Robert Mundell, the Renminbi, and Misapplied Theory

Robert Mundell has spent the past three years as the most visible academic opponent of renminbi revaluation. At conferences in Beijing, in op-eds, and in testimony that the Chinese government has been glad to cite, he has argued that the dollar-yuan peg provides stability, that revaluation would not correct the US current account deficit, and that American pressure on China over the exchange rate is economically illiterate politics dressed in the language of economics. On the first two points he is largely right. On the overall conclusion — that the peg should be maintained — I think the theory he helped build actually points in the opposite direction from the one he is drawing.

Mundell’s 1961 paper on optimum currency areas established a framework for asking when two or more regions benefit from sharing a currency. The answer depends on factor mobility, on the symmetry of shocks, and on the presence of fiscal transfers that can substitute for the exchange rate when regions face asymmetric downturns. The canonical application is the eurozone debate: southern European economies that faced different business cycles and different productivity trajectories than Germany were giving up a stabilisation tool when they joined the euro, and whether the gains from trade and price transparency outweighed that loss was a genuinely empirical question. Mundell himself was more enthusiastic about European monetary union than many of his colleagues, partly because he weighted the transaction cost reduction from a single currency very highly and partly because he thought labour mobility across European states would improve faster than it did.

Apply the OCA criteria to China and the United States and the picture is the opposite of what Mundell is claiming in public. The two economies are not a natural currency area by any criterion he specified. Their business cycles are not synchronised — China is growing at roughly 10 percent while the US is decelerating. Their productive structures are radically asymmetric. Labour mobility between them is essentially zero in both directions. There are no fiscal transfers. The exchange rate is the primary remaining adjustment mechanism, and it has been prevented from operating by a capital account regime and a sterilisation programme that the People’s Bank of China has been sustaining at increasing cost. The foreign exchange reserves, which have now passed $900 billion, represent the accumulated cost of this sterilisation. That is not a sign that the peg is costless. It is a sign of how hard the PBOC has had to work to maintain it.

Mundell’s public arguments against revaluation tend to rest on a different claim than OCA theory: that a revalued renminbi would not close the US current account deficit because the deficit is a savings-investment imbalance, not a relative price problem. Bernanke has made a related argument with his “global savings glut” framing. These are serious arguments and they contain truth: a 10 percent renminbi revaluation will not eliminate a $800 billion current account deficit. But that is not the argument for revaluation. The argument for a more flexible renminbi is that the current peg is generating distortions — in Chinese domestic resource allocation, in the composition of Chinese investment, and in the recycling of reserves into US Treasury securities at below-market rates — that are creating fragility in both economies. The question is not whether revaluation fixes the US current account. It is whether maintaining the peg at current levels is the right policy for China’s own development trajectory. China revalued by 2.1 percent in July 2005 and moved to a managed float in principle. The practical effect has been negligible: the renminbi has appreciated a further 0.8 percent in the eleven months since, against an economy growing at double digits. The peg is operating in all but name.

Mundell’s Nobel Prize is deserved and his analytical framework remains among the most useful in international macroeconomics. What I find strange is the gap between the framework and the policy conclusion he is drawing in this particular case. OCA theory says an exchange rate peg is appropriate when you have symmetric shocks, high factor mobility, and fiscal transfer mechanisms. None of those conditions hold between China and the United States. The same theory that underpins the case for European monetary union, applied without political gloss to the current situation, supports a more flexible renminbi. That Mundell has become the cited authority for the opposite conclusion is a reminder that in economics, as in most fields, the authority and the argument can come apart.