The global BoP surplus is dominated by China, Russia and the Gulf, proxied by Saudi Arabia --
They collectively run a huge surplus. I can account for most of the associated financial flow, but not all --
Compared to the past, a relatively low share of their combined surplus is showing up as central bank reserves ($100b of a bit over $800 -- which would rise to a trillion if all of the GCC is included), and thus the direct flow into global (and US) bonds is modest
The Russian and Saudi surplus is almost all going into the international banking system (with some equity purchases by the Saudi PIF) --
But at least in q3 the large bond purchases of the Chinese state banks seem to appear in the global data ...
So, if my adjustment for the euroclear account (Belgian treasuries = China) is correct, I can account a portion of the large financial flow from the world's big autocracies to the world's biggest democracy.
That is the net flow needed for the '22 global BoP to balance!
• • •
Missing some Tweet in this thread? You can try to
force a refresh
China's current account surplus (which generates an inflow of fx) is huge (in dollar terms). FDI inflows have weakened over the last 4qs, but over that period they roughly offset the $100b in foreign sales of Chinese bonds.
(thread)
So, absent $400b in foreign asset accumulation (i.e. capital flight) by private Chinese citizens, one would expect to see:
a) foreign asset accumulation in the state sector (PBOC, banks, etc)
b) a bit of appreciation pressure on the CNY ...
Zooming in to Q3. The current account surplus soared. But there was an unusual net outflow of FDI ($20b) + ongoing foreign sales of Chinese bonds ($40b),
leaving a gap of about $80b (that is the expected foreign asset accumulation of the state sector, absent capital flight)
China's detailed q3 balance of payments data only sheds a little light on a central foreign exchange puzzle of 2022, namely why did the CNY depreciate alongside other Asian currencies when China's BoP position was much stronger?
1/
China's initial BoP disclosure for q3 showed a puzzling $40b or so increase in China's reserves (when other central banks were selling) --
The detailed BoP shows a (small) reduction in the external assets of the state banks, but very strong purchases of foreign bonds
2/
For many countries, bond outflows would be driven by private flows - but China historically has been different.
The bond flow has come primarily from the state banks (at times it reflected reserves shifted to the banks). About half the q3 outflow is also from the banks.
Motor vehicles are one important industrial sector where Chinese exports are relatively modest.
The WSJ notes that this is changing, largely thanks to the global shift towards EVs: "China’s new-energy vehicle exports have more than doubled year-over-year"
1/
The FT by contrast is arguing that foreign marks aim to reduce their dependence on China as a source of supply (for parts, but also for vehicles)
"Consequently, foreign manufacturers aim to make parts and cars inside China exclusively for use within the country"
2/
The FT article is framed around the forward looking intent of auto manufactures.
But this is still a case where some familiarity with the actual data is helpful.
China exports of finished autos (& for that matter auto parts) have soared since the start of the pandemic.
the big reported fall in the market value of foreign holdings of Treasuries has masked a far more important fact: foreign demand for US bonds has in fact been strong this year.
The US balance of payments data shows net foreign purchases of US long-term debt of around $200b a quarter -- a sum roughly equal to the (large) US current account deficit.
And it isn't coming from reserve managers! Bond demand and reserve growth have decoupled.
High absolute US rates have pulled in foreign funds (unhedged) no doubt.
The only real mystery is who exactly is doing the buying.
Hedged Japanese investors have sold on net. China (SAFE + state banks) still seems to be buying but modestly.
Tracking the evolution of global reserves from the IMF's COFER data is harder than usual this year -- reported reserves are down by $1.3 trillion through q3, and reported dollar reserves are down over $600b.
But, well, big adjustments are needed to the headline numbers.
1/x
Some countries (Japan for example) hold most of the reserves in bonds and mark those bonds to market. Fed's data indicates that a typical central bank Treasury portfolio would have lost 11-12% of its value through q3, more than accounting for the fall in headline reserves.
2/
But some countries don't follow IMF guidelines and mark bonds to market: I discovered this year that Korea and Taiwan don't typically mark to market for example. Others hold a large share of their reserves in fx deposits ...
The result: uncertainty about true dollar sales
3/
It is kind of surprising -- after all, the market value of safe US bonds fell by around 10% in the first three quarters of the year -- but, per the US BoP data, foreign demand for US bonds has remained robust ...
1/x
Japan, as expected, sold in q3 ...
But China has continued to buy (surprising, but that is what is in the US data if Belgium/ Euroclear is included) ...
And the mysterious bid from the world's financial centers remains robust
2/
Part of the story is -- per the detailed Bertaut-Judson data -- that the fall in reserves in emerging Asia has not translated into Treasury sales
That is in part b/c India drew down its deposits rather than selling securities