This is the conventional reply of trade theory to the "infant industry" argument. Subsidies a) do not distort consumer decisions between imports and import substitutes and between protected and nonprotected goods b) do not distort producer decisions between domestic and foreign markets. They do "require" a low deadweight loss source of revenue, like a progressive consumption tax.
Interesting, how critical is the low deadweight loss tax? Is it sufficient for government revenues to come from slightly-lower-DWL sources or do they need to be much better than the DWL from tariffs?
I would assume current government taxes have lower DWL than tariffs, but I'm not familiar with the literature.
Sam, I have written extensively about this. Tariffs are a poor tool if the goal is economic growth. However, we may be able to justify them in rare circumstances.
In my view, we should design a super simple tax system that minimizes economic impact. Here, LVT and VAT come to mind, with the latter being low and broad and the former being high.
Some of this revenue can go to subsidies, but rather than subsidizing specific goods or services, we should subsidize ideas. Fund particle colliders, space telescopes, new material science, fusion energy….whatever it is! The more ideas we produce, the better.
DBCF tax is basically a VAT not applied to some kinds of firms. No redeeming social benefit compared to a VAT. It's possible it would be better than a corporate income tax that raised the same revenue, but that is a pretty low bar.
You're in for a treat... I have had an essay on DBCFT written, but unpublished, for two years now (yes...really). I never felt very confident in my understanding, hence why I never published. After doing a lot more research, I completely revised it a couple of weeks ago, merging the discussion of DBCFT with other consumption taxes. I generally have come to the conclusion that while DBCFT is better than the current CIT, may as well just go all the way to VAT.
I hope. The concept is still rather amorphous. It weird how a DBCFT is considered a "consumption tax." When it functions a lot like the corporate income tax. In that sense, VAT is much more straightforward.
You should not be so afraid of being wrong! :) [Although I confess to being in something of the same place [ut weeks, not years :)] with an explainer of New Keynesian economic models.:
This post seems to rely on an unstated and questionable assumption: that the deadweight loss and other negative distortions of the foregone tariffs will exceed the opportunity cost of the revenue needed to provide the subsidy. One can gesture in the direction of low-deadweight-loss taxes (as you do), but why are we assuming that such taxes are available? Rationally they will already be levied and the revenues used for optimal purposes, since the world is not short on market failures. So the question is whether using revenues to replace tariffs is a better use than something else, which is not obvious: it requires some idea about what the marginal costs and benefits of government spending and taxation are. For this reason, in the real world we live in, it is not at all clear that raising taxes or cutting spending elsewhere in order to subsidize industries is actually more efficient than a tariff.
Yeah, perhaps I should make this more clear: I think the deadweight loss of the average tax dollar is far lower than the deadweight loss of the average tariff dollar. I feel pretty confident this is the case, though I haven't looked at the literature much. The typical tax dollar in the US comes from income tax, and as I understand it, income taxes are far better than tariffs.
As you point out, tariffs generate government revenue while subsidies reduce it. In a world where governments balanced their budgets, switching from a tariff to a subsidy would mean sacrificing some other government program. Assuming that these government programs produce more value than they cost, that would mean subsidies have an opportunity cost.
But if we're assuming there's a government program that passes a cost-benefit test, we should fund it with the lowest-DWL taxes we can find. So if switching to subsidies caused a program to lose funding, we should fund it by raising other taxes, not by bringing back tariffs.
(You mentioning fiscal opportunity cost anticipates where I'm going in future posts btw, one where congress explicitly allocates budgets produced with low DWL taxes to different projects such as R&D, onshoring industries, welfare, etc.)
This is the conventional reply of trade theory to the "infant industry" argument. Subsidies a) do not distort consumer decisions between imports and import substitutes and between protected and nonprotected goods b) do not distort producer decisions between domestic and foreign markets. They do "require" a low deadweight loss source of revenue, like a progressive consumption tax.
Interesting, how critical is the low deadweight loss tax? Is it sufficient for government revenues to come from slightly-lower-DWL sources or do they need to be much better than the DWL from tariffs?
I would assume current government taxes have lower DWL than tariffs, but I'm not familiar with the literature.
Sam, I have written extensively about this. Tariffs are a poor tool if the goal is economic growth. However, we may be able to justify them in rare circumstances.
In my view, we should design a super simple tax system that minimizes economic impact. Here, LVT and VAT come to mind, with the latter being low and broad and the former being high.
Some of this revenue can go to subsidies, but rather than subsidizing specific goods or services, we should subsidize ideas. Fund particle colliders, space telescopes, new material science, fusion energy….whatever it is! The more ideas we produce, the better.
Totally agree. One of the next posts in this (loose) series on policy will be about tax systems!
One tax that people also seem to like is a destination-based cash flow tax, but I haven't really figured out how it works.
https://en.wikipedia.org/wiki/Destination-based_cash_flow_tax
DBCF tax is basically a VAT not applied to some kinds of firms. No redeeming social benefit compared to a VAT. It's possible it would be better than a corporate income tax that raised the same revenue, but that is a pretty low bar.
You're in for a treat... I have had an essay on DBCFT written, but unpublished, for two years now (yes...really). I never felt very confident in my understanding, hence why I never published. After doing a lot more research, I completely revised it a couple of weeks ago, merging the discussion of DBCFT with other consumption taxes. I generally have come to the conclusion that while DBCFT is better than the current CIT, may as well just go all the way to VAT.
What luck! Reading your post is going to bring me up to speed much faster than I could have done myself.
I hope. The concept is still rather amorphous. It weird how a DBCFT is considered a "consumption tax." When it functions a lot like the corporate income tax. In that sense, VAT is much more straightforward.
You should not be so afraid of being wrong! :) [Although I confess to being in something of the same place [ut weeks, not years :)] with an explainer of New Keynesian economic models.:
I look forward to it
Channeling the essence of Hanson in all the best ways. I found this educational despite basically having grokked this pattern many times before.
but... excuse me what:
> "Governor DeSantis recently banned cultured meat in the state of Florida."
I'll come back with cleverer things to say once I'm done crying myself to sleep. 😢
...
Don't lose hope!
This post seems to rely on an unstated and questionable assumption: that the deadweight loss and other negative distortions of the foregone tariffs will exceed the opportunity cost of the revenue needed to provide the subsidy. One can gesture in the direction of low-deadweight-loss taxes (as you do), but why are we assuming that such taxes are available? Rationally they will already be levied and the revenues used for optimal purposes, since the world is not short on market failures. So the question is whether using revenues to replace tariffs is a better use than something else, which is not obvious: it requires some idea about what the marginal costs and benefits of government spending and taxation are. For this reason, in the real world we live in, it is not at all clear that raising taxes or cutting spending elsewhere in order to subsidize industries is actually more efficient than a tariff.
Yeah, perhaps I should make this more clear: I think the deadweight loss of the average tax dollar is far lower than the deadweight loss of the average tariff dollar. I feel pretty confident this is the case, though I haven't looked at the literature much. The typical tax dollar in the US comes from income tax, and as I understand it, income taxes are far better than tariffs.
As you point out, tariffs generate government revenue while subsidies reduce it. In a world where governments balanced their budgets, switching from a tariff to a subsidy would mean sacrificing some other government program. Assuming that these government programs produce more value than they cost, that would mean subsidies have an opportunity cost.
But if we're assuming there's a government program that passes a cost-benefit test, we should fund it with the lowest-DWL taxes we can find. So if switching to subsidies caused a program to lose funding, we should fund it by raising other taxes, not by bringing back tariffs.
(You mentioning fiscal opportunity cost anticipates where I'm going in future posts btw, one where congress explicitly allocates budgets produced with low DWL taxes to different projects such as R&D, onshoring industries, welfare, etc.)
The key question, as you point out, is whether there are low-DWL taxes that produce enough revenue. I think so! I'm particularly fond of land value taxes (https://splittinginfinity.substack.com/p/some-thoughts-on-using-auctions-for) and Pigouvian taxes (https://www.col-ex.org/posts/pigouvian-compendium/) but VAT and income tax also seem fine to me.