Michael Hicks, a professor of economics at Ball State University, tweeted on Monday:
For everyone who thinks trade policy (tariffs) will affect the trade deficit (instead of just slowing economic activity in general, as taxes without better public services tend to do, I offer Exhibit #1: US Trade Deficit Grows for months after Trump’s Tariffs.
Instead of reproducing the trade deficit chart used by Professor Hicks, my figure shows exports and imports separately. The increase in the trade deficit (the difference between the two curves) since June can be seen clearly.
Many people seem puzzled by the fact that the U.S. trade deficit does not necessarily decrease as tariffs are increased. But this is not surprising when you realize that the trade deficit is influenced by many other factors, including the value of the dollar, the net inflow of foreign investment, total domestic savings, and the federal budget deficit, to name the obvious suspects.
Consider the following propositions:
-
Ceteris paribus (other things being equal), a tariff will decrease the trade deficit (or increase the trade surplus).
-
Ceteris paribus, a higher U.S. trade deficit will push down the value of the dollar.
-
Ceteris paribus, a lower dollar will push down the trade deficit.
-
Ceteris paribus, the attractiveness of investment in the U.S. will push up the trade deficit.
-
Ceteris paribus, lower domestic savings will push up the trade deficit.
-
Ceteris paribus, a higher federal budget deficit will push up the trade deficit.
All of these are true, but only ceteris paribus. In the broader economy, the variables they represent influence each other. And variables such as the trade deficit, the exchange rate of the dollar, and foreign investment are determined simultaneously. The economy is a general equilibrium system, where all variables are determined like in a system of equations.
I previously discussed an analogous case: supply and demand–“A Frequent Confusion and the Yo-Yo Economic Model,” January 29, 2018. Quantity is determined by price and price is determined by quantity. These two propositions are not contradictory because price and quantity are determined simultaneously by two equations, one representing demand (quantity demanded as a function of price), the other one representing supply (quantity supplied as a function of price).
Back to the trade deficit. Consider for example propositions (2) and (3) above (in isolation from the other propositions). These two propositions are not contradictory because both the trade deficit and the price of the dollar are determined simultaneously, given other autonomous factors such as the attractiveness of the American economy. More generally, we can say that the trade deficit and the exchange rate are determined simultaneously with all other variables in the economy
What presumable happened over the past few months is the following. The federal government imposed new tariffs, which ceteris paribus would have pushed down the trade deficit and the dollar. But other factors exerted upward pressure on the currency. A trade war and deteriorating prospects for the world economy made the American economy relatively more attractive (the dollar is considered a refuge value). The increasing federal deficit attracted more savings to America. On balance, the dollar increased, pushing imports up (in conjunction with the brisk pace of the American economy), exports down (in conjunction with foreign retaliation), and the trade deficit up.
A fixation on the trade deficit (which has been typical of protectionists for centuries) as the source of all evils fails to consider the economy as a system. Tariffs don’t determine the trade deficit, except in a ceteris-paribus sense.
READER COMMENTS
Jon Murphy
Oct 12 2018 at 6:48am
I want to push back a little on your #1. The Lerner Symmetry Theorem indicates a tax on imports would also reduce exports. So, I see why a tariff would potentially reduce a trade deficit, by having both exports and imports converge toward 0, but would it necessarily increase a trade surplus?
Thaomas
Oct 12 2018 at 11:37am
He is right if the tariff is so small as not to affect any other variables except the single marginal item on which the tariff is levied and the decrease in the deficit is likewise marginal. But the larger point is still valid. It is just silly to model a macroeconomic result like the trade deficit except with a GE model.
Warren Platts
Oct 12 2018 at 1:55pm
As Krugman once said, it is a dirty little secret of economics that a PE model will get you in the right ball park.
Jon Murphy
Oct 13 2018 at 8:55am
It’ll get you in the right ballpark in terms of observing behavior. It’ll get you nowhere near the right ballpark in terms of policy design (toward his point that the optimal tariff model is an intellectual curiosity but practically useless, a point made several times in his textbooks and columns).
When discussing policy, one needs Public Choice. In a knockdown fight between public choice and PE/GE, public choice wins every single time.
Jon Murphy
Oct 13 2018 at 8:53am
Ok. Both these comments are irrelevant to my point, though
Pierre Lemieux
Oct 13 2018 at 10:09pm
That’s an interesting point, Jon, but I am not sure that it is correct or that I understand. Do you care to elaborate?
Jon Murphy
Oct 14 2018 at 8:45am
Let me try to explain my (admittedly off-the-cuff) thinking like this:
The Lerner Symmetry Theorem argues that an ad valorem tariff acts the same as a similar tax on exports by altering the relative price in the same manner.
By increasing the price of imports (and thus reducing the amount demanded of imports) and subsequently by increasing the price of exports (thus reducing the amount demanded of exports), it’ll reduce a trade deficit by having the number of imports and exports coverage toward zero (obviously, you’d only reach zero with a sufficiently high tariff, but the movement would still be the same, I would think).
The way to get a trade surplus from a tariff would be if imports decline quicker than exports decline (no?). I don’t see a reason why this would necessarily be the case.
Warren Platts
Oct 12 2018 at 2:03pm
Pierre, you have got my mercantilist brain fired up. Yes, tariffs alone are probably not enough to unwind the trade deficit (unless we are prepared to levy an across-the-board tariff of 30% on everything). There should be taxes on capital inflows as well imo.
But why not, as a part of renegotiated NAFTA/USCAM bring back the idea of the Amero: a common currency for USA, Canada, and Mexico (and the rest of Latin America if they want in on it)? The Euro works great for Germany: it artificially devalues the Deutschmark. An Amero would probably have a similar effect for the U.S. dollar!
Jon Murphy
Oct 12 2018 at 8:31pm
A weakened dollar would reduce the US terms of trade (by reducing the price received for exports and increasing the price paid for imports) and make Americans poorer by eroading real income.
Pierre Lemieux
Oct 13 2018 at 10:07pm
Yes, it’s like an “optimal tariff” in reverse.
Warren Platts
Oct 15 2018 at 12:25pm
Respectfully disagree. A currency devaluation (such as would be caused by a single, North American currency) should have a similar effect on prices as an optimal tariff, but without the tax revenue: imports get more expensive, and thus there is a tendency to buy less of them.
Especially if the traded goods are high value-added manufactures, suppliers of imports would accept lower profits and lower their prices in response to the cheaper “dollar” in order to maintain U.S. market share. Meanwhile, U.S. suppliers of exports would have leverage to raise their prices in order to capture the foreign consumer surplus. In the logical extreme, the price action would completely cancel the effects of the devaluation in real terms! Concerted action by the central bank would probably be required to maintain the devaluation in the long run.
As for making Americans poorer, the main effect would be to increase U.S. savings and reduce unemployment. As a result, total real wages paid out and average wages would increase and thus stimulate aggregate demand. Desired investment would be pushed up. Short run consumption of imports would be reduced, but overall GDP growth should improve, making Americans better off in the longer run.
Comments are closed.