Submitted electronically to the Department of Commerce on June 27, 2019
I am pleased to enter my comments in support of the proposed modification of CVD regulations to allow Commerce to treat currency undervaluation as a countervailable subsidy subject to CVD regulations, subject to the comments and issues noted below.1
The Economic Policy Institute (EPI) is a nonprofit, nonpartisan think tank located in Washington, D.C. I am a Senior Economist with EPI, and am the Director of Trade and Manufacturing Policy Research. There are a number of issues that arise with respect to the measurement of equilibrium currency values and how they should be assessed that I wish to note, based on my review of the draft regulations.
Currency manipulation and currency misalignment are distinct, but closely related, problems
Currency manipulation consists of government purchases and sales of foreign exchange reserves that are designed to persistently depress the value of the domestic currency for commercial advantage. Currency manipulation by about 20 countries—including China, Japan, and a number of other countries in Asia and Europe—resulted in large, persistent trade and current account surpluses for those countries, especially during the “Decade of Manipulation” from 2003–2013.2 Currency manipulation has distorted global trade and capital flows, to a greater or lesser degree, for much of the past two decades.
Since 2014, the central governments of China and other currency manipulators have begun to reduce their purchases of United States and other foreign exchange reserve securities. But the dollar is still rising, thanks to overseas investors now pouring huge amounts of private capital into bonds, stocks, and controlling interests in American companies, as well as real estate, bank loans, bank deposits, and currency.3 It is important to note that some of these recent, “private” purchases come from quasi-state actors or state-influenced institutions (such as social security trust funds in Japan and Korea, and private insurance funds in Taiwan).
These private purchases have driven up the real value of the dollar by more than 20 percent since mid-2014, continuing the trend of making American-made goods less competitive globally. As a result, the Congressional Budget Office predicts that America’s trade deficit is on track to exceed $700 billion by 2021, a full 3 percent of gross domestic product.4
Thus, the dollar is currently “overvalued” relative to the trade-balancing, equilibrium exchange rate. I assume here that the equilibrium exchange rate is the set of exchange rates that would produce balanced trade for the United States and some or all of its major trading partners.
The trade-balancing, equilibrium exchange rate concept is completely distinct from that of “currency manipulation,” as defined by Treasury, the IMF, and other agencies. Currency manipulation refers to government intervention in foreign exchange markets for the purpose of gaining commercial advantage. The CVD regulations as published, and under discussion here, refer only to the concept of a “subsidy in the form of currency undervaluation,” without direct reference to currency manipulation as a cause of that undervaluation. Thus, my remarks will focus only on the concept and estimation of equilibrium exchange rates.
Currency undervaluation can result from currency manipulation, from excess private capital inflows, or both. These comments are concerned only with addressing the consequences of undervaluation.
The definition and determination of equilibrium exchange rates should be reconsidered
The draft regulations refer to various definitions of the equilibrium “real effective exchange rate” (REER), including estimates from the IMF and from the Peterson Institute. The particular Peterson paper—referred to in footnote 28 of the Commerce Department notice of proceeding5—is a 2017 estimate of “fundamental equilibrium exchange rates” by William R. Cline.6 In these comments, I will distinguish between the particular results that Cline obtained in his 2017 paper using the model that he developed, in a series of related papers, with his Peterson Institute colleague John Williamson.7 That model (hereinafter referred to as the “Cline model”) has been used to estimate equilibrium exchange rates under a number of different assumptions about data inputs and parameter values.
None of the methods cited in the draft regulations will generate reliable estimates of trade-balancing exchange rates. The Cline estimate cited in the draft regulations, in particular, allows for current account deficits and surpluses equal to plus or minus 3 percent of GDP, which are far from trade-balancing exchange rates.
The Cline model has been used by Bergsten to estimate a set of true trade-balancing exchange rates, based on generating zero current account balances for Japan, China, and the European Union, which are responsible for the vast majority of global current account imbalances at the present time.8 These three countries/unions are responsible for the development of large and persistent global trade imbalances over the past 10 to 20 years.9
For purposes of illustration, Bergsten estimates (using 2016 data, and projected to 2019) that the euro, the Chinese yuan, and the Japanese yen must rise by 40.9 percent, 37.1 percent, and 50.2 percent, respectively, in order to eliminate the persistent current account surpluses of these countries. In this scenario, the real dollar would depreciate by 26.5 percent, while the real (price-adjusted, trade-weighted) value of the euro, yuan, and yen would rise by 15.0 percent, 9.9 percent, and 22.6 percent, respectively. The U.S. current account deficit would be eliminated as well, as a product of these realignments.10
Data quality issues must be considered in the estimation of equilibrium exchange rates
The results of the Bergsten model illustrate the fact that estimates of trade-balancing equilibrium exchange rates (which should be used to determine the extent to which a currency is “undervalued” in the proposed CVD proceedings) depend critically on model parameter values used in the Cline model, or any other, alternative model used to estimate equilibrium exchange rates.11
The quality of data used to estimate equilibrium exchange rates is also critical to the determination of the degree of undervaluation of any particular bilateral exchange rate. The Cline and Bergsten models are both based on estimates of each country’s total current account balance as a share of GDP. The current account is the broadest measure of the total flows of trade in goods, services, and income. Having accurate estimates of each country’s trade and income flows is thus essential to the development of reliable estimates of equilibrium exchange rates.
There are substantial problems with the quality of data used to compute equilibrium exchange rates. Such calculations typically depend on estimates of current account flows as reported by the International Monetary Fund.12 The IMF typically relies on self-reported trade data in its reports. There are substantial and well-known problems with self-reported trade data. For example, recent research by the Federal Reserve has shown that China is overstating its tourism trade deficit by roughly 1 percent of GDP (roughly $145 billion in 2018).13 Thus, current Chinese self-reported trade data understate that country’s current account surplus by at least that amount. Use of underestimated parameter values for China’s current account balance will lead to underestimates of the equilibrium dollar–yuan exchange rates needed to achieve current account realignment.
Likewise, it is well known that U.S. estimates of the bilateral U.S.–China goods trade deficit are substantially larger than those reported by China. Since China’s own self-reported current account estimates are based on China’s own self-reported goods trade flows, the goods trade account is also likewise underestimated. Further research is needed on this issue.
Who should be responsible for estimating equilibrium exchange rates?
These questions lead naturally to the issue of which agency or agencies should be responsible for estimating the equilibrium exchange rates that are used to determine the “benefit resulting from a subsidy in the form of currency undervaluation” in the proposed modifications to CVD regulations. The proposed regulations state that Commerce “intend[s] to seek and defer to the Department of Treasury’s (Treasury’s) evaluation and conclusion as to whether government action on the exchange rate has resulted in currency undervaluation.”14
It would be a mistake to assign the job of determining trade-balancing equilibrium exchange rates to Treasury. Although that agency has legislative responsibility for producing the semi-annual reports on the macroeconomic and foreign exchange policies of major trading partners of the United States,15 it has failed miserably in carrying out this responsibility for the past two decades.16 In particular, during the “Decade of Manipulation,”17 China and about 20 other countries engaged in massive, unprecedented currency manipulation that decimated U.S. manufacturing, resulting in the loss of roughly 5 million manufacturing jobs and 90,000 manufacturing plants.18 Yet Treasury did not once find any of these countries guilty of currency manipulation. This despite widespread agreement among professional economists that China and these other countries were acting in clear violation of IMF and WTO norms regarding prohibitions against currency actions taken “to gain unfair competitive advantage over other members.”19
In plain English, there are clear reasons why Treasury has refused to call out currency manipulators or intervene in currency markets. Simply put, Wall Street loves a strong dollar, and Treasury has historically been run by and for U.S. financial markets.20 The strong dollar has made imports artificially cheap (and U.S. exports less competitive on world markets), resulting in large and growing trade deficits. The resulting flood of cheap imports has driven huge profits for the companies that sell them, especially firms like Walmart, Amazon, Nike, and Apple. It has also led to reductions in wages for 80 million American workers competing with millions of workers in countries where labor has been cheapened by these undervalued currencies.21 All of this has filled the coffers of multinationals who outsource production, as well as the pockets of financiers on Wall Street who brokered these investments and the hedge fund managers who presided over the dismemberment of domestic manufacturing firms. It is no mistake that the current Secretary of the Treasury came from Wall Street, just as it is no mistake that most of his predecessors over the past three decades also came from Wall Street—and returned there when their time in government “service” was done.
The Federal Reserve should be charged with estimating equilibrium exchange rates for the purposes of determining the effects of currency undervaluation in CVD proceedings
As these comments have illustrated, there are serious technical issues involved both in modeling equilibrium exchange rates and in identifying the most accurate trade data and underlying parameter values. In addition, there are serious conflicts of interest that would arise if Treasury were tasked with making these assessments, for reasons outlined above. Furthermore, the Federal Reserve Board of Governors retains perhaps the most extensive team of economists and statisticians familiar with international data and modeling issues: These economists are intimately familiar with the functioning of international trade and financial markets, and they are best qualified to perform the objective analysis called for by these regulations.
I am well aware of the historic role played by Treasury as the lead agency for the U.S. government for international financial issues. However, those are policy issues largely having to do with the United States’ relationships with the International Monetary Fund, the World Bank, and other international financial institutions. The question at issue with respect to the determination of equilibrium exchange rates is a technical matter better suited to the expertise of the professional staff of the Federal Reserve.
Furthermore, historic roles and tradition must be subject to periodic evaluation and reconsideration as goals and policy environments change. For the past 46 years, since the end of the gold standard and the Bretton Woods Agreement in 1973, the United States has both adhered to and advocated for a system of market-determined, flexible exchange rates. This slavish adherence to financial market forces, come what may, has come at tremendous cost to communities across the United States. It is time to consider new ways to manage exchange rates, and these new approaches to managing currencies may require new institutional arrangements as well.
The Commerce Department’s proposed regulations to consider currency undervaluation as a part of countervailing duty determinations are an important step in this direction, and they should proceed in implementing these regulations, after due consideration of the issues raised here.
1. Modification of Regulations Regarding Benefit and Specificity in Countervailing Duty Proceedings, 84 Fed. Reg. 24406 (May 28, 2019).
2. C. Fred Bergsten and Joseph. E. Gagnon, Currency Conflict and Trade Policy: A New Strategy for the United States. (Washington, D.C.: Peterson Institute for International Economics, June 2017). See especially chapter 4, “The ‘Decade of Manipulation’ (2003–13).”
5. Modification of Regulations Regarding Benefit and Specificity in Countervailing Duty Proceedings, 84 Fed. Reg. 24406 (May 28, 2019).
6. William R. Cline, “Estimates of Fundamental Equilibrium Exchange Rates, November 2017,” Peterson Institute for International Economics, November 2017.
7. William R. Cline and John Williamson, “New Estimates of Fundamental Equilibrium Exchange Rates,” Peterson Institute for International Economics, July 2008.
8. C. Fred Bergsten, “Time for a Plaza II?,” in International Monetary Cooperation: Lessons from the Plaza Accord After Thirty Years, ed. C. Fred Bergsten and Russel Green (Washington, D.C.: Peterson Institute for International Economics, 2016).
10. C. Fred Bergsten, “Time for a Plaza II?,” in International Monetary Cooperation: Lessons from the Plaza Accord After Thirty Years, ed. C. Fred Bergsten and Russel Green (Washington, D.C.: Peterson Institute for International Economics, 2016), Table 14-5.
11. C. Fred Bergsten, “Time for a Plaza II?,” in International Monetary Cooperation: Lessons from the Plaza Accord After Thirty Years, ed. C. Fred Bergsten and Russel Green (Washington, D.C.: Peterson Institute for International Economics, 2016).
12. International Monetary Fund, World Economic Outlook Database, April 2019: By Countries (Country-Level Data).
13. Ana Wong, “China’s Current Account: External Rebalancing or Capital Flight?,” Federal Reserve Board of Governors, International Finance Discussion Papers no. 1208, June 2017.
14. Modification of Regulations Regarding Benefit and Specificity in Countervailing Duty Proceedings, 84 Fed. Reg. 24406 (May 28, 2019).
15. U.S. Department of the Treasury, Macroeconomic and Foreign Exchange Policies of Major Trading Partners of the United States, May 2019.
16. U.S. Department of the Treasury, Macroeconomic and Foreign Exchange Policies of Major Trading Partners of the United States, May 2019.
17. C. Fred Bergsten and Joseph. E. Gagnon, “The ‘Decade of Manipulation’ (2003–13),” in Currency Conflict and Trade Policy: A New Strategy for the United States (Washington, D.C.: Peterson Institute for International Economics, June 2017).
18. Robert E. Scott, “What’s Good for Wall Street is Often Bad for American Workers and Manufacturing: The Overvalued Dollar,” Working Economics Blog (Economic Policy Institute), June 27, 2019.
20. Robert E. Scott, “What’s Good for Wall Street Is Often Bad for American Workers and Manufacturing: The Overvalued Dollar,” Working Economics Blog (Economic Policy Institute), June 27, 2019.