Print
Email
Share
Facebook
Twitter
LinkedInIn a comment in the Financial Times about ten days ago, David Hale argues that there is little need to worry about either the dollar or the US current account deficit. The US has no shortage of financial assets to sell the world, and the world has no shortage of savings to lend the US.
True enough.
The question is whether the rest of world – or, more precisely, private investors in the rest of the world – want to lend that savings to the US and in the process buy US financial assets in the process.
Right now, in my judgment, the data says that they don’t. That is what currently worries me.
Best that I can tell, there has been a very significant fall off in private demand for US assets over the past two quarters. By private demand I mean net private demand – that is foreign demand for US assets relative to US demand for foreign assets.
The US hasn’t noticed this fall off because foreign central banks have stepped up their purchases of US assets dramatically, offsetting the fall in private demand.
The difficulties distinguishing private and official flows, along with lags in the data, make it hard to point to a single data point that illustrates the fall off in private demand. It is easier, in fact, to demonstrate the surge in official demand, since the US TIC significantly overstates private flows and understates official flows (The growth in official holdings reported in the last survey suggests that official flows were about $140b larger than reported in the TIC data).
We do know that central bank reserve growth rose to around $250b in q4 ($1 trillion annualized), stayed around there in q1 and look similar in q2. Total dollar flows are harder to measure, but there is no doubt foreign central banks have increased their purchases of US financial assets over the past two quarters. The Fed’s custodial holdings rose at an annualized pace of close to $400b in the fourth quarter of 2006, and then an annualized pace of over $500b in the first quarter of 2007 (April has been a bit slower). I would bet that Fed’s holdings understate annualized inflows by somewhere between $100-200b ($25-50b a quarter).
In all probability, central banks will continue to provide the financing the US needs that the private market no longer wants to supply. The market certainly doesn’t seem all that worried by the shift in the composition of flows to the US.
But it seems to me that the risks that the official sector will balk at providing the US financing on the scale it now needs – while small — are once again growing. Most central banks have made it clear that they now have more reserves – and certainly more dollar reserves — than they need. More central banks seem to be finding it more difficult to reconcile fast reserve growth with their domestic goals at a time when changes in global markets have increased –not decreased – the scale of reserve growth needed to finance the US deficit.
David Hale argues that those who focus on asset markets are less worried by the US current account deficit than those who focus on the size of the US trade and current account deficit, because “the US has a vast capacity to attract capital.” In some sense that is a truism – you cannot run a current account deficit without the ability to attract capital.
But I would argue my rising concerns about the financing of the US deficit are very grounded in concerns that the relative attractiveness of US assets has slipped a bit, making it hard for the US to attract the kind of private flows needed to sustain a large current account deficit.
On a backward looking basis, US assets haven’t performed all that well over the past five years or so. Euro and pound denominated bonds have done much better that US bonds, in dollar terms, because of the currency gains. Foreign equity markets have outperformed the US equity market. <!–[endif]–>
US equity markets have rallied recently – but the rally seems at least partially due to a fall in supply. Private equity firms increase the supply of debt and reduce the supply of equity, helping to match the world’s supply of financial assets to central bank demand. I don’t think there is much evidence of a surge in foreign demand for US equities – and certainly not for a bigger surge in foreign demand for US equities than US demand for foreign equities.
Sure, US firms have euro denominated revenues that are worth more in dollar terms. But European firms have even more euro and pound denominated revenues that are worth more in dollar terms. No wonder foreign equity markets have done better.
Hale notes that foreign investors own a relatively small share of US equity markets — both absolutely and in comparison to the share of foreign ownership in other markets.
Very true.
But Americans also have a relative small share of their wealth in foreign equities. What matters is the relative attractiveness of foreign equities and US equities. To generate the net inflows needed to finance an external deficit, foreigners need to find US equities more attractive than Americans find foreign equities. That hasn’t happened for a while.
Most inflows to the US have come into the US bond market. Those buying bonds are making one of three bets: that the US dollar will rise, that long-term US interest rates will fall or that credit spreads (if you are buying corporate bonds) will fall.
But US long-term rates are already fairly low (and lower than short-term rates), credit spreads are already low and the dollar is still kind of strong.
The last point needs a bit of clarification. The dollar, obviously, is not strong v. Europe, of course. But folks are currently betting on the euro, not betting against it. And Europe isn’t the region of the world financing the US. Asia and the Middle East are. The dollar remains strong v Asia. The dollar isn’t so much strong v. the Middle East so much as oil has risen and their currencies haven’t.
It isn’t obvious to me that the US bond is an attractive destination for private flows from these regions at current rates and at the current exchange rate.
Mike Larson wrote recently that U.S. Treasury bonds are “quite possibly one of the least attractive investments on the planet.” Bill Gross might quibble a bit. He thinks the Fed may eventually need to cut rates to prop up the housing market and the US consumer, helping bonds. But Larson’s general point stands. US markets “have gotten trounced” recently by foreign markets recently.
The cut in rates that Bill Gross expects would hurt the dollar, so any gains on the bonds would be offset by currency losses. I certainly suspect that the dollar has further to fall v. many Asian currencies.
David Hale raised one point that was meant to reassure: total savings outside the US is about $7 trillion, far more than the roughly $1 trillion (a bit less) the US needs to borrow. No need to worry, per Hale. There is a lot of savings out there.
Alas, Hale’s statistic had the opposite impact on me. The United States $1 trillion external borrowing need seems large relative to the world’s $7 trillion in total savings. There is lot of investment to be done in Asia, Latin America, Eastern Europe and the Middle East. Most of that savings will stay at home.
Moreover, Hale’s statistic highlighted another point: The increase in the “external” savings of governments – the growth in their foreign exchange reserves and investment funds – is likely running at an annual pace of around $1.1 trillion. That seems large relative to the rest of the world’s total savings, not just large relative to cross border flows.
You can easily make the case that the state has once again occupied the commanding heights of the global financial system.
Good thing too, at least for the US. Right now the United States’ $1 trillion borrowing need is $1 trillion more than private investors (on net) want to lend to the US.
<>The world unquestionably saves enough to finance the US deficit, but right now only other governments are willing to lend their savings to the US . That kind of puts the world's leading champion of private markets in a rather awkward position.
Myth I: The dollar is safe because the US has ample assets
Some say the US current-account deficit that requires foreigners to arrange for more than $3 billion of capital inflows every business day just to keep the dollar from falling does not matter. These pundits say a deficit of 6.5% of gross domestic product (GDP) is sustainable because the deficit is only about 1% of all private assets held in the United States; as a result, deficits could be carried a long, long time.
This argument is one about the dollar going to zero, an extreme case of the dollar losing relative to other currencies. However, the current-account deficit and its affect on the dollar are about cash flow: putting it in the context of GDP is reasonable, as GDP is a cash-flow measure of production. Comparing it to private savings is mixing apples with oranges.
(1) “Foreign investors” suggests private sector investment rather than official purchases. CBs loading up on Treasuries are remarkably insensitive to considerations private sector buyers usually take into account. CBs, enjoy your haircut care of that famed barber Uncle Sam [a little off the top, a little off the side, and pretty soon you've been "Argentinaed"]. A sucker central banker is born every minute.
(2) You can easily make the case that the state has once again occupied the commanding heights of the global financial system.
Not just any old state, but Asian exporters and petrostates too.
I remain of the view that the US current account deficit would be, while large, nowhere near the size it is currently had CBs not purchased so many dollars (and by extension, extended cheap credit to US actors.) To some degree, the US deficit and foreign financing of it are an identity.
I would, however, take issue with the notion that the US has not noticed the fall off in net private sector demand. It has noticed it…in the one asset class where CBs are not price setters- equities. The reduction in the US home bias for equities is a phenomenon of relatively recent vintage, but it has had a substantial impact on net flow. Moreover, US equities have noticeably underperformed the ROTW in local currency terms, let alone single currency terms, over the past several years.
Would the CBs step back from US markets, the private sector would eventually fill the gap- at the right price. This is partially a currency issue- though most metrics have the USD substantially undervalued against virtually all free floating, deliverable currencies other than the yen (and Cassandra would dispute that the yen is purely free floating). Rather, the price of bonds in the US is simply too high- remarkably, 10year yields in the US never rose more than 60 bps above the level prevailing the day before the Fed put up rates for the first time- a level almost exactly matching today’s yield. Econometric models generally peg 10 year yields 50-100 bps too low, a discrepancy explained by CB purchases. Give us our 75 bps more, and we’ll happily buy US bonds (and oh, by the way, the trade deficit would likely be lower as well.)
Fascinating post, but the semantics of the debate can be frustrating:
” In some sense that is a truism – you cannot run a current account deficit without the ability to attract capital “.
This is technically important, I think. The issue is not about availability. It’s about price. There’s a difference between the absence of private demand, and the absence of private demand at a price. Private flows have dried up because official flows have intervened through the conduit of FX reserve intervention – at a price. A sudden interruption of this intervention would cause FX prices and asset prices to gap down, but would not stop capital inflows.
” The United States $1 trillion external borrowing need seems large relative to the world’s $7 trillion in total savings. There is lot of investment to be done in Asia, Latin America, Eastern Europe and the Middle East. Most of that savings will stay at home “.
No doubt. But assuming it’s an accurate number, its noteworthy that $ 7 trillion does not constitute a mathematical or theoretical limit on gross global capital (or current) account surpluses, or gross global capital (or current) account deficits. It doesn’t even constitute a theoretical limit on a single bilateral imbalance. (Such is not the case gross global investment, which can only be the sum of positive numbers.) If it’s not a mathematical limit, then this probably nuances the interpretation of the numerical relationship – i.e., take the $ 7 trillion with a grain of salt as a constraint on $ 1 trillion of U.S. borrowing. (Not to say that a $ 1 trillion deficit isn’t a real concern.)
One shouldn’t underestimate the primary role of the banking system in settling the outcome of global flows. It is also a truism that a current account deficit creates its own first points of capital inflow through the banking system (i.e. interbank vostro accounts). The question becomes how these banking inflows are then converted to some other type of asset. This is the point of default if it comes to the stage where asset prices gap down before money chooses to go there.
There’s an analogy with the ongoing morality debate about who is more to blame for the U.S. requirement for capital – the U.S. (i.e. savings behavior a la Roach) or its trading partners (e.g. China with a manipulated currency). It’s a debate without a resolution, because the trade requires two counterparties. Similarly, the current account deficit creates its own default capital inflow through the banking system. The rest is asset reallocation from bank flows to other assets, at some price.
It’s a little different in total when compared to the margin. If central banks wanted to unwind their existing positions, they’d require private sector bids in the FX market – probably forthcoming at some price, barring a doomsday gapping down scenario. Once that’s done, the corresponding capital flow converts from official to private.
Required capital inflows are a foregone certainty via the banking system. The macro risk is asset pricing – not availability of capital.
Aside from this point, which may be largely semantics, your description of the range of substantial risks makes a lot of sense to me.
Emmanuel – “A sucker central banker is born every minute”
I’m far less sanguine in my outlook for the US. Following on from the previous thread on dutch disease, maybe it’s appropriate to compare the current situation to “english disease” in the late 19th and early 20th centuries.
Back then, Britain was losing manufacturing export competitiveness to Germany, the US, and Japan. Sterling was the defacto reserve currency, and a peg was widely maintained through the gold standard.
Like the US today, Britain had ample financial assets upon which to sell debt, and like the US today, this helped mask the effects of emerging industrial economies.
Along with the competition for markets in which to sell their goods, the emerging economies also increasingly competed with Britain for basic resources. That competition became nasty at times, and arguably culminated with two world wars and a severe depression.
I’m not arguing that the same thing is inevitable today, but I do think that the risks are rising.
If, for example, China were to redeploy her accumulating USD reserves to buy significant parts of the Mexican and Canadian resource industries, she might quite reasonably expect to receive the fruits of those endeavors, even if that means American gas tanks run dry.
What If Foreign Money Shunned the U.S.?
In recent years, economists have hotly debated the potentially catastrophic consequences for the U.S. of a Japanese and Chinese retreat from the U.S. bond market. So in February, 2005, then-Federal Reserve Chairman Alan Greenspan surprised a lot of people when, addressing the U.S. Senate Finance Committee, he opined that the fallout from such an event would be minimal. “We’ve looked into that question and I think that we’ve concluded that the effect of foreign purchases of U.S. Treasury instruments has lowered long-term interest rates a modest amount. And therefore, if they were to choose to stop buying or to sell, it would raise interest rates, but again, by a modest amount,” Greenspan said.
This view was a complete departure from the conventional wisdom that a foreign pullout from the U.S. bond market would send U.S. interest rates spiking skyward. (It’s worth noting that current Fed Chairman Ben Bernanke shares Greenspan’s view.)
Guest – Do you think that, even if he was absolutely certain of it, the Chairman of the US federal reserve would tell the world that if the Chinese hit the sell button, we’re pretty much done for?
Given the enormity of the stock of bonds owned by Japan and China, there is a massive difference between stepping away from US fixed income markets by not purchasing a single new bond, and selling the entire existing stock of debt.
In the former case, the Greenspan prescription is probably right- rates would go higher, but probably to the degree I alluded to above: 50-100 bps.
Should J + C decide to sell their $1.7 trillion or whatever of Treasuries…they’d probably find that they couldn’t, because the price would move so much that it wouldn’t make economic sense to complete the transaction.
Macro Man,
I don’t disagree with your comments on bond values, with one additional observation on the history: The U.S. Treasury curve incline out to 10 years was roughly 350 basis points when the Fed started its tightening cycle from 1 per cent. This was an enormously positive yield curve for the situation – absolutely Himalayan when normalized for the low level of the funds rate at the time – i.e. a slope equal to 350 per cent of the funds level. While not changing the thrust of later analyses of the ‘conundrum’, the unusual severity of this starting position has often been forgotten.
I am not a econ major, so I have a probably very stupid question to ask: why does it even matter?
In any society, there are savers/lenders and there are spenders/borrowers, and we have financial institutions such as banks to match the needs. Everyone is either in capital surplus or in deficit. Yet everything works beautifully.
why can’t this work at the country level? Why does a country have to be self-sufficient in capital?
some countries love to save for whatever reason, some countries love to spend, as long as the global financial system can handle the load, why can’t this work?
4 CONTRADICTORY STATEMENTS
(Guest, not econ major, you’re almost right)
QUOTE1 “The question is whether the rest of world – or, more precisely, private investors in the rest of the world – want to lend that savings to the US and in the process buy US financial assets in the process. Right now, in my judgment, the data says that they don’t.”
QUOTE2 “The market certainly doesn’t seem all that worried by the shift in the composition of flows to the US.”
QUOTE3 “But US long-term rates are already fairly low…” [So where's the US bond crash that the dollar slide was supposed to generate?]
QUOTE4 “The world unquestionably saves enough to finance the US deficit, but right now only other governments are willing to lend their savings to the US.”
What?
“right now only other governments are willing to lend their savings to the US” ?? Uh??
COMMENT
Right now, there are trillions of “dollar denominated assets” bought every year by private investors.
Right now, there are MORE private buying US assets than selling US assets (About 200 bn. in 2006, a bit less now still positive; but that’s NOT the point)
When/if CBs stop buying US assets the dollar will lose 10% of its REER value and there private buying of US assets will fully substitute CB’s fading demand.
As usual, you overestimate the relevance of net flows, even when they are a small fraction of gross flows. Net flows are relevant for some macroeconomic analyses, but are no indication of the market’s willingness to finance the US. You should look at other indicators!
In the past, the sum of all CuA deficits rose repeatedly to one fifth of global savings, and that meant no crash per se. “So what” if the US get (only) one seventh of global savings?
When you will accept that your 4th comment is false (since it confuses net flows with gross flows), then you will accept a little more the views of David Hale and Yours Truly. In the meantime, say it clear:
“I, Brad W.Setser, think that half of the world, holding 4 trillion US assets or so, is completely stupid”.
Not just central bankers!
————-
PS Guest: Greenspan is totally right on the small effect in the US of an Asian sale of US bonds. I say “sale”, not just “stop buying”.
I have only one problem with your view. Economists did not “hotly debate” the effects of a sudden Asian sale of US bonds, only journalists did. Economists know it’s a stupid debate. Furthermore I think it’s Greenspan who represented the (economists’) conventional wisdom. Maybe a storm is coming, but not yet, and not so bad… it’s just that in this blog they like to do a bit of Italian drama… Sorry folks, it’s my view!
No one is contending that there isn’t rote demand for USD assets. I know a certain “dead pet trust” for example that is such a policy buyer of US bonds. And admittedly, even the 12-digit annual PBoC purchases of USD are on the margins of aggregate demand for USD assets. But prices ARE set by the marginal buyers and sellers. The bias of that real marginal demand both supports and emboldens ambivalent holders, a range of speculators, and numerous flavors of feedback traders. Even in a presumably broad and deep market, the relationship between marginal demand and price is NOT linear. Has it not taken a decade for Asian currencies to percolate back from 98 overshoot? Maybe -10% IS the right number, but markets – especially those unnaturally restrained or overly frothed – have the odd tendency to take it there via -25% and 30 dog-years of digestion.
American investors have done pretty damn well with equity investments in foreign securities. For instance, Billionaire Warren Buffet is the largest private investor in PetroChina under his Berkshare Hathaway holding company. Personally, I invested earlier in PetroChina shares than Warren Buffet, but I believe he has at least tripled his money in the stock to at least $10 billion. Since Buffet is donating his entire PetroChina holdings to a US based charity organization, that represents a huge positive financial gain for the US economy. Fidelity is the second largest investor in PetroChina shares with a $1 billion equity investment. The rising holdings of foreign equities by Americans has partially offset the current account deficit with foreigners receiving significantly less return on their US investments. On a personal note, I also donate most of my interest income from my investments to various charity organizations that benefit education, medical technology and the environment. And as the major supplier of natural gas in China, PetroChina is helping the environment by shifting Chinese energy consumption from coal to cleaner natural gas with less greenhouse gas emissions.
Guest:
” I am not a econ major ”
A good question, really. It matters as a factor influencing exchange rates and interest rates for different currencies. Were the world economy unified under a single currency and single monetary policy, it would matter to an understanding of regional contributions to a single closed economy.
‘PS Guest: Greenspan is totally right on the small effect in the US of an Asian sale of US bonds. I say “sale”, not just “stop buying”.’
Economists might “know” that this is a stupid debate, and they might “know” that an Asian sale of the entirety of their Treasury holdings would have a modest impact on bond prices…which is why they are economists, not professional investors.
If the private sector got wind that Asia was dumping, which they would, what do you think would happen? Would everyone get out their buy tickets to fulfill the responsibility to fill the gap left by the public sector…or would everyone get out their sell tickets and front run the CBs?
The latter would appear far more likely, in which case there would be a substantial price move before the private sector covered their shorts/underweights. 1994 is a reasonable example of what happens when there are few bids and no offers in the bond market. Overshoots, as Cassandra alludes, are a fact of life in financial markets, but it is very difficult indeed to “know” how far they will extend.
Sorry, the above should say “few bids and no end of offers” in the bond market…
Gheorghius — right now, there isn’t enough net private buying (net = what matters, gross private inflows – gross private outflows) to cover the current account deficit,or even come close. the situation could get worse if there were ever to be net private selling …
some private exposure, incidentally, is dollar hedged/ or denominated in euro (US corporate debt is about 20% in euros).
As for Koos’ point, the BoJ/MoF stopped intervening in the spring of 04, but they were still buying treasuries through the fall of 04 (they had built up big bank accounts during their intervention peak). and when japan stopped, China and the EM picked up their purchases (big purchases in q4 04). Both Banque de France and Warnock and Warnock (links provided) find a substantial impact from these purchases, and i agree!
If the fed offset central bank sales with purchases of its own (i.e. monetary expansion), it could mitigate some of the impact on long-term rates, but that likely would lower the dollar …
” As usual, you overestimate the relevance of net flows, even when they are a small fraction of gross flows. Net flows are relevant for some macroeconomic analyses, but are no indication of the market’s willingness to finance the US. You should look at other indicators! ”
Good grief!
Central banks are in buying the dollar as a function of price.
That means buying it where others aren’t.
Who else does that?
Could the effect be any more obvious?
How can Greenspan think China and Japan dumping their debt denominated in USD will have a modest effect on interest rates? They would be selling about 2 years worth of our current account deficit and on top of it, the 2 biggest purchasers wouldn’t be buying it.
It would be like if the US government decided to stop buying stealth bombers and also decided to sell their existing stock. I would guess the price of stealth bombers would drop quite a bit.
In reality, if China and Japan did this, I suspect the treasury would print a bunch of money and buy up at least the tbills.
Charlie,
No one is talking about China and Japan dumping their US debt; that would amount to economic suicide should the US economy have a hard landing. With the state-owned investment trust, the PBoC has proposed an exit strategy for an orderly diversification of foreign exchange reserves. By signaling its intentions to the global market, the Chinese Central Bank in recent statements has reiterated that they will no longer in the near future stockpile US Dollar reserves.
The debate on Asian dumping bonds is “stupid” for a number of reasons. First of all, Asia (China) would never dump violently its $-bonds to cause sharp imbalances between (notional) supply and demand: it would be a totally self defeating strategy. (But alas some political commentators and journalists have been so stupid as to put forward that idea as credible: if US-China relations worsen, they wrote, China might decide to sell suddenly and massively so as to create havoc in US mkts)
Second, if Asia (China) either sells suddenly or dumps gradually, this would definitely have an effect on bond prices and long term interest rates, as Macroman argues – I fully share his argument. This would be a relevant event for (mainly, short term) bond traders, with the potential to determine great fortunes and great individual losses.
But the macroeconomic impact of this event would be small. Since economists are no traders, they care for macro outcomes, not for what happens to individual portfolios in this zero-sum, unproductive, sad game of “trading” in these modern casinos called financial markets.
The macro impact would be small because either Asia (China) in a raptus dumps everything in a few days, so the bond mkt collapse would be shortlived; or Asia (China) dumps gradually, so at a moderately lower bond price private demand would balance Asian supply. Interesting BWS’s remark that these Asian sales would imply a lower dollar; precisely so!! A lower dollar would further enhance private demand of bonds! So Greenspan is right, if only you understand him. He’s not a speculator, he’s looking at the broader picture.
this may be simplistic, but i just took the TIC data totals for net foreign purchases of treasuries, agencies, corporates and stocks, got quarterly numbers, and just compared them to current account data; as of the fourth quarter net foreign purchases of US securities covered the CAD with about $100bn to spare, close to previous “peaks” in third quarter 2005, second quarter 2003 and fourth quarter 1996… so so far at least it looks like the US is having no trouble financing its CAD through asset sales!
i agree that in the long run a high and rising CAD can create problems, but i don’t think it’s really “worth” worrying too much about it until net foreign purchases of US securities (either ‘private’ or ‘official’) no longer cover the US’ CAD; that is, CB intervention may be hugely distortionary, but — for financial market participants at least — it is what it is and, unless it changes, “fighting the fed” (or the collective will of emerging market CBs and SWFs) will be a losing proposition for the “little guy”
guest — i just rechecked the us q4 data (bop data on BEA web page, foreign purchases of long-term securities. US private investors bought around $50b of foreign securities, and foreign private investors around $160b of US securities,for a net flow of $110b mex. official flows produced another $90b –
the CAD is now a bit above $200b a quarter, so it was financed, but i wouldn’t say it was comfortably financed. moreover, the data likely understates actual foreign central bank purchases (as it did from q2 05 to q2 06)
which do you think is better 3-mo trailing TIC data (official + private) or bop from BEA? cuz from TIC, net foreign purchases have been consistently running over $200bn since 2004…
there shouldn’t be much of a difference between the tic flows and the bea flows if you match everything up closely (i.e. bea long-term to the relevant tic lines). i checked for q4 tho, and the net tic long-term flows were in the $130-135b range (see line 21) and net tic flows were in the $140b range (line 30). the tic long-term net purchases data needs to be adjusted for agency/ mbs principal repayment to get the bea data, and line 21 does that, so it should match up reasonably well (allowing for revisions).
http://www.treas.gov/press/releases/hp317.htm
oic, tks! those short-term flows make a real difference tho
How Big Could Sovereign Wealth Funds Be by 2015? The world’s SWFs could grow from US$2.5 trillion now to nearly US$12 trillion by 2015, and could exceed the total size of the world’s official reserves within five years, i.e., by end-2011.