Werner's Blog — Opinion, Analysis, Commentary
Will tariffs on critical mineral imports help or hinder the energy transition?

A report in The Globe and Mail, on March 7, 2024 suggested that Nickel from China, Indonesia could face tariffs over market manipulation concerns. Federal Natural Resources Minister Jonathan Wilkinson was quoted saying that Canada and other Western countries should consider imposing tariffs against these two countries because of their stranglehold on the global nickel market.

In general, import tariffs do not make us better off—they simply make products more expensive while protecting (often less-efficient) domestic industries. Can the proposed tariffs be justified or is this just another case of protectionism? Can tariffs be effective instruments to address global production concentration?

There is in fact economic theory that points to potential domestic welfare gains from import tariffs under special circumstances: when the exporter has a monopoly or market dominance. In such cases there is the notion of an "optimum tariff"—a topic pioneered by Canadian economist Harry G. Johnson in 1954.

For the benefit of our students of economics and business, let me write down a simple model of domestic demand, monopolistic foreign supply, and domestic welfare. Let us assume that domestic demand is linear $p= a - b \cdot q$ where $$p$$ is the price, $$q$$ is quantity demanded, $$a$$ is the price intercept and $$b$$ is the slope of the demand curve. The foreign producer faces marginal costs $$c$$ and maximizes profits $$\pi$$ but is subject to an import tariff $$\tau$$ that is applied in excise form per unit of product rather than as an ad-valorem (percentage) tariff: $\pi = (p-\tau)\cdot q -c\cdot q$ The producer's profit maximum is then given by $q^\ast= \frac{a-c-\tau}{2b}$ along with market equilibrium price $p^\ast=\frac{a+c+\tau}{2}$ Domestic welfare is composed only of consumer surplus and tariff revenue, as producer surplus only accrues abroad. Thus the home country's government wants to maximize welfare $W = (a - p)\cdot q/2 + \tau\cdot q + \phi\cdot q$ which is composed of the usual consumer surplus triangle (the first summand) and the tariff revenue rectangle (the second summand). However, there is twist here because the intention is to use critical minerals to build electric vehicles that reduce carbon emissions. So every unit of imported minerals will generate an environmental benefit $$\phi$$ (the third summand). The existence of this positive externality from imports can change the outcome significantly.

Knowing how the producer reacts, total welfare becomes the following expression: $W^\circ =\left(\frac{a-c-\tau}{2b}\right) \cdot\left(\frac{a-c+3\tau+4\phi}{4}\right)$ We can now find the optimal tariff $$\tau^\ast$$ by differentiating $$W^\circ$$ with respect to $$\tau$$ to find the first-order condition for a welfare maximum. This yields the optimum tariff $\tau^\ast=\frac{a-c-2\phi}{3}$ and thus the optimal welfare $W^\ast=\frac{(a-c+\phi)^2}{6b}$ What does the result tell us about public policy? The tariff reduces the imported quantity, but the producer is forced to share some of the monopoly rent with the importing country. When an importing country such as Canada faces a foreign export monopoly, it may be beneficial to impose an import tariff, roughly a third of the gap between maximum price $$a$$ and marginal cost $$c$$. In percentage terms, this can be quite substantial. Note that the optimum tariff does not depend on the slope of the demand when demand is linear.

But is the optimum tariff always positive? It depends critically on how large the environmental benefit $$\phi$$ is. The condition for $$\tau^\ast$$ to be positive is that $\phi<(a-c)/2 \quad\Longrightarrow\quad \tau^\ast>0$ If the environmental benefit from the imported good is large, it could reduce and even eliminate the benefit from the import tariff. The size of $$\phi$$ varies because of potential substitution channels. So how large could $$\phi$$ be? Let us try a back-of-the-envelope calculation in order to establish an upper bound and a lower bound.

Let's look at the upper bound first. We need about 0.7 kilograms of nickel for each kilowatthour of capacity in a Lithium-NMC battery, so about 50 kilograms per EV battery. Over its lifetime of operation, replacing a gasoline-powered vehicle with an EV reduces emissions by about 30 tonnes of carbon dioxide (Buberger et al, 2022). Evaluated at a social cost of carbon of CAD 250/tonne, one kilogram of nickel not available for manufacturing a BEV could preclude realizing environmental benefits of CAD 150 per kilogram of nickel. This is much more than the kilogram-price of nickel, which is about CAD 25 per kilogram. With linear demand we can work backward from the equilibrium price to figure out $$a$$ if we know $$c$$, i.e., $$a=2p^\ast-c$$. If the marginal cost is around CAD 13/tonne, then our $$a$$ parameter is about CAD 37, and the cut-off for the optimum tariff is an environmental benefit smaller than CAD 12/kg. Considering this maximum value of $$\phi$$, an import tariff may not be the right solution to the problem of dealing with near-monopolistic critical minerals supply.

The lower bound of the environmental benefit $$\phi^\ast$$ is zero, in which case the reduced amount of nickel imports will not affect the energy transition. There are several reasons that could justify this assumption. First, if only imported nickel was used for EV batteries, manufacturers could make EV batteries a bit smaller, or perhaps manufacturers might produce more plug-in hybrid-electric vehicles instead of battery-electric vehicles, and still realize the bulk of the environmental benefits. Second, there are domestic manufacturers of nickel that would benefit from the protection effect of the tariff. They would increase production (albeit at a higher price). Third, there are other global producers who also benefit indirectly from the import tariffs on their competitors, and they might also increase exports to Canada. Both mechanisms would provide relief and lower the parameter $$\phi^\ast$$. So when $$\phi^\ast$$ is zero or near-zero, we could indeed utilize an import tariff to extract rent from oligopolistic foreign suppliers of critical minerals.

It is fair to assume that $$\phi^\ast$$ is at least somewhat positive, and I presume that one can estimate the effect empirically. What we can say from the above calculations is that the environmental benefit needs to be below about CAD 12/kg for nickel in order to make the case for an import tariff. And this number would obviously need to be adjusted for the actual market structure: oligopolistic rather than monopolistic supply.

According to the International Energy Agency, Indonesia accounted for 49% of the world nickel supply in 2022, and 43% of the processing. Indonesia has indeed reached a dominant market position.

There are several caveats. The most important, already noted by Harry Johnson, is that the exporter country may also retaliate against Canadian exports, and this may ultimately hurt the economy. Trade policy is rarely straight-forward; it's a strategic game.

The avenues through which import tariffs can be used are also limited by international trade law. Unless they are applied across all countries equally, import tariffs on individual countries must be justified either by identifying price-dumping by individual exporter firms, by identifying government subsidies that boost foreign exporters, or by sudden import surges.

So what is to be done to lessen dependence on oligopolistic foreign critical minerals supply, while protecting overall supply to advance the energy transition? This vexing dilemma feels like squaring the circle. Broad tariffs may have limited use because they can slow down the energy transition, but targeted tariffs (compatible with international trade law) may help address market distortions.

There are two broad answers to the dilemma. First: diversification. This means building up new suppliers that dampen market dominance and promote more competition. In turn, this means investment domestically or with partners abroad. Second: innovation. If a particular material's supply is too concentrated, can we substitute to an alternative?

Lithium-ion batteries that rely on nickel-manganese-cobalt (NMC) chemistry are facing competition from lithium-iron-phosphate (LFP) alternatives, which use cheaper and more abundant materials, although this chemistry comes with slightly reduced performance. My sense is that it is likely more effective to promote substitution through innovation than to meddle in international trade relations.

This said, where there are clear cases of foreign governments subsidizing local production to secure market dominance, such market distortions call for countervailing duties. The methods for that are well established: Canada's Special Import Measures Act and an investigation by the Canadian International Trade Tribunal, which can make a determination that the foreign subsidy injures Canadian producers.

The high concentration of critical minerals supply in a limited number of foreign countries also poses vexing geopolitical questions in addition to the aforementioned economic concerns. Addressing both will require patience and perseverance, and innovation and investment. I do not see any quick fixes.