Alex Muresianu of the Tax Foundation has written a long paper comparing the Tax Cuts and Jobs Act (TCJA), the tax reform that Republicans passed in 2017, with the CHIPS Act and the so-called Inflation Reduction Act (IRA). He concludes that “the TCJA caused a substantial increase in investment, as we would expect. In contrast, the targeted subsidies of the IRA and CHIPS Act have not led to a broad increase in private investment outside of subsidized sectors.”
Of course, the TCJA has had more time to work, and evidence from the IRA and CHIPS is only starting to roll in, a limitation Muresianu acknowledges in his paper. But we would expect the TCJA to work better than the IRA or the CHIPS Act based on economic theory.
Muresianu explains that theory well. He contrasts what he calls “supply-side orthodoxy” with industrial policy. The former is based on policy reforms that are sector-neutral. The TCJA, for example, cut the corporate tax rate from 35 percent to 21 percent for all corporations. It allowed 100 percent bonus depreciation for investment in machinery and equipment in all industries.
Industrial policy, on the other hand, selects certain industries believed to be important, either for foreign-policy or domestic-economic reasons, and targets reforms to them. These would include policies such as subsidies for semiconductor firms in the CHIPS Act or for green-energy firms in the IRA.
It’s important to draw this distinction because otherwise it’s easy to get into weird arguments where everything government does is an industrial policy. Muresianu gives a facetious example:
For instance, in the course of the government’s operations, it needs office supplies, such as staplers. It purchases staplers to facilitate the government’s provision of a variety of public services. By purchasing staplers, it redirects resources to stapler production, thus transforming the structure of economic activity (towards staplers). Thus, government procurement is a stapler industrial policy.
Everything government does affects some industries differently than it might affect others. Everything isn’t infrastructure, and everything isn’t industrial policy.
Supply-side orthodoxy and industrial policy often have a similar goal: to increase capital investment. But they go about pursuing that goal in very different ways. Muresianu writes:
The orthodox view is focused on the price of investment relative to consumption. Under that view, reducing marginal tax rates on investment will increase capital investment, and capital investment will spur greater productivity growth and economic growth. This view leaves the allocation of investment across sectors up to markets as much as possible, trusting markets to know better than policymakers where the most productive investments are.
The industrial policy view is more focused on the relative price of investment in different industries. The core premise in this case is that markets are not efficient at optimally allocating capital across sectors to maximize economic growth. Instead, government policy should push investment towards certain sectors with greater growth potential.
The supply-side view sees a tax code biased against investment relative to consumption and seeks to make it more neutral by reducing the tax burden on investment. The industrial-policy view sees certain industries struggling and seeks to benefit them with intentionally non-neutral policy changes.
In practice both approaches are tricky, since there are already plenty of non-neutral aspects of the tax code, as Muresianu writes. Also, “Tax changes that improve the treatment of investment as a whole benefit some industries more than others because some industries are more investment-intensive, and/or the status quo treats them unfairly.”
But these two approaches seek to increase investment through very different mechanisms. Muresianu explains:
When considering the orthodox view, lowering the cost of capital across the board makes countless projects across the economy that were just barely not worth pursuing on a risk-adjusted basis viable, changing investment decisions on the margin. Conversely, in the narrow sector- or industry-specific investment tax credit case, the policy makes many projects in that subset of the economy viable, even ones with expected negative returns without the tax subsidy.
Now, perhaps those projects with negative expected (private) returns have an untapped social benefit or potential. But if not, then the sector-specific policy has dragged investment and resources away from more productive sectors to the subsidized sector. The subsidized sector could still grow, but its growth is coming at the expense of foregone greater growth elsewhere in the economy. Another possibility is the reason the expected returns of projects in the subsidized sector are negative to begin with is they face some kind of hard constraint beyond just the capital cost, such as regulatory hurdles or a mismatch with the existing workforce, that make investment still infeasible.
The mechanism by which supply-side reforms translate into economic growth is straightforward: a lower cost of capital increases capital investment, which makes labor more productive and increases economic output. It’s up to private capital markets to figure out where investment is most needed, so the tax code should seek neutrality with respect to sectors.
The mechanism by which industrial policy translates into economic growth essentially requires a series of bank shots. The government has to pick the right sector at the outset of the policy that will spill over to the broader economy in the future. It has to hope that the spillover will be large enough to overwhelm the initial loss of pulling capital and labor out of other sectors and pushing it into the chosen sector. It has to actually move the capital around effectively, something government is not well equipped to do. It has to do all of this in the absence of reliable price signals and persist through relentless pressure from advocacy groups that seek to attach their own often contradictory goals to the policy.
The evidence so far points in the same way as the theory. A growing body of economic literature supports the claim that the TCJA increased investment in the U.S. economy. Muresianu does a crude counterfactual analysis in his paper, comparing pre-TCJA 2017 CBO projections for investment to what actually happened after the law passed. He found a 4.34 percent annualized growth rate in nonresidential private fixed investment in the two years immediately after passage of the law, compared to the 2.38 percent projected growth.
Compare that with the CHIPS Act, which is sending billions in subsidies to already profitable companies and saddling projects with numerous progressive policy goals that have little to do with semiconductor production. Or the IRA, the cost of which has exploded well beyond initial estimates because of subsequent changes to the EPA’s vehicle policies and other energy tax credits. And the rollout has been slow, so in many cases the series of bank shots required for the policies to work have not even been attempted yet, nearly two years after the bills became law.
Of course, it’s impossible to fully disaggregate the effects of policy changes from other factors affecting investment. But at the very least, if the government makes capital investment cheaper by reducing taxes on capital investment, it is very obvious why that would increase capital investment. If the government targets certain industries with policy to increase their capital investment at the expense of other industries’ investment in the hopes that it will net out to a positive for everyone, it is less obvious that will increase capital investment. And given what we know about how government will likely choose which industries to help — i.e., ones with political connections, rather than ones that would actually benefit the public — sticking to sector-neutral tax reform seems like the better choice.