Scott Sumner  

Confirmation bias?

Does a measly quarter point ma... Question-Begging and Victim-Bl...

I agree with David Henderson's new post, which starts off with this Bob Murphy quote on the recent stock market decline:

As shocking as these developments [drops in stock prices and increased volatility] may be to some analysts, those versed in the writings of economist Ludwig von Mises have been warning for years that the Federal Reserve was setting us up for another crash.
I find Murphy's statement rather puzzling. What was "shocking" about the stock market decline? Isn't the US stock market normally pretty volatile? Hasn't this been a relatively non-volatile year until last week? Isn't it normal for stocks to decline 10% or 20% every so often? I'm not being sarcastic here, I honestly don't know what we are supposed to find "shocking" about an occasional period of stock market volatility.

Murphy seems to suggest that the fact that Austrian economists were not surprised by the volatility is a point in their favor. But why? Who was surprised? If you had asked me a year ago "Do you expect occasional volatility, up and down?" I would have said yes, and also that I had no idea when that volatility would occur, or in which direction the market would move. (Actually in periods of extreme volatility price declines are somewhat more common.)

Now perhaps Murphy is making a stronger claim, that the Austrians weren't just unsurprised by the fact of volatility, but also the timing. That they saw it coming last week. That would be impressive. So let's go back to August 2011, and see what Murphy had to say:

Investors the world over are still reeling from last Thursday's massive plunge in the US equity markets, in which the major indices all gave up more than 4 percent. It was the worst day for the US stock market since December 2008.

None of this should surprise those conversant with Austrian economics. The "fundamentals" of the economy have been and remain awful because the government and Federal Reserve are consistently doing the wrong things. The apparent recovery, fueled by Bernanke's sheer money creation, has been bogus all along.

Bubble, Bubble, Bubble

For some reason, people still cling to the vague hope that -- at least if we wait long enough -- the market always goes up, and "buy and hold" is a great strategy.

This sounds pretty similar to the recent comment, but also equally vague. If Austrians want to claim they can predict markets, they'll need to be more specific. As far as I can tell Murphy seemed to be rather pessimistic about the prospects for stocks, and the early August 2011 drop was a sort of "I told you so" for Austrian economics.

A commenter named Charlie directed me to this older Murphy post, and added this helpful information:

Since the article:

S&P 500 is up 60%
Gold ETF (GLD) is down 33%
Silver ETF (SLV) is down 64%

Now in fairness to Bob, the earlier post did include this disclaimer:

Knowledge of Austrian economics doesn't render someone an expert investor, but it certainly gives advance warning of the major trends in the economy.
But Charlie also found this warning about the economy:
In this environment, someone relying on fixed-income investments (such as private annuities or, heaven forbid, government retirement checks) could be wiped out by massive price inflation.
Austrians aren't the only ones who think they have something useful to say about future trends in asset prices. Keynesians and others also like to talk about "bubbles", which I take as an implied prediction that the asset will do poorly over an extended period of time. If not, what exactly does "bubble" mean? I think this is all foolish; assume the Efficient Markets Hypothesis is roughly accurate, and look for what markets are telling us about policy.

One other point. If the sharp decline early in the week told us something deep and meaningful about the wisdom of ABCT, then does the recent rebound weaken that argument, or was that rebound also consistent with the model.

Comments and Sharing

COMMENTS (24 to date)
ThomasH writes:

Why direct so much attention just to Murphey. What about other "Grifters and Godbugs?"

Jon Murphy writes:

With respect to Bob (and I do love him, and not just because of the shared last name), I think looking at the stock market is a fool's errand. The market is volatile and activity in any given day can be driven by many non-economic factors. I'm currently an economic forecaster for a company in New Hampshire, and in the research we've done, we've found the stock market to be a rather poor indicator of economic fundamentals.

I think it would be far better to look at larger indicators, like Industrial Production or Labor Force Participation, or Retail Sales, to get a better gauge of economic activity, rather than the stock market.

Scott Sumner writes:

Thomas, Would you expect a conference like that to invite me to be a speaker? If the answer is "no" then that might be a clue that one shouldn't always take DeLong too seriously.

Jon, In my view no indicator can predict the business cycle. The stock market is no worse than most of the others. Of course the recent drop was much too small to indicate a major turning point.

Brian Donohue writes:

Great stuff, Scott.

95% of this bubble thing is, I think, a pervasive cognitive bias, pattern-recognition software run amok.

Supposing you were, in Bob Murphy's eyes, a complete naif in the wisdom of bubbleology.

So you bought the S&P 500 on August 7, 2011, right before that 'bubble' popped.

Then, you sold on August 25, 2015, right after THAT 'bubble' popped.

You would have earned a compound annual return of 13.9% over a period of just over four years. The horror!

Jon Murphy writes:

Jon, In my view no indicator can predict the business cycle.

I agree 100%. At my firm, we rely upon about 20 to give us a clear(er) picture of the US economy. But, as always, there are issues with the data and how it's collected.

The stock market is no worse than most of the others.

Here I'll disagree, but that might be some of my own bias talking (finance was one thing I never really understood beyond the basic theories of it).

About bubbles, I've started reading Vernon Smith's stuff. He found lots of bubbles in his experiments, bubbles yet awaiting explanation as far as I know.

About Bob Murphy's projection of confidence, such a display is a job requirement, is it not, for people who hope to get paid for investment advice? David Henderson and Scott Sumner on the other hand work in academia which has different standards of expression. Right?

E. Harding writes:

I think there are a number of indicators which are highly useful at predicting the business cycle (interest rate spreads, bank lending standards, temporary help services employment), but it's very likely none of them have been 100% perfect for the past century. It may be because different recessions have different causes. And predicting the business cycle is hardly the same thing as predicting individual asset prices.

BTW, suppose I was to find an indicator (almost certain to be a composite one) which has been 100% perfect at predicting the business cycle for the past century. How would I know I wasn't cherry-picking?

Greg G writes:

All possible results confirm Austrian theory.

A drop in stock prices proves an unsustainable prior rise due to a monetary expansion.

An increase in stock prices proves a continuation of unsustainable inflation in stock prices due to a monetary expansion.

It is not really a theory so much as an alternative set of definitions.

Jon Murphy writes:

It is not really a theory so much as an alternative set of definitions.

Isn't that what a good theory is? :-)

Just teasing you, my friend

Floccina writes:
In this environment, someone relying on fixed-income investments (such as private annuities or, heaven forbid, government retirement checks) could be wiped out by massive price inflation.

I do not want to be mean, I like Bob Murphy, but how can government retirement checks be wiped out by massive price inflation since they are indexed for inflation. Maybe he means the SS rules will have to change as they run out of money but that problem goes beyond inflation.

Henri Hein writes:
If the answer is "no" then that might be a clue that one shouldn't always take DeLong too seriously.

I have never thought I needed to take DeLong too seriously, but I clicked through to Nate Heckmann's take-down of Peter Schiff. I have no idea how representative Schiff is, but reading Heckmann's piece and the discussion in the comments, it doesn't commend the Austrian perspective.

Greg G writes:


Good theories do often change the definitions used for the phenomena they pertain to. But good theories are falsifiable.

By the way, if Scott believes that "no indicator can predict the business cycle" then I'm going to go out on a limb and predict that than he'll be even more skeptical that you have 20 indicators that can.

ThomasH writes:


So if we take DeLong less seriously are you suggesting we take the guys he calls "grifters and goldlbugs" more seriously?

Of course American Principles Project people could benefit a lot from listening to you, but for exactly that reason you wouldn't be invited.

Hmm, that applies to the Jackson Hole conference, too. :)

Levi Russell writes:

Does every economic theory have to be specifically about when something happens as opposed to about how it happens? If so, why aren't we all spending our days with VAR and ARIMA models? If a theory that explains how but not when something happens is worthless, there's not much point in doing anything else.

As a side note, I don't think many MM folks read Roger Koppl's comments on David Henderson's post, but it would have been good if they did.

Scott Sumner writes:

Brian, Good point.

Jon, Perhaps my previous comment was too strong. I suppose yield spreads outperforms stocks, for example.

Richard, Experiments often don't carry over to the real world, especially when there is serious money involved.

E. Harding. Here's one thing that is beyond dispute, economists as a group can't predict recessions. So if there is an effective model, economists don't know about it.

Floccina, Good point.

Thomas, But I was invited.

Levi, I don't see how your comment relates to this post.

Bedarz Iliaci writes:

Greg B,
But good theories are falsifiable.
The falsifiability criterion pertains more to hard sciences. Many good theories are not falsifiable e.g. Darwin's theory of evolution.

A good theory should be explanatory. Prediction is not essential. For instance, prediction is hardly a suitable criterion in cosmological theories.

Thus, theories are ordered to understanding, and not essentially to prediction.

Greg G writes:


Darwinian theory is falsifiable. Dawkins often cites a pre-Cambrian rabbit fossil as something that would falsify it. His point is not that that is the only thing that could falsify it.

His point is that the theory predicts we should find simpler forms of animals and organs earlier in the fossil record than more complex forms of those same things. And that is what we have found. Over and over again.

Even if you think that all theories don't need to make predictions, the Bob Murphy article that this post by Scott refers to did claim to have made a successful prediction.

Jose Romeu Robazzi writes:

I don't think Bob Murphy is applying correctly the Austrian view on economics. If one take ABCT, what it predicts is not a sharp fall in asset prices. It predicts (in the late stage of the cycle), a gap between consumer prices and upstream prices (which the current rout in commodities prices seem to be inline with) and it predicts malinvestments going bust, which we can't exactly see directly, apart from the energy sector, but that cannot be dismissed so quickly given the slow productivity growth and apparent reduction in potential RGDP ...

Jon Murphy writes:

I'm with Greg G on this:

A good theory should be falsifiable, otherwise it is illogical.

I think you do need some level of predictability, too. I am a follower of the ABCT because, in my educated opinion, it provides the most logical explanation of the causes of recessions and economic growth. And it does have some predictability to it: malinvestment does create bubbles and will, eventually, lead to those bubbles popping.

The hard part is quantifying those predictions. Saying exactly when a bubble will pop is very difficult, especially in the moment. It's also very easy to declare victory in the moment without all the data. I've done it many times, and still do it from time to time. What we have to remember, no matter where you fall on the line, is that we are all social scientists, trying to predict something that has random components (ie free will). That makes it very difficult.

David Michael Myers writes:

Re: Robert Murphy's “gross assumptions.”

Perhaps Robert Murphy's error was in assuming that readers would recognize his enthymematic argument.

Perhaps he thought it was obvious that the situation wasn't a one-to-one relationship between fiat-currency-inflation, its consequent “boom,” and an immediate or instantaneous “bust.” And that he should have made it obvious that there were possibly many, many other things involved.

Perhaps he should have included the Latin phrase “ceteris paribus” [all things being equal]. In other words, “taking into account all the other possible variables, mainly time and governmental market-manipulation, this 'boom' will generate an inevitable 'bust.' “

Maybe Murphy's comment should have read something like this:

As shocking as these developments [drops in stock prices and increased volatility] may be to some analysts, those versed in the writings of economist Ludwig von Mises have been warning for years that [ceteris paribus] the Federal Reserve was setting us up for another crash.

His statement seems perfectly valid to me when one considers ALL the elements that surround such situation. I view it as a straightforward syllogistic argument with many possible propositions and one conclusion.
For example:

Prop#1> In a truly free-market economy, the value of currency is determined SOLELY by market forces. Government does not intervene or mandate a “legal-tender” and set the value of it.

Prop#2> When governments intervene in the economy, they usually issue “fiat-currency” and set its value by mandate of the government that issued it. Its value is NOT determined solely by the transactions of the market-place.

Prop#3> The USA's present-day fiat-currency is not “backed by” [redeemable at face-value in] gold or any other commodity.

Prop#4> Governments may change the value of their unbacked fiat-currencies at any time. The world right now is in a “currency war” in which many countries are “devaluing” their currencies [issuing more and more] in an attempt to make their products supposedly “more affordable” on the worldwide commercial stage.

Prop#5> The financial and fiscal policy-makers in the USA apparently believe they can defeat or ignore the laws of economics and “stimulate” the economy out of any “recession” by QEs [“Quantitative Easings” ― otherwise known as “creating fiat-currency out of thin air.”]

Prop#6> Each and every unbacked increase in the total fiat-currency-stock contributes to the further debasement (loss of purchasing power) of that fiat-currency. Loss of purchasing power results in increasing prices of all commodities, goods, and services. Those agencies, organizations, and people who first-receive the “new currency” benefit greatly. Their currency “buys more” because prices have not yet responded to the increase in currency-supply. As time goes on and the effects “trickle down” those at the “end of the chain” benefit the least if at all. The “favored ones” benefit at the expense of the “commoners.”

Prop#7> Fiat-currency is a “fiduciary” currency. Its value is based entirely upon the faith of the general public in accepting it at face value. However, as the general price level rises and each dollar buys less and less, people begin to lose that faith in the worth of the currency, Consequently they begin to save less and spend more as quickly as they receive it (“before it loses even more purchasing power.”) This ultimately leads to a general rejection of the currency and eventual “sovereign default.”

Prop#8> Carmen M. Reinhart's & Kenneth S. Rogoff's treatise entitled: “This Time It's Different: Eight Centuries of Financial Folly” documents the inevitable “sovereign default” (governmental bankruptcy)
of regimes that incessantly debase their fiat-currency.

Prop#9> The “ruling elite” [“banksters'” and their plundering, bought and paid-for colleagues in government] apparently want us to believe that the general public (the “sheeple”) do not perceive what is going on and will do nothing about the present situation.

[One may insert here as many additional propositions as required to fully-describe the situation.]

Prop#NN> It seems highly-probable that the USA may be subject to the laws of economics and in the not too distant future suffer the inevitable “sovereign default.”

Conclusion> Everything is “hunky-dory;” our economy is on the road to recovery, and all's well that ends well.

Philip George writes:

"Jon, In my view no indicator can predict the business cycle."

The Austrians' misfortune is that they have no indicator to predict the business cycle. But to say that there is no indicator is going a bit too far.

I would say with some confidence that if the current monetary contraction proceeds the way it has been doing, then somewhere along the middle of next year you will see the first signs of financial distress and 2017 will be a repeat of 2008.

See the graph of Monetary Contraction for an indicator of what's coming.

Greg G writes:


Thanks for the support above. Even so, I can't resist pointing out that a theory can easily be logical and still falsifiable. And you don't need to invoke free will to find good reasons the economy can't be predicted very far into the future.

Jon Murphy writes:


I agree with everything you just said, and it is my fault if I implied otherwise. Logical can be falsifiable, but unfalsifiable cannot be logical.

Regarding free will, I just picked that because it was the first thing to come to mind :-)

Nathan W writes:

Stocks go up and down. Generally more up than down.

But some people love to proclaim that their pet theories have always been true, depending on whether up or down would have partially verified their pet theory.

All I can really observe is that stocks go up and down, sometimes more up than down.

Were stocks in bubble territory? That depends entirely on whether estimates of corporate abilities to earn profits were overestimated or not, but most analysis tends to deal with other matters which may be only peripherally related to corporate profitability.

If corporate America is sitting on hundreds of billions of unused profits, then how much do the other typically variables really matter in estimates of their abilities to earn future profits, or the NPV of those estimated protits?

Comments for this entry have been closed
Return to top