Another Slight Correction to a Boudreaux Anti-Tariff Argument
In my never-ending quest to get everything perfect, I once again must quibble with one of Don Boudreaux’s free trade (or anti-tariff) arguments. Now for context, Don publishes three of these a day, so the fact that I object once a month means I endorse just about everything the guy writes…
Anyway, here’s Don’s piece (in the form of a letter to the editor) that I think is a bit off:
Jeff Jacoby superbly analyzes the Trump administration’s proposal to slap punitive tariffs on softwood lumber imports from Canada – that is, the administration’s proposal to punitively tax Americans who buy softwood lumber from Canadians (“Trump’s tariffs will hurt Americans,” May 3). The goal, of course, is to reduce the amount of Canadian lumber that we Americans receive in exchange for our dollars.
Mr. Trump boasts about his mastery of the “art of the deal.” Given his administration’s approach to trade, we can therefore conclude that, in Mr. Trump’s mind, a truly masterful deal-maker is someone who, with each deal, commits himself to give to his trading partner as much as possible while he himself – the masterful deal-maker – gets in return as little as possible.
You’ll forgive me for being skeptical that such ‘artful’ deal-making will increase our prosperity and make America great again. [Bold added.]
So to reiterate, obviously I agree with Don that slapping tariffs on Canadian lumber makes Americans poorer, but I think his actual argument is wrong. (And so if Don hopes Trump sympathizers will come around, he might understand why they would reject his particular attempt.)
It’s the part I’ve put in bold that is wrong. And since that particular error is the whole crux of this letter, I think the whole thing blows up in Don’s face.
When the United States government puts a tariff on Canadian lumber, that doesn’t make Americans give more to Canadians for Canadian lumber imports. If you think the U.S. is a small player in the world market, it will have no effect on the price. Or, to the extent that reducing U.S. demand for Canadian lumber has an impact on price, it will actually cause the Canadians to give us more lumber per dollar we give to them.
Now the source of the error here is our usual notion of looking at the particular American importer. From his perspective, a tariff makes him pay more for Canadian lumber. But clearly Trump (and Don, since he’s trying to argue on Trump’s own terms) is thinking of it in terms of “us vs. them.” And clearly, the U.S. levying a tariff on Canadian lumber means that either the U.S. gets the same amount of wood per dollar we send them, or we get more wood per dollar. It’s true, the particular American importer gives up more dollars per unit of wood received, but it’s not all going to Canadians; some is getting redirected to Washington. To repeat, the amount “getting through” to Canada is at most the same (per cord of wood imported), or if anything lower.
In the big picture, the problem with a tariff on Canadian lumber is NOT that it forces Americans to give more resources (per unit of wood) to Canadians for the Canadian lumber that Americans continue to import, even after the tariff is levied. (To repeat, in general it’s possible that on this dimension the tariff IS beneficial.) Rather, the problem with this tariff is that it leads Americans to partially reduce the amount of wood obtained from Canada, and to replace it with domestic sources, even though the total resources expended “internally” vis-a-vis other Americans is more than what we’d have to send to Canadians for the same amount of wood.
Yes, I totally agree, the Canadians are unequivocally worse off with their market diminished, selling less commodity at a lower price … there’s a secondary effect beyond that, if Trump can bring the US balance of trade back to zero then I would expect that to be bullish for the USD. That’s got a bit IF in it, but it all comes down to belief. I really don’t see how Don Boudreaux can possibly expect the Canadians to be happy about this.
On the US side of the equation (as I’m sure I’ve said before in similar situations) it is silly to imagine the USA as some collective entity with a single purpose. I’m sure the US lumber producers will be better off because they can sell more commodity at a higher price and never need to pay that tariff. Good for them. The flipside is that the builders who need timber for houses would pay a higher price (effectively their tax rate has gone up) which they in turn pass on and it reduces supply in the whole housing industry… probably not by much once it’s all averaged out, but a large number of people are affected by it.
Then there’s future taxpayers facing the burden of massive US government debt, anything that helps raise taxes on the present generation is a good thing for those future generations.
OT but related.
Andrea Ferrero is a serious economist, and has published papers to the effect that nations with large trade deficits see exploding house prices.
https://www.newyorkfed.org/medialibrary/media/research/staff_reports/sr541.pdf
Property zoning no doubt plays a role. Foreign capita pours into an artificially scarce good.
Whether or not one “believes in free trade” (or supports the status quo, which is managed trade between nations all of which heavily and variously regulate, tax and subsidize their economies) is seems to me academics should face this truth.
Why are house prices $1 million in Sydney? Vancouver, London? Los Angeles?
So, should we tell Americans the truth?
“Yes, we can run huge trade deficits and we will see exploding house prices”?
Or “Due to the imperatives of globalism, you can no longer zone your neighborhood as you see fit. No more property zoning, as it conflicts with global free trade.”
BTW, I really enjoy Robert Murphy’s blogging and what I see of his personality.
So far, the globalists and free traders to whom I pose this question much prefer to talk about how lumber prices and labor shortages will make housing more expensive.
BC, thanks for the kind words. I’ve seen you hit this topic a lot at Sumner’s blog. Are you just wanting economists to more frequently say, “Local zoning keeps up property prices and should be abolished”?
“Or, to the extent that reducing U.S. demand for Canadian lumber has an impact on price, it will actually cause the Canadians to give us more lumber per dollar we give to them.”
No, no. Given the preferences of U.S. individuals, a higher price (tariff) reduces the amount of labor bought from Canadians.
The Canadians would see no point in selling us lumber, since the goal of a tariff is to price out foreign competition – the same way the Minimum Wage works.
“In the big picture, the problem with a tariff on Canadian lumber is NOT that it forces Americans to give more resources (per unit of wood) to Canadians for the Canadian lumber that Americans continue to import, even after the tariff is levied.”
He didn’t say it was. He said:
“The goal, of course, is to reduce the amount of Canadian lumber that we Americans receive in exchange for our dollars.”
That is obviously the goal, since, in Trumps mistaken view, imports reduce jobs that he thinks Americans are entitled to – as if American consumers aren’t entitled to spend their money how they wish.
Americans would have to pay more for Canadian lumber if they chose to keep buying Canadian, but implied in a tariff is the recognition that it imposes opportunity costs on unchanged demand.
So, the goal isn’t to affect the price of Canadian lumber – that’s just a means to and end.
guest, you keep flipping back and forth between “total” and “per unit” arguments.
A new U.S. tariff on Canadian lumber will reduce the U.S. demand for Canadian lumber, from the Canadians’ point of view. So that will have two effects:
1) It will reduce the total amount of lumber Americans buys from Canadians.
2) It will (perhaps only slightly) reduce the unit price Americans pay for each cord of Canadian wood.
Nice catch. The real cost is that it undoes the real benefit of trade: more efficient allocation of resources.
But as ever, that requires a global efficiency metric… which does not exist when some people are better off (i.e. US lumber producers) and others are worse off (i.e. Canadian lumber producers). There is no metric which can objectively trade one person’s gain for another person’s loss.
” There is no metric which can objectively trade one person’s gain for another person’s loss.” That’s what money is for.
More to your point, it’s what Kaldor-Hicks Efficiency is for: to give a precise meaning to “this outcome is better than that one”. In a Kaldor Hicks improvement, a set of (unimplemented) voluntary exchanges would result in a Pareto improvement. (With Pareto efficiency it’s pretty clear what better means, but Pareto improvements are rare.) That’s a perfectly coherent notion.
So your claim there is no metric which can objectively do it is mistaken. You can argue that since the KH side deals don’t actually occur it’s a bad rationale for actually making the change, but it is a perfectly sensible criterion for assessing it.
What is your experimental technique for measuring something that never existed ?
Huh? My neighborMr Hicks enjoys using my lawnmower. His gain, my loss. Incomparable you say? But, and did you see this coming? he pays me money to do so.
My other neighbor enjoys using my hedge clippers. However she just won’t pay me. She’s 12. But, and did you see this coming? her father pays me money! The money comes from someone else, that nice Mr Kaldor.
In both cases we are all happy. And before it happened we agreed on doing it that way — something you insist is impossible.
You have just shifted the ground.
In your story all parties get paid, and no one is left worse off. That’s clearly not what happens in a Trump tariff situation. The US lumber producers are better off, government collects additional revenue, while other parties are worse off.
You try to pretend that one type of outcome is really some other type of outcome, but it won’t work, I won’t believe you. Show me how the Canadian lumber yards get compensated by that nice Mr Kaldor. How do they negotiate a price with him? What about the US lumber worker who gets a job because of Trump’s policy? He is better off with that tariff in place.
Oh my gosh this is taking forever…
Tel, you agree that if a suitable exchange of money would make a move unanimous, then everyone is better off (in an important sense), right?
OK, so the Kaldor-Hicks criterion says that a move promotes efficiency if the people in favor of it would be willing to pay enough dollars such that the people who opposed it would be happy, after compensation out of that pool of money. I.e. the gains to the winners outweigh the losses to the losers. There is no need to measure subjective utils because we’re expressing everything in dollar terms, just like in the unanimous case.
Now, if you want to object to that, you can, by saying, “Without it actually being voluntary, we don’t really know how much the losers would have needed to be compensated. That’s pure speculation.” So is that what you are saying? If so, fine, nothing wrong with that objection. But it doesn’t seem to be the objection you are making.
Presuming the parties come to an agreement and the compensation does get paid you would have a Pareto improvement with no need for any Kaldor-Hicks theory.
I’m sure we all understand that this situation never happens in the real world.
Kaldor-Hicks is worse than that. Not only does it skip the voluntary negotiation phase, but never does any compensation actually get paid to anyone. The entire exercise is speculation.
Hence a serious epistemological question : how do you measure something that does not exist?
Tel
“ There is no metric which can objectively trade one person’s gain for another person’s loss.”
Sure there is. We call them dollars. This little chart tells you everything you need to know about tariffs. Notice the dead weight losses.
http://farm5.static.flickr.com/4100/4822370246_1921331907.jpg
If you assign dollar amounts to the vertical axis and quantities to the horizontal axis you can calculate each of the effects of a tariff, including government tax revenue and dead-weight losses.
You are saying that dollars represent universal and interchangeable utility for all men? That’s a brave statement on an Austrian blog.
Even the “Progressives” don’t believe it, they say that the marginal difference for a very rich man gaining or losing one dollar is so small the rich man hardly notices… while to the poor man that marginal dollar makes a very big difference. A lot of people seem convinced by that argument, it’s the whole theory behind taxing the rich (personally I vote with Willie Sutton on that score, but perhaps that’s another topic).
Of course, there’s an easy empirical way to demonstrate that not all value is measured in dollars… simply take note of the very large number of people willing to volunteer. If dollars were the universal metric, we should expect zero volunteers, since no dollars change hands.
As for your diagram, if Trump imposes 20% tariff on Canadian softwood, I hope you don’t for a moment believe this would increase the price of softwood lumber right across the USA by 20%. The Canadians of course will reduce their price somewhat and also the US industry can grow to make up at least part of the difference. Shipping costs mean that Canada cannot easily substitute Asian or European buyers instead of US buyers.
And “Societal loss” who is that exactly? Show me this Societal fellow, is he Canadian? Trump has made it clear he does not feel obligated to foreigners, given that they don’t vote (other than in California, but Trump will never win there). What about the tax revenue? Someone ends up paying the government debt… maybe that Societal guy doesn’t want to pay income tax?
Tel
“You are saying that dollars represent universal and interchangeable utility for all men?”
No, I’m not. Of course all value is subjective, and you are correct that a dollar may be more important to a poor person than to a rich person. But what we are talking about here is the effect on supply and demand when prices are artificially manipulated by imposing a tariff. When an exchange does occur, usually involving dollars or some other medium of exchange, we can say that the value of the good or service exchanged is worth MORE than some number of dollars to the buyer, and LESS than that same number of dollars to the seller.
From that we can determine an approximate value of the good or service in dollars, based on the number of dollars actually exchanged in a transaction.
“Even the “Progressives” don’t believe it, they say that the marginal difference for a very rich man gaining or losing one dollar is so small the rich man hardly notices… ”
Yes, marginal utility is a real thing, as is diminishing marginal utility.
“Of course there’s an easy empirical way to demonstrate that not all value is measured in dollars… simply take note of the very large number of people willing to volunteer. If dollars were the universal metric, we should expect zero volunteers, since no dollars change hands.”
That’s correct. Not all exchanges of value involve dollars. There are psychic values and profits from volunteering that don’t involve dollars, but we can often compare volunteer activities to similar activities performed for dollars to get a rough estimate of their dollar value to others.
“As for your diagram, if Trump imposes 20% tariff on Canadian softwood, I hope you don’t for a moment believe this would increase the price of softwood lumber right across the USA by 20%.”
No, of course not. It’s simply supply and demand. Prices for US buyers will increase, as quantity demanded for US lumber increases, and US production will increase due to the higher quantities demanded and higher profits available. Overall US quantities demanded may fall.
“The Canadians of course will reduce their price somewhat and also the US industry can grow to make up at least part of the difference.”
Yes, that’s most likely true.
“Shipping costs mean that Canada cannot easily substitute Asian or European buyers instead of US buyers.”
On the other hand, lowering their prices may increase quantities demanded by Asian and European buyers, as well as by domestic Canadian buyers. More basic supply and demand.
“And “Societal loss” who is that exactly? Show me this Societal fellow, is he Canadian?”
No the “societal fellow” isn’t Canadian, he’s an American consumer or producer. Basically, the decrease in consumer surplus is greater than the increase in domestic producer surplus plus domestic government tax revenue, by the amount of the dead weight loss. Total net domestic welfare has decreased by that amount. There are also foreign losses so that total world welfare is decreased by the imposition of a tariff.
Here is an excellent video that covers the entire subject very well. Hopefully you will find that your life is much improved for having spent that 12 minutes. 🙂
https://www.youtube.com/watch?annotation_id=annotation_783429&feature=iv&src_vid=S_pvEupKScc&v=80MJngzdDkI
Trump has made it clear he does not feel obligated to foreigners …”
Trump has made it equally clear that he doesn’t feel obligated to American consumers. Obviously.
What about the tax revenue?
What about it? It’s paid by American consumers.
“Someone ends up paying the government debt… maybe that Societal guy doesn’t want to pay income tax?”
It’s not clear what you are saying here. The additional government revenue comes from American consumers, the same people who often pay taxes. I hope you’re not suggesting that this increase in tax revenue will be offset by lower income taxes. Government spending changes in one direction only, except for shell games designed to favor one group of taxpayers at the expense of another.
Here’s another excellent video on the subject of tariffs:
https://www.youtube.com/watch?v=Gr-Ld7DnBZQ
After all your earlier points, doesn’t that stand out as somewhat incongruous?
Let’s suppose the “consumer” here was one very rich white man, by your own argument about “marginal utility” his loss would be considered small, right? Regardless of the actual total in dollars we can do the whole adjustment on the margin and make it into a small “utility” value… I mean he’s rich enough already, isn’t he?
And if those “producers” happened to be a whole bunch of poor, women of colour, their “marginal utility” would be higher, am I right?
Picture if you will a vast crowd of Whoopie Goldbergs in lumberjack outfits and I think you might see where I’m coming from. Possibly a few of them might look a bit more like Elizabeth Warren with her Aztec heritage and all that. All of them are very appreciative of Trump providing them with jobs.
This is where it gets weird, you cannot actually add up either the producer’s total utility or the consumers total utility on any particular unit scale (and you admit this yourself) but strangely you still want to compare one total with the other total. It’s meaningless.
Unless you can find some genuine commonality amongst all humans (or at least the vast majority of them) you have nothing to use as a reference.
All government debt will be paid by someone, that’s a simple accounting tautology. I cannot tell you who will be that sucker, but I can toss out a few options.
So yes it might be income tax, although probably from future generations not born yet. It also might be additional tariff revenue (if Trump can swing it, which remains to be seen). Possibly government will print money which is tax by stealth, but still a type of tax. In the worst case, government might default and stick the bondholders, which is essentially tax by asset confiscation (they may decide to call it a tax, rather than a default, depending on what works better from a legal and propaganda perspective).
Regardless, someone has to pay. One way, or another way.
Now as for increases in government spending. Thank’s to the magic of debt and central banking, this is no longer directly tied to revenue. However, if the present generation were faced with immediate taxation (by whatever method) to balance the budget, then they might think twice about increasing spending. Admittedly, no one has ever figured out how to slow down government spending before it hits total collapse, so perhaps my suggestion will not work. That said, I put forward that open ended government debt is perhaps the greatest moral hazard of all.
I’m open to suggestions if anyone else has a better way to slow government spending.
Tel
No it doesn’t. It is what the tariff chart is telling you , and what both the videos I linked to are telling you. There is a net loss of welfare from a tariff. Did you watch either of them?
We cannot compare subjective value. Only the very rich white man can determine the value – to him – of his loss. We can’t, nor can anyone else. We know that diminishing marginal utility exists, and we can discuss it in general terms, but we can’t quantify it for others except by observing their actions..
I don’t know that, and neither do you. How can anyone decide for others that they are “rich enough”?
We can only assume so, but we don’t know for sure. You can’t quantitatively compare the subjective value of a dollar to a whole bunch of poor women of color to the subjective value of a very rich white man. Nor can you compare the subjective value of a dollar to a whole bunch of poor white men and a very rich black woman. You can only compare the percentage of a person’s income or wealth represented by a dollar to another person’s percentage.
I didn’t understand that at all, but an image of multiple Whoopie Goldbergs in lumberjack outfits was quite a shock to the system.
It’s not at all meaningless. The chart showing the effects on all producers and all consumers of a tariff on an imported good. In this case, Canadian lumber. It is basic economics, and shows the changes in quantity supplied and quantity demanded due to a tariff. You can assign dollar values if you wish to determine actual dollar amounts. Try it with a 20% tariff on imported Canadian lumber to see what I mean. Watch the videos for a better understanding.
The commonality is that all humans act in their own self interest, and in pursuit of that interest they trade with others who have a different hierarchy of subjective values, and different skills, which results in both parties being made better off. Any government interference in that trade, such as by imposing a tariff, results in a net loss of welfare to those in the imposing country.
Yes, those are all possible options, but imposing a tariff on American consumers is probably the worst option, while default may be the best option from a taxpayer viewpoint. Default would destroy the “Full faith and Credit” of the federal government, and pretty much eliminate any chance of future borrowing. Government size and spending would necessarily shrink as taxpayers would not tolerate a near doubling of their tax burden to support the current level of government spending.
Now as for increases in government spending. Thank’s to the magic of debt and central banking, this is no longer directly tied to revenue.
Yes, that is fascinating subject, but it moves us away from the current discussion of the tariff.
I watched them but no mention of subjective utility, nor an explanation of diminishing marginal utility and the fact that once you have that nonlinearity in the system you no longer have ANY aggregate measurements.
Try it yourself: make a ruler where every “inch” marking on the ruler is a fraction shorter than the previous. Now you have a nonlinear ruler measuring diminishing marginal inches. Use this ruler to design and build some furniture. Observe the results.
Subjective utility is even worse because those inch markings are all jumbled around.
Apparently it seems you can have economists who believe in certain ideas (like diminishing marginal utility), but then when inconvenient simply stop applying those rules (like when declaring consumer surplus). In other words, economics is more about telling a good story than about looking for a consistent way to understand what’s going on.
But you just did attempt to compare subjective value in those videos. The claim is that the area under the curves represents “surplus” which is to say it is claiming some type of aggregate utility.
No, the chart shows the aggregate sum of revenue as measured in TOTAL DOLLARS changing hands. It does not measure subjective effects because what unit do we have that represents total subjective utility? It cannot be dollars, as already covered above (diminishing marginal utility of dollars, observation of volunteers, etc).
Actually, the chart does not even measure commercial profits, because it ignores the cost side of things. It’s possible that due to various regulations either the Canadian or the US lumberjacks are more profitable. The tariff could possibly increase total commercial profits (I’m not saying it does; but nothing in the chart you linked to makes that impossible).
Here’s a tangible example of the absolute simplest transaction: Freddy has a ham sandwich, Franky has a jam sandwich.
They voluntarily decide to swap one sandwich for the other, based on individual preferences… how much total utility has been gained here? Show how you measure it.
Let’s go a step further: they decide to use money as an intermediary so Freddy sells his sandwich to Franky for $10 and then Franky sells his sandwich back to Freddy also for $10. How much total utility gain? How much extra utility if they decided to make it $50 instead?
Tel
I see you are struggling with the terms ‘consumer surplus’ and ‘producer surplus’.
You can acquaint yourself with those two concepts at the link below so you can read and understand the simple tariff chart. You may be asking the chart to explain too much.
https://www.khanacademy.org/economics-finance-domain/microeconomics/consumer-producer-surplus
No, there are no mentions of subjective values or diminishing marginal utility in the videos or in the chart because they deal with aggregate amounts, not individual values.
You are confusing individual subjective value with aggregate price changes due to a tariff, and you are misunderstanding the concepts consumer and producer surplus. There is no inconsistencies in a belief in diminishing marginal utility and in understanding the effects of a tariff.
Please watch the series of videos at the link I just provided for a better understanding of this subject.
Where is my edit key?
I once asked Bob for a preview (after all, Nick Rowe has one) but anyway I blame myself for the time I spend earning a living instead of writing better blogging software.
“But as ever, that requires a global efficiency metric… which does not exist when some people are better off (i.e. US lumber producers) and others are worse off (i.e. Canadian lumber producers).”
I see that your effort, here, is to hold Austrians accountable to Methodological Individualism, which is also a concern of mine.
It helps to say that it’s not, ultimately, countries that are trading with each other, but individuals within countries.
Any voluntary trade is seen as efficient to the parties of the transaction.
There is no need for a global metric, to your point.
Even when we’re talking about money, it’s still the case that the individual parties to the transaction believe that trading money for goods is efficient to them as individuals.
Bob
Just a minor nitpick. You are referring to Canadian softwood measured in ‘cords'”. A ‘cord” of wood usually refers to timber cut to length and split for use as firewood. I believe the Canadian wood in question is lumber cut to uniform dimensions such as 2×4, 1×6, 4×8, etc. for use in constructions and sold by the board-foot. The price is usually expressed in $ per 1000 board-feet.
Thanks Ron. But to be clear, is what I said just clunky, or is it actually false? Eg is it like expressing a country’s oil production in gallons per day (rather than the more standard barrels per day), or is it like expressing lumber output in Houses?
Bob
Well, it’s technically wrong. Although somewhat confusing, wood is measured in different units depending on it’s intended use.
Logs headed for the saw mill to be cut into dimensional lumber for construction are measured in estimated board feet, and after cutting, the lumber is measured in board feet.
Wood cut, split and stacked for firewood is measured in cords, one cord being a stack 8 x 4 x 4 ft. Wood destined to be pulped for use in other products may also be measured in cords for some reason.
For discussions of Canadian lumber and the abominable tariff, the commonly used unit of measure would of board feet, and prices expressed in dollars per 1000 board feet.
edit:
“…. the commonly used unit of measure would
ofbe board feetOK great thanks.