Tax Analysts

Click here to see operators for terms and connectors searching.

Economic Analysis: Why Economists Believe the Blueprint Boosts the Dollar

POSTED ON Apr. 25, 2017

A lot of economics is glorified common sense. For example, when economist and Nobel laureate James Tobin was asked to explain his award-winning ideas, he replied, “You know, don’t put all your eggs in one basket.” Unfortunately international economics is one area of the dismal science where results counter to our intuition are common. In 1936 economist Abba Lerner wrote a landmark theoretical paper showing that a tax on imports was equivalent to a tax on exports (a result supported by empirical research). Based on results like Lerner’s, economists make assertions about trade and international finance that are a puzzlement not only to the general public but even many of the world’s most sophisticated business leaders.

Politicians and people who try to influence them are usually quick to cite and support economic research when it aligns with their objectives. But in the current debate about the effect of border tax adjustments contained in the House tax reform blueprint, despite unusually widespread agreement inside the profession, economists’ assertions are often written off as academic nonsense with little application to reality. The economists’ view is commonly stated as: exchange rate changes will offset any effects of border tax adjustments on trade. That is, the extra burden on imports from a tariff will be offset by the extra buying power of a stronger dollar, and the boost to exports from an export subsidy will be offset by the reduced buying power of weaker foreign currencies.

A far more accurate and complete statement of the profession’s viewpoint is: real exchange rate adjustments will offset the effect of border tax adjustments on trade if the rate of tariff on imports and the rate of subsidy on exports are uniform and equal. Real exchange rates depend on nominal exchange rates and on domestic and foreign price levels. In the United States, the Federal Reserve controls the price level. If the Fed does not allow the price level to increase, only then will market forces cause the dollar to appreciate in response to border tax adjustments. The markets’ adjustments (price level changes or nominal exchange rate adjustments or a combination of the two) are expected to completely offset the border tax adjustments only if export subsidies and tariffs apply at an equal rate to all U.S. goods and services. Border tax adjustments will deviate from the ideal because of intractable design and administrative issues and because of carve-outs for politically favored sectors.

There are many excellent recent papers discussing this topic. Some of them are listed at the end of this article. The most complete and authoritative references on the economics of border adjustments and trade are the outstanding series of articles by Alan Viard of the American Enterprise Institute. What follows is an attempt to summarize as simply as possible the key findings of all this work.

Export subsidies increase exports. Tariffs reduce imports. But a combination of the two does not change the balance of trade. Border tax adjustments are a combination of tariffs and export subsidies both imposed at an equal rate. Given that the balance of trade is the excess of exports over imports, it might seem that border tax adjustments could deliver a one-two punch that would improve the trade balance. And so when economists make the argument — the argument that is so central to the case for border tax adjustments, which in turn is so central to the House blueprint — that border tax adjustments do not affect the trade balance, the idea understandably creates a great deal of puzzlement.

One thing that is indisputable, though by no means obvious to the non-economist, is that in any one year the accounting identity must hold that exports minus imports equals net U.S. saving. In other words, if we sell more than we buy from the rest of the world, net U.S. assets increase. Although there are many aspects of tax reform that may affect U.S. saving and investment — like a net decrease in public saving from a larger federal deficit or more incentives for private saving — economists generally believe that border tax adjustments on their own do not affect net saving. Accordingly, it follows that the trade balance cannot change as a result of border adjustments. Admittedly, this explanation is not intuitive, and it provides little insight about prices (and ultimately about exchange rates), so let’s move to the next point and take a different approach.

Border tax adjustments affect prices between countries but not within countries. To drill down to the essential economics, let’s assume there is trade between two countries in a world with a single nameless currency. Suppose that before any border adjustments, Cadillacs are bought and sold in the United States for 100 and exported and sold in Germany for 100. And similarly, Audis are bought and sold in Germany for 100 and exported and sold in the United States for 100. Also assume zero shipping costs. If this is a free market outcome, this is an optimal overall state of affairs for producers and consumers in both countries.

Now suppose the United States imposes an import tariff of 20 percent. One thing that is indisputable is that as a result of the tariff, whatever amount Germany is receiving for the Audis, the U.S. customer is paying 20 percent more. (The government pockets the difference.) Naturally, the tariff would put upward pressure on U.S. prices. Similarly, with an export subsidy it is indisputable that the U.S. seller is receiving 20 percent more than the German consumer is paying. (The government shells out subsidy payments to the U.S. seller.) And clearly, the subsidy provided to exports from the United States would put downward pressure on German prices. Thus, both components of the U.S. border tax adjustment put upward pressure on U.S. prices relative to German prices.

How far will this relative price increase go? If the original state of affairs was optimal, market forces will adjust to restore the original optimal state if they can. But can they do so with this large artificial wedge of border adjustments inserted between buyers and sellers from different countries?

The answer is yes. They can restore the original optimal state by leaving relative prices within each country unchanged. (If there were relative price changes, the mix of production and consumption and the balance of trade would change for the worse.) And they can offset the export-stimulating, import-repressing effect of border adjustments by simultaneously allowing the relative prices between countries to adjust.

We can show this by returning to our example. Although the following statements would be true for any absolute price level, let’s suppose for mathematical simplicity that German prices — what German sellers receive and German consumers pay — remain at 100 after the imposition of border tax adjustments. Because of the tariff component of the border tax adjustment, German sellers receive 100 for Audis that must sell for 120 in the United States. Because of the export subsidy, U.S. sellers will receive 120 for Cadillacs that sell for 100 in Germany. As before, all cars in Germany sell for 100. What is new is that all cars in the United States sell for 120. Because production and consumption of automobiles comprise the entire U.S. economy in this example, all prices (including wages) will rise by 20 percent, and so nothing — that is, no quantities of goods and services consumed or produced — has changed except the nominal price level. The figure illustrates this outcome that sometimes can be lost in a jumble of words.


Real exchange rate appreciation equal to rate of border adjustment results in trade neutrality. In the previous example, there was only one currency. So the only adjustment possible to restore trade neutrality was though a change in the relative price levels between two countries. More generally, when each country has its own currency, trade neutrality can be achieved if real exchange rates adjust to offset the border tax adjustment.

The real exchange rate is the nominal exchange rate times the ratio of the domestic price level to the foreign price level. If the nominal exchange rate is one, as it was in the last example, trade neutrality can only be achieved through an adjustment in the price levels in each country. At the other end of the spectrum, if the price levels are frozen, trade neutrality can only be achieved through an appreciation of the exchange rate. Of course, there can also be a combination of the two.

Price level changes depend on Fed policy. It is a fundamental tenet of macroeconomics that the domestic price level is largely set through the Federal Reserve’s control of the money supply. Basically, if the Fed expands the money supply faster than the real growth of the economy, there is inflation. This means that if the Fed does not increase the money supply in response to imposition of a border-adjusted tax — in economics lingo, if the Fed does not accommodate the tax — there will be no increase in the domestic price level. In this case, attainment of trade neutrality must occur through appreciation of the dollar. The often unstated assumption by economists who argue nominal exchange rates will fully offset the effects of border tax adjustments is that price and wages are too “sticky” (especially in the short run) to offset large border adjustments or that the Fed will not accommodate upward price pressure from a border-adjusted cash flow tax.

Fed accommodation is more likely — and nominal exchange rate appreciation is less likely — with certain taxes like VATs and retail sales taxes. This statement requires a bit of economic reasoning that non-economists are unlikely to think of on their own. It goes like this: labor markets will clear when real wages — that is, wages divided by the price level equate supply and demand. In a world with no price adjustments, the imposition of retail sales tax or a VAT on business would lower profits. To restore profitability to a sustainable level and prevent unemployment, real wages must decline. This can occur either through a nominal wage decline or an increase in the price level. It is a time-proven economic phenomenon that nominal wages are sticky in the downward direction, especially in the short run. So we cannot depend on the nominal wage declines to prevent unemployment. Given that one of the Fed’s jobs is to fight unemployment, we can expect the Fed to raise prices in response to a retail sales tax or VAT imposed on employers.

Fed accommodation is less likely — and nominal exchange rate adjustment more likely — with border-adjusted cash flow taxes. The business cash flow tax in the House blueprint is equivalent to a subtraction method VAT plus a deduction for wages. Because of this deduction, the cost to business of this new tax does not require a reduction in the real wage rate to reduce labor costs. Therefore, the threat of unemployment posed by an unaccommodated retail sales tax or VAT is absent. The Fed does not have to increase the price level to save jobs. Therefore, it is reasonable to expect the Fed to remain committed to price stability and not allow a one-time increase in inflation in the presence of this border-adjusted tax. Without an increase in the domestic price level, it is up to nominal exchange rates to adjust through dollar appreciation in order to preserve trade neutrality. Many economists assume this scenario is most likely and that is why they emphasize that dollar appreciation is the most likely outcome of a U.S. border tax adjustment.

Imperfections in real-world border adjustments can deter real exchange rates from achieving trade neutrality. If border tax adjustments are not evenly applied, the net effect may be that of a growth-reducing national tariff or of a growth-reducing export subsidy. For example, if business cash flow taxes are not fully refundable, net exporting firms are likely to be in a permanent loss position and will be unable to receive the full export subsidies to which they should be entitled. Concerns about tax fraud and the bad politics of profitable corporations receiving refunds is the reason we are likely to never see full refunds for negative cash flow. In this case, the border tax adjustments are not uniform. Import duties are greater than export subsidies. And so, on net, the United States would be imposing a tariff on overall trade.

Conversely, as foreign businesses sell directly to domestic consumers, there may be no effective method of enforcing an import tariff, especially for sales over the internet of digital products. With this leak in the system of import tariffs, the net overall effect of a border adjustment system on trade will be an export subsidy on trade. In these cases, there will be real exchange rate adjustments, but they will be incomplete and produce uneven and inefficient effects internally in the economy as well as distortions in international trade. The magnitude and the importance of these imperfections are probably impossible to ascertain in advance. But they are likely to grow over time as international and electronic commerce expands and tax planning techniques are developed and refined.

Empirical research supports the theory that real exchange rate adjustments can neutralize trade effects of border adjustments, but the applicability of this research to the House blueprint is uncertain. In a recent empirical study by Caroline Freund and Joseph E. Gagnon of the Peterson Institute for International Economics, the authors found that changes in VAT rates (adjusted to take into account exemptions and other forms of preferential treatment) were generally matched by one-time changes in the price level with little or no change in the trade balance. Exactly as theory predicts, real exchange rate adjustments (mostly through price level adjustments) were found to offset border tax adjustments and VATs (unless they were replacing excise taxes on a similar magnitude). Because it takes time for prices and wages to adjust, these offsets could take two or three years to complete, though the bulk of the adjustment occurred up front.

However, because no country has ever implemented a border-adjusted cash flow tax with a deduction for wages as in the blueprint, there is no direct empirical evidence supporting (or denying) that a border-adjusted cash flow tax will maintain trade neutrality through exchange rate adjustment. In general, it seems reasonable to expect nominal exchange rates to adjust more easily than all of a nation’s domestic prices and wages. On the other hand, because of the unique role of the U.S. dollar in world markets, with some currencies pegged to the dollar and with some countries resistant to devaluation of their currencies relative to the dollar, the necessary appreciation to offset border adjustments could also take some time.


Most of the theoretical work and empirical evidence confirms the assertion that real exchange rates will adjust to offset the effects of border tax adjustments. But this research, especially the empirical research, is not sufficiently complete to provide a high degree of confidence that the nominal exchange rate will adjust promptly to a destination-based cash flow tax for a country as large as the United States and for a rate change as large as the House Republicans are proposing.

There is good reason to suspect adjustment could take years. There is good reason to believe the adjustment will be incomplete because of imperfections in the design and administration of the border tax adjustment and because of politically necessary relief that will certainly move final legislation away from uniform application to all imports and exports. It is likely, even if border tax adjustments do not directly affect trade, that they will be part of a large tax bill that will indirectly affect trade through that legislation’s effects on national saving and investment. It is difficult to predict how foreign governments will respond to U.S. border adjustments, to the associated tax reform, or to any change in the dollar and how those foreign policy changes could affect the trade balance. Finally, one important factor to keep in mind is that the Fed (or Congress indirectly through a mandate on the Fed) can promote or deter price level adjustments through changes in monetary policy.

Further Reading

Alan J. Auerbach and Douglas Holtz-Eakin, “The Role of Border Adjustments in International Taxation,” American Action Forum, Dec. 2, 2016.

Alan J. Auerbach, Michael P. Devereux, Michael Keen, and John Vella, “Destination-Based Cash Flow Taxation,” Oxford University Centre for Business Taxation Working Paper, January 2017. See especially pp.17-21.

Alan J. Auerbach, “Border Tax Adjustments and the Dollar,” American Enterprise Institute Economic Perspectives, Feb. 2017.

Martin Feldstein, “The House GOP’s Good Tax Trade-Off,” Wall Street Journal, Jan.5, 2017.

Caroline Freund and Joseph E. Gagnon, “Effects of Consumption Taxes on Real Exchange Rates and Trade Balances,” Peterson Institute of International Economics, April 2017.

Goldman Sachs, “What Would the Transition to Destination-Based Taxation Look Like?” Dec. 8, 2016.

Gary Clyde Hufbauer and Zhiyao (Lucy) Lu, “Border Tax Adjustments: Assessing Risks and Rewards,” Peterson Institute of International Economics, Jan. 2017.

Kyle Pomerleau, “Exchange Rates and Border Adjustments,” Tax Foundation, Dec. 15, 2016,

Alan D. Viard, “Border Adjustments Won’t Promote Competitiveness,” Tax Notes, Oct. 4, 2004, p. 122.

Alan D. Viard, “Border Tax Adjustments Won’t Stimulate Exports,” Tax Notes, Mar. 2, 2009, p. 1139.

Alan D. Viard, “Tax Increases and the Price Level,” Tax Notes, Jan. 6, 2014, p. 115.

Alan D. Viard, “The Economic Effects of Border Adjustments,” Tax Notes, Feb. 20, 2017, p. 1029.

Alan D. Viard, “The Brady-Ryan Plan: Potential and Pitfalls,” Tax Notes, Apr. 10, 2017, p. 249.