Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


Does China’s (de facto) dollar peg limit China’s capacity to diversify its reserves?

by Brad Setser
November 26, 2006

The dollar hasn’t recovered from the Thanksgiving sell-off.   At least not in Asia.   The won is not as strong as it has been since Korea's own crisis back in 1997.   

Those with dollar exposure – particularly those who expected the dollar to remain in its comfortable band of this summer and fall (somewhere between 1.25 to 1.29 — See Menxie Chinn's graph) — have to be considering their options.   At least those who were surprised by the dollar's recent move.  Bloomberg:

The dollar's decline “caught a lot of the market napping,'' said Gibbs (currency strategist at ABN Amro Holding NV in Sydney). “A lot of people are not positioned as they would like to be.''

No one has more exposure to the dollar than the PBoC. But do they have any real alternative to adding to the dollar exposure rapidly so long as they basically peg to the dollar?  

I have long thought that this constrained China’s options.   But I was starting to get nervous.   The last really good data point that I have on the composition of China’s reserves comes from June 2005 (the US survey of foreign portfolio holdings).   It is getting a bit stale.    Everything else since then is an (educated) guess.

And with all the recent Chinese talk of diversification, I was starting to wonder if I might have under-estimated China’s ability to quietly reduce the share of dollars in its reserve portfolio.   So I was relieved to hear Guan Tao of SAFE complain of the difficulty making in major moves. Reuters:

"It is very difficult for these countries to make significant adjustments in their reserve asset portfolios," Guan told a forum, adding that that was his personal view and not a statement of SAFE policy.  (Hat tip tmcgee)

China presumably now holds more dollars it really wants.  But it also may have no choice but to add to its (already bloated) dollar portfolio at even faster pace that it has been so long as the dollar is under pressure.  At least so long as China wants to keep the RMB from appreciating against the dollar.  Remember, right now keeping the RMB from appreciating (by much) against the dollar requires pushing the RMB down against the euro.  

Of course, diversification can mean a lot of different things.

China almost certainly holds a more diverse set of currencies as part of its reserves than it did a few years ago.  And it also holds a more diverse set of financial instruments – think mortgage backed securities and emerging market bonds – than it did a few years back.   

So in some sense, China certainly holds a more diverse portfolio now than it did a couple of years agto.    On the other hand, China could have added more currencies and more instruments to its (rapidly growing) portfolio without reducing the dollar’s overall share of its portfolio.  And if the share of China’s reserves in dollars hasn’t fallen, in some deeper sense it hasn’t diversified. 

Which leads to the second point Guan Tao raises.   He argued that central banks would continue to be able to dampen volatility in the foreign exchange market.

"The U.S. buys cargo, and countries with a trade surplus buy U.S. treasuries — they actually have no other choice," he said. "If I have a surplus on the trade account, if I have foreign exchange income, I can certainly invest part of it in non-dollar assets, but the size of the market for non-dollar assets is very limited so the great majority of foreign exchange reserve assets has to be invested U.S. financial markets."

Therefore, it was very unlikely that currency markets would see any massive selloff of dollars by countries with large reserves, Guan said.  "Given that U.S. dollar liquidity is currently mainly held by a few monetary authorities, we feel that the possibility of a big fluctuation in the dollar is very small — at least for now," he said

If Guan Tao is talking about the risk of big fluctuations in the $/ RMB, he certainly is right.  If the PBoC wants to limit the possibility of big fluactions in the $/ RMB, it knows what to do. 

But what about the dollar/ euro?   Lots of folks in the market think central banks have played a role in dampening down volatility in the dollar/ euro, not just volatility in the dollar/ RMB.  Highly concentrated central bank positions are — in this view — a source of stability for the rest of the market.

Here is the logic – as I understand it. 

The The PBoC (and most other central banks) tend to buy dollars for local currency when they intervene.   Keeping the dollar share of their portfolio from rising consequently requires the ongoing sale of euros for dollars.   When the dollar rises against the euro, central banks often use the dollar's rally as an opportunity to sell dollars.  That dampens the dollar's move.   One example: the available data suggest that the world’s central banks bought a lot of euros in the first and second quarter of 2005.

Now consider what happens if the dollar is falling.   Those central banks that peg (more or less) to the dollar have to buy more dollars to maintain their peg.   And other central banks tend to intervene more.  Last week, both Korea and India reportedly bought dollars to keep their currencies from rising.

They could turn around and sell some of those dollars for euros.  But that would add to the pressure on the dollar.    Or they could hold on to the dollars they are buying, removing one source of demand for euros (the ongoing demand created as the PBoC shifts some of the $20b of dollars it buys every month into euros) from the market.   That too would tend to dampen the dollar’s move.   

But to dampen volatility in the dollar/ euro, central banks both need to buy more reserves when the dollar is under pressure and to hold more of those reserves in dollars.The aggregate data — while incomplete — suggests that central banks have generally done so.    

Call it waiting to sell dollars until the market conditions are favorable.   Or call it targeting a constant dollar share in your portfolio.  The result is similar.

The market certainly has come to expect that central banks will act as stabilizing speculators, buying dollars when others won’t. 

The world's central banks were probably sitting on more dollars than they wanted at the end of 2004.   But the combination of the Homeland Investment Act, rising fed rates and the defeat of the European constitution allowed many central banks to scale back their dollar holdings in 2005.  Central banks increased their euro and poud holdings dramatically in the first half of 2005.

At least that is what the IMF data suggests.   THe IMF data, though, is famously incomplete.  We know what central banks that report have done.  But we don’t know if the central banks that don’t report are doing something similar.   

I am doing some work with Christian Menegatti that uses the BIS data to try to fill in the gaps in the IMF data.  It suggests that a tiny bit of diversification.  There was a spike in pound deposits in q4 2005, and there was a spike in euro deposits in the second quarter of 2006.   Spikes that cannot be explained by looking at the countries that report data to the IMF.  

One note: the BIS data, though, doesn’t cover China, as the SAFE isn't a monetary authority.  The surge in pound/ euro deposits recently is coming from someone else. 

To be clear, the sums involved are still small … too small to change the overall story.  The main story line is still that central banks are adding to their dollar reserves at an unprecedented pace, and providing nearly unprecedented financing for the US. 

But these side plots do hint that some central banks may now be holding more dollars in their portfolio than they really want.   And consequently they may not be terribly pleased to find that acting as “stabilizing speculators” requires that they add to their already large dollar portfolios.

I think Shinichi Takasaka (manager of foreign-exchange and financial products trading at Mitsubishi UFJ Trust and Banking Corp. in Tokyo) is only 1/2 right.  He argues: 

“Central banks in Asia are more concerned about their own currencies strengthening than they are about the dollar weakening, so this prevents them from selling the dollar aggressively,'' said Takasaka.

For most central banks, though, the question isn't about selling.  It is about buying.   The real issue is whether central banks will once again be willing to buy dollars agressively.  

The 5 trillion dollar question though is whether any of this prompts a major reevaluation of policy.    After all, a major shift in central bank policy is one potential source of a Wile. E. Coyote moment.  The 5 trillion dollar question though is whether any of this prompts a major reevaluation of policy.    After all, a major shift in central bank policy is one potential source of a Wile. E. Coyote moment.  


  • Posted by Anonymous

    Brad- Have you researched passed periods of “$ glut” as was described in financial litterature in the lat 1960s?

    Western central banks in early 1970s and Opec central banks in mid 1970s…we’ve been there before.

  • Posted by Dave Chiang

    The Paulson & Bernanke Show will soon be airing in China. Bernanke getting into dangerous waters outside his domain. As an independent central banker, it’s debatable whether Bernanke should be traveling to China with Bush’s cabinet members. I had the same reservations about Greenspan sitting next to Rubin in meetings with Asian counterparts in the late 1990s.

    While the U.S. is still the major trading partner in Asia, there’s a growing opinion that China’s 10 percent annual expansion is more important. Chinese officials know this. They also are realizing the clout that comes with playing a key role in controlling U.S. monetary policy. China’s central bank governor, Zhou Xiaochuan, may now be the most important person in currency markets. The U.S. certainly wants to avoid a massive dollar-selling binge in Asia.

    Yes, China deserves grief for holding down its currency. Yet when it comes to imbalances, there’s plenty of blame to go around. Bush’s massive tax cuts weren’t matched with spending reductions, and the U.S. needs to save more. The result: huge and unsustainable current-account and budget deficits. If the U.S. wants the yuan to rise, China in return will want the U.S. to stop living beyond its means — and to stop blaming China for U.S. excesses.

    By William Pesek

  • Posted by Guest

    re: “Last week, both Korea and India reportedly bought dollars to keep their currencies from rising.”

    Wondering if we should be paying more attention to possible internal and external affects of stresses on India and other Asian economies which, presumably, must also keep the value of their currencies depressed to compete with China.

    “…Despite widespread claims that China’s economy is overheating, actually India’s shows more signs of boiling over…”

  • Posted by a

    “The Paulson & Bernanke Show will soon be airing in China. Bernanke getting into dangerous waters outside his domain. As an independent central banker, it’s debatable whether Bernanke should be traveling to China with Bush’s cabinet members. I had the same reservations about Greenspan sitting next to Rubin in meetings with Asian counterparts in the late 1990s.”

    DC – I couldn’t agree with you more. Presumably the Chinese want Bernanke there to make promises he shouldn’t be making (such as keeping interest rates high…)- because they infringe on US sovereignty.

  • Posted by bsetser

    Korea has allowed its currency to appreciate more than most, and its exporters are feeling the pressure. Many have said — and I agree — that China’s willingness to appreciate is the key to broader Asian appreciation.

    India has strong consumption growth, unlike China, and a current account deficit — no doubt there are interesting policy issues there, but i don’t know them as well.

    That said, rather than changing the topic, I would be interested in hearing what people think of the argument on Central Banks impact on volatility in the euro/ $ market and China’s capacity to shift the composition of its reserves while targetting the dollar.

  • Posted by Gheorghius

    “Gheorghius — you clearly belong on a trading floor, not at RGE”.

    Why? I thought that economic analysis, when/if it is good, can be used to either trade, or to advice on policies, or to produce academic work. I haven’t published in academic journals in recent years, but I bet I could if my job was doing research again. Yes, I was advicing forex traders in 2004-05, and I still do it somehow, but I don’t spend my life in a trading floor because it does not contribute to the common good and wealth creation (it’s only redistributing from losers to winners). I’m a rather a policy, macro, instutional consultant. But in my opinion, assessing one’s macro views by comparing one’s forecasts with actual long term mkt moves is a good way to measure the quality of one’s ideas, and correct them if needed. (For ex.: “Financial mkts are larger now and can sustain for years the large US deficit, until the US debt rises to above 40% of GDP” is a view that implies many things: so forecasting a steady dollar at end 2004 was one way to check that view). But I find that broad “vision” is an under-rated asset in today’s world. Maybe at RGE too?

    On $ reserves – What if developing world CBs keep their pegs and keep buying dollars (slightly increasing the share of $ reserves over time), but the private sector decides that the dollar is just undervalued (and increases its speculative dollar sales)? Could it not be enough to produce a major downward dollar move against the Euro? Would a rapid move of the $/Euro to 136-140 against the CBs not prove that the private sector is relevant enough to move the major crosses against CBs, thus that it has decisively contributed to the dollar steadyness in 2005-06? Would it not calm down the fears that an irrational private forex mkt helps global (US) imbalances to rise above orderly sustainability levels?

    PS. And by the way: I thought I remembered a recent post of yours supporting someone’s view that the Yen was 55% overvalued… If I am wrong I apologize. Also, please keep in mind that I have fun in producing polemics, but I do appreciate your work.

  • Posted by Guest

    re: “Central Banks impact on volatility in the euro/ $ market”

    was thinking about the build up of presumably related stresses which may force sudden adjustments – those could be a real stretch – and in what way, difficult to guess.

  • Posted by Anonymous

    Why are you surprised with Bernanke’s visit to China?
    He is doing his job of touring the 13th district of the Fed system?

    The anomaly is that Zhou is not sitting on the FOMC as a regional president. lol

  • Posted by HK

    Brad–I think China’s ability to diversify its foreign exchange reserves is seriously constrained by its policy of the virtual dollar peg.

    Suppose China purchses the dollar with the renminbi in order to maintain the dollar peg, and at the same time sell the dollar so acquireed to diversify to the euro. If that amount is small, the impact on the euro-dollar rate might be limited. But, the amount is large, so that the euro would appreciate against the dollar, and against the renminbi, which is pegged to the dollar. This would entail two things; 1) the effective exchange rate of the renminbi declines, thus further increasing the current account surplus, which would necessitate further more currency market intervention, and 2) the dollar value as well as the renminbi value of foreign exchange reserves increase, while the euro value would decrease.

    All of these show that China’s reserve diversification would result in further accentuating the imbalances and amplifying the eventual loss (interms of exchange loss and resource misallocation loss). I wonder if the Chinese authorities are willing to diversify its foreign exchange reserves now.

  • Posted by Maverick

    Dr. Setser,

    Your blog is truly “copacetic”. (Mr. Rubin has indeed a knack for words)

    Talking about current $ depreciation; what would it take for a single player, eg. PBoC, to intervene in the FX markets and defend its enormous $ reserve exposure from furher decline in value? I know that the issue is far more complex than the question implies but does anyone have a ballpark number? I am not refering only to the peg per se.

    Gheorghious always look forward to your perspective. Do you reckon that the current Eur/$ rate is the highest we are going to see for 2006?

  • Posted by bsetser

    Re: “Zhou is not sitting on the FOMC as a regional president”

    If I were Zhou, I wouldn’t settle for anything less than Vice-Chairman …

    China has been buying almost as many (US) Treasuries in the conduct of its monetary/ exchange rate policy as the Fed has been buying in the conduct of its monetary policy.

    p.s. the source for the argument that Bernanke shouldn’t be going to China (b/c of the political overtones) is Bill Pesek, not DC.

  • Posted by Dave Chiang

    Hi Brad,

    In my earlier posting, the listed source for the argument that Bernanke shouldn’t be going to China was clearly stated “written by William Pesek”. However, I fully concur with Pesek’s argument that Bernanke’s visit to Beijing is beyond the economic and political jurisdiction of the Federal Reserve. Bernanke’s visit to Beijing almost reminds me of the Wall Street-Treasury complex under Robert Rubin when the Treasury department de facto assumed the foreign policy administration of the Department of State during the 1990’s Asian economic crisis. By contrast, the Chinese Central Bankers have always taken a far more reserved attitude; China’s central bank governor, Zhou Xiaochuan, rarely attends World Bank or IMF meetings in Washington, and wouldn’t be caught dead partying late night with Greenspan and the Hollywood crowd in the Swiss resort of Davos.


  • Posted by bsetser

    Maverick — the answer is about $25-30b a month, assuming that China can continue to convince the banks to add $5-10b to their own dollar exposure a month …

    China isn’t really exposed to swings in the euro/ $. they matter to be sure — at the margin china would rather have a capital gain (in $ terms) by holding a rising euro than a loss on a sinking dollar. but China’s big exposure is to changes in the $/ RMB and euro/ RMB, since the PBoC finances its $ and euro holdings by issuing RMB debt. And right now, with a currnet account surplus in the $250b range and $50b in net FDI inflows and $40b or so in net portfolio equity inflows (I think) from the bank iPOs plus whatever additional hot money is coming into china, someone in China needs to add about $30b in dollars/ euros to their external portfolio every month for the BoP to add up.

  • Posted by FTX

    Interesting article in UK’s Daily Telegraph today. Suggests that EU’s political masters aren’t going to stand idly by if dollar depreciates further against the euro.

    Airbus could trigger ‘nuclear option’ of currency controls

    “My hunch is that Airbus will bring matters to a head. I was told by an Airbus official last year that if the euro exchange rate went above $1.30 for long, the company was “cooked”. He said the chances of this happening were almost nil.

    “Well, “nil” may be here. While Airbus has an order backlog of 2,177 aircraft worth $220.3bn, these delivery contracts are in dollars while costs are in euros. “This is the nub of the problem,” said Louis Gallois, the Airbus chief.

    “In 2004, the group was shielded by currency hedges at an average rate of $0.98. This year the rate is $1.12, and the hedges are expiring fast. Soon Airbus will face the full violence of the spot market. The aerospace champion is so deeply tied up with Europe’s sense of industrial self-worth that it will not be sacrificed lightly on the altar of free currency flows. When the French premier vowed to do whatever it takes to save Airbus, I believed him.”

  • Posted by Anonymous

    “Interesting article in UK’s Daily Telegraph today”??uh?

    Guess who gets the stick when a ceditor area goes deeper into savings?
    FTX don’t worry Europe savings rate is big enough to compensate for such doomsday scenario.

  • Posted by moldbug


    I know you know what you mean, but I still wish you or anyone would invent a new glossary for market intervention in which stability was not an antonym for volatility.

    Unless intervention agencies have more information than the market about what the “correct” exchange rate should be, a proposition I doubt even the most diehard neo-Keynesian would defend, any action that “dampens fluctuations” tends to create disequilibria.

    Of course any disequilibrium is stable, in some sense, by definition, but since the demotic bisyllabic synonym for disequilibrium starts with b, ends with e, and floats rapidly upward in a glass of beer, it’s hard to find this consistent with the political mission of the aforementioned agencies.

  • Posted by FTX


    The article seems far-fetched to me, but I think it serves as a reminder that the fate of the dollar won’t necessarily only lie in the hands of China, Japan and the oil exporters. The political response of the eurocrats to a sustained rise in the euro also needs to be factored in. The protectionist mindset is still pretty endemic over here.

  • Posted by Guest

    The US dollar is still used as the currency of account and denomination by the International Monetary Fund. So the world system is based on US Dollar hegemony, Despite the fact that some parts of the world are trying disengage from heavy implication in dollars, to other currencies, that’s no escape either. Because the entire world system is based on the ability to collect on the U.S. dollar! If you can’t collect on the U.S. dollar at parity, then you are bankrupt, too! Every nation in the world: China goes into a crisis. India goes into collapse. Every part of the world goes into a collapse, if the dollar goes down by 30%!

  • Posted by moldbug

    On another linguistic note, it distresses me that people still feel the need to use terms like “diversify” and “exposure” in relation to the PBoC.

    This terminology is only meaningful with respect to a private actor whose goal is to maximize profit. The PBoC is not a private actor. It is a political actor. Its goal is to support the PRC in whatever way that institution’s leaders see fit.

    Because both the RMB and the dollar are paper currencies, barring a shortage of wood pulp, the exchange rate between them can be fixed indefinitely by either party. Right now the party doing the fixing is the PBoC, and the RMB/dollar peg will shift when the PBoC wants it to. The PBoC can print as many RMB as it wants to buy as many dollars as it wants, and if there is anything that prevents it from holding $2 trillion, $5 trillion, $10 trillion, or $100 trillion, the proximate cause can only be found in Chinese political considerations.

    If you assume that an exchange-rate shift between the dollar and some other asset class, such as euros, mold, Hummel figurines, or whatever, would cause some political trauma to the PBoC or PRC, as they calculated how many Hummels China could have had if they’d gone on eBay Wednesday instead of today, words like “diversify” and “exposure” may be meaningful.

    But this is a political assertion which strikes me as improbable. If anyone has any evidence to justify it, I’d be very curious. (Jawboning from central bankers is not evidence.)

    If we didn’t want exchange rate prediction to be a continuation of politics by other means, we shouldn’t have built a system of pseudo-floating paper currencies. Since we do have such a system, I think it’s obtuse to pretend that we are discussing economics rather than politics. The analogy to Clausewitz is precise.

  • Posted by Guest

    Review of Economist Fred Bergsten new book on China’s Economy
    By Shih Smith

    Well, what can I say about this book? With a due respect to the authors, it’s the same usual American bias toward developing countries. As an American living and working in China for the past 2 years:

    1) It’s hard for Americans/Westerners to understand how business is done in China? Well, my answer it’s because China is NOT in the West and Chinese people are not westerners. Try learning the language first, be respectful, and don’t be so obnoxious.

    2) This is the same background noise that I hear from overeducated ideologues: If a country is not like the US, then it must be wrong. I guess the Chinese should have a 4th world healthcare system like the US?

    3) I personally feel that some authors wish China to stay poor so Madonna can have a concert there and pretend that the rich countries care. Well, maybe Africa, but not China.

    4) Just because you watch CNN, FOX, and NBC, it does not make you an expert.

    5) Just because you went to an Ivy League school, it does not make you a genius about other countries and cultures. As matter of fact, it does not make you a genius (period).

    6) US should comment less on China and worry about the pointless war in Iraq, the death of our soldiers for a pointless cause, broken US healthcare, outsourcing of the American jobs, broken borders, broken education system, children killing each other at schools…and the list goes on.

    7) China has many societal problems; please tell me something I don’t know. China is working on it, it takes time, and they are not magicians.

    8) If China is so awful please don’t do business there nor go visit there. For the fact that there are SO many foreign investments there, makes all business executives and companies from the West idiots.

  • Posted by Gheorghius


    We’re going to see new highs in the Euro/$ before year-end. I think at least 134. CBs are unable / unwilling to intervene to halt the dollar slide.

  • Posted by MCS

    Hmmm. Sounds like it’s possible that we may get the Europeans and the Chinese competing to support the dollar – essentially pegging the Euro to the dollar at around 1.30.

    I think the euro/yuan exchange rate will be the fly in the ointment, but trying to get my head around this may make my head explode.

  • Posted by gillies

    “economic analysis, when/if it is good, can be used to either trade, or to advice on policies, or to produce academic work.” (gheorghius)

    the traders get up every morning, go to the bourses, and do their trading, and the politicians do their politicking, but are the academics doing their job ?
    for instance – can anyone tackle these simple but significant theoretical questions :

    1. “can the printers of a fiat currency that is used as a reserve currency, print more currency than is actually required to lubricate world trade, and hold interest rates at a lower level than is justified by the current market conditions, without creating a surplus of savings somewhere – in the economies of exporters of manufactures, oil, or other commodities ?”

    2. “can the effect of a surplus of a currency that is used as a reserve currency be other than circular – in that the surplus will find its way back into the country of origin and – by adding to liquidity – further push interest rates below their otherwise natural equilibrium, generating more lending, more resulting imports, and yet more reserves in foreign central banks ?”.

    this is a formal academic way of asking the theoreticians –
    “what the hell else did you expect ?”.

  • Posted by Gheorghius

    Yuo wrote: “I did think the $ was massively overvalued v the majors in 01 — but at the time, I was at the TReasury/ IMF, not writing a blog.”

    Yeah, and so did I. In 2001 I advised an institution similar to the one u were serving at the time to switch all the country’s public debt into dollar bonds if/when the dollar would fall below 0.89 agaisnt the Euro (as it did a few months later). But that institution didn’t dare to do it. On the contrary, they even hedged the few dollar bonds they had issued with derivatives… halas for that country!

    How was it at the US / UK? Treasury, were they open to new ideas? Do you think it is dangerous to redenominate a country’s public debt in a major foreign currency (under the condition that there are no domestic inflation risks, the country’s domestic exrate is heavily undervalued and/or the major foreing currency is clearly very overvalued on a UIP basis)? Public sector speculators have had no great performance to my knowledgge in the past – CBs that tried forex speculation on major foreign currencies, derivatives, etc., generally lost (I have in mind some Asian CBs in the “90s). So I believe that they should not do it as a rule, unless… the misalignment is absolutely great and out of question.

  • Posted by gillies

    at least one poster to this site believes that a hegemonic reserve currency that is a fiat currency,confers the de facto right to print ‘free’ money. i do not agree. the hegemon has access not to free money – but to free credit.

    and even a forger of hundred dollar bills good enough to fool the authorities worldwide, would have to put a self imposed limit on his own activities. eventually the ‘victims’ of the scam also accumulate massive wads of $100s and the cat is choked with cream.

    take it to absurd extremes: at some point the market in some asset – gold, oil, tulip bulbs – will rise exponentially. it will rise faster than he can print. capitalism is ultimately a self correcting system. that is the good news . . . . . . and the bad news . . .

  • Posted by bsetser

    one point of information: the IMF does not use the $ as its unit of account. it famously uses the SDR. folks borrow in sdr and repay in SDR (SDR = basket of big currencies). It dedollarized a long time ago. The private sector never followed suit!

  • Posted by bsetser

    gheorghius –i worked for the US treasury, which, quite appropriately, was not considering borrowing in any currency other than the $ when i was there. hell, when i was there, the treasury was actually repaying its market debt.

  • Posted by Emmanuel

    Buying more Treasuries to forestall a (virtually guaranteed) dollar decline is ultimately futile. If I were China, I’d try to either (1) keep diversification efforts under the radar somehow until such a time that dollar holdings are substantially reduced or (2) precipitate a Wile E. Coyote moment of sheer Armageddon by selling off a large stack of Treasuries and leave the rest holding the bag with these IOUs and their built-in haircut feature.

    Get it on, China. Show ’em who’s boss in the global political economy, and it surely ain’t Sammy pimping his Treasuries. Dr. Roubini suggested as much sometime ago.

  • Posted by gillies

    gillies’ rule :

    the capacity of the values of a favoured asset class (for example tokyo property 1980s ?) to rise exponentially is theoretically sufficient to absorb, overwhelm, and exhaust all funds that are globally available to that market.

    gillies’ second rule :

    – and if it is not theoretically sufficient – they will prolong their exuberance until it is.

    gillies’ third and utterly final rule :

    – the capitalist system, or non-system, is able to absorb, and eventually correct, more liquidity than can be printed and airlifted by all the printing presses and helicopters in the world.
    – reality wins.

  • Posted by gillies

    gillies’ fourth rule :

    “the capitalist system is stronger and more resilient than any other economic system – including its own (current) rules.”

  • Posted by Guest

    No Chinese yuan revaluation for Bernanke or Paulson. LOL

    November 24 – Bloomberg (Yanping Li): “The People’s Bank of China…is focusing policy on withdrawing yuan from the economy after purchasing dollars, said Vice-Governor Wu Xiaoling. The current key monetary policy objective is so-called sterilization, Wu told a press conference… She said China’s trade surplus is largely caused by manufacturers building factories in China and a stronger local currency would cause them to relocate to other Asian nations.”

  • Posted by OldVet

    Interesting question: can/should CB’s diversify out of so many dollar bonds? As a mechanism, how about rapidly swapping dollars for home-country debt, and paying down Renminbi debt for example, or Singapore dollar debt? This would effectively exchange assets that are likely to depreciate for assets likely to appreciate. Most countries have a stock of domestic or local FX denominated debt, even if they run large foreign FX reserves.

  • Posted by Anonymous

    – Ms Wu deserves a round of applause for her candid clarity; “FDI for Treasuries” program.

    – gillies: “can the printers of a fiat currency that is used as a reserve currency, print more currency than is actually required to lubricate world trade..?”

    What is reasonable lubrification?

  • Posted by Gheorghius

    “…when i was there, the treasury was actually repaying its market debt.”

    Yeah, the good ol’ times when the US nation was growing stronger and was respected…

  • Posted by Gheorghius

    OldVet “how about rapidly swapping dollars for home-country debt, and paying down Renminbi debt for example, or Singapore dollar debt?”

    this would crush the domestic exrate peg

  • Posted by moldbug

    Gillies, anon,

    Since denomination does not matter, the only restriction on “reasonable lubrification” is the intrinsic divisibility of the currency. Probably the equivalent of about a US dime is somewhere near your minimum reasonable divisibility.

    Or did someone mean that more new currency has to be printed, minted, or just double-counted, on each and every day, lest heaven and earth collide, etc, etc? Is there… gasp… an inflationist in the room?

  • Posted by moldbug


    But I suppose it would be more fair if, rather than resorting to expletive and innuendo, I actually tried to answer your questions.

    For question 1, again, you appear to want to find a qualitative distinction between “more currency than is actually required” and some presumed alternative “no more currency than…” or even “less currency than…” Presumably you imply some function whose derivative is zero at this magic point of optimal printatiousness. Unfortunately I do not know how to design such a function, so I will try to answer the question without it.

    Consider Carl Menger’s theory of money. People hold money because they want to trade it for something else. In other words, money is an indirect medium of exchange. Fiat currency is the simplest and most elegant form of currency from this perspective, because almost no one is involved either in creating it directly, or in using it directly. (Except when the K Foundation burned a million quid.)

    So in other words there is an initial exchange of stuff for money, and then a second exchange of money for stuff. Call these E1 and E2.

    The demand for stock on hand of any commodity that is used as a medium of indirect exchange in this way will depend, of course, on how many people use that good, and on how often they use it. Since this is integrated over time, however, it depends to a considerable extent on the time period between E1 and E2. Though there is no quantitative or qualitative boundary that can be drawn, we can think of exchanges in which the time period between E1 and E2 is short as cases of “spending,” and ones in which it is long as “saving.”

    “Saving,” of course, in this sense, is the same as what some people call “hoarding.” It does not mean that the money is traded for some other instrument, such as a loan note. When money is exchanged for a security, it does not alter the amount of money in existence, unless some kind of double-counting convention is used. Both before and after the transaction, one party has the money and one party has the security. The demand for money – ie, the quantity of said commodity held for the purpose of indirect exchange – is thus unaffected.

    The general effect of a commodity that is being used as a medium of indirect exchange, especially if it is used more as a medium of saving than of spending, is to increase the “reservation demand” for this commodity. Since supply and demand are what they are, this will increase the price it commands in exchange for other articles. In other words, relative to another commodity with comparable production and consumption profiles, it will appear overstocked and overvalued.

    For example, if you look at the current global warehouse stock of gold, you notice that it is much higher not just in physical mass, but in relative price, than the current global warehouse stock of platinum. In other words, at current exchange rates, you could trade a small percentage of all the gold in the world for all the platinum in the world. Similar ratios apply when we compare, for example, US dollars to Confederate dollars.

    There is no objective way to look at the physical properties of gold versus platinum – never mind the dollars – and tell which is more likely to be used as a medium of exchange. Clearly both fit the usual criteria of portability, stability, divisibility, etc. But the decision is purely path-dependent. It is dependent, in other words, on historical events.

    An important motivating factor in choosing a commodity to use as a medium of exchange, therefore, is one’s own expectation of who will join you in the use of said commodity.

    This has two advantages. One is obvious, which is that you and everybody else can pay each other in this commodity. A sort of network effect. The importance of the effect varies as to the general efficiency of moneychanging, etc.

    The other effect is that if you are injecting indirect demand into some commodity, you want either nobody else to be using the same commodity as a medium of saving, or everyone else to be using it. Since it is always hard to be sure you are the only one following your own strategy, the system is in equilibrium when, and only when, everyone picks the same commodity and sticks with it.

    The result is typically that the word for this commodity will be the word for “money” in whatever language you may happen to parlay-vous in.

    Other commodities will follow the normal valuation rules of supply and demand, in which inventory buildup is a clear sign of any kind of mispricing. Money, on the other hand, cannot be mispriced by definition, since it provides the reference frame for all other calculation.

    This is basic game theory. It is very stable for your market to use either gold or platinum, or even a standardized alloy of gold and platinum. It is not stable for some people to use gold and some to use platinum, because any perceived imbalance will make it rational to flock to the winner. If they get in before the rest, the people who are inevitably forced to follow will drive the value of their savings up further. A classic drinking-game scenario. This instability was responsible for the fate of the bimetallic Ag/Au monetary system of the 19th century.

    One important caveat in this model is that a commodity will be disfavored as a medium of indirect exchange if is not hard – that is, to the extent that its supply curve is less than vertical. For example, the proprietors of the US dollar are very diligent about deploying their various items of military hardware in the direction of anyone who independently augments the supply of that item. If they became lax in this duty and allowed anyone to print dollars, the value of the dollar would fall to the cost of physical production, and dollar holders would instantly trade their dollars for any good that could be eaten, burned, or used as a weapon.

    So the reason people are buying commodities and stuff is that because the proprietors of said instruments, and its rectangular brethren, are not too good at stopping themselves from producing them. Nor do they seem inclined to employ their admittedly impressive armaments to this end.

    Therefore, indirect demand tends to find its way into other goods. If these other goods are natural currencies, if they can absorb total global indirect demand without springing a huge leak of their own (transmutation, asteroid mining, etc, etc), they will tend to develop a feedback cycle in which they form a new monetary equilibrium. Otherwise, they are simply likely to go up and back down as people notice that by traditional measures of commodity pricing they are overpriced. If price appreciation in these goods tends to generate significant new supply, the effect will be especially nasty (think Thai skyscrapers, Florida swamp condos, etc).

    However, all of this stuff is dependent on the vagaries of public fashion, and none of it is anywhere near beyond the influence of government.

    For example, it is clear that residential housing has absorbed a heck of a lot of savings in the US today, because it has often been priced far above its production cost. In some places, restrictions on new property building make real estate an almost plausible “pseudocurrency,” acting as a sort of anti-counterfeiting protection. Of course the various tax subsidies do not exactly hurt either. Specie, on the other hand, gets the short end of just about every tax stick.

    So why is the price of specie going up and the price of residential housing going down? Perhaps just because it’s more common to borrow money to buy the latter than the former, and the price of money has been rising. Perhaps because of the ugly deflating effect of overproduction. But really, it could just be a matter of fashion.

    Hopefully all this should make it obvious that my answers to questions 1 and 2 are both “no.” No fiat currency is safe when alternative natural currencies exist. And any currency can be analyzed on a global basis – borders are not qualitative.

  • Posted by moldbug

    Sorry, I meant “may tend to develop a feedback cycle” rather than “will tend to…” Obviously the point is that there are multiple such natural currencies, and “there can only be one” – so they can’t possibly all be involved.

  • Posted by bsetser

    Old vet — China cannot swap $ for domestic debt. It can use its dollars and euros to finance either a) a domestic capital outflow or b) a current account deficit. If china’s government started to run large fiscal deficits, drawing down domestic savings to the point that China started running a current account deficit, that deficit could be financed by running down china’s existing external assets (rather than selling new external debt).

  • Posted by Maverick


    ” CB’s unwilling/unable”? Why? One would think that protecting the value of their massive $ holdings will be a paramount factor in any of their strategies. I would guess that both PBoC and the Gulf States are feeling the pain as we speak.

    What about the ECB? Will it not be absolutely compelled (political pressure included) to intervene and protect EU’s exports/growth? It seems to me that an EUR/$ rate of 1.34 will be a harbinger of carnage to come in EU exports (hedging can only get you that far). Markets seem to agree since shares of all major EU carmakers lost about 2% on Monday.

  • Posted by Rueff


    “This is basic game theory. It is very stable for your market to use either gold or platinum, or even a standardized alloy of gold and platinum. It is not stable for some people to use gold and some to use platinum, because any perceived imbalance will make it rational to flock to the winner”

    insightful: replace gold by RMB and platinum by USD.

    “It is very stable for your market to use either RMB or USD, or even a standardized alloy of RMB and USD. It is not stable for some people to use RMB and some to use USD, because any perceived imbalance will make it rational to flock to the winner” => the constant inflow into the RMB leaves no doubt which is the winner.

    Is the “dirty RMB-USD peg” the ultimate Maginot line of the financial system?

  • Posted by Gheorghius


    – We’ll see where currency mkts go
    – You seem to have a mercantilistic approach. In other words you seem to miss the following. An appreciating currency reduces net exports but boost domestic investment and consumption, provided the country has sound investment/macro environment. (Hint: K inflows). Witness the US in recent years with all its bubbles.
    – The ECB is more concerned about inflation right now, certainly they wont enslave monetary policy to exrate targets – unless the exrate move becomes truly large.
    – The ECB is also concerned about globalimbalances – that boils down to preventing the US foreign debt from becoming too large – and wants Asia to appreciate, but is prepared for a higher Euro/$ rate.
    – Countries who hold massive dollar reserves are unable to prevent the dollar from sliding against all currencies but their own, because their firepower is minuscule compared with private participants in the forex mkts.

  • Posted by Maverick


    Thank you for your valuable input.

    I do wonder about this elusive creatute called the European consumer, although its existance is assured, its potential for consumption does leave a lot to be desired. Unlike the US consumer, the dream of every company, the EU consumer seems almost sedated by comparison. I doubt if a stronger Euro promotes much higher consumption. Factor in unemployment, over exposure to credit via mortgages (eg. Spain, Ireland the two prime examples although Britain and Italy not far behind) and the forecast looks somewhat grey.

    What do you think of the British Pound? Any chance of a further increase against major currencies?


  • Posted by moldbug


    No, I don’t think the RMB/USD peg or dirty float or whatever you want to call it is the ultimate Maginot line.

    Fiat currencies, because they are really just commodities whose manufacture is under political monopoly control, are not exceptions to a Mengerian analysis. But because savers in fiat currencies have to consider the actions of these political actors, as opposed to just the actions of other savers like themselves, the game theory is very different.

    Take your example with RMB/USD. Clearly the undervaluation of RMB makes RMB a good place to store wealth for dollar holders – that is, if we accept the common perception that RMB/USD is likely to rise over time.

    But this perception is based on “free-market fundamentalism” which assumes that the PBoC’s interventions will have to decrease over time. As we’ve seen, there is no reason for this to be true, and the PBoC has a specific political incentive to keep the rise in RMB below a level which would overwhelm its capital controls.

    Suppose this were not true. If the PBoC destroyed its printer and could somehow credibly renounce all future production of RMB, capital controls would be pretty useless. I suspect that RMB would, as you describe, become the new global currency within weeks. And they would be some pretty exciting weeks.

    Of course this is impossible, and this is where your Maginot line analogy breaks down. At least from the perspective of competitive devaluation, if fiat currencies are Maginot lines, they are Maginot lines that work.

    A more interesting possibility from the Chinese perspective, although still politically unlikely to say the least, would be an instant conversion of RMB to gold. Suppose each RMB in the Chinese money supply was converted into a redeemable receipt for an equivalent fraction of the Chinese gold reserve.

    Like the above, this would force the entire planet onto the gold standard. Unlike the above, it is a practical plan that could be instituted tomorrow. And especially unlike the above, its effects on productive activity in China would be stimulative rather than narcotic.

    China has a lot of RMB and the rest of the world has very few. So forcibly switching the planet from dollar hegemony to RMB hegemony, by making the RMB an ultra-hard, gold-like currency, would tend to convert China into a US-like nation of rich, lazy, neurotic consumers. China’s political system could probably not withstand this, which I think is one reason they tend to pursue the opposite policy of aggressive undervaluation. I don’t think PRC officials see the matter in at all this cynical of a way, but the fact remains that ascetic, hard-working, hard-saving citizens are much easier to manage.

    But China, at least by US and European standards, has very little gold. The rest of the world has a lot. A global snap conversion to gold, therefore, would tend to decrease the RMB/USD rate rather than increasing it.

    So a snap conversion to gold, which the US, Japan and Europe would have no choice but to follow, would accomplish three Chinese goals. It would be a substantial devaluation of the RMB, stimulating productive activity in China. It would result in an intrinsically stable international monetary system. And it would clearly establish the PRC as the planet’s most powerful and prestigious nation.

    Remember, you heard it here first 🙂

  • Posted by gillies

    “What is reasonable lubrification?”

    reasonable lubrification maximises trade without fuelling asset bubbles. now you will ask me what a bubble is. anything which bursts can be presumed to have, formerly, been a bubble.

  • Posted by gillies

    moldbug, your offering is above my head. i cannot engage at that level of economic expertise for i am not an economist. i have, however, printed and issued a currency. this was an experiment for a one day fair, some ten years ago.

    my method was to issue an equal amount to each participant. this was for appearances sake. as ‘hegemon’ i might just as well have spent all the money into circulation, but people would have seen this as a ‘rip-off’ just as people do with the real world – (or imagined, if you insist) – dollar hegemony.

    the amount of currency that we issued – you can call it about 50 dollars per participant – was based upon common sense. as you point out, a certain measure of divisibility was desirable. the fair would not have worked with a single note of 2500 units/dollars. nor would the fair have worked with wheelbarrow loads of small denominations, a la weimar republic. i denominated the currency in 1s,2s,4s,and 8s – which is logical. much more rational than 1s, 5s, 10s?

    but surely most money is lent/borrowed into circulation ? printed cash is a minor issue. so it is the depressed interest rate, not a hyperactive printing press, which inflates the bubbles.

    the ponzification of asset markets leads to a society in which the heroes are not warriors or nation builders – but gamblers.

    if i had cheated and made an extra printing on the side – i would have had a brief free run – but rapid inflation would then have occurred, and those who sold to me at escalating prices would have in turn created further inflation in something at the fair – or sat upon increasing reserves.

    it seems to me that america is possibly stuck up the euphrates economically as well as militarily. keep interest rates depressed, and the bubbles will resume. raise the interest rates and the chinese and other cash rich players will get even richer ?

  • Posted by moldbug


    I am not an economist either. I am just a computer scientist who finds this stuff interesting. Everything I know I learned on the Internet.

    But in my own personal opinion, the idea that economics is some kind of esoteric exegetic doctrine, like the Talmud, the Hadith, or the Spiritual Exercises, is one of the most pernicious ideologies of the 20th century. Or at least, if you look at any such ideology, you will find this idea resting calmly, like a white Persian cat, on its lap. Economics is just a matter of clear thinking, and if you do lack the intellectual confidence to pick a school of economics, learn it, and follow it, you should blame not yourself, but whoever were the teachers who chose to install this inexplicable inferiority complex in an otherwise healthy and intelligent young brain.

    That said, I don’t disagree with anything you say and I like your binary currency idea. You might find it interesting to follow this idea and see where it takes you.

    Note that when I talked about new money being created, I said “printed, minted, or double-counted.” Lending money into circulation counts as either printing (the Fed prints dollars and lends them) or double-counting (the Fed authorizes lenders to keep their dollars and lend them too, ie, mismatch maturities, as part of the normal practice of fractional-reserve banking).

    Also, raising interest rates will lower the market value of any bonds held by the Chinese, not raise it. Though if you believe this consideration is not particularly important to their decision-making process, I’d certainly agree.

    In any case, one, there are worse consequences of raising rates than enriching the Chinese. And two, short-term interest rates do not by any means effectively describe the amount of liquidity that is entering the system. Despite his eccentric, James Frey-like approach to initial punctuation, Doug Noland’s Credit Bubble Bulletin is an excellent weekly overview.

  • Posted by gillies

    “Also, raising interest rates will lower the market value of any bonds held by the Chinese, not raise it. Though if you believe this consideration is not particularly important to their decision-making process, I’d certainly agree.”

    yes, i understand that. but the chinese seem largely disinterested in capital losses / gains ? i was thinking that if they invest $x billion every month they will get more bonds, and more interest, as the bond prices fall.