from Follow the Money

Call me Ahab (q3 IMF COFER data)

More on:

Monetary Policy

The data on global reserve accumulation is my great white whale.   I suspect I am about the only person who cares that the IMF just released its q3 data on global reserve accumulation.  

What is the story?  Simple: strong reserve growth and not much evidence of diversification, at least among those countries who report data to the IMF.

Countries that report data on the overall composition of their reserves to the IMF added $162.5b to their reserves in q3.  I usually add the non-reserve foreign assets of the Saudi Monetary Agency to that total.  SAMA's non-reserve foreign assets increased by $22.5b or so in q3.   That brings the global total up to $185b, give or take.

Valuation wasn't much of a factor -- the dollar rose marginally against the euro in q3, going from 1.2779 to 1.2687 (based on the St. Louis Fed data series).  That reduced the value of the world's existing stock of euros and similar currencies.   My rough cut puts those valuation losses at around $9b -- implying a valuation adjusted reserve increase of $194b in q3. 

Back in November (warning: RGE premium subscription required), Christian Menegatti and I esimated a q3 global increase of around $172b (valuation adjusted) based on the countries we track.   We were a bit on the low side. I suspect a lot of the smaller oil exporters added to their reserves at a very rapid clip.

$194b is particularly impressive because Russia made big payments -- over $23b in payments -- to the Paris Club out of its reserves in q3.  Moreover, Chinese reserve growth was relatively low -- about $47b -- in q3 for reasons that i don't fully understand.   The $194b global increase came without the two biggest recent sources of reserve growth firing on all cylinders, so to speak.

The q3 total is higher than my estimates for the valuation-adjusted increase in q1 and q2 ($177b in q1, $168b in q2).  $194b is within striking distance of $200b.   Indeed, I would be surprised if the q4 total isn't above $200b after adjusting for valuation.  

I consequently think it is fair to say that reserves increased at a $800b annual pace in the second half of 2006, with emerging economies accounting for the lion's share of the increase. That would be a record pace for a year -- though it doesn't quite match the record quarters generated by Japan's heavy intervention at the end of 2003 and in early 2004.

Were central banks buying dollars or something else with their growing reserves?  The honest answer is that we don't really know.  

Countries that report data on the currency composition of their reserves accounted for $101b of the $185b raw increase, and $107b or the $194b valuation adjusted increase.   China and Saudi Arabia explain the gap.

But we do know that the countries that report data to the IMF increased their dollar holdings by $75b or so.   That works out to about 70% of the valuation adjusted increase from that set of countries --  enough to push the dollar's share of their reserves up ever so slightly.   At least if one compares q3 to q2 ... the dollar's share is down a bit from the end of 2005, but much of that is simply from valuation changes.

For all the talk of diversification, there isn't much evidence of it in the aggregate data -- especially when one considers that Russia almost certainly diversified its reserves over the course of 2006.  Then again, there are also huge holes in the aggregate data.

More later.  Guaranteed.

Update: Here is the Bloomberg story on the COFER data. I think it gives a bit too much emphasis to the overall dollar and euro share of global reserves while glossing over the data on total reserve growth and flows.  I was also struck by the quote from Wang Jian of the China Society of Macroeconomics.  Wang argued about 70% of China's reserves are in dollars.  I wonder if Wang has independent information, or that is an echo of my/ other estimates? China obviously matters, as it alone accounts for about $1 trillion of the 1.6 trillion in reserves from countries that don't report the currency composition of their reserves to the IMF.  That total would be closer to 1.8-1.9 trillion if you include Saudi foreign assets/ reserves shifted to the Chinese banks (my preferred measure). 

More on:

Monetary Policy