Global reserve accumulation

An upfront warning: this is a wonky, data-intensive kind of post

In the fourth quarter of 2004, Malaysia’s reserves grew by $9.9 billion. In the first quarter of 2004, its reserves grew by $6.2 billion. Seems like its pace of reserve accumulation slowed.

Hold on, though. Malaysia just reported that currency moves added $2.6 billion to its reserves in the fourth quarter, and subtracted about $1.2 billion from its reserves in the first quarter (valuation gains that, incidentally, are consistent with quite substantial non-dollar reserves). Net those changes out and Malaysia’s bought $7.4 billion in new reserves in the fourth quarter, and $7.3 billion in the first quarter.

Kind of interesting.

If China’s reserves grew by $19.5 billion in March -- bringing its end q1 reserves up to $662 billion, then the reserves of eleven major emerging economies (China, Taiwan, Korea, Hong Kong, Singapore, India, Malaysia, Thailand, Mexico, Brazil and Russia) grew by about $104 billion in the first quarter.

That is a substantial sum: it works out to over $400 billion annually. It is a particularly substantial sum for a group of countries that includes only two oil exporters (Russia and Mexico). The OPEC countries clearly saved much of their recent oil windfall as well, so they have money to play with too. But it is still well below $207 billion these countries added to their reserves in q4 (by my calculations).

Or is it? I made a bunch of assumptions about the composition of these countries reserves -- assumptions influenced by observation of how changes in the euro/dollar seem to have influenced the value of their reserves during the first quarter. Overall, I assumed that about 70% of these countries reserves were in dollars. But I also assumed that there was substantial variation across countries. Malaysia and Thailand were assumed to hold the fewest dollars (a bit over 50%), followed by India, Russia and Singapore. China and Taiwan were assumed to hold the high proportion of their reserves in dollars (80%).

These are just guesses; informed guesses, but still guesses.

I then estimated q4 valuation gains from the dollar’s fall ($38 billion), and q1 valuation losses ($20 billion) from the dollar rise. Remember, the dollar fell by around 9% against the euro in q4, and then rose by about 4.2% in q1. Once you adjust for these valuation changes, the fall off in these countries’ reserve accumulation in q1 is much smaller. I estimate that these countries added $169 billion in new reserves to their portfolios in q4, and $124 billion in q1.

The $45 billion fall off, incidentally, is almost entirely explained by the fall in China’s reserve accumulation (net of valuation changes) from $86 billion in q4 to an estimated $57 billion in q1.

Of course, by any normal measure, $57 billion in reserve accumulation by a singel country in one quarter is enormous: it works out to about $230 billion a year, enough to finance a quarter of even more enormous estimated 2005 US current account deficit. And unless something changes, I would still expect to see the pace of China’s reserve accumulation pick up as the year progresses.

As you all know well by now, China doesn’t have to sell its existing reserves to cause problems for the US. All it has to do is to stop adding to its dollar reserves at the current pace.Three other points:

1. China is the biggest player in the reserve accumulation game, but it is not the only player. Ignoring valuation changes, Taiwan’s reserves grew by 9.5 billion in q1, Malaysia’s by $6 billion, India’s by $10 billion, Russia’s by $14 billion (Russia paid off a lot of debt too) and Brazil’s by $9 billion.

2. By my (rough) estimates, these eleven countries added $90 billion to their dollar reserves in q1, only slightly less than the $122 billion they added in q4. $90 billion is enough to finance about 1/2 half of the likely q1 US current account deficit. And remember, that total leaves out smaller emerging economies and the big OPEC countries, who clearly added lots of petrodollars (and petroeuros) to their external portfolios in the first quarter. Central bank reserve financing of the US is far from over, even if Japan is no longer actively playing the game.

3. By my calculations these countries added $46 billion to their non-dollar reserves in the fourth quarter (independent of valuation gains), and $33 billion in the first quarter. That works out to an annual increase of between $120 billion and $180 billion. Most of that presumably went into euros. And the petrosheiks are adding to their external assets at a very fast clip too, and I don’t think anyone would be surprised if they were investing a relatively large share of their oil windfall in non-dollar assets. All in all that amounts to a rather substantial inflow into a region that has a roughly balanced current account; an inflow that would tend to bid up asset prices in "euroland" until "euroland" investors decided to shift some of their assets out of the euro, create offsetting outflows.

Low real interest rates globally are the prime piece of evidence for a "global" savings glut. Real interest rates in Japan have been low for a while, so that is hardly news. The real news is low real interest rates in both Europe and the US. I suspect central bank reserve inflows -- and petrosheik asset building -- could be playing a role keeping real interest rates low in both regions.