Arnold Kling  

Schools of Macro

PRINT
Is Repo Lending Inherently Uns... Test the Predictions - Or Chec...

Simon Wren-Lewis offers a take on this phenomenon from a Keynesian perspective. Pointer from Mark Thoma.

My latest essay begins to offer a different take.


There are now so many versions of "what's wrong with the economics profession" that, with apologies to V.S. Naipaul, I could describe the state of economics as one of a million mutinies.

Be sure to read the conclusion.


Comments and Sharing


CATEGORIES:



COMMENTS (10 to date)
Rick Hull writes:

It's a bit unclear:

> Simon Wren-Lewis offers a take on this phenomenon from a Keynesian perspective.

Which phenomenon? That different schools of macro exist?

Chris Koresko writes:

@Arnold Kling: I really enjoyed your article. Nice mix of solid info with labeled, properly caveated opinion.

The Wren-Lewis was good too, with some interesting insights. But there was one line that made me squirm:

"...synthesis models clearly show..." Yikes! models show? "Models predict" would be more correct.

PJR writes:
This is important, because the robust version of the EMH implies that even though individuals and markets may behave irrationally at times, policymakers are not in a position to detect and correct the mistakes of the market. Like everyone else, policymakers cannot reliably outperform the market.

(1) Individuals can reliably outperform the "market": invest in value, small cap and/or low volatility stocks.
(2) Policymakers (and their staffers) reliably outperform the market because they exempted themselves from the insider trading rules.

Harrison Searles writes:

What's the diagnosis of the stubborn monetarists?

J Oxman writes:

@ PJR:

"(1) Individuals can reliably outperform the "market": invest in value, small cap and/or low volatility stocks."
Unfortunately, there isn't a lot of evidence that these anomalies are reliable sources of excess returns. The small cap phenomenon, for example, disappeared after people noticed it.

One has to keep in mind that apparent sources of excess returns may not be arbitrageable. For example, small stocks may appear to have a value premium associated with them, but one finds they are very difficult to trade. In fact, they are often so illiquid that the mere trading of them is sufficient to move the price and wipe out the excess returns.


"(2) Policymakers (and their staffers) reliably outperform the market because they exempted themselves from the insider trading rules."

Indeed this is the case. If insider trading were perfectly legal, markets would be even more efficient becaue insider information would then be impounded in the price.

PJR writes:

@J Oxman.
(1) Fortunately, there is quite a bit of evidence for the persistence of many of these anomalies including value and momentum (which I forgot to mention in the original post). Momentum in particular remains robust even since it was first noticed. These "anomalies" are the result of behavioral biases. I suspect these anomalies are more durable than academics want them to be; the anomalies are a product of human behavior, which will not be changing anytime soon.

Even Paul Samuelson, the Godfather of EMH, concedes that the market is micro efficient with regards to individual securities but macro inefficient as a whole.

(2) As to insider trading, making it legal cuts to the very heart of efficient markets and perfect competition. One of the basic assumptions of perfectly competitive markets, everyone has the same information and can act upon it at the same time. Legalizing insider trading would put most investors at an even greater disadvantage.

J Oxman writes:

PJR

(1) The value premium does not appear to be arbitrageable. Have you some papers indicating that it is? Keep in mind the Fama-French value and size factors may actually be proxies for macro risks, so it might be a confounding factor problem. Work is still being done on this. (see Aretz et al. in J of Banking & Finance 2010, 1383-1399)

I know everyone likes to point at Warren Buffett and Seth Klarman and Tweedy Browne as evidence that the value premium is strong and persistent. My own work indicates that the value premium is actually driven by lack of analyst followings and volume, so value stocks tend to exist in pockets of inefficiency.

I have seen many papers on these issues and it's not clear that momentum strategies generate enough abnormal profits to cover the frequent trading. (see Korajczyk and Sadka, Journal of Finance 2004, 1039 - 1082). So there might be paper profits available, but if you tried to actually implement the strategy, it would be too costly.

This is all to say that market values can be in a range and still have essentially efficient market pricing.

I never thought of Samuelson as being the godfather of EMH. In financial economics, it's really Eugene Fama that has done the specific groundwork for defining EMH and testing it.

(2) I completely disagree. That notion of perfect competition is the typical starting point in a study of economics, but is nonsense in the real world.

Do you not think that increasing the information that is impounded in the price makes the market more efficient?

Arnold:

I don't think your "bumper sticker" for stubborn monetarists fits Scott Sumner at all. I doubt he would ever say that the problem is "the Fed" and not "economists"...after all, he believes that Fed policy represents more or less the consensus of the economics profession.

PJR writes:

J.O.
(1) Samuelson or Fama as the Godfather, who cares? Their seminal papers related to EMH were both published in 1965. At the time, everyone was using Bachelier's work from 1900 to think about stock price formation. Samuelson was an established authority in economics, while Fama was a snot-nosed kid writing a thesis.

The Fama-French factors could very well be macro risks, macro inefficient risks. With regards to zero-cost, "zero-risk" arbitrage portfolios, most investors cannot easily access them. In fact, it is difficult for even skilled investors to access simply because the short side is so expensive. Building long only value and momentum portfolios work. The debate is still ongoing. You are on one side, and I am on the other.

(2) Insider trading fails on ethical, economic, and legal arguments. Insider trading leads to inefficient allocations (non-pareto optimal) of capital since some people are trading on privileged information. From a legal perspective, insiders have fiduciary responsibilities and such trading is ripe for market manipulation.

Do you not think that increasing the information that is impounded in the price makes the market more efficient?

Yes, if it is done quickly, efficiently, and without abuse/manipulation. Human nature leaves me skeptical on all counts.

Why not just require disclosure of all information immediately so everyone can judge and act upon it simultaneously (reg FD)?

Bob writes:

Behavioral economics might not have any prescriptions for making markets work better, but I think their main contribution has been providing clear undeniable evidence that there is a problem with the efficient markets hypothesis.

Comments for this entry have been closed
Return to top