from Follow the Money

"Not quite so SAFE": This week’s Economics Focus column

April 24, 2009

Blog Post

More on:

Monetary Policy


This week’s Economist covers two topics that regular readers know very, very well: China’s reserve growth and China’s holdings of US debt.

The estimates for China’s US holdings in the Economist are based on the methodology laid out in a paper that Arpana Pandey and I did in January. But that paper was written before the Treasury released the results of its June 2008 survey of foreign portfolio investment and consequently the estimates in the actual paper are now a bit out of date. The last survey showed fewer Treasury purchases than we expected, and more equity purchases. * The estimates in this week’s Economist reflect the results of the last survey, and therefore differ from the estimates in the underlying paper. We should have an updated paper out soon as well.

The most recent US survey of foreign portfolio investment also showed somewhat smaller Chinese purchases of US assets from mid-2007 to mid-2008 than I would have expected -- something I plan to discuss in more detail. That could be evidence of (modest) diversification away from the dollar from mid-2007 to mid-2008, but it equally could represent greater use of private fund managers. If China handed some of its reserves over to a private manager, the US survey data wouldn’t attribute those funds to China. The increase in China’s Treasury holdings last fall was far larger than can be explained by the underlying growth in China’s reserves -- and one explanation for the strong recorded inflows would be that China was pulling funds out of privately managed accounts. Tracking China’s portfolio is an art not a science.**

The Economist closes with a key point: the main constraint on China’s reserve management is China’s own policy of managing its currency against the dollar.

China is trying to have it both ways. It wants to lessen its dollar exposure, but it also wants to hold down the yuan. The picture has been temporarily clouded by shifts in “hot capital” flows, but so long as China runs a large current-account surplus, its reserves will rise. In order to keep the yuan weak against the dollar, a large chunk of those reserves will end up in greenbacks. Beijing’s appetite may not match Washington’s growing need for cash. But China cannot sour on the dollar without letting its own currency rise.

China’s reserve growth has been temporarily held down by speculative outflows, but there are (tentative) signs those outflows have slowed -- which implies that China’s reserves are likely to resume their steady upward march so long as China maintains a large current account surplus.

And that isn’t necessarily a good thing. Any abrupt change, of course, would be bad. The past several months have illustrated why gradual transitions that allow needed economic adjustments to take place over time rather than all at once are important. But I personally don’t think it is in China’s interest to continue to invest so much of its savings in US assets, especially on terms that imply likely losses for China’s taxpayers. And I equally don’t think it is in the interest of the United States to rely so heavily on a single country’s government for financing. A smaller US current account deficit, financed by a more diverse group of creditors, would be far healthier.

UPDATE: China’s disclosure that it has increased its holdings of gold from 600 tons at the end of 2003 to 1054 tons (worth $31b) now has attracted a bit of attention. But the rise in China’s holdings needs to be put into perspective. From the end of 2003 to now, I would estimate that China’s holdings of US Treasuries and Agencies rose from about $275 billion to $1250 billion, a rise of close to a trillion dollars. China reserves have increased enormously over this period, so it now holds more of everything. The suggestion that China has recently stepped up its gold purchases is far more significant than the total rise.

* The Setser/ Pandey methodology is pretty blind -- it uses the pattern of revisions in the most recent available survey to estimate future revisions. It thus will miss changes in the composition of China’s portfolio due to changes in reserve management that occurred subsequent to the last published survey. I sometimes have a sense that things have changed, but I want my published estimates to be based on numbers that are replicable (with a fair amount of effort).

** If an informed reader thinks my estimates are off -- or that I am missing something -- do let me know. Feedback helps me improve my estimates.

More on:

Monetary Policy