Are the foreign exchange losses of central banks real?

The accounting losses central banks will soon report as a result of foreign exchange market moves has all the markers of an issue that is about to explode. Richard McGregor found that this issue was so hot that his usual range of Chinese sources declined comment.

A number of prominent Chinese economists, contacted for this article, declined to discuss the issue of potential foreign exchange losses on the record because of its sensitivity.

Why?

Such losses, were they to be credibly tallied, could easily become a significant political issue in China, where there is perennial and often sharp-edged debate over policies blamed for a loss of state wealth.

Other central banks are likely to report large losses from currency moves when they put out their annual reports. I expect the Reserve Bank of India, Brazil's Central Bank and the Bank of Thailand to post losses for 2007. These countries are intervening to offset large capital inflows; they generally do not have large surpluses (India has a small deficit). China is a bit different: it has a large current account surplus and is attracting large inflows.

Everyone agrees that if a central bank pays more on its sterilization bills and the funds the banks have on deposit than it gets on its reserves it will incur a loss. The rest of the government will have to write the central bank a check to cover its interest payments.

There isn’t agreement on what it means, though, for a central bank to lose money in the foreign exchange market.

Remember, central banks with lots of foreign assets have domestic currency liabilities. China’s central bank has three basic kinds of liabilities – RMB cash, money it owes to the domestic banks on their mandatory reserves and its bills (short-term IOUs). It hold mostly dollars and euros as assets. When the RMB rises against the dollar (or the dollar falls v the RMB), the value of its assets, expressed in RMB falls.

A private financial institution would incur a loss. But does a central bank? There is a real debate. Prominent former central bankers – Ted Truman and many others – argue that central banks should only care about protecting the external purchasing power of their foreign assets, not their domestic purchasing power. Accounting losses don’t matter much – as a central bank doesn’t really need any equity capital as it is backed by the broader government. Accounting losses can be covered by a check from the government – it all just sort of nets out. The Treasury bonds the government hands over to the central bank to cover its accounting losses (allowing the central bank to report assets equal to its liabilities) pay interest, but the central bank generally just hands its interest income back to the government...

Others – like me -- are not fully convinced by this argument. Countries that are holding a reasonable level of reserves for prudential reasons shouldn’t worry to much about the domestic purchasing power of their reserves. They hold reserves precisely to meet a surge in demand for foreign currency, and thus they have to worry about the external value of their currency. The fluctuation of the domestic currency value of their foreign exchange reserves is simply a cost of holding needed prudential reserves.

But countries that are adding to their reserves even though they already hold way more than they need to hold the value of their currency down are in effect using the central bank to mobilize domestic savings and invest them abroad. They are effectively subsidizing the consumption of China's exports. A government whose finance ministry borrowed in domestic currency to provide export financing in a foreign currency would incur a loss in the event the value of its foreign currency loan fell relative to the value of the money it borrowed domestically. And that in some loose sense is what China’s central bank is doing, just very indrectly.

The domestic purchasing power of the savings the PBoC has borrowed (through bill issuance) and invested abroad is likely to fall over time. The IMF believes best practice in the event of central bank losses is "support .. in the form of a budget appropriation by the central government in either cash or government securities to recapitalize the central bank." I think it is fair to argue that bonds that will eventually be handed over to the Chinese central bank to cover its currency losses could have been sold to the public and used to finance spending and investment in China. But I would also appreciating hearing dissenting views.

What matters most, though, is what China’s own citizens think. As James Fallows notes in the Atlantic monthly (also listen to his NPR interview), the creation of the CIC has made the process of mobilizing domestic savings to invest abroad very very explicit.

"The Chinese public is beginning to be aware that its government is sitting on a lot of money—money not being spent to help China directly, money not doing so well in Blackstone-style foreign investments, money invested in the ever-falling U.S. dollar. Chinese bloggers and press commentators have begun making a connection between the billions of dollars the country is sending away and the domestic needs the country has not addressed."

The Ministry of Finance bonds issued to finance the CIC’s investment abroad – investments that could well return less in RMB terms than what China owes on the bonds issued to finance the investment – could be used to finance investment in domestic infrastructure. Or to finance more spending on health care – with a bigger fiscal deficit as a result.

Of course, more investment or more spending would tend to stimulate China’s domestic economy, and put more pressure on domestic prices. Then again, China could offset the inflationary pressure by letting its currency appreciate not just against the dollar but also against the euro and a host of emerging Asian currencies.

Fundamentally, issuing bonds to raise funds to invest abroad rather than to invest or to spend at home is a choice. It is clear that China's public did not like the government's loss on its investment in Blackstone – though it isn’t yet clear if they disliked the fall in the dollar value of China’s investment or the fall in its RMB value. China's central bank isn't likely to report dollar losses. It might though soon start to report RMB losses.

p.s. the 5% loss on an accrual basis that is sourced to me in the Richard McGregor article comes from assuming that the increase in China's foreign assets (counting funds in the state banks) is about 15% of China's GDP, and assuming that the RMB will eventually appreciate by about 33% against a basket that matches the currency composition of China's reserves. It is a very rough estimate.