Sunday, October 5, 2014

Salmon Roasted

I confess to being shocked by the exchange between Martin Wolf and Felix Salmon in this week’s New York Times Book Review.  First Salmon had criticized Wolf for using the terms “net external liability position” and “real unit labor costs” instead of “national debt” and “wages”.  Wolf pointed out that the first two terms and the second two mean different things and concludes, “[if] Salmon is unaware of such distinctions...., one must ask whether he was qualified to review the book.”

Salmon’s reply actually confirms that he doesn’t understand.  And this is really basic stuff, that anyone writing for a national audience ought to know.  How can Salmon explain this?

When was the last time we’ve seen such an embarrassing, high-visibility economics fail?


Unknown said...

Peter, you seem to think that

1) I don't understand the difference between “net external liability position” and “national debt”, or between “real unit labor costs” and “wages”.

2) Martin Wolf believes that (1) is true.

3) My reply to Martin Wolf demonstrates that (1) is true.

I think all three assertions are false, but your post is rather short on specifics so I don't really understand where you're coming from. Could you clarify?

Unknown said... fail: This is Felix, in case that wasn't obvious. said...


I do not know what you know or do not know, but I read your review, and you did criticize Wolf for doing exactly what Peter reports you did. If you in fact know what these terms mean, why in hell were you criticizing Wolf for using the proper terms when appropriate?

Barkley Rosser

Unknown said...

Barkley, thank you for appreciating that the question here is not the meaning of the terms, but whether their use was appropriate.

Let's say I'm reading a book where the author insists on talking about "undetached rabbit parts", and in every situation where he talks about "undetached rabbit parts", the sentence would be equally true if he just said "rabbits". I would be within my rights to complain, in a review, that the author uses the term "undetached rabbit part" where "rabbit" would be just as good, if not better, because it is easier to understand.

Now suppose that the author responds by saying that I clearly don't know the difference between a rabbit and an undetached rabbit part. You can see why that might be annoying. I do know the difference, but that doesn't mean I'm wrong. The complaint misses the point.

So what I said in my reply is that in the passage I was talking about, the term "national debt" would have been just as good as "net external liability position" -- if not better, because it's easier to understand.

Now you can respond, if you like, that I'm wrong -- and that replacing "net external liability position" with "national debt" would actually change the truth-value of the sentence. If that's the case, then Wolf is right and I am wrong. I maintain, however, that in this particular context, the truth of the sentence would not be changed, and that in fact it would just become easier to understand.

So, here's the sentence in question: "By 2011, the net external liability position of several countries had become too large to manage". (He's talking about Portugal, Ireland, Spain, and Greece.)

Certainly Wolf's sentence is true -- I say as much in my review. But would it become "something that is false", per Wolf, were it to use the term "national debt" instead? said...


Well, "national debt" usually is used to refer to the net indebtedeness of the national government of a nation, irrepective of whether or not it is held domestically or abroad. Thus, Japan has a "national debt" that is over 200% of its GDP, but this is overwhelmingly held domestically, and it is in a net external creditor position.

So, the terms are not interchangeable, and it looks from the sentence you have provided that Wolf used the correct term, although I have not read his book, and it may be that looking at the broader context where the sentence you quote appears, that in fact "national debt" would have been accurate and more appropriate.

Peter Dorman said...

Hi Felix. You are perfectly entitled to criticize Wolf for stilted writing. I think his columns for the FT, much as I like most of them, read as though he simply wanted to get his thoughts down and didn't rework them for grace or clarity of expression. I don't get much pleasure from reading him -- but I do get enlightenment.

Now as to the terms. Really: I can't understand why you are defending what you wrote. As BR said, "national debt" in common use refers to what governments owe bondholders, while "net external position" refers to the value of foreign assets held by people/institutions in your country minus the value of domestic assets owned by foreigners. One results from government budget surpluses or deficits, the second from trade surpluses or deficits (or offsetting international capital flows). They are completely different animals. Granted, Wolf characteristically explains them with little concern for comprehensibility.

And the difference between wages and unit labor costs is also quite significant. Wages don't take into account productivity, ULC's do. That's pretty major when you compare, for instance, developing and developed countries. I *might* give you a pass on gliding over the "real" part, but only if the context makes it clear that we are adjusting for changes in the value of the currency labor costs are being measured in.

The most charitable thing I think I can say is that you were preoccupied by the writing effectiveness issue and simply didn't notice the chasm between his terms and your suggested replacements. But having had it pointed out to you, the right thing to do would be to fess up.

Cameron Murray said...

"When was the last time we’ve seen such an embarrassing, high-visibility economics fail?"

I've always thought this was a pretty big fail

A bunch of professional economists couldn't answer a multi-choice econ101 question about opportunity cost any better than chance.

Here's one response from icons

Though in the journals the argument seemed to arise that opportunity cost can mean whatever we want it to mean!

I'd say that's quite an embarrassment.

Peter Dorman said...

I remember this episode. Yes, it's embarrassing, but it's a bit ambiguous just what concept is being tested. My guess is that respondents correctly understood that op cost refers to the value of the best alternative forgone but didn't get that value in this case means consumer surplus. So I'm not sure I would file it under "people who don't understand what opportunity cost means". Perhaps the problem is that these masters of simultaneous equations can't cope with two verbal concepts at the same time.

Sandwichman said...

"When was the last time we've seen such an embarrassing, high-visibility economics fail?"

Ha! Much of what big-name economists crank out routinely, without objection, amounts to "an embarrassing, high-visibility economics fail." But I guess it's O.K. as long as they use the approved terms to refer, ceteris paribus, to the hopelessly muddled concepts.

I would count this as a trivial gotcha-gotcha. Meanwhile, carry on with the inanities of the Lucas critique... Say's so-called Law... the "law" of supply and demand... marginal utility theory...

Unknown said...

Barclay, I never accused Wolf of using an incorrect term. And neither would I say that the national debt of Japan is too large to manage, even if its debt/GDP ratio is larger than that of, say Greece or Venezuela. I *would*, however, say that there more to "too large to manage" than net external liabilities. (Just look at the USA.) So using net external liabilities rather than just plain "national debt" doesn't somehow get you to a magic number which tells you whether a country is in trouble or not.

Peter, Wolf was not comparing developing and developed countries. And neither was he comparing countries with different currencies. Given that I was writing about a very specific sentence in a very specific context, it confuses me that you feel free to say that I'm wrong while clearly knowing nothing about the sentences I'm referring to.

What's more, there isn't a "chasm" here. Net external liabilities are one way of measuring a nation's indebtedness. Real unit labor costs are one way of measuring the amount that workers get paid. So it's hardly surprising that sometimes what's true of one will be true of the other. What *is* surprising is that economists like yourself and Wolf seem to think that what's true of one can never be true of the other.


Peter Dorman said...

Felix, the comment thread quickly exhausts its usefulness for hashing out disputes of this kind. Other readers will have made up their minds by this point. If you think that Barkley and I are either mistaken or nit-picking, please feel free to ask any other economist you feel would qualify as competent to adjudicate this.

Incidentally, I have to thank you for calling attention to the confusion over "national debt" vs "external balance". I tried to anticipate misunderstandings like this when I wrote my textbook, but somehow I overlooked this one, even though I've encountered it a lot in the classroom. Next revision, it's definitely going in. said...


I have made clear that since I have not read Wolf's book, more precisely the section surrounding the quoted passage, I am not in a position to judge ultimately if he was either misleading or wrong or simply too wordy or confusing. Certainly if one is concerned about the financial condition of a nation, one should be looking at multiple indicators, including certainly both the national debt and external net indebtedness, along with some other indicators as well.

Unknown said...

Barkley, first, apologies for misspelling you earlier.

Second, in case it's not clear, I never for a minute said or meant or intended to imply that what Wolf wrote was either misleading or wrong. It is neither. And maybe in this one particular passage it's OK that he gets a bit wordy. My point in the review is that the *whole book* is like that. If you have a copy lying around, look at page 104, for instance: it's basically just a list of numbers. Which says to me that Wolf doesn't care about his readers. He wants to write down something that is true, and he's much more interested in that than he is in *communicating* what he thinks.

At the beginning of the book, Wolf thanks his wife, saying that she "forced me to explain what I mean to a highly intelligent reader who does not live in the wold of international macroeconomics and global finance". I fear she might have tried, but she didn't succeed. To a first approximation, no one outside those two worlds will actually read and understand this book.


JW Mason said...

I'm a huge fan of Felix Salmon in general. But I'm afraid his critics are right in this particular case. The specific distinction here is between debt incurred by government, which is what "national debt" invariably means, and debt incurred by all actors, private and public, which is what Wolf is referring to. With the exception of Greece, the Euro crisis countries did not have large public debts; the shift toward net borrowing corresponding to their current account deficits took place entirely in the private sector. So Felix's translation distorts the meaning in an important way.

Peter Dorman said...

Thanks, Josh. Also, the productivity side of unit labor costs actually matter within the eurozone. It's true that the main action during the 00's was Germany's wage repression, but the background is the much higher level of economic mobilization in the export-oriented economies of north-central Europe. Germany is more productive than Greece et al. due to its remarkable set of industrial policies, and this further adds to the effect that wage repression has had on imbalances.

Rajiv said...

Here's the exchange in a nutshell:

Wolf: you can't replace A by B in a sentence unless A = B.

Salmon: you can if neither sentence is false

Disingenuous at best. His attempts to defend it above are just bizarre.

Alan said...

I do not think Rajiv's summary fully characterizes Salmon's position. As I understand it, Salmon is saying that, in writing for a lay audience, it is proper to substitute Rajiv's B for A if (1) either proposition makes the larger point the writer seeks to make, and (2) B will be intelligible to the lay reader while A will not. While I do not know if, in this particular case, (1) holds true, the overall claim does not strike me as bizarre.

Thornton Hall said...

Felix filled in his Sudoku puzzle wrong. If there wasn't a right solution it wouldn't be so much fun!