Monday, December 6, 2010

Oh My

. Monday, December 6, 2010

I literally cannot believe this:

One thing I’m hearing, now that all hope of useful fiscal policy is gone, is the idea that trade can be a driver of recovery — that stuff like the South Korea trade agreement can serve as a form of macro policy.

Um, no.

Our macro problem is insufficient spending on U.S.-produced goods and services; this spending is defined by

Y = C + I + G + X – M

where C is consumer spending, I investment spending, G government purchases of goods and services, X is exports, and M is imports. Trade agreements raise X — but they also lead to higher M. On average, they’re a wash. ...

There is a case for freer trade — it may make the world economy more efficient. But it does nothing to increase demand.


That is international economist and Nobel Laureate Paul Krugman, arguing that trade does not matter for GDP or employment, but only for "efficiency". The post is entitled "Trade Does Not Mean Jobs". But trade markets don't operate in a fundamentally different way than other types of markets... does Krugman also believe that those markets have no effect on GDP or employment? Maybe he does, but it's wrong. Efficiency gains from markets leads to an increase in overall production. This is the logic of comparative advantage that Krugman used to defend so powerfully.

Let me break this down. GDP ('Y') can go up through trade in two ways: 1. If 'X' is greater than 'M', so that the economy is producing more than it is consuming; 2. If efficiency gains from trade leave people with more disposable income overall, which they then spend part of on domestically-produced goods ('C').

To see why number 2 is important let's think through a stylized example. Suppose we used to have to buy kim chi from Korea at $10/pound, because we had to pay a 100% tariff on the import. Suppose that tariff goes away through a free trade agreement. We can now buy our kim chi for $5, and have $5 left over to buy haircuts from Americans. The quantity demanded of Korean kim chi and American haircuts will both go up, as will GDP and employment in both countries. That is how gains from trade happen.

That is also why this makes no sense:

And there’s even an argument to the effect that increased trade reduces US employment in the current context; if the jobs we gain are higher value-added per worker, while those we lose are lower value-added, and spending stays the same, that means the same GDP but fewer jobs.


But why should we expect spending to stay the same? If we gain higher-value added jobs, we are also gaining income. If we are gaining income, we will spend more. That means higher GDP and more jobs.

I have Paul Krugman's textbook on international economics. It is very good. It covers standard trade models nicely. Maybe he should read it. Or any of his columns from the 1990s.

3 comments:

Latinamericanist said...

So you believe trade agreements have a short-term aggregate demand effect? Any evidence for that?

Thomas Oatley said...

But won't G fall by $5 per unit of pickled cabbage purchased? And isn't most of the tariff cut therefore pure redistribution from G to C (with some reduction in deadweight losses too)?

Kindred Winecoff said...

LA -

Depends on what you mean by "short-term". Monetary policy operates on long and variable lags, fiscal stimulus multipliers are estimated 12-24 months out or longer. I don't think it's controversial to say that trade can have an effect on demand (and supply) over a 2-3 year time horizon.

Let me be clear: I don't think a bilateral FTA with Korea will have a large enough effect to shock a $14tn economy back to full employment. I'd have no problem at all if Krugman had written "It's nice that Obama finally wrapped up the Korea FTA, but we need to be realistic about what the effects will be. They will be moderate. In the right direction, but moderate nonetheless." That's actually what I was planning on writing today before I saw his post.

But it's just disingenuous to say that trade has no effect AT ALL on employment or GDP.

Thomas -

Yes, but gains from efficiency and reduction of deadweight loss is where all the action is. That was Ricardo's whole point. The pie gets bigger, the deadweight loss gets smaller. Those gains have to go somewhere.

Oh My
 

PageRank

SiteMeter

Technorati

Add to Technorati Favorites