China's Exchange Rate Regime
by Morris Goldstein, Peterson Institute for International Economics
Testimony before the Subcommittee on Domestic and International Monetary Policy, Trade, and Technology Committee on Financial Services
US House of Representatives
October 1, 2003
Mr. Chairman, thank you for the opportunity to testify before this committee on the important issue of China's exchange rate regime and its effects on the US economy. My Institute colleague, Nicholas Lardy, and I have recently been analyzing China's currency regime and this afternoon, I would like to share with the committee five of our main conclusions.1
First, so long as China maintains controls on capital outflows, runs surpluses on both the overall current and capital accounts in its balance of payments, and accumulates international reserves in large amounts, there is a compelling case that the Chinese currency, the renminbi (RMB), is significantly undervalued. Our preliminary estimates suggest that the undervaluation of the RMB is on the order of 15 to 25 percent. These estimates of RMB misalignment can be obtained either by solving a trade model for the appreciation of the RMB that would produce equilibrium in China's overall balance of payments, or by gauging the appreciation of the RMB that would make a fair contribution to the reduction in global payment imbalances, especially the reduction of the US current-account deficit to a more "sustainable" level. Both approaches produce similar answers.
Those who argue that the under-valuation of the RMB is much larger-30 to 40 percent or more—often confuse China's large bilateral trade surplus vis a vis the United States ($100 billion in 2002) with its much smaller overall current-account surplus ($35 billion). Adjusting for the overheating of the economy and other factors, China's "underlying" current account surplus in the first half of this year was probably about 2½ or 3 percent of GDP. Equally relevant, China's surplus on foreign direct investment (4 percent of GDP on average during the 1999-2002 period) has been considerably larger than the surplus on the overall capital account (1½ percent of GDP). It is the overall current and capital-account positions that matter for judging the extent of exchange rate misalignment—not bilateral trade balances or components of the current and capital accounts. In a similar vein, those who conclude that the degree of under-valuation of the RMB is small—10 percent or less—often ignore the fact that China makes extensive use of imported inputs to produce exports. China's role as a regional processing center means that it takes a larger revaluation to change the trade balance in China than it does in economies where imported inputs figure less prominently in exports. Admittedly, when China does decide to liberalize significantly capital outflows, it won't take a huge degree of international diversification of household savings deposits to put strong downward pressure on the RMB—but that doesn't alter the conclusion that right now the RMB is undervalued.
Second, a revaluation of the RMB is in China's own interest as well as in the interest of the global economy. If China does not revalue the RMB, net capital inflows and the large accumulation of international reserves (about $135 billion over the past 18 months alone) will continue. With its mountain of bad loans, China should not permit capital inflows and reserve accumulation to exacerbate the already excessive expansion in bank lending, money supply growth, and investment. In the first half of 2003, the increase in bank loans outstanding relative to GDP rose to 38 percent—an all time high, while the investment share of GDP rose to 42 percent—also an all time high. Although the process of "sterilizing" the effects of reserve increases on the base money supply has so far been less onerous in China than in many other emerging economies, experience shows that sterilization typically becomes more costly and less effective the larger it is and the longer it goes on. The primary domestic risk of an undervalued exchange rate is that it will handicap China's efforts to rein-in the pace of credit expansion and to improve the quality of bank lending decisions.
An appreciation of the RMB is likewise in the interest of the United States and the wider global community. Unless China permits the value of its currency to rise, it will be much more difficult to obtain the broader realignment of key exchange rates in Asia and elsewhere needed to produce a marked correction in global payments imbalances. The uncomfortable truth is that the US current-account deficit (projected to be about $550 billion this year) is much too large, that a further decline in the broad trade-weighted value of the dollar is necessary to reduce our deficit to a more sustainable level (say, $300 billion or so), and that a rise in the value of the RMB is an important catalyst for achieving both a broader-based decline in the dollar and a better global "sharing" of the adjustment burden.
Third, urging China to adopt a flexible exchange rate regime and to open its capital markets—as US Treasury Secretary Snow and others have suggested—is sensible advice for the longer term. But that advice is not appropriate for China's current circumstances—especially its weak banking system, and therefore is not likely to be heeded anytime soon, providing little relief for current exchange rate and payments problems. China is justifiably concerned that if it floated the exchange rate and opened its capital markets today, the weakness of the domestic financial system could generate large-scale capital flight and sharp currency depreciation in response to bad news. In addition, the government still dominates foreign-exchange transactions to a degree that precludes the market functioning properly. These obstacles should recede over time but they are binding at present.
A more practical approach is to urge China to reform its currency regime in two steps. In the first step, China would immediately revalue the RMB by 15 to 25 percent, it would widen the currency band (to between 5 and 7 percent, from less than 1 percent), and it would switch from a unitary peg to the dollar to a three-currency basket peg (with roughly-equal weights for the dollar, the euro, and the yen). This would remove the incentive for further speculative capital inflows and reserve accumulation. No longer would the foreign component of the money supply be working at cross-purposes with the needs of domestic stabilization. A three-currency basket would increase the stability of China's overall trade-weighted exchange rate, and the basket would also permit the US dollar to depreciate further against the RMB without a series of RMB parity changes. This can't happen if China retains its present unitary peg to the dollar; instead, the RMB would follow the dollar down, much as it has done since early 2002. Step two should be adoption of a managed float, after China takes further reforms to put the domestic financial sector on a sound enough footing to permit significant liberalization of capital outflows. A managed float is the preferred long-run regime because capital mobility in and out of China will increase over time (making a publicly announced exchange rate target more vulnerable), and because China will want to exercise greater monetary policy independence for stabilization purposes.
The advantage of this two-step approach is that it neither asks the rest of the world to live too long with a misaligned RMB, nor does it ask China to ignore a key lesson of the Asian financial crisis by prematurely opening its capital account. The key to reconciling China's desire for exchange rate stability with the need for the RMB to play its proper role in global balance-of- payments adjustment is to recognize that a fixed rate for the RMB need not be at the present parity, that "stability" of China's currency should be viewed against a wider set of reserve currencies than the US dollar alone, and that the transition from "fix" to "flex" for the RMB need not occur all at once, since liberalization of China's capital account will almost surely proceed in stages.
Fourth, the United States should take a multilateral tack in persuading China to alter its exchange rate policy and should eschew unilateral trade measures directed against China's exports. Other countries—including members of the European Union and many emerging economies in Asia—also have a strong interest in seeing the RMB rise closer to the level implied by fundamentals. If China resists a rise in the RMB, too much of global exchange rate adjustment will fall, for example, on the euro, exacerbating Europe's anemic growth performance. Conversely, if China does allow the RMB to rise, its emerging-market neighbors would be able to more easily accommodate a rise in their own currencies since then, the loss of competitiveness would be smaller than if each acted alone. The US Treasury should therefore continue to enlist the support of other countries in convincing the Chinese authorities that a more appreciated exchange rate for the RMB is in the common interest. Recall that China received plaudits during the Asian financial crisis for conducting a responsible exchange rate policy and for taking the interest of the region as a whole into account. At that time, China allowed the RMB to appreciate on a trade-weighted basis. China now has another opportunity to demonstrate responsible leadership.
As the institution charged with overseeing the international monetary system and with exercising "firm surveillance" over the exchange rate policies of its member countries, the IMF should take a more activist stance in monitoring exchange rate misalignments and in applying a mix of persuasion and pressure—both private and public—to reduce the duration of such misalignments. It is not useful to identify large-scale, prolonged, exchange market intervention in one direction, and behavior of the exchange rate inconsistent with underlying fundamentals, as implicit pointers of a "wrong" exchange rate and then do little when these pointers signal a problem. In this connection, it is regrettable that the Fund has very rarely made use of a provision to hold (or even discuss holding) special consultations with countries that are undergoing exchange rate problems. Endorsing a vague G-7 call for more exchange rate "flexibility" is not exercising "firm surveillance."
Multilateral does not mean everybody but the United States. The United States also needs to do its part to contribute to global adjustment by improving our savings-investment balance, and in particular, by adopting a workable plan to reverse the now-projected long string of US budget deficits. If we want other countries to act in a cooperative way on exchange rate policy and in implementing macroeconomic polices to promote growth, we too must be willing to put something positive into the package.
What the United States should not do is impose a unilateral import surcharge on China's exports. This would be both misguided and ineffective. As suggested above, bilateral trade imbalances are a bad indicator of overall payments imbalances. China is not the only country to have used—or now using—prolonged, large-scale, unidirectional exchange market intervention to maintain an undervalued exchange rate. China's imports more than quintupled between 1995 and 2002 and its import ratio (to GDP) now stands at a level three times higher than Japan and twice as high as in the United States. An import surcharge directed exclusively at China's exports might well invite retaliation against US exports to China and could risk a wider upsurge in protectionism when the opposite is needed to support global growth. Note too that much of China's exports to the United States compete primarily with exports from other emerging economies—and not so much with US domestic producers. An improvement in US competitiveness calls for a broad-based decline in the value of the dollar (among other measures)—not for a tax on one side of one developing country's foreign trade.
Fifth, the impact of a medium-size revaluation of the RMB on the external accounts of the United States should not be exaggerated. Even if China did revalue the RMB by say, 20 percent, and even if other Asian emerging economies and Japan followed China's lead by allowing their currencies to appreciate by say, half that amount, the trade-weighted value of the dollar would only decline by roughly 5 percent. This in turn might produce an improvement in the US current account on the order of $50 billion dollars; to reduce the US current-account deficit by say, almost half ($250 billion), would require a much larger and more broadly based further depreciation of the dollar (in the neighborhood of 25 percent on a trade-weighted basis). The long-running decline in US manufacturing employment started well before the Chinese currency became undervalued and has a much wider set of origins (including rapid productivity growth) than exchange rate factors. Given that the political campaign leading up to our presidential elections is now in full swing, it is perhaps not surprising that the China exchange rate issue has taken on increased visibility. But overblown expectations are apt to produce disappointment. An appreciation of the RMB will be helpful but not sufficient for reducing concerns about US external imbalances and their effects.
To sum up, as one of the four largest trading countries in the world, it is important that China take seriously its obligation (under IMF rules) to avoid manipulating exchange rates to prevent effective balance of payments adjustment or to gain an unfair competitive advantage over other countries. There is also a powerful domestic reason for seeking an early end to the existing medium-size under-valuation of the RMB: exchange rate under-valuation will increasingly handicap China's efforts to achieve and maintain long-term financial stability—to say nothing of potential protectionist threats against its exports. By adopting a two-stage reform of its currency regime, China can become part of the solution to the unsatisfactory global pattern of payments imbalances-not part of the problem. The currency regime that served China well over most of the last decade is not the currency regime that will serve China best today or in the future. If the United States wants to persuade China to reduce the serious undervaluation of the RMB, it should drop its doctrinaire insistence that China move in one great leap forward to a free float and to completely open capital markets in favor of a currency and capital market regime that China could actually adopt now with good effect and that could contribute now to reducing serious global imbalances. I believe the Goldstein-Lardy proposal for a two-step currency reform passes that test. The United States can push the ball forward by working in a multilateral fashion with other concerned countries to convince the Chinese authorities that RMB revaluation is seen as a legitimate remedy for a widely perceived imbalance, by refraining from unilateral protectionist measures, and by demonstrating a readiness to address its own longer-term saving-investment imbalance in the public sector. It's also long past time for the IMF to do more seriously one of the things it was created for—namely, to monitor the agreed international "rules of the game" on exchange rate policy, to provide objective assessments of exchange rate misalignments for important currencies, and yes, to help mobilize some pressure on countries that resist correction of undervalued or overvalued exchange rates. With some compromise by all parties and with the right sequencing of China's currency reform, a workable solution can be achieved.
1. In drafting this testimony, I have drawn heavily on two recent op-eds done jointly with Nicholas Lardy: (i) "A Modest Proposal for China's Renminbi," Financial Times, August 26, 2003; and (ii) "Two Stage Currency Reform for China," Asian Wall Street Journal, September 12, 2003.