Before I read Deirdre McCloskey, I had a rather primitive view of methodology. Economists should develop hypotheses and then test these theories using real world data. Models were mathematical and empirical tests used regression analysis.
Today, I have a more eclectic view. A macroeconomist uses basic economic concepts, stylized facts and financial market reactions to news in order to develop a framework for understanding important macro phenomena. Influential economists utilize more than just formal tests; they develop coherent stories that use powerful metaphors to persuade those with an open mind (i.e., the young.)
In my book on the Great Depression, I challenged the prevailing view that tariffs have an inflationary impact. Free trade advocates tend to see tariffs as a negative supply shock, and assume the impact will be inflationary. Keynesians admit that tariffs have modest efficiency costs, but also argue that they can boost demand for domestic goods, especially in a severely depressed economy. They also see tariffs as having an inflationary impact, for both supply and demand-side reasons.
I argued that the Smoot-Hawley tariff had a deflationary impact. This tariff interacted with monetary policy to reduce aggregate demand, which more than offset the inflationary impact of reduced aggregate supply. The response of asset markets to news about Smoot-Hawley further supports that view. Deflation accelerated after Smoot-Hawley was enacted in June 1930.
I don’t think my book will convince many people, partly because hardly anyone read it and partly because I don’t use standard methods. Nonetheless, I believe my explanation will eventually be accepted.
A recent issue of The Economist had a special essay on the causes of the global trend toward low inflation. This caught my eye:
Ironically, the recent incremental reversals of globalisation provide good examples of the importance of global financial conditions to inflation. In theory tariffs should boost inflation in the country that sets them. But as the trade war between America and China heated up during 2019, it sparked fears about global growth and triggered a rush into safe assets such as Treasury bonds. Long-term bond yields fell to new depths and the dollar surged. In response the Fed has cut rates and the ECB has restarted QE.
The deflationary impact of a change in global risk appetite has proved far more significant than the modest inflationary impact of the tariffs themselves.
Just to be clear, tariffs do not always have a deflationary impact; it depends on how they interact with the monetary regime. It is possible that the Fed’s recent rate cuts will prevent any significant fall in inflation (although TIPS spreads have declined). But the fact that they needed to cut rates is itself telling. In 1930, we were on the gold standard and central banks had less ability to engage in monetary offset.
READER COMMENTS
James
Oct 23 2019 at 10:03pm
Isn’t every tax increase inflationary in itself in the sense that all taxes reduce everyone’s budget constraint, lowering the denominator on the right hand side of P=MV/Y?
If monetary authorities respond to a tariff or other tax in some way, that is a separate policy decision which may be inflationary or deflationary but it separate from the tariff.
More simply, deflation increases the purchasing power of the currency unit. Are you claiming that a tariff can increase purchasing power?
Scott Sumner
Oct 24 2019 at 11:36am
James, Your first point is incorrect because any reduction in private budget constraints is offset by an equal rise in the government’s budget constraint. Yes, real GDP will likely fall, but for reasons unrelated to budget constraints.
On your second point, that gets into philosophical discussions of “causality”. Think of my claim as “tariffs are likely to be followed by lower inflation, in the real world.”
Your third point mixes up the purchasing power of a currency unit with the total purchasing power. The deflation of the early 1930s increased the purchasing power of the dollar bill, but reduced total purchasing power because NGDP fell so sharply.
James
Oct 27 2019 at 3:52pm
Scott,
I take it as given that all economic behavior is chosen in response to preferences and constraints. All changes in economic behavior must therefore come from changes in preferences or changes in constraints or changes in both.
Have I erred so far?
This leaves only a finite number of ways to understand how tariffs affect the economy:
Tariffs affect behavior because they affect preferences.
Tariffs affect behavior because they affect constraints.
Tariffs affect behavior because they affect preferences and constraints.
Unless you believe tariffs affect preferences, you are left with constraints as the link between tariffs and behavior.
Now to your other suggestion that changes in private and government budget constraints must be equal and opposite, this could only be true if RGDP were constant. In the real world, we know that is not the case.
Jon Murphy
Oct 27 2019 at 4:56pm
I don’t follow you on this point. If taxes go up by X dollars, the government’s budget constraint is loosened bu X dollars while the private budget constraint is increased by X dollars.
Thaomas
Oct 24 2019 at 11:43am
It depends also on whether the tax increase (leaving aside any effects through changes in animal spirits/expectations of consumers and investors) affects the deficit which in turn interacts with monetary policy which policy in turn may or may no be optimal.
Philo
Oct 24 2019 at 10:42am
“I don’t think my book will convince many people, partly because hardly anyone read it and partly because I don’t use standard methods. Nonetheless, I believe my explanation will eventually be accepted.” This is not at all unlikely. Eventually some more prestigious economists will come around to advocating many of the views expressed in your book-—perhaps arriving at them independently, perhaps having picked up on faint echoes originating with you–and the prestige of these advocates will make the views popular in the profession.
Scott Sumner
Oct 24 2019 at 11:37am
Let’s hope so.
John Arthur
Oct 24 2019 at 11:28am
So are the Trump Tarrifs playing a major role in the current low inflation, or is this more of the Federal Reserve missteps?
Scott Sumner
Oct 24 2019 at 11:37am
Some of each.
Thaomas
Oct 24 2019 at 12:21pm
Meaning that the slightly lower deficit (if that’s the result) stemming from the tax collections on imports adds to the “stock” of accumulated below-target NGDP (not an “Okun gap” but a “Sumner gap?) that the Fed is not addressing forcefully enough?
Stewart Bovi
Oct 24 2019 at 12:23pm
The pros and cons of tariffs and how they affect the country vary from one implementation to another, but I believe that despite short term effects they can bolster the country in the long term. The increase in the prices of foreign goods leads to a decrease in their demand. The decrease of their demand can lead to an increase in the demand of domestic goods, especially in the case of necessary goods such as light bulbs or computers. The negative short term effects of tariffs cannot be ignored, especially in terms of stocks and investment. Though these mostly seem to affect larger companies who import luxury goods such as toy companies like Hasbro, who recently suffered a loss in their stock value due to the tariffs.
The tariffs may also bolster competition between domestic good producers and foreign good producers, as increased prices for imports will potentially lead to foreign good producers to try and find methods to lower the prices of their goods to minimize the effect on profit caused by the new fixed cost. I agree that tariffs will have a deflationary impact but only in the long term, in the short term they will almost certainly lead to inflation. Looking at how tariffs have affected the country in the past is the best method to try and compare how these tariffs may affect our country today.
Matthias Goergens
Oct 25 2019 at 6:16am
You just have a shift in demand: there will be less demand for exports. (And the whole rest of your analysis then runs in reverse for export goods.)
Jon Murphy
Oct 25 2019 at 8:50am
Ok, be careful here. Yes, increasing the price of foreign goods can increase demand for domestic substitutes, but there is also an income effect going on. Such an increase in prices necessarily means domestic consumers have fewer dollars to spend on other goods. So: 1) the increase in domestic demand is offset by lower demand for other domestic goods, 20 the increase in domestic demand is less than the reduction of foreign imports, ineffectively offsetting the decline in GDP and reducing the domestic standard of living, or 3) some combination thereof.
Further, as Matthias rightfully points out, a decrease in imports necessarily means a decrease in exports as well, which contributes more to a reduction in domestic GDP.
Hard to see how restricting competition will increase competition, especially given your mechanism here. Foreign competitors were already the low-cost producers. If tariffs are designed to increase their domestic prices, then domestic firms have no incentive to compete; you’ve removed that incentive. Indeed, by giving them an implicit property right in future profits, you’ve incentivized them to fight for higher tariffs and other competition-reducing measures. In all likelihood, we’d see just the opposite (which is what the empirical evidence shows us)
Lorenzo from Oz
Oct 25 2019 at 6:40pm
The 1973 oil price shock had a deflationary impact, despite being seen as inflationary. Strikes me as a somewhat similar case.
Warren Platts
Oct 26 2019 at 4:34pm
Fascinating post Scott! So the neoliberal fear that tariffs cause inflation can be laid to rest! Recent events evidently confirm your theory: although one can cherrypick prices for individual items that have gone up, the overall CPI has not budged since Trump doubled U.S. tariffs.
Jon Murphy
Oct 26 2019 at 6:30pm
Why use the overall CPI? That doesn’t make sense; it’d be like claiming minimum wage has no effect on jobs because the overall job numbers rise. Rather, it makes more sense to look at the items that the tariffs have been imposed upon.
Thomas Hutcheson
Oct 26 2019 at 7:49pm
Inflation is anywhere and everywhere a monetary phenomenon. [Milton Friedman, PBUH] Neoliberals think that tariffs affect relative prices and real incomes.
Jon Murphy
Oct 26 2019 at 9:36pm
Point of clarification: it’s more than just “neoliberals” (I hate that term because of how amorphic it is, but we’ll punt on that). Most microeconomists will argue tariffs, by design, affect relative prices as opposed to the price level as a whole. Macroeconomists, like Scott, are concerned about inflation as well, but they’d still discuss that the tariffs affect relative prices.
Comments are closed.