Paul Krugman writes that trade does not equal jobs and concludes that trade restrictions cannot even in principle trigger a depression. After all, restricting trade means restricting exports (less jobs!) but it also means restricting imports (more jobs!) so everything washes out.
Well, let’s try an extreme example. Suppose I prevent everyone in America from trading with anyone outside their own households. We’d eat only what we could raise in our own gardens, burn only the fuel we could gather from our own backyards, and wear only the clothes we could make for ourselves. In other words, we’d all be living pretty much at the subsistence level. Would you be willing to call that a Depression? I would. Krugman, apparently, would not.
Krugman is absolutely correct that under these extreme trade restrictions, you’d be just as employed as you ever were — you’d be getting up at 5AM to milk your cows and tend your garden before going out to gather firewood and climbing up to patch the holes in your roof. Krugman even acknowledges that you might be poorer as a result when he says “There is a case for freer trade — it may make the world economy more efficient”. But he insists that this wouldn’t count as a depression because you’d still be working.
That’s at best idiosyncratic; everyone else defines “depression” in terms of prosperity, not in terms of employment. But let’s accept his language and agree that, yes, you could impose a bunch of trade restrictions without affecting employment — we’d all still be working just as hard, or even harder. The right response is so what? Most of us care not about how much we work but about how much we earn.
It’s very bad practice for an economist to make policy recommendations without telling you what he’s trying to maximize. Krugman is frequently guilty of this particular sin. But now he’s given us an insight into what he thinks is important. By relegating “may make the economy more efficient” to an essentially parenthetical remark and putting all the emphasis on how much we get to work, he’s revealed a lot about his priorities. Those priorities are nuts.
“Would you be willing to call that a Depression?”
I’d be willing to call it many things, decimation for certain: we stopped being an agricultural country years ago, and I strongly doubt that we have enough arable land for 400 million people to sustain themselves and their families purely through subsistence agriculture on their own small plots of land, nevermind the obvious problems of redistributing the population over that land in order to let everyone have their backyard farms…
…but a depression? Once the mass deaths were over and those that were capable of self-sufficiency were planting their crops and hunting their game? Were the Native Americans in a constant depression from 14,500 BCE through 1492 CE? I rather suspect it would have come as news to them.
(And yes, I’m aware that various of the first nations in north america actually had functional trade systems, but I’ll assert that on average it was much closer to your hypothetical zero-trade scenario than to any modern economy.)
Doctor Memory:
In the standard usage of the term, a depression refers not to the current level of output, but to the current level compared with recent past levels. So the living standards of 14,000 BC were not depression-level in 14,000 BC, but would be today.
Likewise, the 1930s would have been times of previously-undreamed-of prosperity had they occured 1000 years earlier. That doesn’t stop us from calling them a depression when they occur a third of the way into the 20th century.
see my comment on your excellent post here: http://www.rationaloptimist.com/blog/self-sufficiency-another-word-poverty
“let’s try an extreme example.”
I read Mr. Krugman a bit different; I didn’t read his column to apply to non-depression settings. In a depression setting he wants to maximize employement to maximize demand.
There are two reasons why I think it is more correct to read the column in this way. First of all, in his discussion on Keynes he admitted as much that only when full employement is established do the classical postulates come into play. Second of all, at full employement, he has argued for more free trade. When he discussed child labor and free trade in 1992 for instance he didn’t argue that it was good that children were employed by multinationals for the sake of being employed, he argued that it was good because it was the path to wealth. Without it, those children would have had less options and would have been considerably worse off.
What I am trying to say here is that Mr. Krugman’s column should be read given the context. Or as economists would put it: evaluated at the margin ;-).
with ‘the context’ I meant the depression setting, not his other columns. The other columns were merely an illustration of that he stated that his opinion would change and that it fact has when the context changed.
Interesting post.
I find Paul a bit hard to follow at times. Large leaps between sentences, but that seems to be his sweeping style. Having said this, I don’t read Paul’s article in quite the same way. This article is no exception where it seems he trying to say two things simultaneously – trade alone cannot cause huge shifts in employment levels and trade policy today should aim to rectify what amounts to unilateral trading.
His primary point, in my view, is that any trade policy employed in efforts to turn around economic performance should encourage the account surpluses to be spent, i.e. China buying something from the US with the $US it’s acquired – that is, trading back US dollars for US goods. From this, employment may lift to service the additional demand. From that extra employment is extra consumption, which, in turn, will help resolve the depression (increase GDP).
Paul is very much a free trade man as I understand it, and this column is an effort,albeit a odd one, to move the discourse away from protectionist based approaches towards something that would satisfy both efficiencies and the depression. Though I’ve no idea what the specifics of these policies would look like.
As I understand it Paul is very much a free trade man
ignore the last sentence
Paul Krugman is certifiably nuts. He uses his notoriety and espouses nonsense. A perfect example of a condescending liberal who only cares for himself and his well being. A pathetic human being who is ignorant in simple economics. He needs to go back to school to learn – not at Princeton, but somewhere where they teach sense and not nonsense.
Perhaps someone who knows more about macro can help me with something. Paul says . . .
—
“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.”
—
I guess Paul means for C to be consumer spending on domestic goods and services, G government spending on domestic goods and services, and M consumer and government spending on imported goods and services. But, beyond that, his point is that M and I are dependent.
But when Paul thinks about government spending and consumer spending, he treats them as independent. Personally, I would be comfortable if he had said something like “a rise in government spending G is canceled by a drop in consumer spending C and investment I, because the government takes its money from consumers.” But, knowing Paul, I’m surprised he didn’t attach a multiplier to G and say “each dollar spent by the government increases overall spending by a factor of X.” But he didn’t say either. Can anyone help me understand why?
And, while I’m on this theme, I’m waiting for the day when someone writes the formula like this:
Y = G + T
where T is the amount of money that citizens ‘spend’ to support their government, and G is the amount that the government spends to support its citizens. Then we can finally achieve that special time when the citizens pretends to work and the government pretends to pay them.
Thomas Bayes:
, his point is that M and I are dependent.
But when Paul thinks about government spending and consumer spending, he treats them as independent.
This is a spectacularly good comment.
I read Landsburg first and wondered if he was misquoting Krugman. How in fact could he not be? But …
“This, by the way, is why claims that the Smoot-Hawley tariff caused the Great Depression are nonsense. Yes, protectionism reduced world exports; it also reduced world imports, by the same amount.”
Imagine you went to a basketball tournament, and the organizers forbid teams playing? Sure the number of wins would go down but so would the number of losses, by the same amount, so no-one would be worse off.
Putting aside Krugman’s assertion that a countervailing duty on Chinese imports would create jobs, it seems to me that you and he are pretty much in agreement. Trade does not create jobs, as your example shows. We work harder for less real income under protectionism. The “jobs” issue is a red herring. Let us promote free trade for what it does, not what it doesn’t.
Wasn’t Krugman recognized for finding a convoluted way of making protectionism work? I thought I read Tayler post something on that on Marginal Revolution.
@Neil: No I think SL is pointing out a Krugman non-sequitur. Which you identify.
Ken B. It is probably not what Krugman has in mind but I think he and Steve agree that protectionism increases employment, because we have to work more to maintain our consumption, and free trade decreases employment, because we can use some of the increase in consumption to take more leisure. Krugman has it right, but perhaps for the wrong reason.
@Neil, yes, I do believe Steve and Paul are agreeing (at least in these two posts) that protectionism increases employment – or leaves it unchanged. However, Paul says “therefore, protectionism is good” – the unstated assumption is that employment is good.
Steve points out that trade affects what kind of employment is available – a fact that Paul glosses over or ignores. Steve points out that not all forms of employment are equally desirable, again a point ignored by Paul.
In fact, Steve has argued in the past that employment is not *a priori* good – contradicting Paul’s unstated assumption. Rather, it is improvements in ones quality of life that are “good”.
While it is true that unemployment often results in a loss of quality of life – especially involuntary unexpected unemployment. On the other hand, if “unemployment” comes in the form of an early retirement into a life of leisure – now that’s something many people aspire to.
The rebuttal Steve has attempted here of Paul’s argument is to attack Paul’s unstated assumption that all employment is necessarily good, by giving an extreme example.
@Thomas_Bayes
If A sells good to B in exchange for pieces of paper that represent a claim on B’s future production, then A has to increase its imports if A wants to consume. This means that B should start producing.
If B buys goods from A in exchange for pieces of paper that represent a claim on B’s future production; B only has to start to produce if A wants to consume.
B’s imports therefore are dependent on A’s willingness to accept those pieces of papers. If A doesn’t want to consume then B does not have to buy back its pieces of paper in exchange for its production and therefore B does not have to start to produce.
Exports are therefore necessary for Imports, but Imports are not necessary for Exports. What’s necessary for Imports is the willingness of the Exporter to accept those pieces of paper; the Importer has no problem with accepting goods.
It is perfectly possible to Import as long as the other expects you to start exporting in the future, you however do not have to expect that you will start exporting. It is however not possible to Export without increasing your own expected Imports. That would hardly be rational. If your own expected Imports do not increase you are essentially selling your goods for free. If your own expected Exports do not increase you are buying goods for free, which to me seems like a good deal.
The same goes for Consumption and Investment. It is perfectly possible to Consume as long as the other expects you to invest in your future production; it is however not necessary for you to expect that you will. On the other hand if you Invest in your future production you expect to Consume. And for you to expect to consume you have to expect that others will 1) Consume your production 2) you can consume their production.
I think, that this is the logic that underlies the observation that some variables are dependent and some are independent. In the end however they are accounting identities and as long as expectations line up it does not really matter what increases or decreases for the new equilibrium to be established.
So I guess the distinction you were looking for was divergent vs. non-divergent expectations?
Krugman is correct when he states that Korean Free Trade will likely have little effect on net job gains (at least in the short term). Its a pity he doesn’t divulge who gave him that particular straw-man to destroy.
With regards to Krugman’s statement that “This, by the way, is why claims that the Smoot-Hawley tariff caused the Great Depression are nonsense.”, Steve’s analysis shows why that claim is stupid.
I was especially interested in Mr Bayes comment about independent and dependent variables within Krugman’s equation. I would fancy that Krugman would claim that really none of the variables in that equation are entirely dependent. However, he would (probably) claim that government spending (in a liquidity trap) is more independent of consumer spending than imports are likely to be independent of exports.
I think it’s insane to say that “On average, they’re a wash”, re. X and M in the equation Y = C + I + G + X – M. X and M aren’t independent of C, I, or G.
Let’s say I export 10M tons of grain to China at $100/ton, and I import 500M ball bearings from China at $2 each. Now you tell me that I can’t import or export. First, I may not be able to consume 10M tons of grain, which will then glut the domestic market and drive the price down to, say, $20/ton. Second, I now need to buy my ball bearings domestically at $5 each. I can’t afford to buy 500M ball bearings, I can now only afford to buy 40M ball bearings. At this point I have to choose between borrowing to buy more ball bearings or reducing capacity and laying off staff. Sounds like a depression to me.
“It’s very bad practice for an economist to make policy recommendations without telling you what he’s trying to maximize.” He’s trying to maximize the vote for the Democrats, don’t you think?
P.S. Is there any particular reason why you omitted “purported” before “economist”?
From the first time I read a Krugman column years ago, the thought occurred to me that he was running an experiment.
The experiment was to see how long he would stay employed writing nonsense. I predicted that when he would eventually be fired because nobody read him, he would have a big laugh and say that took way too long.
It’s amazing to me how a Nobelist in economics can so consistently ignore the most basic and irrefutable insights economics has to offer.
Reading some of the reactions to I realized that quite a few readers believe that protectionism played a major role in causing the Great Depression and even believe that this is what all the experts believe. Bear in mind that what protectionism does according to textbook economics is to cause a misallocation of resources reducing the economys efficiency.