VoxEU Column Exchange Rates

The renminbi since 2005

Much of the debate over China’s exchange-rate policy has focused on the effect on the US and other western economies. This column provides a comprehensive summary of China’s exchange-rate policy over the last five years and argues that it would also be in China’s interest to let its currency appreciate – and now is as good a time as any.

For better or worse, the topic of China’s currency policy, and in particular the exchange rate of the renminbi against the dollar, is now perhaps the most salient of controversies in international monetary economics. The US Treasury’s bilateral report to Congress on countries that may have been unfairly manipulating their exchange rates is awaited with far more eagerness than usual.1

The US Treasury reports

In a 2007 paper for Economic Policy, Shang-Jin Wei and I examined econometrically the factors that have apparently determined US Treasury findings regarding manipulation, in the years since Congress first mandated these reports in 1988. We found some variables that an economist would recognise as legitimately relevant to the possible undervaluation of foreign currencies to have effects on the decisions in the Treasury reports. Particularly significant statistically was the partner’s overall current account/GDP ratio. Less significant are the partner’s reserve changes and the undervaluation of its currency as judged by purchasing power parity (PPP). Some variables that an economist would not recognise as the legitimate responsibility of the foreign countries also matter in the decision. The bilateral trade balance is the most consistently and strongly significant, generally at the 99% level. The US unemployment rate is sometimes also significant, particularly when interacted with a dummy variable for a presidential election year.

Overall, three aspects of the regression results suggest that the domestic political variables are as important determinants of the Treasury decision as the legitimate global manipulation criteria: the absence of a clear role for reserve accumulation by the partner country as mandated by IMF criteria, the significance of US unemployment (especially in election years), and the very high significance of the bilateral balance. If it was the IMF interpreting the criteria in its Articles of Agreement, rather than the Treasury interpreting the criteria in the 1988 US law, then consistent unidirectional intervention in the foreign exchange market would receive a lot more emphasis, and the US-specific variables such as the bilateral trade balance and US unemployment would not appear at all.

Using this equation to predict whether the Treasury in the end will go all the way to naming China as a “manipulator” in 2010 does not give a clear answer. Relative to the last dozen Treasury reports, two major factors have changed, but in opposite directions. On the one hand, China’s overall trade surplus in 2010 has dramatically – if perhaps temporarily – dropped to zero. On the other hand, US unemployment is unusually high.

The remainder of this short paper addresses two questions:

  • What has been the actual exchange rate followed by the Chinese authorities over the last five years?
  • What exchange-rate policy is in China’s own interest?
What has been China’s exchange rate policy since 2005?

In July 2005, the Chinese government announced that it was changing its official exchange-rate regime. As US politicians had been demanding, the renminbi would no longer be pegged to the dollar. Rather the authorities would:

  • set its value with reference to a basket of foreign currencies (with numerical weights unannounced), and
  • allow a margin of fluctuation in the exchange rate that, though small in any given day, could cumulate substantially over time.

What has the exchange-rate regime been in practice, as opposed to the official announcement? It would not be surprising if the two differed. Many currencies show a discrepancy between de jure and de facto regimes. Accordingly, statistical techniques were developed some years ago to discern the true exchange-rate regime.

The standard techniques show that, in practice, the renminbi initially continued to maintain a tight peg to the dollar after July 2005. Gradually, in 2006, the relationship loosened. Statistical analysis suggests that the People’s Bank of China did indeed begin to assign a little weight within the anchor basket to a few non-dollar currencies, perhaps beginning with the Korean won during a period centred on January-March 2007. However most of the weight remained on the dollar (Frankel and Wei 2007).

The addition of another few years of data and the use of a new, more sophisticated, statistical equation revealed that during the course of 2007 the effective anchoring basket began for the first time to assign substantial weight to the euro (Frankel and Wei 2008). For a period that ran up to approximately May 2008, the anchor was in effect a basket that put virtually as much weight on the euro as on the dollar. There was also some limited flexibility around that anchor. When high or low international flows were working to push the currency away from the basket, the authorities would intervene, or “lean against the wind,” to push the currency back (Frankel 2009).

In mid-2008, however, the renminbi virtually returned to a dollar peg. By September 2008 all the weight once again had come fall on the US currency. The regime had come full circle, virtually back to what it was in late 2005. Evidently the motivation for China’s return to the dollar was complaints from the exporter provinces that had lost competitiveness in 2007 as the currency appreciated against the dollar. (Barry Naughton 2008, gives a glimpse inside Beijing politburo politics.)

Has the policy switch of 2008 led to a weaker renminbi than otherwise? US politicians don’t really care whether the renminbi is fixed or floating. What they want, of course, is for it to be stronger against the dollar rather than weaker, so that US firms don’t find it so difficult to compete against Chinese exports. In 2007, when the renminbi was loosely tied to a basket that put heavy weight on the euro, it appreciated against the dollar because the euro was appreciating against the dollar. Indeed from mid-2006 to the end of 2007, the overall value of the renminbi did not fluctuate much if one defines the value in terms of a yardstick that assigns half-weight to the euro and half-weight to the dollar.

Figure 1 shows the foreign exchange value of the renminbi, in terms of three different measures. One can see around 2007:

  • the steadiness of the currency measured in terms of a euro+dollar average (the green line in the middle), and
  • the resulting observed appreciation of the renminbi against the dollar (the magenta line on top). The appreciation appears to be due to the presence of the euro in the basket, and not in fact to appreciation against the basket as usually implied in the press.

Figure 1 Exchange rates, Jan 2005 - March 2010

In a sense it would be wrong to complain that the return of the dollar link in mid-2008 gave American firms an additional price disadvantage in world markets. Over the last two years, the euro has depreciated against the dollar. (In late 2008 and early 2009, this was due to a flight to the perceived safe haven of the dollar; more recently, in the winter of 2009-2010, it has been due to the Eurozone’s troubles with Greece and other members on its periphery.) In other words, at precisely the moment when the renminbi changed horses in mid-stream, jumping back on the dollar horse, the dollar horse and the euro horse changed directions. If the Chinese authorities had kept the loose basket policy of 2007 instead of switching back to the dollar peg in 2008, the value of the renminbi would be even lower today, not higher, and dollar-based producers would be at more of a competitive disadvantage, not less.

Those worried about undervaluation of the Chinese currency will not be much impressed with these arguments. Just because the renminbi would be even more undervalued if the authorities had stuck with their policy of 2007-08, doesn’t mean it isn’t unacceptably undervalued as it is. Just because the RMB has followed the dollar upward over the last two years, does not mean that it should not appreciate even more. The more relevant alternative is to introduce true flexibility, not just against the dollar, but against all foreign currencies.

Is it in China’s interest to appreciate?

The phrase “undervaluation” has no precise definition in economics. Or, in fact, it has many possible definitions. In the case of most currencies, most of the time, it is impossible to say with any confidence what is the correct value for the exchange rate. The renminbi is unusual, however, in that a wide variety of different definitions and criteria give the same qualitative answer: the currency is undervalued. An appreciation would improve economic welfare.

Whose economic welfare? Here I consider China’s economic welfare, narrowly defined.

It should be conceded from the outset that China has been pursuing an exchange-rate policy that is effectively fixed to the dollar for the last decade or so. A fixed exchange-rate regime has a number of advantages. Two advantages are perhaps the most important. First is the provision of a nominal anchor to prevent inflationary monetary policies and expectations thereof. Second is the facilitation of trade with those countries that use the dollar, or at least are pegged to the dollar. Other advantages of fixed rates include facilitating financial integration, forestalling competitive appreciation or depreciation, and preventing the sort of speculative bubbles that seem occasionally to afflict floating exchange rates. (There is of course a corresponding list of advantages of floating rates.)

There are five different reasons to believe that it is now in China’s interest to allow increased flexibility of the renminbi. Increased flexibility, under the conditions prevailing today or during most of the period since 2003, would imply appreciation. The five reasons, which will be considered in turn, are:

  • Overheating
  • Excessive reserve accumulation
  • Long-run need to use the exchange rate for expenditure-switching
  • Avoiding future crashes.
  • Low price level by international standards.
Overheating of the economy

The target for internal balance is traditionally defined as output equal to potential output, or as employment equal to the natural rate of employment, or as an absence of inflationary or deflationary pressures.

Let us say that the country in question finds itself with output above potential, Y>Y , i.e., there is excess demand for goods. As a result the economy is overheating. In terms of a standard Swan Diagram, we are talking about an economy that finds itself to the right of the Y= line. (See slide 28 of Frankel 2010). To cool off the economy, the government can either raise interest rates or undergo a real appreciation.

In 2002, it was possible to argue that China was in the zone of excess supply of goods. At that point, some easing of monetary policy was called for, to increase spending and restore internal balance. By 2007, however, China had clearly moved to the other side of the line. As a result, some combination of an increase in interest rates and a real appreciation was called for.

On what are such judgements based? After all China’s growth has been uniformly high by the rest of the world’s standards for three decades, but so has unemployment. The explanation of that paradox is that the rate of growth of potential output has also been high, perhaps as high as 10%, but that there is always a lot of frictional unemployment due to the rapid pace of structural change, especially rural urban migration. But given the large uncertainties in pinpointing the rate of growth of potential output or the natural rate of unemployment, how can one judge when the economy has entered the zone of excess demand? The expansion in 2007 was running into input bottlenecks and other supply constraints. Inflation, running at 6-7% in 2007 and provoking social unrest and price controls, was a clear sign of overheating. A bubble in the Shanghai stock market sent the same message. An important lesson of the 2008 global financial crisis is that excessively elevated asset prices should be accorded a role, in addition to inflation, as manifestations of an overheated economy.

In 2008-09, the overheating danger abruptly disappeared, as the global crisis reduced exports by 40% in a 12-month period. By early 2010, however, China had come roaring back, through a combination of domestic demand stimulus and recovery in the export market. The economy returned once again to the zone of excess demand, as reflected in some labour market shortages and new price acceleration in goods markets and real estate. A real appreciation would help cool off the economy more efficiently than any other policy.

Excessive reserves

Central banks in Asia and other emerging markets drew the lesson from the East Asia crises of the late 1990s that it was wise to take the precaution of accumulating large levels of international reserves. They spent the next decade doing precisely that, via balance of payments surpluses. The levels of reserves has tended to strike western economists as inefficiently high, particularly in the case of China – which now has amassed an unheard of $2.5 trillion in reserves – notwithstanding some recent evidence that the countries that held more reserves came through the global financial crisis of 2008 in better shape than others. A majority of the reserves are typically held in the form of dollars (an estimated 70%, in China’s case), especially US Treasury bills. These pay a low rate of return. Meanwhile, China is paying to foreign investors on their inward investment a higher return than it is earning, which means that the arrangement is a losing deal for the country on aggregate.

Another disadvantage of rapid reserve accumulation, for a country worried about excess demand, overheating and inflation, is that it can swell the money supply and thereby add to these pressures. One option for central banks in emerging markets facing large inflows is to try to sterilise them – that is, to prevent the growing reserves from expanding the money supply too rapidly. The central bank offsets the rising reserves via reductions in net domestic assets, into negative territory if necessary. Historically some have tried this, but have only succeeded for a year or two. Over time, it gets harder to sterilise an inflow. Sterilisation keeps domestic interest rates high. This prolongs the capital inflow, while in the meantime producing a “quasi-fiscal deficit” on the books of the central bank: a gap between the interest rate earned abroad on the reserves and the higher interest rate that it must pay domestic investors to hold its sterilisation bonds. Eventually the central bank gives up, and either allows the currency to appreciate or allows the money supply to increase.

During the years 2004-2006, the People’s Bank of China was remarkably effective at sterilising the record reserve inflow. There was little sign of either a quasi-fiscal deficit, money growth in excess of real GDP, or inflation. One reason was domestic financial repression; domestic banks were forced to absorb sterilisation bonds at interest rates below what a market economy would have yielded and to increase their domestic reserves as a proportion of liabilities. In 2007, however, the longstanding warnings of foreign economists began to come true; higher domestic interest rates, an incipient quasi-fiscal deficit, rapid money growth, and inflation. This may have contributed to China’s decision to allow the renminbi to appreciate against the dollar at that time. As we have seen, the global financial crisis hit China as a large but temporary reprieve from concerns of overheating; but as of 2010 we are back to 2007.

Attaining internal and external balance

Most countries have at least one goal for the domestic macroeconomy (such as potential output) and at least one goal for the international sector (such as a balance of payments objective). The principle of goals and instruments says that to achieve two independent goals, you need two independent instruments. More specifically, the point of the Swan Diagram is that to achieve both internal and external balance, it is not enough to vary spending (monetary or fiscal contraction). One must also have an expenditure-switching policy. Usually this is the exchange rate. In a small open flexible economy like Hong Kong, one can hope to achieve equilibrium by adjusting domestic wages and prices, and to preserve the advantages of a fixed exchange rate. An economy as large as China, however, needs its own currency. There was a time when capital controls could play the role of second policy instrument. But as China becomes increasingly integrated into the world economy, it becomes both less desirable and less practical to block capital flows. Eventually the second policy instrument will have to be the exchange rate. The only question is when.

Avoiding future crashes

Some have argued that if China wants to minimise the probability of future crises such as those that afflicted its neighbours in 1997-98, it should avoid appreciation, so as to keep the current account as strong as possible. It is true that overvalued currencies played a role in the East Asia crisis of 1997, even though some westerners had urged appreciation for surplus countries in the past. It is also true that real appreciation is likely to lead to lower trade balances and raise net borrowing from abroad, and that countries that borrow from abroad are more likely to have crises. But there is another respect in which moving to a regime of increased flexibility now might reduce the chance of future crises rather than increase it. If and when inflow turns to outflow, as part of the cycle that so many developing countries have gone through so many times before, it is important not to cling to a peg for too long. Many countries procrastinate, postponing adjustment either through devaluation or expenditure reduction.

One lesson from the past experience is that of the exit strategy. If an exit from a peg, to a regime with greater flexibility, is likely to occur eventually anyway, it is better to do it at time when the balance of payments is strong and the initial movement is likely to be appreciation. The alternative of waiting for a time of balance of payments deficit often turns out to mean exiting the peg under strong downward speculative pressure, with the result that confidence is undermined and the national balance sheet is weak. Hence the argument for being safe, and increasing flexibility before any cut-off in capital flows.

RMB undervaluation judged by the Balassa-Samuelson relationship

Purchasing power parity (PPP) is often calculated as a guide for what the exchange rate should be, for China as for other countries. Most are estimates of relative PPP, that is, based on price indices. They do not necessarily show the renminbi to be strongly undervalued. But they use the past as the benchmark, and the standard hypothesis is that the renminbi may have been undervalued in the past as well. Comparisons of price levels across countries are difficult, because such absolute PPP data are much less available than relative PPP data (for which one only needs price indices and exchange rates). But some data are available. They show prices lower in China than in advanced countries. But this does not in itself answer the relevant question. Even if we thought that markets in internationally traded goods were perfectly integrated, there is no mechanism to arbitrage disparities in prices of non-traded goods. There is abundant empirical evidence, along both the cross-section and time-series dimensions, that prices of non-traded goods, and thereby of general price levels, rise with levels of productivity, real wages and real income. This robust empirical regularity is called the Balassa–Samuelson effect.

In Frankel (2006) I have estimated the renminbi to have been undervalued by 36% in 2000, if equilibrium is defined by the Balassa-Samuelson relationship. The latest update comes from Subramanian (2010), who estimates a gap of around 30%.2


Given the imprecision of the statistics, one should not rely overly on any one indicator such as the Balassa-Samuelson calculation. But it is striking that a wide variety of criteria give similar answers: the renminbi is undervalued and China would probably be better off if it were allowed to appreciate.3

At the same time, an appreciation of the Chinese currency, if it went into effect immediately, would help promote exports, output and growth in the US and other countries still suffering the after-effects of the 2008-09 recession. It is rare to have such a clear answer as to the appropriate value for a currency, even in qualitative terms. Typically some criteria point in one direction and others in the opposite direction. This is a good argument in general against setting up formal legal mechanisms that attempt to judge whether countries’ currencies are correctly valued and whether their exchange rate policies are “fair.” In the case of the renminbi in 2010, however, an appreciation would bring benefits both for China and for the rest of the world.

The author acknowledges research assistance from Daniel Xie.


Cheung, Yin-Wong, Menzie Chinn and Eiji Fuji (2009), “China’s Current Account and Exchange Rate,’ in Rob Feenstra and Shang-Jin Wei (eds.), China’s Growing Role in World Trade (University of Chicago Press).

Cline, William, and John Williamson (2008), “Estimates of the Equilibrium Exchange Rate of the Renminbi?” in Morris Goldstein and Nicholas R Lardy (eds.), Debating China's Exchange Rate Policy, Peterson Institute for International Economics, Washington DC, 155-165.

Frankel, Jeffrey (2006), "On the Yuan: The Choice Between Adjustment Under a Fixed Exchange Rate and Adjustment under a Flexible Rate", in Gerhard Illing (ed.), Understanding the Chinese Economy, Oxford University Press, 246-275.

Frankel, Jeffrey (2009), “New Estimation of China’s Exchange Rate Regime", Pacific Economic Review ,14(3):346-360, August.

Frankel, Jeggrey (2010), “China, the US, and Currency Issues”, presented at China Future Leaders, 29 January.

Frankel, Jeffrey and Shang-Jin.Wei (2007), “Assessing China’s Exchange Rate Regime”, Economic Policy, 51:575-614.

Frankel, Jeffrey and Shang-Jin Wei (2008), “Estimation of De Facto Exchange Rate Regimes: Synthesis of the Techniques for Inferring Flexibility and Basket Weights”, IMF Staff Papers 55:384-416.

Goldstein, Morris, and Nicholas R Lardy (2009), The Future of China's Exchange Rate Policy, Policy Analyses in International Economics 87, Peterson Institute for International Economics: Washington DC, July.

McKinnon, Ronald (2010) “A Reply to Krugman: What Manipulation?, The International Economy, 37-39, Winter.

Mundell, Robert (2004), “Adjustment in China’s exchange rate regime,” remarks at Inaugural Seminar on Foreign Exchange System”, Dalian, China, 26- 27 May.

Naughton, Barry (2008), “A New Team Faces Unprecedented Economic Challenges”, China Leadership Monitor, No.26.

Subramanian, Arvind (2010), “New PPP-Based Estimates of Renminbi Undervaluation and Policy Implications”, Policy Brief 10-8, Peterson Institute for International Economics: Washington DC, April.

1 As of the date of writing, the Treasury has announced that the report will be delayed well past the technical April deadline, widely hypothesised to be part of a face-saving arrangement that will give China enough political space and time to allow some small appreciation of the currency.

2 More recent estimates should be somewhat more reliable, not so much because they incorporate data from more recent years, but because they use more reliable measures of Chinese prices, which used to be based on highly inadequate raw data. Other estimates are offered by Cheung, Chinn and Fuji (2009) and Cline and Williamson (2008). Goldstein and Lardy (2009) base their assessment of yuan undervaluation on the adjustment necessary to produce an appropriate trade balance – a line of argument that has been weakened in 2010 by the disappearance of China’s trade deficit.

3 It must be acknowledged that a few highly reputable economists, notably Ronald McKinnon (2010) and Robert Mundell (2004), do not believe that the renminbi should appreciate. Their beliefs derive from an attachment to fixed exchange rates generically, rather than an evaluation of the desirable level of the Chinese exchange rate.


1,820 Reads