Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


$1.5 trillion, not $1.1 trillion – and rising fast

by Brad Setser
March 20, 2007

China's formal reserves were a bit under $1.1 trillion – think $1.066 trillion – at the end of last year.  No doubt they are close to $1.1 trillion, if not above it, now.    No wonder China’s central bank government  just made it quite clear that the PBoC doesn’t really want any more reserves.   Zhou, quoted by Reuters:

“Many people say that foreign exchange reserves in China are (already) large enough … We do not intend to go further and accumulate reserves

Zhou’s statement is probably yet another sign that the growth in China’s reserves this year has been kind of fast — and that the PBoC doesn't really want to continue to absorb the lion's share of the all the dollars that are now piling up in China.

But even now, the PBoC (though SAFE) is not the only Chinese institutions that have been adding to their foreign assets.   James McCormack of Fitch just released a quite interesting new report (registration with Fitch required) that estimates that China now holds $1.55 trillion in foreign assets.    And by China, Fitch effectively means China, Inc.   Most foreign assets not held by SAFE are held by various state companies and state banks.   Especially the state banks – Fitch, drawing on data from IMF, estimated China’s banks hold $273b in foreign assets, a $50b increase last year.   And it isn’t clear if that total – or for that matter China’s balance of payments data – counts the funds the banks raised in their offshore IPOs.  What happens in Hong kong may stay in Hong Kong. 

McCormack estimates that China added $342b to its foreign assets last year.  And even before China created the “People’s Investment Company” nearly $100b of the increase was coming from outside the central bank and wasn’t taking the form of reserves.    That squares with my own estimates – Chinese reserve growth, as large as it was, has been well below what it should be, given China’s burgeoning current account surplus and ongoing net inflows of FDI.    Large hot money outflows could explain the difference, but, well, all the evidence suggested that private money wanted to get into China and its booming equity market, not get out.   The other explanation: the state banks, state insurance companies, state pension funds and state companies all helped the PBoC out.

That is what makes Zhou’s comment that the PBoC is through accumulating reserves so interesting.   I don't think that the pace of growth in China’s foreign assets has slowed from last year.  If anything it probably picked up on the back of the still growing trade surplus.   So keeping something like the status quo in place – and by status quo I mean the current  exchange rate regime and ongoing Chinese openness to greenfield FDI – requires that someone in China add $400b or more to their stock of foreign assets.

Private Chinese citizens don’t seem very keen on adding to their foreign assets.  Not when those assets yield even less — in RMB terms — than domestic deposits.

And if the PBoC doesn’t want to add to its foreign assets either, well, the People’s Investment Company will get really big fast.   It would need to go from zero to $400b in a year to absorb all the foreign exchange coming into China.   Or if the bureaucratic battle between the Ministry of Finance and the PBoC for control of the new investment authority is resolved by creating two competing investment authorities, both would need to go from zero to $200b fast.

Three other interesting points emerge from McCormack's analysis, especially when that analysis is considered together with some of the points that have emerged out of the discussion surrounding plans for China's new investment authority.

1. McCormack notes — drawing on IMF data — that the reported foreign assets of China’s banks rose $50b to $273b (last data point, end-December) while Chinese banks’ deposits in BIS reporting banks fell by about $10b to around $100b (last data point, end-September).   That implies that China’s banks hold around $175b in foreign securities, and added up to $60b of foreign securities to their portfolio in 2006 – though the banks also could have fleshed out their external balance sheet by making external loans.   I suspect a lot of those purchases were funded by swaps with the PBoC – swaps where the PBoC lets the banks buy its dollars (for RMB) but promises to buy the dollars back at a pre-set price, liming the banks foreign exchange risk.   I wonder if shifting dollars – and I mean dollars, not euros – to the banking system, both through the transfer of assets to Central Huijin and swaps – is one way that the PBoC has held the dollar share of its reserves down.     If my suspicions are right, the overall share of dollars in China’s total external portfolio may be larger than the dollar share of China’s reserve portfolio (Richard McGregor of the FT has recently put the dollar share at 75% — which seems right to me). 

2. Simon Derrick of the Bank of New York has made explicit something that has long been rumored, namely that the Europeans have asked China not to add to their euro portfolio, or at least not to increase the euro’s share of their portfolio.    Derrick notes that “Market News International was reporting back in September that “Germany has told China that it would be undesirable for Beijing to sell its existing reserve holdings of USDs to buy currencies such as the EUR.”  The question is whether the Chinese have listened.    It is possible that they have.  The last thing the PBoC wants right now is even faster reserve growth.  And if shifting into the euro pushed the dollar and the RMB down, well, that is what it would get.  Premier Wen’s comments about China’s continued interest in dollar-assets may have been directed, in part, at Europe.  

3. There is a lot of chatter about the impact of the new investment authority on equity markets globally.  It will be big after all.  If the PBoC really isn’t going to add to its reserves (which generally are invested in a fairly diverse set of bonds) and China’s foreign assets will flow exclusively to investment authority and then into equity markets, that would be a pretty big shift.    But if China doesn’t immediately shift all of its foreign asset accumulation toward equities, and continues to add a mix of bonds and equities to its portfolio, my guess is that the overall impact of China’s shift won’t be as large as some thing — for a couple of reasons.  First, more rapid growth in Chinese foreign assets in 2007 will in part offset less rapid growth in the foreign assets of the oil exporters, and maybe 1/3 of the overall increase in oil exporters foreign assets came from oil investment funds, not central bank reserves – to a degree, rising Chinese demand for equities will offset a fall in oil state demand for equities.  Second, the huge flow from central banks into the bond markets and the world’s banks already has indirectly supported the prices of a lot of riskier assets.   When SAFE buys a bond from a US pension fund, the fund puts the money to work elsewhere.  And private equity firms have used the dollars central banks have on deposit in the world’s banks to bid up the price of a lot of equities as well … 

But perhaps the biggest and most interesting issue – at least to me – that McCormack raises is the most obvious one.   China now has a very large portfolio of foreign assets.  It now has $1.55 trillion.  That is likely to rise to close to $2 trillion by the end of the year.  Lots of other countries also have large external holdings — such holdings are part and parcel of financial globalization.  But most of the foreign assets of countries like US and the UK and even France are in private hands.  Most of China’s assets, by contrast, are controlled in one way or another by China’s state.   Reserves account for an unusually large share of China’s foreign assets.  And most of the non-reserve assets are held by state firms and state banks.   

Scratch one country, two system.  Think two systems, one global economy.    And two systems, one global financial system.   China Inc's investment decisions are going to have the capacity to move markets for a long time.    


  • Posted by Macro Man

    A couple of points: I wonder if QDII uptake would be better if firms were allowed to offer equity products rather than just bond and money market funds? It doesn’t take a genius to figure out that any positive carry generated by foreign (i.e dollar) securities/deposits is likely to be wiped out by RMB appreciation. If these guys are allowed to buy foreign equities and still refuse, then I’ll be more impressed.

    I don’t think anyone believes that China is still seriously diversifying into euros. But then again, they don;t need to to have a huge impact- when your annual reserve growth is bigger than just about anyone else’s stock, then even maintaining portfolio benchmarks is a big deal.

    Brad, do you think the banks’ foreign security purchases would show up in the TIC (for dollar purchases, obviously)?

    On equities I suspect the impact of CIC or PIC or whatever will be pronounced via regional outperformance. The buzz on the street is that there will be a relatively heavy weight on Asian regional equities, for whom the sorts of sums being bandied about are quite potentially market-moving.

    Any idea of what % of Japan’s foreign assets are currently in private sector hands, and how that percentage has evolved over time?

  • Posted by bsetser

    about a quarter of Japan’s foreign assets are reserves – per fitch … but i cannot quite figure out their methodology for calculating foreign assets. they seem to be looking only at holdings of debt (their US number is around $5 trillion, which matches us lending in the US NIIP). You might want to add the foreign assets of the postal savings system to that total.

    I don’t quite get how China can dump funds into Asian regional equities — that would drive up the local equity markets and please those long asian equities. but wouldn’t it be a nightmare for their central banks. The Thais for example don’t exactly want huge inflows. Most aren’t running current account deficits and don’t want their currencies to appreciate v. the RMB/$. If they intervene, China’s policies effectively shift reserve growth from China to the rest of asia — passing the dollar hot potato around, but leaving it in Asian hands.

  • Posted by Macro Man

    True (re: hot potato), but then again China isn’t exactly in the business of pleasing others, are they? Moreover, if PIC/CIC is ringfenced, then presumably any flows that derive from it are ‘non-official’ policy.

    And somehow, I think even the generals in Thailand would view a strategic investment from PIC somewhat differently from capital inflows from London and New York hedge funds. FWIW, GIC/Temasek have made similar noises re: shifting the emphasis to regional equities.

    As an aside: do you have any idea what the yield is of the bills that PBOC issues to sterilize its FX intervention? And who buys the stuff? I’m just trying to get an idea of what PBOC/SAFE’s carry is on the reserve portfolio and what the likelihood is that it would ever turn negative.

  • Posted by bsetser

    I haven’t checked yields on the sterilization bills recently, but generally they have been in the 2-3% range. Who buys? The banks. Remember, their deposit growth has vastly exceeded their lending growth (in large part b/c of administrative curbs on their lending), so they have tons of cash they have to part somewhere. the bills pay interest — tho not much and the banks don’t make money on the bills.

    as for the overall carry — it is clearly still positive so long as you ignore the capital loss from the RMB’s appreciation — Stephen Green of Standard Chartered did the math and it was published not long ago in the Economist. One thing really helps — a lot of cash (as in bills) circulates in China, so a lot of the PBoC’s liabilities don’t pay interest. So long as you don’t count the capital loss from the xR mismatch and so long as only about 1/2 the PBoC’s assets are offset by interest paying liabilities, it would take a lot to get negative carry.

    on the other hand, the PBoC’s cash flow profits have been one way the government has financed NPL disposal, so a fall in cash flow profitability still has consequences.

    re: the TIC data, it should count purchases by Chinese banks (the TIC data combines both private and official purchases), but I suspect that in practice a lot of the same issues that apply to the middle east apply to Chinese banks — including custodial bias. My guess is that a lot of their purchases are routed through hong kong and register as hong kong purchases or purchases from another financial center (the bank loans $ to its branch in HK, which purchases US securities and counts as a HK resident in the US reporting system, i think). If that’s right, there should be a surge in net purchases from HK starting in late 05, as that is when CHinese bank purchases really started .. another thing to check when i have time.

    In general, TIC flows to CHina from the US have been running at around $100b a year in 06, which is a bit low given the overall growth in China’s foreign assets. So my operating assumption is that the next survey will show a lot bigger increase in Chinese holdings that is implied by the sum of TIC flows (That was the case with the june 05 survey, released in 06).

  • Posted by Emmanuel

    There’s so much rich information here. Thank you. I just wanted to add it’s western finance that’s seeking Chinese citizens, not the other way around as foreign banks set up shop offering RMB accounts locally.

    OTOH, it may not be that Chinese are shying away from investing their money abroad. Rather, the product was defective to begin with as a fixed-income fund at a time when the dollar is on the Great Leap Downward:

    Bank of China’s “U.S. Dollar Return Enhancement Fund” promised to protect returns from currency fluctuations and invested in money-market products such as certificates of deposit.

    The fund yielded 0.48 percent during the holding period from Sept. 18 to Feb. 9, or about 1.3 percent annualized, according to its Web site. The rate was even lower than the 2.52 percent for one-year yuan deposits.

  • Posted by Ricardo Smith-Keynes


    Your line, “Think two systems, one global economy”, is (almost) quite literally the bottom line. We are in a bipolar exchange rate world. The appropriate historical analogue is the 1920s, when, following war-induced fiscal debauchery and inflation, the international monetary system was characterized by some on gold (at pre-war parities), some on gold (at new, devalued parities) and some on floating exchange rates. At the time, Keynes lamented the “mal-distribution” of gold that, under the strictures of the gold standard, forced countries in recession to adopt austerity as the orthodox (or “prudent”) policy response. The reason was the asymmetry in the “rules of the game”. As Barry Eichengreen put it, the international monetary disorder of the 1920s was caused by a failure of leadership: “the Bank of England couldn’t, and the Fed wouldn’t” provide the leadership needed to promote international monetary stability.

    In the post-war period, the IMF was chosen instrument. Today the Fund is in disrepute. The Fund “can’t” and the other players seemingly “won’t” provide the leadership. Meanwhile, discussions of IMF reform focus on everything except the real issue. Against this background, the question is: Do the men and women charged with the responsibility of overseeing reforms to the Fund have the vision of Keynes and Harry Dexter White to avoid another round of international monetary disorder?

  • Posted by bsetser

    Ricardo-Smith-Keynes ….

    Good question. Right now, i would guess no. But i also don’t think it is entirely the fund’s fault: there isn’t much appetite in the big players for policies that would reduce the risk of int. monetary disorder. The amount of opposition to the imf’s modest agenda for strengthening exchange rate surveillance stunned me — all of East Asia seems to be against the idea (hmmm …). The US is more interested in a surge and tax purity than fiscal balance and so on. And in the markets, there isn’t any real demand for action — no one seems all that worried. Bets on the stability of the current system have generally paid out handsomely (as long as they weren’t “go long $ v everything”), bets on instability not so much. So long as the markets don’t see much risk of a new round of monetary disorder, building support for action will be hard.

  • Posted by moldbug


    This is neither here nor there and you probably know it all already. But just in case you don’t, you may find it useful to get a quick contrary view of these icons of financial statesmanship: