what
    happened with autos, radio, TV, computers, and on and on.

    What does cause changes in profits is explained by Cullen in his papers i.e. the
    “Kalecki equations”. These reveal that profits have been stronger in recent years
    because of high deficits. These deficits, as we know, as now declining. The other
    sources of profits are improving. However, it looks like profits will have a difficult time
    growing materially in the next year or so.

  • SS

    Charles, great comment. Does that mean that profit margins don’t matter in the aggregate? Or can you expand on that thought as it pertains to margins?

  • John Daschbach

    This is not very interesting if one understands a little basic mathematics. The figure displayed has some information, but it’s not what the author thinks it is. He is simply comparing the ratio of the normalized integrals of profit over different periods of time. For a stationary function this construction just defines the mean. He is not doing what Shiller does and compare two different quantities, he is just comparing something to itself over different time periods.

    But there is something interesting in this (if the data and analysis are good, which given the lack of understanding what he is doing is questionable).

    For a monotonically increasing function the function value will always be above the moving average and for a monotonically decreasing function it will always be below, and for a stationary function the integral of the area above the average exactly matches the integral of the area below (that is what a mean is).

    So what the graph actually shows (because the function is predominantly above the average) is that on average profits are increasing, which we already know.

    The Shiller approach has a bit more value but I think any scientist would see the linear trend in the over 100 year data set and evaluate relative to that function. Most would look at his graph and say there are two clear features (1929 and 1999). Current valuations look like they are very close to the linear trend. That isn’t saying the trend will continue but there are good reasons to expect it might.

    In the end I find a lot of the graphs and analysis in popular economics to be very weak compared to fields like Statistical Mechanics.

  • Rafael

    Is this really irrational behavior for rational analysts? The goals of the analysts might be to reduce career risk as much as possibly by following the current trends while being bias to the upside.

    It is more irrational for investors, who know that such analysts have poor track records, to use the analysis for making decisions.

  • Andrea Malagoli

    This is a very good point, and I agree very much. I was being a little sarcastic since academic economists would not view it as “rational” behavior just because they assume perfectly automated logical robots as their ideal agents.

  • Old Dog

    So labor costs have nothing to do with profitability?


  • Rafael

    Good point.

    This is also something that happens with certain trading styles. It is quite rational for a trader to buy a security simply because its price is rising as long as they are managing risk as part of a plan.

    The security may end up priced “irrationally” by not only rational, but consistently profitable traders. What is best for each individual market participant might not be what’s best for the market itself.

    An academic economist might think such behavior at the individual level is irrational, but this may be because they themselves never participated actively in an actual securities market. Therefore, their assumptions of what is a rational act is probably incorrect!

  • GreedsGood

    Brilliant post. The building blocks of a critical system at work. It’s only when the hand off between the various market actors is not efficient (think “air pocket”) do we see serious corrections. Some “fundamental” managers are out and have been for the last 25%, whereas the optimists and momentum managers are only now beginning to feel a bit nervous.

    If we break down technically/momentum wise — will the baton be dropped?

  • Hans

    I am sorry to go off topic, but congrats to Mr Roche’s newest ranking !!

  • rwperu34

    The real problem is knowing what mean they revert to.

  • Benjamin Cole

    Nice blogging…but why did US corporate profits surge from $600 billion annually end of the great 1990s, to $1400 billion under Bush II, then fall in 2008, but then ramp up to $1800 billion under Obama (of all people).

    This seems (hate the word, but) structural.

    Really, business is just having a lucky streak? There is a lack of competition out there (ask any business guy if he believes that)? What would make profits pull back from these great levels?

    Seems to me, we have hit a profit plateau of sorts (although in a good economy maybe it will go higher).

    I mean, the US government is not exactly pro-business right now, and this economy is weak…and we still are seeing huge corporate profits….this is not an ideal situation right now…

    If the Fed would just stop suffocating the economy, maybe we could break the $2 trillion in profits barrier….

  • Auburn Parks

    Congratulations Cullen,

    Prag Cap is a new entrant into Onalytica Iindexes’ Top 200 influential econoblogs.
    Coming it at #53.
    Spread the word people, we need to get PragCap, and all other PK blogs really, up into the top 20 for next year! Then maybe we can start to have rational economic debates about policy and markets.

  • Auburn Parks
  • ilya

    Benjamin, pro business is a catchphrase. There are always winners and losers under any policy. If an “anti business” Obama introduces a bunch of regulations but boosts the fiscal deficits to 30% GDP profits of AGGREGATE would go up even though some companies will go out of business due to new regulation (look at coal companies now with S&P at record profits).
    John Daschbach : I appreciate you trying to introduce more math into data analysis, however, sometimes it clouds your understanding. Anyone who thinks that profits have a long term increasing trendline has to assume that the economy is not a closed system over long term (short term you can have international contributions that are substantial). Unless we start trading with aliens long term profits will have a steady level defined by equilibrium in our system (capitalistic or socialistic, whatever people will bare).

  • Hans

    Yes, Auburn, we should all hat tip, Mr Roche, for his
    recognition and his fine accomplishments!

  • Cullen Roche


  • John Daschbach

    Profits probably have some asymptotic limit but it’s probably a long way above where we are now. Equilibrium is often misused in economics. The economy operates far from equilibrium (sustenance) and profits represent the capture of part of the excess work (productivity) by other than the person producing that excess productivity. If you look at feudal Europe, or England, or many other economies throughout history you can find systems where the capture of excess work was far more concentrated. The absolute limit of this has no upper bound, it is possible, though perhaps improbable, that productivity can increase forever. The practical limit to this is how much can one group capture of excess productivity. This is the political stability aspect. However, a system with increasing productivity AND more than half of the productivity growth captured by one group (corporations) will see profits grow in both real and relative terms forever.

    I’m not saying that is a good system, or even likely to be the most productive, but that is the system we have. The political trends in the US have been pushing us in this direction, favoring reduced productivity growth and increased productivity capture by one group. But with one party in the US favoring greatly reduced growth and maximizing capture by one group (Republicans) and the other party favoring a reduced version, a bit more growth and a bit less capture by one group (but still more than half) (Democrats) I expect the trend to continue for some time. History says these trends end badly (riots, revolution) but is that decades or centuries for the US?

  • InvestorX

    You are spot on!

  • Andrew P

    Think a bit about the mechanism of margin mean reversion. What pushes margins lower? One thing that could do it is increased competition. Another is politics, as demands for wealth redistribution increase until a breaking point is reached. I think we can see the latter in play right now, with the fast food strikes and the like.

  • Andrew P

    Obamanomics is not pro-business, but it is very pro-big business, and very very pro big-finance.

  • CharlesD

    Sorry been away. Regarding profit margins “mattering”: Profit margins are a function of the sources of profits. The four (and only four -as the Kalecki equations explain) sources of profits are net investment, net exports, consumers spending more than their wages, and
    government deficits. When these four inputs in total are high relative to GDP, margins will
    also tend to be high (profits as a % of GDP will be high by definition).
    Yes, labor costs have nothing to do with profits. That is incredible given the conventional
    wisdom. But the logic is simple. One company’s lower labor costs is another company’s lower
    revenue. So if I get fired, I will no longer have wages to spend and overall corporate revenues will drop by the same amount that corporate costs drop – with profits unchanged.. Now
    I can borrow to help me get by – that will show up in the consumer spending relative to
    wages. I might get an unemployment check – that will show up as an increase in the
    government deficit. But the lower labor cost caused by me getting fired had no direct impact
    on profits. Hope helpful.

  • CharlesD

    Regarding why profits changed refer to the sources of profits above. During the Bush 43 era, consumers borrowed heavily, investment (which included housing) was very strong as we know. Profits fell in 2008-09 as private investment and borrowing collapsed. Profits
    recovered strongly due to the trillion dollar deficits in 2009-10. Today the deficits are declining but the other sources of profits (investment, consumer spending relative to wages, and net exports) are improving somewhat, resulting in overall profits being flattish. Because of buybacks, profits per share grow a bit faster than profits.
    Also to John D: You are not spot-on. As I explained in earlier comment in this thread,
    productivity has zero impact on profits at the overall level. Think of a closed society where
    we all worked for the same company which produced all of the goods and services we used.
    Assume we spend all of our wages. The company’s revenues will be equal to wages paid and their is no room for profits, regardless of how productive we may be. Therefore, profits have to come from “outside” the system – where revenues are created without an associated cost.
    Such revenues can come from investment (which is funded by past savings or borrowing),
    , consumers spending more than their wages, net exports (foreigners spending more than their “imports”) and/or governments spending more than their income (deficits).
    Hope helpful.