The Atlanta Fed's macroblog provides commentary on economic topics including monetary policy, macroeconomic developments, financial issues and Southeast regional trends.

Authors for macroblog are Dave Altig and other Atlanta Fed economists.

« Wolfowitz's Defenders -- Bono Next? | Main | On 1st Quarter GDP, Hold That Thought »

April 29, 2007

What Are You Going To Believe -- Theory Or Your Own Lying Eyes?

The blogger epicenter of the free-trade debate is rumbling at Harvard, with Greg Mankiw and Dani Rodrik engaged in a terrific -- and important -- conversation about winners, losers, and how (or whether) economic theory divides the two.  You can check-in on the state of the debate at Angry Bear, where pgl provides the appropriate links.  It is highly recommended reading, but I think it ought to come with a few warning labels.  For example, Professor Rodrik responds to Professor Mankiw with this claim:

... there is no theorem that guarantees that the partial-equilibrium losses to import-competing producers “are more than offset by gains to consumers from lower prices.”

In a related vein, pgl opens his post with:

As we were applauding Dani Rodrik, Greg Mankiw was defending the Dan Drezner lower prices from free trade benefits everyone fallacy.

Let's be perfectly clear:  There are no theorems in economics that guarantee anything about the real world.  Economic models are not descriptions of physical realities but formalizations of stories about how social interactions deliver particular outcomes.  Different, equally coherent, stories deliver different predictions about the world.  The claim that "free trade benefits everyone" is not a fallacy, but a particular outcome based on a particular model.  Different models deliver different answers, so theory alone does nothing beyond eliminating stories that are internally inconsistent.

Or, perhaps, unconvincing.  The missing ingredient in this most recent installment of the free-trade discussion is evidence in favor of one story or another, a task that is a good deal messier than writing down models.  What makes matters worse is that adjudicating the issue is not a mere matter of counting up winners and losers.  In the court of determining what is "good" or "bad", economists have standing to address one question, and one question only:  Can someone be made better off without making anyone worse off?  That too depends on the model at hand, and in fact it's even worse than that.  The Rodrik-Mankiw debate revolves in part around a result known as the Stolper-Samuelson theorem. Greg Mankiw does a good job explaining Stolper-Samuleson and its relevance to the subject at hand, but I'll note one item from the Wikipedia description of the theorem

If considering the change in real returns under increased international trade a robust finding of the theorem is that returns to the scarce factor will go down, ceteris paribus. A further robust corollary of the theorem is that a compensation to the scarce-factor exists which will overcome this effect and make increased trade Pareto optimal.

In simple terms, there are losers, but the winners can win enough to more than match those losses.  All would be well with the world if the winners and losers could be easily identified, and an appropriate compensation scheme implemented.  But what if that is not feasible?  What is the right move then?  To protect the losers at the expense of significant opportunity cost to potential winners?  The other way around?  I've yet to encounter an economist trained to answer those questions, and you should be very suspicious of any who speak as if they are.

April 29, 2007 in This, That, and the Other, Trade | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference What Are You Going To Believe -- Theory Or Your Own Lying Eyes?:


As always, I'll have to complain about the use of the appellation "free trade" in reference to a trading system in which China actively pegs the yuan against the dollar at one-fifth its purchasing-power-parity value, and Japan actively manipulates the yen by a zero-interest-rate policy (no longer using obvious direct intervention as in 2003-2004, when it openly bought $320 billion and budgeted for $1 trillion of additional intervention).

This exchange-rate manipulation creates trade barriers just as real as tariffs of the same magnitude. If absent the manipulation the yen-dollar exchange rate would be around the 80 yen/dollar level of the mid-90s, when Japan-US trade was starting to come back into balance, then the current manipulated rate is equivalent to a 50% tariff in Japan on imports from the US.

Posted by: jm | April 29, 2007 at 02:49 PM

You make the same point Greg did in his comment to my post. The efficiency gains are such that IF the winners decide to compensate the losers, there is still a net gain. No one denies this. What Dani was saying - and I think he's right - is that SINE compensation, there will be losers. You don't deny this - and now Greg is claiming he does not either. So what's the debate here?

Posted by: pgl | April 29, 2007 at 03:47 PM


That's a good post, taken as a whole.

I go a step further, though. Once trade policy is implemented, the issue is then based on real world outcomes, not theorems. And it's about at that point that many economists appear to get lost if not intentionally disappear if all isn't going well or according to preconceived notions.

Long before Alan Greenspan stepped to the microphone and explained that U.S. offshore production shifted to China-based operations could be impacted with U.S. trade policy but that such offshore production would not return to the U.S. but rather would flow elsewhere to another cheap global production source...well, long before that I had written, forwarded, and blog post my brief Economic Hydrology Theory (EHT) statement and principles. That statement summarized the offshore production growth situation in clear and precise language.

Economic reality and related outcomes are based on real world results, not classroom dogma whether pitched to kids or adults. Theorems only go so far. Economists who can't venture beyond the academic bounds of such devices and evaluate the real world results, recommending appropriate adjustments in the metrics are not economists that I recommend that any corporate clients hire.

Unlike many, you are an exception. Well reasoned thinking is what I am seeking. And you have it.


Posted by: Movie Guy | April 29, 2007 at 08:00 PM

Go tell "pareto optimal" to dead Iraqis why don't we. (if we ever find the documentation we are stealing oil from Iraq)

economist can be dangerously obtuse creatures.

Posted by: andy | April 30, 2007 at 06:23 AM

Had to run an get my glasses after that wallopin, eye-bulgin title Dave...I take it all back --Finnegan's Wake by what'shisIrishface is so Irishly over-rated.

jm, has most of my view of that matter without giving any space to the transnationals who are always neglected in this play ...especially with Paulson's vaulted connections. Same sorta vaulting with energy pricing and this administration...the sacking of Rome by a very few --not the hordes of barbarians as previously thought. [The barbarian critique of History]
Dave writes:
"All would be well with the world if the winners and losers could be easily identified, and an appropriate compensation scheme implemented. But what if that is not feasible?"

And it is hard to believe that the winners cannot be easily identified: those HF managers, those CEOs, those large shareholders.
The appropriate compensation scheme OTOH should be left to the hordes of barbarians because QED the present scheme is FUBAR...such is the inappropriate view of the losers whose eyes as andy points out may be no longer seeing much of anything.

Posted by: calmo | April 30, 2007 at 02:03 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts

November 2014

Sun Mon Tue Wed Thu Fri Sat
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29



Powered by TypePad