Adjusting to the Border Adjustment Tax: Imperfections and Unintended Consequences
American Enterprise Institute
Originally published in the Mercatus on Policy Series
The newly unified Republican government has raised hopes of fundamental tax reform, particularly corporate tax reform. The most concrete and politically relevant plan is the broad outline presented in the House Republican Tax Reform Task Force Blueprint.1 The plan would radically transform US corporate taxation by shifting from an origin-based corporate income tax to a destination-based cash flow tax (DBCFT). Under such a system, the location of consumers, not producers, determines whether activity is taxable. The plan would shift the tax base to being destination-based through a border adjustment that exempts exports from taxation and subjects imports to a new 20 percent corporate tax rate. Because the United States is projected to run a trade deficit for the foreseeable future, the border adjustment is projected to bring in over $1 trillion in net new tax revenue over the next decade.2 The House GOP's corporate tax reform plan uses that revenue to lower rates and otherwise reform the tax code while maintaining something approaching revenue neutrality. This policy brief discusses the transition to a border adjustment.
Because the border adjustment features a tax on imports and a subsidy on exports, it may look, at first blush, both mercantilist and protectionist. The constellation of special interests agitating for and against the policy proposal reflects this first impression. For example, the American Made Coalition, which is made up of export-heavy firms, supports the border adjustment, while Americans for Affordable Products, a coalition of firms and trade associations that rely heavily on imports, opposes it.3
The economic assumptions that apparently motivate the makeup of these coalitions do not reflect the standard economic view of the impact of a border adjustment—a view that is quite different from the aforementioned first impression. This standard economic view, perhaps best represented by Emmanuel Farhi, Gita Gopinath, and Oleg Itskhoki in the academic literature and articulated by Alan Auerbach, Alan Viard, and many others in the policy sphere, holds that the introduction of the border adjustment would lead to an immediate dollar exchange rate appreciation.4 This exchange rate adjustment, in turn, would compensate precisely for the import taxes and export subsidies that constitute the border adjustment tax (BAT), leaving relative prices and trade flows unchanged.
The easiest way to see the deep logic underlying this is as follows. A onetime, unexpected, permanent 25 percent import tax would, in effect, introduce a wedge that makes domestic prices 25 percent higher than foreign prices. A 20 percent export subsidy does the same thing.5 A combination of the two then simply produces a onetime upward shift in the domestic price level relative to the foreign price level. But if the underlying economic fundamentals do not change and exchange rates are flexible, there is no reason for consumption or production decisions to change. Instead, the dollar will appreciate by 25 percent, leaving prices everywhere unchanged in the local currency.
The sizable appreciation of the dollar that would occur according to the standard economic view would have unintended yet important consequences. I will distinguish two types of unintended consequences: those produced by the dollar appreciation's effects on asset prices and those produced by the dollar appreciation's effects on relative price levels in situations in which the border adjustment is not or cannot be administered.
A 25 percent appreciation of the dollar would impose significant wealth losses on US owners of foreign-currency-denominated assets. Although US foreign assets amount to only 140 percent of GDP, compared to 180 percent for foreign liabilities, these losses would not be fully compensated for by wealth gains from corresponding foreign-currency-denominated liabilities. As Farhi, Gopinath, and Itskhoki point out, 85 percent of US foreign liabilities are dollar-denominated, but only 30 percent of US foreign assets are.6 The net loss would amount to about $2.5 trillion, or almost $8,000 per American.7 This is particularly concerning when the US owners of the foreign-currency-denominated assets face liabilities that are almost exclusively dollar denominated, as is the case for pension funds. The flipside of this wealth loss to Americans is a gain to foreign owners of dollar-denominated assets. While certainly not an intended consequence of US tax policy, this gain is less disturbing than the increased burden on emerging-market governments and firms derived from the trillions of dollars worth of dollar-denominated debt they have issued.8
The shift in relative price levels caused by the border adjustment, according to the standard economic view, would have unintended consequences for settings in which the border adjustment is not administered. These include settings in which the import tax is evaded because the goods and services in question cannot legally be exchanged. For example, it would become more attractive to smuggle prohibited drugs into the US because their foreign-currency value would rise. The unintended consequences also affect settings in which the export subsidy is not applied because the foreign counterparty is physically in the United States. For example, the tourism industry would be adversely affected. It is also plausible that individuals would be exempted from the import tax, which in turn raises questions about what the ultimate tax treatment of pass-through entities and small businesses would be.
These unintended consequences all materialize even if the standard economic view of the introduction of the BAT is correct. That said, there is significant reason to believe that the standard economic view is not entirely correct. Instead of instant, complete adjustment of the exchange rate when the border adjustment takes effect, we may see an anticipatory appreciation of the dollar and incomplete adjustment after the fact. This claim comes with a strong caveat: we do not know nearly enough about the drivers of exchange rates, the coming legislative process, what the House Republican plan will look like beyond the current Blueprint, and how the plan would be implemented to predict the transition process precisely. But we do know certain things.
First, the border adjustment will not be completely unexpected. As a result, arbitrage in foreign exchange markets will lead (or may have already led!) to anticipatory appreciation of the dollar. Until the border adjustment takes effect, this makes imports cheaper to US consumers, while US exporters will find a less willing audience for their products. In other words, the trade deficit will increase before implementation, as imports are accelerated to escape the impending tax and exports are deferred to await the impending subsidy.
Second, it is unlikely that the border adjustment will be applied to all companies, industries, and transactions in a uniform and symmetric way. We saw earlier that some industries will effectively be exempted for legal or logistical reasons, but others may be exempted as a result of special-interest politics. To the extent that this happens, the onetime shift in the relative price level will be reduced, as will the corresponding exchange rate adjustment. The lack of cash refunds for exporters in the current plan would have the same effect.9 As a consequence, imports that are subject to tax will be more expensive than before the introduction of the border adjustment, while exports that receive the subsidy will be cheaper. Interestingly, this is precisely the scenario expected by the main lobbying coalitions that have formed so far. If investors expect the border adjustment to be phased out or eliminated in the future because of a World Trade Organization challenge or a shift in the political climate, a similar scenario would materialize in response to anticipatory depreciation of the dollar.
There is a series of other considerations that may lead to imperfect exchange rate adjustment. Active monetary policy focused on maintaining a nominal peg, or even just expectations thereof, could at least temporarily limit exchange rate adjustment. Other countries could respond to the introduction of border adjustment with protectionist measures, especially given some Republicans' decision to promote their corporate tax plan as mercantilist trade policy. The wealth effects induced by exchange rate adjustment will affect saving and investment decisions and thereby alter trade flows and exchange rates. Other elements of the plan, such as the reduction in the corporate tax rate, while technically separate, are facilitated by the border adjustment revenue and will also have real effects. The revenue may have other impacts on the government's financial needs. Teasing out the aggregate impact of these various considerations is quite difficult, and such policy uncertainty is harmful in and of itself.10
It is effectively impossible to avoid the unintended consequences of implementing a border adjustment. For example, gradually phasing in the border adjustment would exacerbate the unintended consequences of anticipatory appreciation. Instead, policymakers need to recognize these imperfections and compare the current system of corporate taxation to realistic alternatives, not to Platonic forms of alternative tax systems.
1. Office of the Speaker of the House, "Tax Reform Task Force Blueprint—A Better Way: Our Vision for a Confident America," June 24, 2016.
2. James R. Nunns et al., "An Analysis of the House GOP Plan" (UrbanBrookings Tax Policy Center, Washington, DC, September 16, 2016).
3. Richard Rubin, "Exporters Tout Benefits of Republicans' "Border Adjustment' Proposal," Wall Street Journal, February 2, 2017; Bernie Becker, "Everyone Gets a Coalition," Politico, February 2, 2017.
4. Emmanuel Farhi, Gita Gopinath, and Oleg Itskhoki, "Fiscal Devaluations," Review of Economic Studies 81, no. 2 (2014): 725– 60; Alan J. Auerbach, "Border Adjustment and the Dollar" (AEI Economic Perspectives, American Enterprise Institute, Washington, DC, February 2017); Alan Viard, "The Economic Effects of Border Adjustment," Tax Notes, Tax Analysts, February 20, 2017.
5. I am expressing tax rates in tax-exclusive terms, that is, as a percentage of the before-tax price. The border adjustment then applies a 25 percent tax rate to imports and a negative 20 percent tax rate to exports.
6. Emmanuel Farhi, Gita Gopinath, and Oleg Itskhoki, "Trump's Tax Plan and the Dollar," Project Syndicate, January 3, 2017.
7. Stan Veuger, "How Border Adjustment Reduces the Value of Your Scottish Golf Course" (AEIdeas, American Enterprise Institute, Washington, DC, January 10, 2017). Note that the dollar appreciation does not benefit post-appreciation buyers of foreign assets. While a stronger dollar makes foreign assets cheaper in dollar terms, the reduced dollar value of income from those foreign assets offsets this gain.
8. Robert N. McCauley, Patrick McGuire, and Vladyslav Sushko, "Global Dollar Credit: Links to US Monetary Policy and Leverage" (BIS Working Papers No. 483, Bank for International Settlements, Basel, Switzerland, January 2015).
9. Viard, "Economic Effects of Border Adjustment."
10. Jason J. Fichtner, Veronique de Rugy, and Adam N. Michel, "Border Adjustment Tax: What We Know (Not Much) and What We Don't (All the Rest)" (Policy Brief, Mercatus Center at George Mason University, Arlington, VA February 2017); Daniel Shoag and Stan Veuger, "Uncertainty and the Geography of the Great Recession" (Research Briefs in Economic Policy No. 69, Cato Institute, Washington, DC, February 2017).