Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


Still going strong (Bretton Woods 2)

by Brad Setser
March 25, 2007

I watch the data on global reserve growth rather closely.  But I was still surprised to see (via Global Liquidity Blog) the scale of the growth in the Fed's custodial holdings so far in the first quarter.   Since December 27, the securities the Fed holds on behalf of foreign central banks have increased by $123.8b.   That is only through March 22 – so it isn’t for the full quarter, and it still is quite an increase.  Annualized, it works out to a stronger increase than in 2004, back when Japan seemed to be buying every Treasury note the US government issued.

And the Fed data just covers the securities — the safe securities — held by the Fed.   Central banks also hold Treasuries and Agencies through private custodians.    They have dollars on deposit in the US.  They have dollars on deposit outside of the US.  They hold "private" mortgage backed securities (especially the PBoC).  They hold corporate bonds (though not very many) and so on.

Stephen Jen puts global reserve growth at $75-80b a month, which seems about right to me.  China has not reported data for January or February, which makes it hard to know.   But Brazil (reserves up $6.5b so far in March), Russia (reserves up $6b in the first two weeks of March) and India (reserves up a billion last week — which is nothing compared to February) are all set to add $20b or more to their reserves in the first quarter.  Japan's reserves should increase by $10b a quarter on interest payments alone.  China is probably good for $20-30b a month even in q1 — or $60-90b for the quarter, though a lot depends on what China decides should show up on the balance sheet of the PBoC and what will be placed elsewhere.   

Throw in the growth in the assets of the oil investment funds, and official institutions — central banks and oil investment funds — probably added $250b to their assets in q1.  That is roughly a $1 trillion annual pace, nearly all from the emerging world.    There shouldn't be much doubt left over who finances the US current account deficit.

Mohammed El-Erian and Michael Spence noted the uphill flow of capital in their Saturday Wall Street Journal oped as well (Hat tip, Mark Thoma).  They argue that the rise in US consumption is a natural consequence of the rapid increase in the value of US assets over the past few years — along with the emerging world's willingness to finance the resulting US deficit.   The uphill flow of capital, in turn, explains most recent asset market conundrums (“excessive compression in risk spreads, the unusual collapse in market volatility, the inverted shape of the U.S. yield curve”).  

I basically agree.  But I am not sure, though, that the story really starts with an increase in the value of US housing stock, which in turn leads US households to naturally want to consume more and thus borrow from the rest of the world – something that the rest of the world accommodated.  El-Erian and Spence write:

“In an open economy like ours however, we can consume and invest more than output by importing more than we export — and we did. Hence the trade deficit. However, this ability is dependent on the ability and willingness of the rest of the global economy to accommodate the US desire for higher consumption by investing in the US.”

To the extent  extent low rates in the US and Europe have stimulated the increase in home prices — and to the extent these rates reflect an increase in the pace of central bank reserve growth in the emerging world — the world’s central banks didn’t so much “accommodate” higher US consumption as “induce” higher US consumption.

Consequently, my telling of the story would put a bit more emphasis on exchange rates – and how the impact of dollar pegs on emerging economies changed when the Fed cut US policy rates and the dollars started to tumble in 2002.   Asian currencies — as Jon Anderson has demonstrated – still track the dollar closely.   So the change in the dollar's trajectory against Europe had a big impact on Asia, not just on the US.

There is a debate about what changed to propel the big surge in Chinese savings — and the resulting surge in China's current account surplus.  Chinese investment growth has been very strong; savings growth was just stronger.   But something clearly did change.  In 2000, China wasn't a big source of financing for the world.   In 2006, it was.   That shift presumably explains a lot of different financial market conundrums.  


  • Posted by Qingdao

    rather than “solve” the Chinese savings puzzle, I’d like suggest another. David Dollar recently wrote:”. . .the reservoir of $1T in reserves means that there is virtually no chance of a serious crisis.” By which (I assume) he means domestic Chinese crisis. John Makin writes: “If China continues on its current path, we may expect to see a crisis erupt sometime in the next twelve months comparable to the Asian currency crisis that erupted in 1997.” (

  • Posted by Dave Chiang

    Thanks to Alan Greenspan, the US Economy has been living on a Financial Credit Bubble for far too long, creating massive global economic imbalances. The US Economy “too big to fail” thesis will soon be challenged by the looming financial crisis in subprime and Alt-A mortgages that will turn into a serious liquidity problem for the US consumer. The heavily indebted US consumer is toast.

  • Posted by Macro Man

    “I basically agree. But I am not sure, though, that the story really starts with an increase in the value of US housing stock, which in turn leads US households to naturally want to consume more and thus borrow from the rest of the world – something that the rest of the world accommodated.”

    That is it. China, Russia, Korea, Gulf States et. al. have lowered the supply curve for credit. Thus, with no shift in the demand curve, US consumers have rationally borrowed more than they otherwise would have. This is also the reason risk premia/spreads are so low, because this lower-than-private-sector supply curve is not just about lending to governments.

    Yes, the superaccomodative monetary policy from the Fed in 2003-2004 is also to blame, but Voldemort and co. pervented the Fed from withdrawing it when they wished to do so.

    There is little to suggest, however, that the D Chiang Armageddon scenario will come to pass. Debt service/income ratios are lower than they were a few years ago, and net worth/consumption ratios are at their highest point in the last 50 years, with the exception of the equity bubble period. The subprime “crisis” is being felt in areas with relatively high unemployment; default rates in the areas with the largest house price appreciation over the last several years have remained low.

    There remains little evidence other than faith that housing can, by itself, cause a recession.

  • Posted by Macro Man

    As an aside, Brad, do you know how Chinese savings (as a % of GDP) is split between the public and private sector?

  • Posted by RebelEconomist

    John is not Makin much sense to me either!

    Qingdao points out the inconsistency of warning of an Asia-crisis-style RMB currency crisis while complaining about China’s reserve accumulation. That is not the only inconsistency in John Makin’s testimony to the Senate Finance Committee.

    John Makin also complains that the nominal appreciation of the yuan against the US dollar has been partially offset by lower inflation in China, but later warns that inflation in China would be destabilising.

    He explains why it is rational for the Chinese people to save a lot and that they do so in bank accounts, makes the link between this and government saving abroad in reserves, and then argues that M2 balances represent suppressed inflationary pressure.

    He argues that if China removed capital controls, RMB would flow out of China and pressure on the yuan to appreciate would ease. Many would disagree that liberalising the capital account would lead to a net outflow, but presumably if it did, the yuan would depreciate.

    He worries that disruption in China would jeopardise Japan’s recovery, but calls for China to end a policy that he claims has led to a 15% undervaluation of the yen.

    One despairs that such people are invited to advise politicians. Or is it that the politicians know what their constituents would like to believe, and just want someone with some academic economic respectability to confirm it?

    As Gheorghius says, China can determine its bilateral exchange rate with the dollar, but not the dollar’s overall value. If the American government do not like that, then they are free to intervene themselves in other currency pairs. But they have not been so unhappy with low interest rates and booming consumption.

    Americans should be reminded that their own (not quite English) word for debt repayment is “honor”.

  • Posted by Dave Chiang

    From Bill Fleckenstein, why has the Federal Reserve, pandering exclusively to Wall Street speculative Hedge Funds, repeatedly ignored the damaging role of Asset bubbles to the US Economy. Bill Fleckenstein writes, “Asset bubbles are harmful for the same reason high inflation is: Both create misleading price signals that lead to a misallocation of economic resources and sow the seeds for an inevitable bust. The unwinding of today’s housing bubble is not merely an academic question: It is likely to inflict real hardship on millions of Americans. To reduce the risk of a similar outcome in the future, it is important that policymakers, economists, and policy analysts correctly diagnose the root cause of the current housing bust, not just its symptoms.” The rising global uncompetitiveness of US Economy is directly correlated to the misallocation of capital and overinvestment in the Housing market. If the trillions of dollars squandered on Housing speculation were invested in Industrial R&D, Education, and Capital investment improving real US productivity, there simply would not be the endless debates over Bretton Woods 2 and the “China threat”.

  • Posted by Gheorghius

    “Thanks to Alan Greenspan, the US Economy has been living on a Financial Credit Bubble for far too long, creating massive global economic imbalances.”


    Relative to what? To what benchmark? To global GDP? To the GDP of debtor countries? GDP corrected by the dimension of global financial mkts? And: are they greater today than in 1981? Or 1975?


  • Posted by Dave Chiang

    Saudi petrochemicals group eyes GE plastics for $12 billion

    Saudi Basic , the largest public company in the Middle East, is lining up a bid for General Electric’s plastics division in a deal that could be valued at up to $12 billion.

  • Posted by bsetser

    macroman — the best data on the composition of Chinese savings comes from kuijs. the biggest contributor to the recent growth in chinese savings is the corporate sector — but that includes a lot of state firms.

    dollar v makin. dollar is i think arguing that $1 trillion means there is little risk of an external run a la Asia 97. he is right. makin is arguing that the expansion of domestic credit, domestic stock bubble and overinvestment in the external sector will create a different kind of crisis. he too might be right. i recommend — surprise, surprise — my own paper on the topic which was published be CESIfo (title “the Chinese conundrum: External strength, domestic weakness”)

  • Posted by Emmanuel

    Macro Man–I line up with Dave Chiang, Nouriel Roubini, and many others in being less sanguine than you are. To be fair, do give us your source for the assertion that American consumers are doing better now than in the past. From the Federal Reserve data, both the debt service ratio (DSR) and financial obligations ratio (FOR) are at their all-time worsts as of Q4 2006.

  • Posted by moldbug

    It takes two currencies to make a currency run.

    The EM crises of the past were the result of governments diluting their currencies to the point at which they became uncompetitive with the dollar as a medium of indirect exchange. This historical pattern is responsible for the usual mental association between expansionary monetary policies in emerging markets, local boomlets in the classic inflationary pattern, and eventual busts with capital flight back to the dollar.

    I suspect that anyone who is waiting to see this pattern reemerge (except perhaps in Iceland!) will have to wait quite a while.

    The old crisis pattern occurred because EM fiscal and monetary policy was relatively loose compared to its US equivalent. Today, EM fiscal and monetary policy is absolutely loose – in terms of dilution rates (liquidity growth) – but, compared to the US and Europe, it is relatively tight. Money supply growth in China certainly still exceeds the US dilution rate, but the growing prosperity of China (= increasing demand for money) seems to more than make up for this.

    In other words, for most of the second half of the 20th century, the fiscally sound monetary systems of the First World – mainly the US – provided the financial equivalent of a fire sprinkler for the emerging world. When local monetary policy got out of control and entered an ascending inflationary spiral, the currency-competition fuse melted down, capital flight ensued and the country got a cold, wet lesson in the dismal science.

    Now that political sclerosis and generational imbalances have placed the First World currencies on an apparently unstoppable path to indefinite monetary expansion ($50 trillion in unfunded US entitlement liabilities, etc), the sprinkler system is greatly weakened. The roof tank is not entirely dry – dollars can still be traded for many delightful goods and services – but it is, arguably, no longer the primary constraint on the expansion of money and short-term credit in emerging markets.

    If this is true, it is interesting to speculate on what that primary constraint might be. Is it “inflation” in the modern sense of the word – that is, political opposition due to visible price increases in various consumer goods? Or could it be competition from an entirely different class of currency – for example, a natural one?

    Certainly, the argument that the direct demand-reducing effect of a recession in the US will be sufficient to cool China strikes me as very unsound. Suppose US demand for Chinese goods falls by 5%? Suppose the PBOC decides not to let the RMB appreciate by that same 5%. Or even depreciates it! The point being: because they are not under the gun of currency competition, they have all the flexibility they need.

    Basically, the PBOC is in the business of creating RMB to buy the products of Chinese industries. As far as it affects anyone in China, the fact that these goods are shipped to Long Beach and exchanged for US dollars, then for dollar bonds, is irrelevant. They might as well be dumped into the ocean. Or the dollars, or the bonds, could be burned. The comparison to a wartime boom, in which the goods produced are never used in the producing economy, is precise.

    Now, if the severity of a US recession was such that dollar liquidity actually fell – an effect that, since there are far more dollar claims in the world than actual dollars, could be easily and quite dramatically produced by “throwing away the printer” – the force of currency competition would return. A credible promise of zero future dollar creation would send everyone in the world, China included, rushing head over heels for the Benjamins.

    But it would also make Manhattan look like Sadr City, which is why it’s probably not going to happen. In fact, given political realities, a recession in the US seems more likely to lead to looser US policy and a relative weakening of the dollar. If China deflates, it will be for Chinese reasons.

  • Posted by bsetser

    sure, if China deflates, it will be for Chinese reasons (i.e. overcapacity). but isn’t the current bout of inflation and the like in the emerging world different from inflation in the past in the sense that it is a component of real exchange rate adjustment in the context of pegs that have left many EMs undervalued? that seems different than inflation that comes from running the government printing presses to cover a big fiscal deficit. right now, the pressure for inflation seems to be coming from money growth tied to rapid reserve growth, not money growth tied to big fiscal deficits (see China, the oil exporters, etc)

  • Posted by moldbug

    Sure – the effect of monetary dilution can be very different depending on which direction the firehose is pointed in. Which is another way to say “where the pressure is coming from.”

    But it is perhaps easy to overstate the difference between using new money to peg an exchange rate and using new money to fund your new Ministry of Sport. The one seems very 21st-century and the other is very ’70s, but in both the apparent profitability of an enterprise is dependent on the subsidy.

    You could calculate the subsidy that the RMB peg has provided to Chinese exporters if you could calculate the free-market exchange rate between RMB and the dollar. Of course this is not calculable, which is the whole point of intervention. But if China refrained from intervention and instead printed new RMB, courtesy of a big fiscal deficit, and used it to fund something like an export tax credit, the effect (within China and excluding PBOC dollar holdings) would be the same.

    If you think of the Ministry of Sport as a (presumably quite unprofitable) business, you can approach the same analogy from the opposite direction. Ministry of Sport, Ministry of Export – same deal.

    The key difference is just that, when you spread diluted money around a large, growing economy and condition it on actually producing something of value, you get a much healthier and more helicopter-like distribution of the new funds, than if it all goes to the former revolutionary comrades now manfully manning their new desks at the Ministry of Sport.

    From the perspective of “inflation” in the sense of a consumer price index – a political indicator rather than an economic one, but no less important for that – this improved distribution, which is both more Hayekian and more egalitarian, may actually be more problematic than if the money just goes to the aforementioned fat cats. Nobody riots when the price of Bordeaux goes up.

    But in both cases you’re doing strange and awful things to the market for savings. A monetary system is stable when there is no other asset whose forward price curve exceeds the cost of forward money. Keeping X under your mattress should not yield a predictable profit relative to keeping money under your mattress. Otherwise, holding money is an irrational act. When you have 20% more money every year and your long-term rates are 5%, you have a lot of goods whose dilution-adjusted price has to drop very steeply for a very long time to avoid looking like obvious candidates for X.

    Money is the bubble that doesn’t need to pop. As long as there is demand for indirect exchange, at least one asset will be stockpiled by hoarders, hence experience demand that is not a consequence of any direct utility, hence be overvalued. As long as the storage cost for this asset is zero and the supply in existence is fixed, you have a perfect Nash equilibrium – using any other asset as a medium of indirect exchange provides no advantage, and runs the risk of buying into a bubble which will subsequently pop as punters revert back to the stable standard. If the forward price of an asset which is not the current monetary standard, but which is theoretically capable of replacing it, rises to exceed the forward price of money, arbitrageurs will buy it now and store it, increasing the present price and closing the gap between the yield curves.

    Even bimetallic (eg, gold and silver) monetary systems are unstable in this simple model – even without any attempt to fix the bimetallic ratio. It is a pencil balancing on its point. If demand shifts for some random reason toward one of the metals, it creates a feedback loop which will eventually demonetize the other, leaving it priced by normal commodity criteria.

    Of course this is a simplistic free-market approach and does not even begin to describe the complexity of the various official forces acting on monetary systems. It just provides some indication of the pressures lurking under the surface.

    Besides of course the “export subsidy,” I suspect that some of the industrial overcapacity in China (not to mention residential overcapacity in the US!) is a result of this blind search for a replacement currency, a savable good whose price will not underperform the dilution rate. Factories, especially, make a very bad monetary system, because they are not inelastically supplied (unless you have a very large supply of environmentalists). But the supply of shares on a stock market may not respond instantaneously to an influx of demand, temporarily mimicking an inelastically supplied commodity, and tricking the unwary.

  • Posted by Macro Man

    The degree to which Chinese exports will be impacted by a US consumer recession (in the uncertain even that one should come to pass)is, in all likelihood, dependent on the ‘ownership’ of those exports. This data, of course, is diffficult to get. However, if a substsntial portion of US imports from China are essentially intra-firm imports (i.e, US firms outsourcing production to a lower cost area), then the impact could be quite substantial, currency be damned. I believe this to be the case, both anecdotally and via the scant data that exists, which suggests that 40-50% on US goods imports are intrafirm.

    In a sense, outsourcing means that the US might finally be able to export something to China- a slowdown.

  • Posted by bsetser

    China’s direct exposure (intra-firm, not — I am not sure it matters who owns the supply chain)to the US market is rising, as Chinese exports to the uS are growing faster than its economy. But over the past few years, Chinese exports to europe and the emerging world have grown far more rapidly than Chinese exports to the US. That certainly was the case in 06. So if the pace of Chinese export growth to the US slows to say 5-10% and exports to the oil exporters are growing at 50%, China might pull through … and of course if China started exporting to China, that would help. The biggest potential resevoir of consumption growth right now has to be China itself, given how low a share of GDP consumption now is (40%, roughly)

    the scenario that worries me is one where US imports from China rise substantialy (China is moving upstream, into autos, into chips, into machinery …. i.e. sectors where there is still industrial production in the US) during a slowdown. The politics there would be brutal. And China might not be willing to take policy steps to slow its exports if a US slowdown combines with an investment slump inside China. Investment is a huge share of Chinese GDP, and it can be volatile. China may want to try to export its way out of an investment slump. That strikes me as politically impossible given the size of China’s pre-existing surplus.

  • Posted by moldbug

    Whether inter-firm or intra-firm, the dollars have to be converted into RMB before they’re spent in China. Because the absence of meaningful currency competition gives China all the room it could possibly need to devalue the RMB or suppress its ascent, it controls this lever and can (in theory) compensate for the impact on its own economy of any conceivable slump in US demand.

    Of course this is theory, not practice. I defer to bsetser’s description of the actual politics. “Brutal” is probably right.

  • Posted by Guest