Scott Sumner  

Protectionism is not inflationary

Final Reply to Dolan on the UB... The Psychiatry Museum vs. the ...

Protectionism is clearly a bad policy. But it's not bad for the reasons that many people assume. If you put barriers on the import of a specific good, it tends to raise the relative price of that good. But an across the board import tax does not raise the overall cost of living. Recall that the Fed determines the rate of inflation, which is likely to be about 2% regardless of whether policy is protectionist or not.

Here is the Washington Post:

The GOP plan to reform corporate taxes -- which the Trump administration last week claimed could make Mexico pay for President Trump's border wall -- would effectively charge companies 20 percent on their imports. To consumers, it would feel a lot like a hefty new sales tax on foreign-made products. And, like any sales tax, it would put the most strain on the poorest households.
That would be the case if the Fed allowed the tax to boost inflation. More likely, they Fed would push the dollar 25% higher to prevent an inflationary impact, and consumers would not see any increase in import prices. I can't guarantee that would be the case, but it is what has occurred in other countries under similar circumstances.

I am particularly skeptical of this claim:

Research shows, furthermore, that when the prices of imports increase, the prices of domestic goods also increase because there is less competition. Russ's estimates represent a range: On the low end, only imports become more expensive; on the high end, the price of domestic goods also increases, by 5 percent.
Let's say the policy was not the tax/subsidy scheme proposed as part of corporate tax reform (which is not protectionist), but rather was simply a border tax---which actually would be a protectionist policy. What then?

In that case, there would still be no inflationary effect, as long as the Fed kept inflation close to 2%. But the relative price of imports would rise, and the relative price of domestically made goods would fall. Lower prices of domestic goods would lead to lower nominal (and real) incomes. Consumers would be worse off, but not because of higher prices. Instead, they would see smaller pay increases, and the same rate of inflation.

The following graph is also somewhat misleading:

Screen Shot 2017-02-01 at 7.02.33 PM.png
The poor might be hurt more than the rich if they are more likely to buy imported goods. But they are not worse off by virtue of the fact that they consume a greater share of their income. Any policy that makes consumption 10% more expensive also makes everyone 10% worse off, because even wealthy people who are quite miserly see the purchasing power of their wealth decline by 10%.

Just to be clear, I do think it likely that the poor are more likely to buy imported goods. My only point is that this is the only reason why import tariffs are regressive. A VAT is not regressive if it applies to all goods, as all wealth eventually ends up in consumption.

HT: Tyler Cowen

COMMENTS (17 to date)
andy writes:

Rising productivity is deflationary. How can lower productivity (i.e. result of protectionsism) not be deflationary? Ceteris paribus, obviously?

John Hall writes:

This is why I think your emphasis on aggregate supply and demand is so important. Import tariff causes a supply shock. The Fed offsets this partially through monetary policy. The net effect depends on the Fed's monetary rules. If they are inflation targeting, then the whole impact is on the real economy. If they use a Taylor rule, then it is a mix.

Effem writes:

Unless of course the Fed decides protectionism is a "temporary supply shock" and is willing to tolerate the ensuing inflation.

It has long been my suspicion that the inflation mandate will prove to be wildly asymmetric (with a Fed willing to tolerate quite high inflation under the right circumstance and/or political pressure).

Hazel Meade writes:

Many questions:
What if the Fed didn't act to counter the inflation on the theory that the whole point is to make imports more expensive and thereby drive consumers to purchase domestic goods? (Let's pretend the Fed could be influence by Trump to do such a thing. )

What policies would the Fed have to pursue to drive up the dollar by 25% and what impact would that have on the economy? Wouldn't that entail some sort of fiscal shock, such as a big jump in interest rates, with dollars becoming more scarce and hence slowing down the economy in general?

Hazel Meade writes:

Sorry, I mean a monetary shock not a fiscal one (although maybe that could happen too).

Thaomas writes:

If the Fed followed a NGDP policy protectionism (a negative supply shock) would be deflationary. If the Fed followed a price level policy it would be neither inflationary or deflationary. There is probably some kind of exchange rate policy that if followed after protection would be inflationary. Nothing can be said without specifying the policy the Fed would pursue.

Scott Sumner writes:

Andy, The key is ceteris paribus. If the Fed targets inflation at 2%, then they offset any inflationary tendency.

I'm not sure there would be 100% offset, but it's a good baseline assumption to start the analysis.

John, Good point.

Effem, You may be right, but of course 2009 is when they should have been willing to "tolerate the ensuing inflation".

Hazel, The point is not to raise the nominal price of imports, it's to raise the relative price of imports. Protectionism is not intended to raise the cost of living.

If it was in response to the proposed border tax/subsidy, it would not have to do very much to raise the dollar by 25%. I'm not sure that interest rates would have to rise at all.

Thoamas, That's true.

bill woolsey writes:

With strict inflation targeting, the border tax will cause a recession.

With flexible inflation targeting, the border tax probably would cause higher inflation for a time, and then a return to 2% inflation at a higher price level (growth path.)

It is interesting that the quasi-micro analysis given by the reporter assumes a desirable policy--nominal GPD level targeting. And under that regime, the result of protectionism will be inflation and no change in nominal income growth.

Fralupo writes:

I think you're wrong about this Scott. You assume that the Fed won't let inflation go past its target and then include the Fed response to exogenous shocks in your analysis of the proposed policy change. You recognize that the protectionism will be inflationary, but then foresee the Fed will pursue deflationary policy to stay on target.

The border adjustment/tariff is going to make imports more expensive. In general equilibrium each market for these goods is going to experience a decline in quantity sold, an increase in price, and a dead-weight loss.

This has has a feel of a post hoc ergo propter hoc fallacy. Even if you're right that the Fed can perfectly accomplish its inflation target and make the forex market eat the entire adjustment they're still responding to an inflationary shock. The lack of increased inflation in the future doesn't mean there were no inflationary events in the present or the past.

I guess what I'm saying is that "inflation won't happen because the Fed has inflation under control" is not the same as "the proposal isn't inflationary".

Michael Savage writes:

Agree on inflation, but curious about the claim 'VAT is not regressive if applied to all goods as all wealth is consumed'. Aren't poorer people typically going to consume more and invest less than richer people, so shifting from income tax to consumption tax would be regressive? Lifetime and inter-generational impact would benefit richer people.

And isn't the assumption too unrealistic to be a guide to policy? Does any jurisdiction tax property on a par with other goods? And it is too easy to evade on high value items like jewellery that can be bought elsewhere, to relative gain of the wealthier in both cases.

Neither point argues against consumption tax, just questioning claim about regressive effect.

Mike W writes:

I'm still unclear as to how this currency adjustment is not going to negatively impact consumers. This explanation is from Vox:

"But most economists think the impact on retail is greatly exaggerated. That’s because they expect that currency markets would adjust in reaction to the change. The new tax on imports would reduce demand for imports from US consumers. When they buy imports, consumers provide foreigners with dollars; doing less of that means fewer dollars being exchanged, and a lower supply of dollars means the value of the dollar, relative to other currencies, will increase."

Why would consumers want a tax system that will result in higher prices that "reduce demand for imports"?

Alex Z writes:

I'm confused. I would expect that such a tax would raise the price of goods relative to income. The Fed could stabilize the overall inflation rate, but wouldn't people's incomes still buy fewer goods?

Alex Z writes:

Sorry, let me clarify. What I mean is that I would expect the above to raise the price of goods relative to the price of labor. As a result, people's real incomes would fall.

Gabriel Cormier writes:

Protectinism under NGDP level targeting would actually be inflationary and not deflationary as Thomas proposed.

AD would still grow at the rate set by the fed (lets say 5%)
We would face a supply shock wich would reduce real supply relative to a stable level of A-Demand.
Depending on the strenght of the supply shock, we could easily assume in this situation to face rouglhy 0% real output growth and a 5% price inflation due to the 5% increase in NGDP

Scott Sumner writes:

Bill, Yes, it does seem to be assuming NGDP targeting.

Fralupo. Let me use this example. I believe that the Smoot-Hawley tariff of 1930 hurt US consumers. But not because it increased inflation, which was roughly negative 8% to 10% in 1930, 1931 and 1932.

You said:

"I guess what I'm saying is that "inflation won't happen because the Fed has inflation under control" is not the same as "the proposal isn't inflationary"."

In my view it is. It's saying that in a counterfactual case where the tariff is not enacted, the cost of living would be lower.

Michael. It's possible that the rich can evade consumption taxes more easily, but it's also possible that the poor can evade them more easily. That's an empirical question. My point is that it has nothing to do with the rich spending a smaller faction of their income than the poor.

Mike, Consumers would not see any net price change in the border tax/subsidy example. That's not the example I was considering in this post.

Alex, That's right.

Gabriel, Yes, I read Thaomas's comment too quickly, it would be inflationary, as you say.

Maurizio writes:

if you have protectionism and all other variables stays the same (and therefore the Fed does not do any offset), protectionism is inflationary. If you assume the Fed does the offset, then you are changing two variables at once.

fralupo writes:

When you put it like that I'm not sure why I thought otherwise!


Return to top