from Follow the Money

A frustrating TIC data release

January 17, 2007

Blog Post
Blog posts represent the views of CFR fellows and staff and not those of CFR, which takes no institutional positions.

More on:

Monetary Policy

I guess all TIC releases are frustrating, but I found today's release especially frustrating.  Total net monthly inflows -- $75b -- are just enough to cover the current account deficit (which is running a bit under $900b).  Though, consistent with recent trends, net long-term inflows ($58b) weren't quite large enough to cover the deficit.  Foreign demand for US long-term debt and equities was strong ($107.5b), but so was US demand for foreign assets ($39b in November).   Incidentally, net inflows are a bit different than the difference between foreign and US purchases because of an adjustment for principal repayments on agencies and mortgage backed securities

But that isn't the frustrating bit.  The frustrating bit is that the split between private/ official flows simply doesn't make sense, nor does the geographic breakdown.

Even if you use Morgan Stanley's estimate for global reserve growth ($600b) rather than my estimate ($750b), the world's central banks added, on average, about $50b a month to their assets in 2006.  Oil investment funds are getting another $10b a month.   So there is -- roughly speaking -- at least $60b a month in official funds flowing into either the US, Europe or Japan.   That isn't really in question.  If you don't trust me, look at the statistical data tables in the IMF's WEO.

November was a reasonably strong month for global reserve growth.  That also isn't really in question.  Look at the reserve data from the BRICs in November ...

So if the world added $60b plus to its official assets in November, how come only $9.1b of that shows up in the US data?  The US data shows $6.5b in net official purchases of US long-term securities (Temasek seems to have been selling some US equities, accounting for the net official equity outflows while China clearly accounts for most of the $3.6b in official purchases of "corporate debt).    Official institutions bought $7.7b of Treasury bills, but cut their holdings of other short-term securities by $1.8b and (implicitly) ran down their US bank accounts by about $3b.

I don't think anyone realistically thinks official institutions only put $10b or so into the US and US dollars while putting $60b into other markets and other currencies.   So official funds -- whether directly (custodial bias) or indirectly (dollar bank accounts lent out to hedge funds and private equity funds and others seeking leverage) -- are behind some of the private inflows.

The coutry data supports this point.   The flows reportedly coming from places that we know have a lot of funds to invest are just way too low.

China supposedly bought only $3.1b of US long-term debt in November.  It did may $2.4b of short-term Treasuries, but it reduced its other short-term claims on the US by $2.7b.   Net it out, and that works out to under $3b in inflows.   Do you think that accurately reflects the increase in China's dollar reserves during the month of November (valuation adjusted reserves were up by $20b).  I don't.  Even if you throw in Hong Kong's $2.8b in long-term purchases, the inflow is too small.

Asian and African oil exporters (think the Gulf and North Africa) reduced their long-term holdings of US debt by $2b in November.  The Gulf's short-term claims also fell, while Africa's rose  a bit.  But on net, they still reduced their US claims by $1.5b or so.  Does that make sense, given the size of the oil exporters current account surplus?   Clearly not.  Some of those funds are in the international banking data, but not all.   The Gulf in particular has been buying securities, not building up its bank deposits.

Russia reduced its long-term holdings by about $0.8b in November, but it did -- by contrast -- significantly increase its short-term holdings.  They rose by about $7.2b in Novembe, with most of the increase in bank deposits rather than short-term custodial holdings.   That at least makes sense.    For the first 11 months of the year though, Russian short-term holdings are down by $48b ... a fall that hasn't been offset by a commensurate rise in Russia's long-term holdings.   Russia has not only diversified; it has moved funds offshore.  The huge surge in Russian reserves (almost $110b on a flow basis) has coincided with a fall in total Russian claims on the US.

Fortunately, the US data does a very good job of picking up flows from Norway, and especially its government pension fund.  China wasn't buying Treasuries in November and the Asian and African oil exporters were selling.   So net official purchases of long-term Treasuries likely would have been negative but for Norway's purchase of $6.2b.   We don't know for sure that those purchases came from its government fund, but they likely did.    Norway very clearly ran down its long-term holdings and increased its short-term holdings in the first half of 2006, and then reversed course in the second half. 

I wish the US data also told us what other countries were doing, with a lag.  Unfortunately, it doesn't. 

And apologies for dragging you into the weeds of the TIC data.

(all my data comes from the TIC homepage, both the long-term data and the short-term data)



More on:

Monetary Policy