Wednesday, March 9, 2016

On the Distributional Consequences of Protectionism

It is good to see Paul Krugman turning from the macroeconomic aggregate effects of protectionism to the distributional consequences:
But it’s also true that much of the elite defense of globalization is basically dishonest: false claims of inevitability, scare tactics (protectionism causes depressions!), vastly exaggerated claims for the benefits of trade liberalization and the costs of protection, hand-waving away the large distributional effects that are what standard models actually predict. I hope, by the way, that I haven’t done any of that; I think I’ve always been clear that the gains from globalization aren’t all that (here’s a back-of-the-envelope on the gains from hyperglobalization — only part of which can be attributed to policy — that is less than 5 percent of world GDP over a generation); and I think I’ve never assumed away the income distribution effects.
I want to pick up on what Mark Kleiman noted:
So when the modern Republican Party (R.I.P), in the name of “small government” and opposition to “class warfare,” set its face against policies to redistribute the gains from economic growth, it destroyed the theoretical basis for thinking that a rising tide would lift all the boats, rather than lifting the yachts and swamping the trawlers. Free trade without redistribution (especially the corrupt version of “free trade” with corporate rent-seeking written into it) is basically class warfare waged downwards.
Imagine an economy with two sectors. One benefits from free trade and the other loses. Let’s assume that the first received a 15% increase in income while the second saw a 5% decrease in income. Let’s also assume each sector has a 50% weight in overall income. Aggregate well being may have increased by 5% but when the elites say everyone gains from free trade – they display their total ignorance of international economics. Mark is noting that it is theoretically possible to impose a tax equal to 8% of the income of the winners and rebate it to the losers. The winners on net gain 7% and the rest of us on net gain 3%. Has this ever happened? Likely not. Will it happen under Republican rule? No – more likely Republicans will cut taxes on the winners and further punish the losers.


Sandwichman said...

Which came first? Free trade or the slave trade? To what extent did modern ideas about trade evolve out of conditions in which colonial plantation slave labor was a cornerstone of international trade?

How can we think about trade with any integrity while refusing to acknowledge the material and intellectual foundations upon which modern international trade was erected? To what extent are the evasions and deceptions about the distributional effects of trade built in to the received discourse about trade rather than an invention or distortion of the Republicans?

Sandwichman said...

"Mark is noting that it is theoretically possible to impose a tax equal to 8% of the income of the winners and rebate it to the losers."

No, it is not.

The reason is that "income" is not a consistent yardstick against which the "15%" and the "8%" can be measured to arrive at those supposed proportions. Because people's bundle of preferences differ from each other's (otherwise there would be no exchange) the total amount of income will change if the same quantity of goods is distributed differently.

This is a paradox that has no solution, so instead of admitting its insoluble nature, economists simply PRETEND that aggregate income remains constant with different distributions. That is they ASSUME that all economic agents have the same preferences... and thus that there should be no basis for exchange.

Of course, if economists say it three times it is true so let's not worry about the egregious flaw at the bottom of the potential Pareto improvement compensation principle (assuming that we have long forgotten that those ersatz Paretian economic agents were already immortal anyway).

Rich C said...

Well, this is more or less how the Scandanavian countries have been doing things since WWII. They have generally been very open to international trade (not always as open to capital movements), and have specialized in high-skill manufacturing and design exports (these are the "winners") while maintaining a comprehensive, universal, and generous system of social insurance funded in part through progressive taxation (quite a lot of the funding comes from taxes on mid-range incomes). This social insurance system includes a dramatically better-funded portfolio of "active labor market policies" than anything the US has ever put together, the purpose of which is to enable workers dislocated by changing markets to shift as quickly and costlessly into new jobs in new industries. I'd just note for the record that the costs we're discussing here are in most respects a consequence of trade, as opposed to just international trade. The argument for increased international trade (that it facilitates greater specialization and thus higher productivity) also applies domestically, as does the argument for a social democratic system to equitably distribute the burdens and benefits from such economic cooperation.

Sandwichman said...

"how the Scandinavian countries have been doing things"

A niche strategy.

ProGrowthLiberal said...

Rich - thanks. David Brooks a while back decided to attack these countries as being too socialistic. Of course Brooks never bothers to check basic facts when he writes his usual weird stuff for the NY Times. Maybe we should pay more attention to how other nations address these matters.

Wallfly said...

Dear S'Man, Though I am pretty sympathetic to critique of econ (and am not part of the tribe myself) something strikes me a wrong about your argument. From how I see it, there is limited use to be gained from arguing the axioms or assumptions are wrong. Rather, the relevant question is whether there is a model that doesn't generate outrageous errors.

How I would state what I think your argument is (though correct me if I butcher it) if I were trying to make it is that the increase in cash from trade is weighed against other preferences such as employment, employment security, working conditions. (cash and stuff related to unemployment are the things that seem likely to change most from trade, though I will also grant you environmental externalities).

I guess my point is, even if lots of stuff changes in the basket of preferences and the income distribution, you can still make a plausible argument if most of the changes are small.

Sandwichman said...

"...even if lots of stuff changes in the basket of preferences and the income distribution, you can still make a plausible argument if most of the changes are small..."

If I catch your drift, your argument is similar to one made, I think, by Hicks. The idea was that if the many small negative changes were randomly distributed, they would tend to cancel each other out. The problem with that argument is that they are not random, they tend to be systematic and cumulative.