And some are perhaps trying harder than others …
``All central banks are trying to diversify,'' he told Reuters on the sidelines of a European Central Bank conference in Frankfurt. ``We have had a very clear diversification plan for several years.''
The data so far suggests that central bankers have been far more successful at diversifying out of Treasuries into other types of dollar denominated debt (China has been a leader here) than at diversifying out of the dollar. Russia is to my knowledge the only major central bank that has diversified the currency composition of its reserves in a significant way in 2006. Countries like India, Malaysia and Thailand have long been diversified. But some others may have done so quietly, though there isn’t much evidence for diversification in the global data.
I suspect the real constraint here has been the fact that a lot of the countries that want to diversify their reserves also want to maintain a peg to the dollar. That limits your options. Zhou presumably knows this better than anyone.
Shifting out of dollars at points in time when private investors do not want dollars tends to put pressure on the dollar. And on the exchange rates of countries that peg to the dollar. The last thing China and the Gulf need are weaker exchange rates …
Indeed, the available evidence suggests that central banks actually increased their dollar purchases in the first half of 2006 relative to their dollar purchases in 2005. Why? It was easier to diversify in 2005 when the market wanted dollars rather than euros and the euro was falling than in 2006, when the dollar has slid against the euro.
I have a paper in the works on this (an update of this paper -- subcription required), so here are a few charts ---
The first shows the currency composition of new central bank deposits in the international banking system, as reported by the BIS. The data for 2006 comes from the first half of the year, and it has been annualized for the sake of comparison. And, critically, the BIS data doesn’t include any Chinese deposits – the SAFE isn’t considered a monetary authority. There was a noticeable surge in euro and pound deposits in 2005 – and something of a resurgence in dollar deposits in 2006.
The second shows my estimate, based on the COFER data, on the currency composition of the growth of global reserves. To get the global estimate, though, I had to fill in a lot of blanks – and to assume that the countries that don’t report the currency composition of their reserves to the IMF (notably China) have acted like the countries that do, only more so. A lot of emerging markets countries that don’t report peg to the dollar, so I have assumed that they hold a higher fraction of their reserves in dollar than the countries that do report …
Both data series show a pick up in dollar purchases in 2006 relative to 2005. If China has been diversifying the currency composition of its reserves in 2006, the green bar would be smaller in 2006 – but it still would be larger than in 2005.
But the flow data doesn’t really provide much evidence that China has been diversifying out of dollars. The flow data has huge problems, but recorded Chinese purchases in 2006 have been stronger than in 2005. China is just buying more agencies and corporate bonds.
That said, I do think the markets have grown rather complacent, and lots of folks have concluded that the world’s big central banks will finance the US no matter what even as the internal Chinese debate (and I assume, the internal debate in the Gulf) about the wisdom of having an ever growing dollar-denominated claim on the US is heating up.
There is certainly a gap between market expectations (continued financing of the US on the current scale) and the talk in some policy circles (how can we reduce our dollar exposure) -- but at least so far, the talk in policy circles hasn't really led to noticeable action. Russia excepted.