Arnold Kling  

Good Reading Recommendation from Tyler Cowen

PRINT
Hamilton on Bernanke on Bubble... Emotions and Decisions...

A while back, Tyler Cowen recommended Perry Mehrling's book on finance theorist Fischer Black. I read it while on vacation this past week.

I strongly endorse Tyler's recommendation.

Black co-developed a formula for option pricing which garnered a Nobel Prize, for which he was no longer eligible, having died before it was awarded. But he was much more prolific than Myron Scholes, his co-author on the option pricing paper. And he had a very strong point of view about economics and finance, which Mehrling brings out quite well.

Merhling also discusses Black's traits and quirks. While certainly not as dramatic as Skidelsky's biography of Keynes, Merhling's work does deserve mention in the same breath.

I plan to write more about Black in the future.

UPDATE: One of many issues on which Black has strong, nonconformist views is accounting. If the market value of the firm is P, and the price-earnings ratio is P/E, then Black says that we want accountants to come up with a value of E that gives the right P. Mehrling writes,


he proposed that the goal of accounting should be to come up with a number that the analyst needs only to multiply by a constant, say 10, in order to arrive at an estimate of value. In effect, the goal is to internalize within the firm all the various adjustments that the analyst sitting outside the firm would do himself if he had access to the detailed information available to the accountant.

This would seem to support accounting that includes a lot of judgment, and even earnings "smoothing." For another viewpoint in support of judgment, Nick Schulz pointed to this piece by Alex J. Pollock.



COMMENTS (4 to date)
Dan Landau writes:

Most "principles" texts (Mankiw, Ekelund & Tollison) distinguish between "accounting profit" which is revenue minus explicit out of pocket costs and "economic profit" which is revenue minus full opportunity costs.

To me that means accounting measures of "earnings" are bound to be inaccurate.

spencer writes:

since 1871 the stock market PE has ranged from around 5 to 30 -- moreover it has almost an equal probablility of falling anywhere between 9 and 19. In the short to intermediate run PE moves account for over 100% of changes in the market.

Given the history of pe changes why should analyst assume that it is a constant?

the correlation between the changes in the market and the change in EOS is essentially zero --so even perfect knowledge of what EPS will do is of no help in telling you what the market will do.

Maestro writes:

Shouldn't Earnings represent how much the Value of the firm went up (or down) in the period?

dsquared writes:

No, no, a thousand times no. Fisher Black might have won a Nobel Prize but he is dead wrong on this one.

A set of accounts is a record of transactions; what was done during the accounting period. Estimating the value of a company is *one* thing you might want to do with a set of accounts but it is not the only thing or even always the most important thing.

Accounting is a record of the immediate past. Valuation is all about the future. Black's "goal" is completely wrongheaded:

"to internalize within the firm all the various adjustments that the analyst sitting outside the firm would do himself if he had access to the detailed information available to the accountant."

Why would this be a good idea? Someone outside the firm might have a completely different view of all sorts of important and relevant matters in the accounts. If you deprive an outsider of the ability to know what's gone into the numbers, then the accounts are a black box and you might as well just let the firm set its own share price.

Comments for this entry have been closed
Return to top