Brad Setser

Follow the Money

Cross border flows, with a bit of macroeconomics

Print Print Cite Cite
Style: MLA APA Chicago Close


And now, the rest of the story: long-term portfolio flows have fallen by more than the trade deficit

by Brad Setser
July 20, 2009

The goods news: the US trade deficit has shrunk. On a rolling 12m basis the trade deficit is down to around $500 billion, and the data from the last few months suggests that it should fall even further.

The bad news: the US trade deficit hasn’t shrunk by as much as foreign demand for US long-term assets.


My graph only showed inward portfolio flows. That isn’t the entire balance balance of payments. But inward and outward FDI flows tend to offset each other. And in general Americans have been adding to their foreign portfolio, not reducing their foreign holdings. That means the (remaining) deficit is increasingly financed by short-term flows, which isn’t the most comfortable thing in the world.

All this is pretty clear if you look at the details of the last TIC data release (already covered in depth by Rachel). Over the last three months, private investors reduced their US holdings by over $100 billion (line 31). That total was offset by the repayment of the Fed’s swap lines — but the long-term flow picture isn’t great. Net portfolio inflows over the last 12ms totaled $188 billion (line 19). After adjusting for repayment of ABS, that total falls to zero (lines 20 and 21).

As the following graph shows, net private demand for long-term US assets — that is gross long-term private portfolio inflows net of US portfolio outflows — started to disappear in late 2006, and then took another down leg after the subprime crisis broke in August 2007. And over the last 9 months, official demand for US long-term bonds also disappeared — as reserve growth slowed (until recently) and central banks moved in mass toward short-term treasury bills.


The split between official and private flows in the chart reflects my adjustments to the TIC data – but my adjustments basically just make the TIC data match the US survey data and the revised BEA data on official flows.* My adjustments change the official/ private split, but not the total.

The usual narrative would argue that the fall in demand for US long-term assets is a reflection of foreigners concerns about the scale of the US fiscal deficit. That though isn’t really the case. Demand for long-term Treasuries has held up better than demand for almost all other kinds of US assets. The reality is that the rest of the world has lost confidence in claims on the US private sector, not in claims on the US government. Consider the following chart …


The higher frequency data tells a similar story. Net (private) demand for US long-term financial assets (private demand for US long-term assets from the rest of the world, net of US purchases of long-term foreign assets) hasn’t been strong for some time. It picked up in the crisis, as flows contracted in a way that favored the US. Americans sold their foreign portfolio faster than the rest of the world sold their US portfolio. But as the crisis has abated, net private demand for US assets has fallen off. Foreign demand for long-term US assets remains weak – and over the last few months Americans resumed buying foreign assets.


Incidentally, this chart is one of many that suggests, at least to me, that the US could not have sustained large deficits over the past few years if the US had been forced to finance its deficits in the private markets. The demand wasn’t there. Not on the scale that was needed. **

For a while, official purchases of long-term assets offset weakness in private demand. But after the crisis, risk-adverse reserve managers have preferred short-term bills – so demand for long-term US assets from central bankers largely disappeared.


There are some tentative signs that central banks are once again buying Treasury notes, at least short-dated ones. My estimates suggest that official purchases of long-term US financial assets are trending up again — largely because of purchases of Treasuries through London.***

Even so, there was — at least in the three months to May — still a sizable gap between net demand for US long-term financial assets and the US trade deficit. The (much reduced) trade deficit is being financed by short-term inflows, and specifically short-term official inflows.

That isn’t ideal. The “quality” of the financing of the US deficit has gone down.

The fall in the trade deficit has, in my view, reduced the risk of a disruptive dollar adjustment — at least relative to the pre-crisis world where the slow growing US was running 6% of GDP or so current account deficits, deficits that private investors showed little sign of wanting to finance. Analysts who emphasize the rise in the fiscal deficit tend to ignore the (even bigger) fall in private sector borrowing and corresponding fall in the United States total external financing need. But finding the financing to cover a period of adjustment is never easy; the US isn’t yet out of the woods.

* Using the methodology described in my work with Arpana Pandey, I attributed some of long-term Treasury and Agency purchases through the UK (and for Agencies, Hong Kong) to official buyers on an ongoing basis. This avoids the jumps characteristic of the annual survey revisions. I also adjusted official equity purchases for Chinese purchases through Hong Kong from 2005 on, using a similar methodology. There was a big jump in Hong Kong’s equity purchases in early 2007, and the survey data subsequently attributed those purchases to China. The overall data should be fairly close to the BEA’s revised data series – though I didn’t adjust official purchases of corporate bonds down to reflect the survey revisions (apparently central banks do not use US custodians for this exposure), as I suspect that the data actually understates the official sector’s true exposure to corporate credit. Private fund managers did a lot of things to get a bit of extra yield, and some central banks made increasingly use of private fund managers when they themselves were reaching for a bit of yield.
**A counter argument is that official demand displaced private demand, but didn’t change the overall equilibrium. This has two components. First, if say China’s government hadn’t been building up claims on the US and Europe, private Chinese investors would have bought the same amount of US and European assets (something I don’t believe). And second, if say China had bought more European assets and fewer US assets, private investors in Europe would have shifted funds from Europe to the US – keeping the US deficit up. That argument has a grain of truth, but it assumes a bit more substitutability between US and European assets (including a willingness to take unhedged currency risk) than I suspect is present in the market.
*** The UK’s Treasury holdings rose by about $30 billion over the last three months.


  • Posted by jonathan

    So US companies also have too much debt – as shown by the huge red camel’s hump – and that makes their debt less attractive.

  • Posted by bsetser

    jonathan — asset backed securities (private ones) are considered corporate bonds, so the red hump reflects over indebted households as much as over indebted companies.

  • Posted by HZ

    The shrinking trade deficit is even more impressive given the exploding fiscal deficit.

  • Posted by FollowTheMoney

    @jonathan…be surprised not to see a spike in both rates of gov & corporate debt within 18 months.

    and then what?:)

  • Posted by Cedric Regula

    Brad:”The reality is that the rest of the world has lost confidence in claims on the US private sector, not in claims on the US government…”

    The least worst argument may have merit again here. Excluding equity investors, who seem to be having a good time everywhere in the world, and just looking at fixed income the global state of affairs is truly a sorry one.

    Some quotes from today’s Mauldlin newsletter…

    “As the International Monetary Fund made clear last week, Britain is lucky that markets have not yet imposed a “penalty interest” on British Gilts, given the trajectory of UK national debt – now vaulting towards 100pc of GDP – and the scandalous refusal of this Government to map out any path back to solvency.”

    “France and Italy have been less abject, but they began with higher borrowing needs. Italy’s debt is expected to reach the danger level of 120pc next year, according to leaked Treasury documents. France’s debt will near 90pc next year if President Nicolas Sarkozy goes ahead with his “Grand Emprunt”, a fiscal blitz masquerading as investment.”

    “The IMF says Japan’s gross public debt will reach 240pc of GDP by 2014 beyond the point of recovery for a nation with a contracting workforce. Sooner or later, Japan’s bond market will blow up.”

    So Japan, England and the Eurozone are goners in the long term. Sounds like Ireland and Eastern Europe will close up shop by the end of the year.

    Of the BRICS, Russia has -10% GDP growth, Brazil -2% and only India and China register in growth territory, and we know that China bought their growth.

    I’m trying to think of any countries that are neither export growth dependant nor credit growth independant. If anyone knows of any, let me know, I’m looking for a place to move to.

  • Posted by Cedric Regula

    oops, typo. I meant …credit growth dependant..

  • Posted by Too Much Fed

    Cedric R, notice it is the high wage countries with all the debt. Is Australia in the same boat too?

  • Posted by Too Much Fed

    “Even so, there was — at least in the three months to May — still a sizable gap between net demand for US long-term financial assets and the US trade deficit. The (much reduced) trade deficit is being financed by short-term inflows, and specifically short-term official inflows.

    That isn’t ideal. The “quality” of the financing of the US deficit has gone down.”

    Is the USA setting itself up for a sudden stop debt deflation with a currency crisis event instead of a sudden stop debt deflation without a currency crisis event (like just happened)?

    Is too much debt the problem, period?

  • Posted by FollowTheMoney

    any thoughts on an opinion i share with a few others:

    “When the trade deficit begins shrinking again, economists will assign it a bullish value when in fact this time it will go hand in hand with the onset of a depression.”

  • Posted by Michael


    I think the wise individual will accept that credit is the lifeblood of all modern economies and the global trade system, and that credit contraction will always be taboo everywhere, and even lack of credit growth is politically unacceptable. Those predicting – and in many cases wishing for – a “return” to a primordial society in which expanding credit is not the basis for economic expansion are both living in fantasy and appparently unaware of how commerce has worked for the last 400 years. That said, the wise individual will neither consume beyond his/her means nor seek bigger, better, faster wealth through leverage, and can flourish in any country in any era.

  • Posted by Brick

    Good thoughtful piece and I wonder at the root causes for the world losing confidence in claims on the US private sector. It could be argued that stress tests that did not stress, changing accounting rules and the likes of CIT who were apparently well capitalised getting into difficulty has caused a credibility problem for US corporations. Though this might be part of the case, I suspect actually it has more to do with a perception that there might be some sort of disconnect between world economies and particularly China and the US. We should also take into account the other side of the flow that confidence might be raising in the Chinese private sector which gives them problems in terms of surplus.
    Are we seeing the decoupling theory re-emerging and will it collapse again as it did before. I suspect elements of the decoupling theory have emerged, but this might have some different aspects to it than previously. Perhaps the flows reflect a loss of confidence in claims on the US public sector by private investors as opposed to central bank views who still appear to have plenty of confidence

  • Posted by Cedric Regula

    Too Much Fed responds:

    “Cedric R, notice it is the high wage countries with all the debt. Is Australia in the same boat too?”

    I think they are still in pretty good fiscal shape. They traditionally have had high taxes, and that has resulted in a low public debt/GDP ratio relative to the rest of the developed world. They did have a real estate boom too, but I think it was more muted than the one in the US and UK. Their central bank did keep interest rates higher all along, and they do not have has easy access to capital like the US(with the GSE system) or the UK.

    Even so, as a commodity exporter they are sure to have some ups and downs. I think the consumer there may be strapped on a cash flow bases, because high taxes and a high interest rate mortgage would do that. But the problem we have here is worse. People have realized they may have spent the next decade or two on vapor housing. A friend in S.Cal just told me of someone he knows in one of the far inland areas of S.Cal. The poor guy bought his house for $360K during the boom, and a neighbor just put the same floor plan on the market for $130K. So here we have a middle age guy that just found out he could be working for a long time to pay off a $230K loss.

    Micheal:”That said, the wise individual will neither consume beyond his/her means nor seek bigger, better, faster wealth through leverage, and can flourish in any country in any era.”

    I’m hoping I can get it to work that way, but I get a little nervous when it becomes apparent that I may get to pay everyone else’s bills, and it is getting increasingly difficult to find reasonable risk/return investments….

  • Posted by ReformerRay

    As a faithful reader of this blog, I have watched Brad look for signs that the large trade deficit will reduce the value of the dollar and automatically correct the trade deficit. Brad was only one of a group of writers dealing with foreign trade that either expected or hoped that the “market” would correct the trade deficit.

    I have argued for years that this is a vain hope. I have been right for the last decade. The decline in the value of the dollar since 2002 was too gradual to really correct the trade deficit (though it did increase exports for 3 years).

    Foreign portifolio investment in the U.S. has sustaned the U.S. dollar value (up until the current recession), preventing the automatic adjustment via SERIOUS decline in value of the dollar.

    Today the value of the real dollar (inflation adjusted by FRB)is near the low point of its history.

    Assuming that portifolio flows into the U.S. do not increase any time soon, the question then becomes whether or not the value of the dollar will decline.

    I think future value of the dollar is unpredictable. Assuming the trade deficit will remain at its current level for a while, the value of the dollar may already be low enough that it will remain at its current level.

    The composition of the decline in imports in the last 9 months suggests that producers needs for foreign Industrial Supplies and Capital goods will determine when imports return to their previous leve (see comments in yesterday’s discusssion).

  • Posted by jonathan

    Thanks for the clarification.

  • Posted by ReformerRay

    “The goods news: the US trade deficit has shrunk. On a rolling 12m basis the trade deficit is down to around $500 billion, and the data from the last few months suggests that it should fall even further”.

    That the trade deficit has fallen is definitely good news. However, the decline is slowing.

    The last two months reported (April and May of 2009) show only a 3 billion dollar decline from the first quarter (goods only).

    The Census Bureau Press release for May shows that Industrial Supplies and Capital Goods accounted for 75% of the import decline between July and August of 2008 and April and May of 2009 (goods only).

    My guess is that U.S. goods producers reduced overseas buying when credit dried up and that they will resume overseas buying as soon as they think they can sell their product.

    The U.S. production system remains an important part of foreign trade.

  • Posted by Greg

    BS: The reality is that the rest of the world has lost confidence in claims on the US private sector, not in claims on the US government.

    Brad, where do you think the government’s revenue comes from if not the private sector (aka the tax base)?

    The government’s promises cannot be any better than the strength of its tax base

  • Posted by Cesar

    @ Cedric:

    … Chile might be an option. Good wines, good food and good countercyclical policies. They depend on exports but they havent ‘bought’ their growth.

  • Posted by Jenn

    The US is financing really long term debts (entitlements and chronic deficits) with ever shorter maturity debt. This creates enormous refinancing risk — the same stuff that brought down the S&Ls and caused many SIVs to blow up.

    How Brad or anyone can call this arrangement “safe” is beyond imagination.

  • Posted by But What do I Know?

    It’s good to see you back from vacation, Brad–I hope you had a nice break. Thanks for this data, which is, as usual, insightful and not found anywhere else. Your point that the foreign central banks are shying away from US long-term debt even as they continue to rollover short-term Treasuries is well-taken and fits in with the thesis that the Fed is the Last Buyer of longer-term GSA and (eventually) Treasury debt at these prices. I read today where Mr. Bernancke suggested that one of the Fed’s “exit” strategies would be to sell the long-term debt on its balance sheet. Good luck with that, Sparky. Private buyers know that the GSE debt is too expensive (yields are too low) and the foreign central banks would have to explain why they were buying what they’ve just finished dumping. Where does the Professor think the buyers will come from? Or does he, in true economics fashion, assume that they will appear?

    Of course, the Fed can extend and pretend for the foreseeable future by continuing to buy up the GSE debt and MBS. The magical illusion of money will continue to work until it doesn’t.

  • Posted by bsetser

    Jenn — I would find your argument more persuasive if you left “entitlements out of it”; soc sec is still running a cash flow surplus last i checked, and will for another few years. it is reducing the current deficit, not adding to it — (the concern about entitlements focused on their future impact,i.e. the expected cash flow deficits after 2040 for s sec and the end of financing for the rest of the government will before that). I am worried about the united states increased reliance on s-term financing for its external deficit. but i also have to factor in the fall in that deficit with assessing the risks. on balance, i think the fall is the more important of the two factors.

    as for the treasury, it actually cut its bill stock and increased the outstanding stock of notes in june; the financing of the fiscal deficit is no longer really coming from bill issuance. i’ll have to post on this …

    follow the money. a trade deficit can shrink b/c of a fall in imports (often a sign of a fall in demand, not a shift in demand toward home production and thus often linked to a slowdown) or b/c of a rise in exports. you didn’t spell out which of the two is driving your scenario.

    greg — sure, the government relies on taxes to cover its debts, so its credibility is a function of an expectation of a return to growth and an expectation that the government (the federal gov at least) will in the end be responsible and match expenditures and revenues. but the fall in confidence in private claims is very much a function of the loss of confidence in American securitization technology. ABS issuance (and demand) collapsed.

    and then there is the story of the agencies — where foreign central banks decided the risks of holding agencies exceeded the rewards, which is kind of strange as the fed had the opposite judgement …

  • Posted by Yoda

    can scumbag Fed move asset to SIV to appear they are not printing dollar and expanding their balance? man, we been cheated, all along they been printing dollars, purchasing garbage and move the garbage to Super SIV in stealth?

  • Posted by glory

    top FT story now…

    China to deploy forex reserves
    Part of $2,000bn to aid overseas growth

    Beijing will use its foreign exchange reserves, the largest in the world, to support and accelerate overseas expansion and acquisitions by Chinese companies, Wen Jiabao, the country’s premier, said in comments published on Tuesday.

    “We should hasten the implementation of our ‘going out’ strategy and combine the utilisation of foreign exchange reserves with the ‘going out’ of our enterprises,” he told Chinese diplomats late on Monday.

    Mr Wen said Beijing also wanted Chinese companies to increase its share of global exports…

  • Posted by glory
  • Posted by Cedric Regula

    Wen will Wen ever get any respect? Nobody ever listens to the little guy.

    Today the big news was Ben said interest rates will stay low for a considerable period of time! Treasuries rallied big on the news.

    He also said his exit strategy is to not dump long term assets on the market. Good for another gain in Ts&MBS

    Didn’t say anything about buying a bunch of T-Bills from the Chinese???????

    Does Wen just make this stuff up?

  • Posted by Nino

    “While it is true, as Keynes intoned, that we are all dead in the long run, I see no reason to die young from orthodoxy-imposed anorexia.”

  • Posted by Cedric Regula


    I read McCulley once and a while, and used to think he was pretty good at econ analysis, but after reading that article, I think he lost his mind.

    Krugman has even renounced Krugman’s Japanese recovery plan.

    But McCulley just embraced it. The only thing I can figure is pimco went short their entire portfoilio of treasuries.

  • Posted by Cedric Regula

    So I guess this must mark the end of China’s dollar peg.

    They can get out of T-Bills at face value, but this should make the dollar drop some. So they lose something in FX. But if you are going on a shopping spree maybe you need dollars? Or maybe you need to trade dollars for all those different kind of Pesos and other commodity currencies? Will they buy anything in Euroland? Office space in London?

    This should be really weird! Maybe we end up with countries that buy things with dollars to spend? Stranger things have happened.

  • Posted by Bill

    Pardon my ignorance. Where does monetized debt get added to these figures?

  • Posted by bsetser

    glory, clever.

    i couldn’t resist commenting on the anderlini story

  • Posted by Pax Americana

    Even so the US Deficit is projected to come in at $350b in 2009,and will add 2.5% to GDP,and as for the Decoupling theory, the world still needs American consumer spending.
    seems to me Brad the chinese as you have mentioned before are reweighting into short term T/bonds, in the eventual desire of exiting their us Assets,and using and activly encouraging the use of the RMB in their international trading.