Don Boudreaux writes:
Unlike you who find Duncan Braid’s May 6th harangue against supporters of free trade “devastating,” I find it to be tendentious. Braid writes triumphantly as if he’s caught us free traders in yet another of our Keystone Kops antics – specifically here, our effort to blame tariffs for inflation. Yet no competent economist or advocate of free trade is guilty of this ridiculous charge. If one were, Braid’s evidence of free-traders’ belief that tariffs cause inflation would consist of more than single link to a piece in, of all places, Vox.
If inflation means an increase in the money supply, then no, tariffs don’t cause inflation.
If inflation means a higher persistent growth rate of the price level, then no, tariffs don’t cause inflation.
But if inflation means an increase in the price level, then yes, tariffs do cause inflation.
One of the equations I’ve found most useful in understanding macroeconomics is MV = Py. (M is the amount of money, whether M1, M2, or some other M; V is the velocity of money; P is the price level; y is real income.)
Adding a tariff makes the economy less efficient, making real income lower than otherwise. Imagine that a lot of tariffs are added and that, as a result, real income is 0.5 percentage points lower than otherwise. That’s a lot, by the way. Then, if M and V are unaffected, the price level, P, will be 0.5 percentage points higher than otherwise. QED.
Does this mean that adding tariffs is a major cause of inflation? Of course not. 0.5 percentage points in a given year (the year the tariffs are added) is a small fraction of the 3 or 4 percent inflation rate. 0.5 percentage points, on top of 3 years of inflation averaging more than 4 percent annually, is rounding error.
READER COMMENTS
Richard W Fulmer
May 13 2024 at 9:27am
Jon Murphy did a nice job of explaining your point in the Facebook discussion of Don’s Cafe Hayek post:
Jon Murphy
From a macroeconomic perspective, it is possible that a sufficiently poorly-designed tariff regime could increase inflation. If tariffs imposed caused the Long Run Aggregate Supply Curve to decline and there isn’t a subsequent decline in aggregate demand (or a relatively mild one), then the price level could rise. With a 10%-across-the-board tariff, I think that outcome could be possible given that most import [tariffs] are on intermediate goods.
Richard Fulmer
So, in your scenario, there is a general price rise as the ratio of goods to money falls – not because of an increase in the money supply, but because of a decrease in the number of goods produced.
Jon
yes
Jon
And, I should emphasize: this would be in the short run
Richard
In your scenario, then, the tariffs would result in an increase in unemployment. Normally, economists argue that tariffs shift, but not reduce, employment. Is that correct?
Jon
It could result in short term unemployment, yes. In the long run, though, people just get reallocated to less productive jobs
Richard
I understand why the impact on employment would be short-lived. As you say, workers would eventually get less productive jobs – less productive because the tariffs reduce the market’s ability to allocate resources (including labor) to their most valuable uses. But I don’t understand why the inflationary effect would also be short-lived. If people are now less productive, doesn’t that lock in the lower goods-to-money ratio?
Jon
Because inflation is a rate of change, not a level. The price level would rise (inflation) but then remain at that higher level (all else held equal)
David Henderson
May 13 2024 at 10:14am
Well done.
Jon Murphy
May 13 2024 at 11:05am
One point of clarification. I said (with your addition in brackets):
The sentence is missing a letter, not a word. I meant to say “most imports are on intermediate goods.” Of course, since most imports are on intermediate goods, then most tariffs would be on intermediate goods, so the bracket isn’t incorrect. But I do want to make sure that the point that US imports are mainly inputs in the production process is understood by readers.
Jon Murphy
May 13 2024 at 10:12am
The exact nature of the tariff will vary the effect on inflation, but I think the inflation point is also something of a red herring. What matters more from an economic perspective is that the tariff regardless of the effect on inflation will reduce productivity, and thus reduce standard of living. Tariffs increase the scarcity of the goods/services on which they are applied (which is why I used to call protectionism “scarcityism”).
David Henderson
May 13 2024 at 1:38pm
I agree that the inflation point is a red herring. That is, it’s a distraction from the much more important issues. I see it with other issues too. I never argue that an increase in the minimum wage from $7.25 an hour to $15 an hour or more will increase inflation, even though it will, because that’s not the most important part of the discussion.
vince
May 13 2024 at 2:41pm
Isn’t it a concern about minimum wage increases that they *might* initiate a wage-price spiral? If the minimum wage goes to $15, then those who were already making $15 would demand more, and so on. But then minimum wage is again too low.
It seems we have little understanding about inflation. It’s certainly not very predictable.
David Henderson
May 13 2024 at 6:14pm
You write:
It’s not one of my concerns. I chose those numbers carefully. Many states have minimum wages well above $7.25, so a jump to $15 isn’t as extreme as it sounds. It is extreme for some states with low wages, but even many of those states have few workers working for under $9 an hour. I would bet that the number of people affected by an increase in the minimum wage to $15 an hour is under 5 million, which is under 3% of the labor force.
You write:
Who is we? I think I have a good handle on it. I predicted low inflation in 2009 and, in fact, won a $500 bet based on my prediction. I predicted high inflation in 2021.
vince
May 13 2024 at 6:19pm
Your track record is better than the Federal Reserve. What’s your forecast for the next twelve months? Will it be on target at 2 percent?
Don Boudreaux
May 13 2024 at 10:23am
David: Nice post.
Someone else e-mailed me this morning about the post of mine to which you here reply. Here’s my response to that e-mail:
David Henderson
May 13 2024 at 1:38pm
Thanks, Don. I agree with what you wrote.
Andrew_FL
May 13 2024 at 11:01am
Interestingly, political debates over tariffs in the 19th century often focused primarily on their price level effects, and trying to prove one way or another how prices had moved after changes in tariff policy motivated Congress to commission the procurement of price data that form the main sources now available to us for reconstructing consumer and wholesale price indexes. In retrospect these reports come across mainly as misguided-monetary and cyclic factors swamped any level effects during the periods they investigated.
I would agree that good competent economists have been careful to distinguish the relative, level effects of tariffs from trend inflation, however the popular press and political polemicists have, as ever, been far more sloppy (and hardly in just one Vox article), unfortunately making for easy targets for rebuttal, supposedly refuting “the” argument against tariffs.
Pierre Lemieux
May 13 2024 at 12:19pm
David: It seems to me that your interesting point–as discussed by Don, Jon, Richard, and Andrew–cannot be addressed without first realizing that y is a vector, not a scalar. So what is meant by a reduction in y? This question is related to Samuelson’s article “Evaluation of Real National Income” (Oxford Economic Papers, 1950), but perhaps we don’t have to go that far to realize that, except for a catastrophic set of tariffs or a trade war, a tariff only pushes society to another point on the production possibility frontier. If that is true, a tariff does not have any impact on inflation.
Craig
May 13 2024 at 1:06pm
If we were to presuppose that all foreign vendors formed OPECUS, Organization of Product Exporting Countries to the US which then 1voluntarily raised their prices by 10% wouldn’t we expect this to have the same impact on the general price level as a relative change in prices?
Kevin Corcoran
May 13 2024 at 1:27pm
This reminded me of a lecture I listened to ages ago by Jeff Hummel on why fractional reserve banking is more libertarian than the gold standard – and thanks to the wonders of the internet, I was able to easily find the audio of that lecture here. At one point, he makes a distinction between inflation and sustained inflation. When most people say inflation, they are referring to what Hummel calls sustained inflation – that is, what David Henderson describes as a “higher persistent growth rate of the price level.” By contrast, some things (like tariffs!) might cause a one-time increase in the price level, which by Hummel’s terminology would constitute inflation, but not sustained inflation.
David Henderson
May 13 2024 at 1:39pm
Agreed.
Ahmed Fares
May 13 2024 at 3:42pm
The equation of exchange is flawed because it doesn’t account for non-GDP transactions. This from Richard Werner:
Jon Murphy
May 13 2024 at 3:52pm
Point of fact: the QTM preceeds GDO by about 400 years. Py can be measured as nominal GDP, but it doesn’t have to. It is simply a measure of nominal output (which can include non-GDP transactions).
Now, all that said: to argue that QTM is flawed because it doesn;t capture non-GDP transactions requires more evidence. Do non-GDP transactions differ fundamentally?
Ahmed Fares
May 13 2024 at 4:08pm
Yes. Sometimes an increase in the money supply means more GDP transactions. Sometimes it means more financial engineering. Same for velocity. Take a look at this chart. Note that huge decline at the end. That’s the result of that financial engineering because of low interest rates caused by the global savings glut (or maybe that was a dearth of investment.. or both):
Velocity of M2 Money Stock
How is a change in the money stock useful if it tells you one thing at one time and something different at another time?
Jon Murphy
May 13 2024 at 5:28pm
The QTM helps answer that. A good theory helps explain changes. Just like how an increase in price can tell us one thing in one set of circumstances (a decrease in supply) and another in another case (increase in demand). Never reason from a change. Always strive to understand the change.
I’m pretty sure it has to do with COVID shutdowns…Can’t spend money if there is nothing to spend it on. i mean, the timing lines up perfectly: the huge decline starts in 2020 and then ends in 2021 as lockdowns were lifted.
Regardless, I do not understand how your comment answers the question: how do non-GDP transactions differ fundamentally from GDP transactions?
One further question: why limit ourselves to just GDP? As I stated earlier, the QTM predates GDP by about 400 years. GDP is just one measure of Y.
Ahmed Fares
May 13 2024 at 6:06pm
I was referring to the huge decline in velocity from 2000 to 2020 which is what my comments are about so far.
One affects goods prices and the other affects asset prices.
Monte
May 14 2024 at 12:55am
It must be exasperating for free traders to continually kick against the goads of articles like Duncan Braid’s and the following:
Tariff Increases Did Not Cause Inflation and Their Removal Would Undermine Domestic Supply Chains
Economic View: Tariffs Have Strengthened the U.S. Economy
Why Trump is Right About Tariffs
Will anyone ever manage to permanently put a stake through the heart of protectionism? As is often the case, bad policies make for good politics.
Jon Murphy
May 14 2024 at 11:30am
Selfishly, I hope not. I have made a ton of money refuting those claims.
Thomas L Hutcheson
May 14 2024 at 12:08pm
Maybe or maybe not, but “tariffs cause inflation” is not the right stake to use. It would just dissolve in the slush of solecisms “justifying” tariffs. 🙂
Monte
May 14 2024 at 6:58pm
I agree. Inflation is a minor tiff. The stake we should be pounding is that protectionism can lead to depressions and wars.
Jon Murphy
May 14 2024 at 7:16pm
Well, let’s not overstate the case either. That can be just as bad (if not worse) to persuade.
Monte
May 14 2024 at 10:32pm
I’m not so sure. Let’s not forget Bastiat’s admonition, “When goods don’t cross borders, soldiers will”, and the Great Depression, aided enormously by disastrous trade policies (in particular, Smoot-Hawley).
See Trade Wars Lead to Shooting Wars and Depressions
Jon Murphy
May 14 2024 at 10:36pm
Bastiat never said that. The doux commerce thesis is that trade reduces the likelihood of war. But it doesn’t logically follow that a lack of trade causes war.
Furthermore, the Depression was well underway before Smoot-Hawley was passed. The tariffs deepened the Depression, but did not cause it.
Monte
May 14 2024 at 11:17pm
Well, there’s some disagreement about whether he originated the phrase (I saw Henderson’s post about Otto T. Mallory possibly being the first). Regardless, I think Marks makes a good case about tariffs leading up to WWII. And I said the Great Depression was exacerbated, not caused, by Smoot-Hawley.
One real concern I have is our on-going trade war with China, which (according to this guy), could be a lead-up to a military conflict:
Monte
May 15 2024 at 1:56am
And this article (When Trade War Threatens Real War) echoes that concern:
Monte
May 15 2024 at 12:11pm
Alongside the prospect of war is the potential for Trump (should he win the presidency) to continue his protectionist trade policies and decoupling from China. This could ignite an inter-country recession that ultimately devolves into a global depression, and protectionist trade policies are one of the primary factors that economists cite for deepening and/or prolonging a depression.
Will these things happen? I hope not. Can they? Absolutely!
Vivian Darkbloom
May 14 2024 at 10:39am
“But if inflation means an increase in the price level, then yes, tariffs do cause inflation.”
This is correct and seems to be in line with both theory and actual experience. An increase in tariffs should cause a temporary increase in the price level, but not a persistent increase in prices (“inflation”).
The best analogy and the best real-world experiment would likely be with an increase in consumption taxes, such as a value added tax. An increase in a value added tax rate should have roughly the same effect on prices as an overall increase in across-the-board tariffs (taking into account that tariffs don’t directly or necessarily increase all domestic prices).
In this respect this reminds me of the recent post by Art Carden who argued that an increase in tariffs would actually be a punishment of consumers (“Tariffs are Sanctions Against Consumers”: https://www.econlib.org/tariffs-are-sanctions-against-consumers/
However, most of the same economists who argue against tariffs also argue *for* “punishing consumers” with consumption taxes rather than income or wage taxes (those advocating consumption taxes never use such rhetorical terminology, though!). It strikes me that a consumption tax in the form of a tariff might actually be easier to collect and certainly more politically feasible to enact. The latter may actually be the point of many of those who argue for tariffs—a back-door consumption tax, if you will. There may be other reasons to oppose tariffs; however, it is not evident to me that “punishing” or “sanctioning” consumers is one of them, any more than, say, a VAT would be.
Economists also differ somewhat on who actually bears the cost of tariffs. At the extreme, such as a VAT, many would argue that is is ultimately all borne by the domestic consumer. However, others, somewhat convincingly, argue that some of the cost of tariffs *is* borne by foreign exporters who cut prices somewhat to maintain their ante hoc competitive advantage.
Thomas L Hutcheson
May 14 2024 at 12:05pm
I think this attempt at distinguishing “inflation” from an “increase in the price level” makes no analytical sense. During the time when the price level is changing, that is “inflation” and inflation during any given period of time in the increase in the price level. Both (except that there really are not two different things) are outcome of wise or not-so-wise monetary policy.
Thomas L Hutcheson
May 14 2024 at 11:58am
No tariffs do not causes an increase in the price level. The cause an increase in the relative price of the tariffed item. That translates into an increase in the average level of prices if the Fed (in this case probably wisely) allows/engineers it rather than trying to reduce other prices enough to prevent an increase in the average.
But please keep your eye on the ball! Inflation and deflation are outcomes of monetary policy not the changes that might motivate monetary policy makers to take action or not.
To say that _tariffs_ cause an increase in the price level is as bad as saying that _deficits_ cause an increase.
Comments are closed.