In his very nice series of election issues, Nominal News today addresses tariffs. He points out the well-known fact that economists generally oppose tariffs. He also has a discussion of the distribution of the costs of the tariff between consumers being led to a mix of goods of lower value and additional cost of producing import substitutes. Correct but not new.
NN does have a less conventional discussion of tariffs aimed at reducing the “leakage” from domestic policies that raise the price of domestic goods with emitted CO2 content. A fee on the emitted CO2 content of imports would indeed be the correct instrument to accompany the domestic taxation of net CO2 emissions. This is hardly a "tariff" as the word is usually understood, but that's nomenclature.
It's worth pointing out that the tariff/fee would be applied only to imports from countries that do not have the same tax on net CO2 emissions as the fee imposing country. And its application woud be as much to encourage others to join the net CO2 tax emissions club as to prevent "leakage" of demand for items with CO2 emission content.
Returning to “tariffs,” economists have been developing and refining arguments for and against import restrictions -- bans and quantitative restrictions are generally worse than tariffs -- for decades.
In general, only two arguments pass muster. One is the "optimal tariff" (which would apply as well, generally better, as an "optimal export tax -- a tariff and an export tax being equivalents macroeconomically). This works when imports/exports of item X by Country A are a large share of the world market for X and X is imported/exported into/from A by many perfectly competitive importers/exporters. IFF the Government of A knowing the supply and demand curves of X can impose a tariff that will make A an optimizing monopsonist of X imports/monopolist in the export of X, this will force down/up the international price of X and A will benefit.
The other is a “Pigou tax” on an externality created by domestic use of the import. Examples would be a risk of disruption of supply that no one importer has the incentive to optimally diversify away from or (similarly) import of a "strategic material" whose supply disruption would cause damage greater than the damage to the importers themselves. The correct response here is a specific tariff on the strategic or risky good from specific sources, but it depends on the government having correct knowledge of the disruption risk from each importer.
NB these "Pigou tariffs" are not the same as "industrial policy" that seeks to increase domestic production of item X for some positive externality reason, including that production of X is an “infant industry. THAT calls for a subsidy on production, not restriction of imports. In the first case the negative externality is located with the _domestic use_ of item X from each source(s). In the second case, the positive externality is located with the _production_ of X.
An aspect of import restriction that is generally not understood is that they "tax" exports and "subsidize" non-restricted imports by raising the value of the domestic currency relative to foreign currencies. Formally this is the “Lerner Equivalence” of tariffs and export taxes.
Image prompt: I’m just lazily re-using an old “trade war” image: dramatic is not particularly relevant to this post.
[Standard bleg: Although my style is know-it-all-ism, I do sometime entertain the thought that, here and there, I might be mistaken on some minor detail. I would welcome comments on these views.]