VoxEU Column Global economy International trade

How to avoid trade war: A reciprocity requirement

With the US threatening to label China a “currency manipulator”, this column presents a plan to address global imbalances without risking a trade war. It proposes a “reciprocity” requirement – if the US can’t buy Chinese government bonds, then China can’t buy US bonds either.

Editors’ note: This column is an updated version of the column first posted on 29 September 2010.

The endless discussions about global imbalances, and China’s supposedly self-serving exchange-rate policy, have for long resembled discussions about the weather; everybody talked about it, but nobody did anything. This is now changing.

The recent move by China to invest heavily in Japanese government bonds has set in motion a chain reaction. The Japanese authorities had little choice but to react to the Chinese move by intervening themselves in the only really liquid market, namely the market for dollars. Japan got the blame for its “unilateral” move, but the end result was the same as if the Chinese had bought US assets themselves. The Japanese are only unwitting intermediaries, who, on top of the blame, have to take on even more exchange-rate risk.

Overall it seems that the rest of the world with free capital markets can do little to stop the Central Bank of the People’s Republic of China to continue "steering” its exchange rate1 by accumulating more and more international reserves – it does not matter whether these are US or Japanese. The US, Japan, or the ECB cannot do the same (as proposed in Bergsten 20102) because China has capital controls and there are simply no significant renminbi assets that foreigners, especially foreign central banks, are allowed to invest in.

The US political system has become so frustrated by this situation that Congress is now seriously considering whether to label the country a “currency manipulator” and impose trade sanctions which would be illegal under WTO rules and threaten to throw the global trading system into turmoil.

But there is another way. The US (and Japan) could easily prevent the Chinese Central Bank from continuing its intervention policy without breaking any international commitment. The US and Japan (and the Eurozone countries) only need to invoke the principle of reciprocity and declare that they will limit sales of their public debt henceforth to only include official institutions from countries in which they themselves are allowed to buy and hold public debt. Instead of the “moral suasion”, tried in vain by the Japanese, the Chinese authorities would just be told that they can buy more US T-bills and Japanese bonds only if they allow foreigners to buy domestic Chinese debt.

Imposing such a “reciprocity” requirement on capital flows would be perfectly legal – although the US (and Japan and all EU member countries) have notified the IMF that they have liberalised capital movements under Article VIII of the IMF. Yet, in contrast to the area of trade, there are no legal constraints on the impositions of capital controls.

This “reciprocity” measure would of course be equivalent to a very specific form of controls on capital inflows. Capital controls are always somewhat leaky, but not in this case because the Chinese Central Bank would find it difficult to hide its huge investments going through western financial institutions. No reputable financial institution would dare to become a hidden intermediary for the Chinese given that no institution bidding for hundreds of billions of T-Bills would take the risk of secretly fronting the Chinese government or central bank as it would have to certify to the US authorities that the beneficial owner is not from a country in which foreigners cannot buy and hold public debt instruments.

As a practical matter the introduction of the reciprocity requirement should provide a grand fathering of the existing stocks of Chinese official assets abroad (already above $2,500 billion). However, the Central Bank of China would still not be able to continue its interventionist policy – and that is what counts for foreign exchange markets. It would still have to invest the proceeds of the continuing current account deficit (running at about a billion dollars a day), and it would face the problem of re-investing the flow of T-bills coming due.

The immediate objection is: “What if the Chinese react emotionally and dump their holdings of T-Bills and US agency debt on the market? Would that not disrupt the US government debt market?” This “dumping” is not as simple as it sounds. What assets would the Chinese Central Bank buy when it sells T-Bills? There are not many choices if the Chinese Central Bank wants to dispose of thousands of billions of dollars. Either it holds cash in the form of bank deposits (this would mean a massive refinancing of the US banking system) or it buys other private US assets which would mean a welcome refinancing of the US private sector.

Brender and Pisani (2009) and Gros (2009) argue that the huge acquisitions of US government debt by Asian central banks during the boom years (2003-2007) were a key factor in the financial crisis. They helped create an excess demand for safe and liquid assets which induced the US financial system to use securitisation to create new forms of supposedly safe and liquid assets out of US mortgage debt. The collapse of the housing bubble then showed that these assets were neither safe nor liquid. Even if the imposition of a reciprocity requirement were to be limited to government debt and if it had only a limited impact on the ability of China to intervene in the renminbi dollar market, it could already make a significant contribution to the stability of the US financial system if it forced the Chinese central bank to buy hundreds of billions of dollars in private US assets.

Moreover, the reciprocity requirement could be extended to private debt instruments as well. Since foreigners, especially foreign official institutions, are not allowed to freely hold renminbi bank accounts in China, the US authorities could introduce a similar rule, or just levy a special tax3 on bank accounts held by foreign official institutions from countries with capital controls. At the very least the US could eliminate the existing exemption of foreign official holdings of US assets from withholding taxes on interest payments.

But extending the controls on capital inflows from China to more and more instruments is probably not necessary as the Chinese Central Bank is unlikely to invest hundreds of billions of dollars (or euros) in private assets. Buying euro assets would of course constitute an alternative, but this does not appear too attractive at present, and would be prevented by the Europeans adopting the same reciprocity requirement.

The US might hesitate to impose a reciprocity requirement for sales of its public debt because (in contrast to Japan) it needs foreign financing for its public sector deficit. But this also constitutes the litmus test for the sincerity of the US position which cannot have it both ways, i.e. Chinese financing of its external deficit and an end to currency intervention. The choice is now up to the US, it can easily stop Chinese interventions without violating any international commitment if it is willing to rely on domestic savings to finance its own fiscal deficits.

Japan should have no problem with the approach proposed here since it can easily finance its budget deficit with domestic savings. The same should hold true of the Eurozone, which has a current account in rough balance, indicating that domestic savings are enough to finance all domestic investment and budget deficits.


Bergsten, C Fred, (2010) “We Can Fight Fire with Fire on the Renminbi”, Financial Times, 4 October.

Brender, Anton and Florence Pisani (2009), “Global Imbalances and the Collapse of Globalised Finance”, Brussels, Centre for European Policy Studies, CEPS Paperback.

Goodhart, Charles and Dimitrios Tsomocos (2010), “How to restore current account imbalances in a symmetric way”, syndicated column available at eurointelligence.com.

Gros, Daniel (2009), “Global imbalances and the accumulation of risk”, VoxEU.org, 11 June.

Mundell, Robert A (1960), “The Monetary Dynamics of International Adjustment under Fixed and Flexible Exchange Rates”, Quarterly Journal of Economics.

Thorbecke, Willem (2010), “The effect of a renminbi appreciation on US-China trade imbalances”, VoxEU.org, 6 October.

This contribution takes it as given that the exchange rate matters for the current account, and hence potentially employment in the US and other slowly growing advanced countries. This is of course disputed by the Chinese authorities. For a recent empirical assessment of the impact of the renminbi exchange rate on trade balances see Thorbecke (2010).

Bergsten (2010) assumes that Chinese capital controls are not effective – it is unclear on what basis. The Bergsten proposal seems to be aimed also at Japan, treating any foreign exchange intervention in the dollar market as an action to which the US should respond. However, as argued here, the recent Japanese interventions were simply a reaction to increased buying of Japanese bonds by China.

The spirit here is different from that of Goodhart and Tsomocos (2010) who propose taxing capital exports. But they emphasise the inefficiency of private capital markets as the main reason for imposing this tax. By contrast the key consideration underlying the present proposal is the desire to “level the playing field” in terms of capital controls and hence the potential for foreign exchange intervention.


1,890 Reads