What’s more, the Council’s website has an interactive tool that lets you play with the charts …

Enjoy the charts over the holiday weekend (at least in the United States). And let me know what you think. Are there any charts that we used that didn’t work? Are there charts that we should have included but didn’t?

“the crisis was caused by ineffective supervision and regulation of financial markets in the US and other industrial countries …. [NOT BY] ….”current account imbalances, particularly by net flows of savings from emerging markets to the US,” “easy monetary policy in the US” or “financial innovation. … the idea that fraud and reckless lending flourished because US financial markets were unable to honestly and efficiently intermediate a net flow of foreign savings equal to about 5% of GDP, while having no problem with intermediating much larger flows of domestic savings, is astonishing to us.” *

The authors of Box 1.4 of the IMF’s Spring 2009 World Economic Outlook also attribute the current crisis to risk management failures in large financial institutions and weaknesses in the regulation and supervision of such institutions.** The role of imbalances are downplayed, as a “disorderly exit from the dollar has not yet been part of the crisis narrative.”

The last point is hard to refute: the dollar rallied during the most intense phase of the crisis.*** Reserve growth stopped, but that was because private money moved out of the emerging world and into the dollar, yen and swiss franc after the crisis – not because the world’s central banks lost confidence in the dollar. The proximate cause of the most recent phase of the crisis was a collapse in private financial intermediation, not a collapse in key central banks’ willingness to finance US.

But the absence of the kind of dollar collapse that many postulated might bring Bretton Woods 2 to an end doesn’t imply – in my view – that there was no connection between a global system marked by large inflows from the emerging world and the current crisis. The key issue is whether or not the large net flow from the emerging world to the US and Europe created conditions that facilitated, directly or indirectly, the failure of private risk management.

Three potential connections come to mind:

A rise in offshore dollar deposits by central banks provided some of the financing for the growth in banks’ dollar balance sheets. Central bank inflows into offshore money market funds had a similar impact.

The availability of dollar funding allowed banks’ balance sheets to swell, with an associated rise in demand for complex financial products that generated a bit of yield.
This was particularly pronounced with European banks. Thanks to the work of the BIS, we now know that a host of large European institutions funded their dollar book – a book that included a lot of synthetic triple A – in the wholesale market.

The fall in real yields pushed real money to seek higher returns.

This happened in US money market funds, who increasingly financed European banks taking a punt on risky US securities in order to pick up a bit of yield.

It seems to have happened among fixed income fund managers, many of whom underperformed their benchmarks in 2008 because they loaded up on riskier assets in good times. The Wall Street Journal:

“What went wrong? Most intermediate funds held far fewer of the safest bonds than were inthe index. Worse, as the credit crisis unfolded and prices of risky bonds collapsed, many managers boosted holdings of low-quality debt.”

It also happened in pension funds. Some turned their fixed income portfolio into something like a credit hedge fund (it is hard otherwise to explain say Ontario Teachers’ 2008 fixed income returns). They needed higher yields than on offer in the Treasury or Agency market to meet their future obligations (in the absence of higher contributions).

The life insurers also likely reached for yield, even if one sets aside the special case of AIG’s financial products group.

The flat yield curve encouraged risk taking by special investment vehicles and other leveraged players

Investment vehicles that borrowed short to lend long couldn’t make money buying “safe” long-term assets so long as central bank demand for longer-term debt helped keep the yield curve inverted. Various vehicles could have closed up shop and said, more or less, current market conditions aren’t conducive to our basic strategy. Most though took on more credit risk and increased their leverage to maintain a high return on equity as the yield curve inverted and credit spreads fell.

A host of financial intermediaries were in the same position as they special investment vehicles. The inverted yield curve put pressure on all institutions that borrowed short and lent long. Many though seem to have offset the natural pressure on their profits by expanding their balance sheets and taking more risk. There is a rough correlation between the inverted yield curve and the rise in ABS issuance in the US –

European purchases of US corporate bonds (including ABS) surged along with issuance from 05 to mid 07. Those European purchases collapsed in the middle of 2007. That marked the beginning of the crisis. The surge reflected demand from a mix of European banks that relied on the “wholesale” market for dollar financing and a slew of offshore vehicles, including some sponsored by US banks. My guess is that this demand was also a manifestation of the market impact of a flat yield curve.

I am not sure this provides the kind of theoretical or empirical evidence Dooley and Garber called for, but it could — I hope — provide a couple of hints that help further future research.

Another argument could be added to the list of potential links between the credit crisis and central bank demand for reserve assets, but it is a bit harder to document.

Central bank inflows helped helped create the illusion that a world with large imbalances could be a low volatility world. In the pre-crisis world, a shortfall in private demand for US assets would, generally speaking, be offset by a rise in central bank demand. Low volatility, in turn, meant that high levels of leverage appeared safe.

Dooley and Garber, I suspect, would argue that the stable flow from the periphery to the center from the emerging world’s central banks – i.e. Bretton Woods 2 –meant that it was rational for private players to bet on its continued stability. The rise in volatility during crisis, they would argue, didn’t reflect a fall in the United States access to external funding.

That though strikes as too narrow a view – especially with the benefit of hindsight. Central banks did step up their financing of the US when private flows faltered in 2006, 2007 and the first part of 2008, taking away one source of volatility. But Bretton Woods 2 – at least after 2004, back in the days when the US fiscal deficit was trending down — required intermediaries willing to sell their safe assets to the world’s central banks and use the proceeds to invest in riskier assets. The gap between the increasingly risky assets (as loans backed by inflated housing collateral are more risky than loans backed by more conservatively valued homes) that the US economy was generating and central banks desire for fairly safe assets that had to be filled by private intermediaries. And as we now know, a failure in this process was a potential source of “volatility” inside the system, especially after the source of the US external financing migrated from the government to households.

To be sure, a surge in central bank demand for treasuries and agencies that pushed real rates down and inverted the yield curve didn’t have to lead to $4 trillion or so in (estimated) losses in the financial sector. Bankers didn’t have to take more risk to try to boost returns; they could have sat out the last dance. And even if they didn’t, regulators should have stepped in earlier, limiting the buildup of risk.

At the same time, absent official inflows, a shortfall in (net) private demand for US assets, relative to amount of external financing the US needed so long as US households were running large external deficits, would likely have cut the housing and lending-against-rising home values bubble off at an earlier stage, before it got so large. The recycling of the emerging world’s surplus back into the US regardless of the dollar’s slide or the relative performance of US assets helped allow financial leverage to build, as financiers came to believe that a world with large internal and external deficits was a stable world and thus high levels of leverage were warranted throughout the economy. That is one reason why the IMF, after initially discounting the importance of imbalances, ends up arguing that “a pattern of low interest rates and large inflows into US and European banks” encouraged “a buildup of leverage, a search for yield and the creation of riskier assets.” (see p. 36 of the WEO)

Back in March, the FT’s Krishna Guha — in a box linked to a Gillian Tett missive on “destructive creation” — wrote: “a more balanced economy might allow the world to live with a less perfect financial system”

Very well said.

* Net foreign inflows of 6% of GDP actually were fairly large relative to the US national savings rate (using the NIPA definition of savings), which wasn’t that fair from 12% of US GDP. Roughly a third of all investment – in a macro sense – in the US was financed by foreign not domestic savings.
** The IMF has changed its tune here. The IMF’s 2007 staff report on the US extolled securitization, arguing that it had left the “core” of the US financial system well capitalized.

All data is in dollar billions, and is presented as a rolling four quarter sum.*

The red line is the United States current account deficit.

The black line is the United States financing need – defined as the sum of the current account deficit plus US outward FDI and US purchases of foreign long-term securities.** The dip in the total US financing need from mid 2005 to mid 2006 isn’t real. It reflects the impact of the Homeland Investment Act, a holiday on the repatriation of the foreign profits of US multinationals that produced a sharp fall in outward FDI.*** The rise in the United States financing need over the course of 2007 by contrast is real; American investors bought the decoupling story and wanted to invest more abroad.

The shaded area represents official demand for US assets. The inflows from central banks that report data to the IMF and Norway are known. The inflows from central banks that don’t report and other sovereign funds are my own estimates. The key countries that do not report reserves are – in my judgment – China, Saudi Arabia and the other countries in the GCC. I have assumed that the dollar share of their reserves is closer to 70% than 60% (supporting evidence). I by contrast have assumed that the GCC’s sovereign funds have a diverse portfolio.

What does the graph tell us?

In my view, three things:

First, the rise in the US current account deficit from 2002 to 2006 is associated with a rise in official demand for US assets. The quarterly IMF data doesn’t extend back to the late 90s – or to the early 1980s. But trust me, that is a change from past periods when the US current account deficit expanded. To be sure, private investors abroad were also buying US assets. But the rise in the overall US financing need associated with the rise in the current account deficit wasn’t financed by a comparable rise in private demand for US assets.

Second, Official demand for US assets soared from the end of 2005 to the end of 2007 – even by the standards of this decade. That surge was hidden, as the majority came from countries that don’t transparently report the currency composition of their reserves (let alone their sovereign funds). But it happened. At the peak of official asset accumulation in late 2007 and early 2008, official demand for US assets (best I can tell, and if my estimates are off – do tell, and explain, with data) exceeded the total US financing need. Central banks and sovereign funds were financing both the US current account deficit and the “diversification” of private US portfolios. Absent that uptick in official demand, the US would have experienced a dollar crisis before it experienced a banking crisis.

Third, official demand for US assets has fallen quite sharply over the last four quarters. Paul Krugman is right. The US depended on China’s central bank – and other official investors – far less over the last four quarters than it did in late 2007 and early 2008. The collapse in gross private capital flows during the crisis rebounded in the United States’ favor, as Americans scaled back their investment abroad faster than foreigners scaled back their investment in the US. The fiscal deficit may be up, but US “dependence” on central banks for financing fell sharply.

At least through the first quarter. The second quarter of 2009 will be to be different. Reserve growth seems to have resumed.

But there is little doubt that reserve growth slowed sharply from mid 2008 to mid 2009. The countries that report detailed data on their reserves to the IMF reduced their dollar reserves from the end of q1 2008 to the end of q1 2009 by about $130 billion. That is a fact not a guess. Guessing the overall total requires guessing what China (and Saudi Arabia did). But there is little doubt that China’s overall reserve growth slowed – so unless it was diversifying into the dollar, its growth in its dollar reserves must also have slowed.

Wait. Doesn’t the US data (including the Fed custodial accounts) tell a somewhat different story? And doesn’t the US data show record inflows into Treasuries?

All true.

My estimates of dollar reserve growth (and the available data from the IMF) are at odds with the US data showing a pick up in official demand for Treasuries.

What gives?

My explanation: central banks reserve managers pulled funds from banks, the US agencies, and private fund managers – producing an uptick in demand for Treasuries.

We know that this happened inside the US. From the end of q1 08 to the end of q1 09 central banks reduced their dollar deposits in US banks by about $200 billion – providing a lot of the funds that flowed into Treasuries. I would bet the same thing happened globally.

From early 2007 to mid 2008, central bank demand – in even the revised US data – lagged my estimated of global dollar reserve growth. From mid 2008 on, central bank demand in the US data has exceeded my estimate for dolllar reserve growth.

I don’t think central banks shifted out of the dollar as rapidly as the US data implies in late 2007 and early 2008 (the IMF data rules out a shift by those countries that report detailed data to the IMF, so such a shift would have to have come from China and the Gulf). And I don’t think central banks shifted back into dollars on the scale the US data now implies. Rather central banks shifted first into riskier dollar assets (and made greater use of private managers) and then shifted back to traditional reserve assets like Treasuries.

That does though mean that there has been an enormous gap between the rhetoric coming from some important holders of reserves — rhetoric that suggests a loss of confidence in the long-term value of US Treasuries — and their actual actions. For all their faults, central bank reserve managers still seem to consider US Treasuries safer than other dollar-denominated assets.

*The underlying data comes from the IMF’s COFER data, the US BEA, the PBoC, SAMA and the national balance of payments of countries with sovereign wealth funds. But I also had to make a few assumptions to flesh out the data, notably assumptions about the dollar share of countries that do not report detailed data about their reserves to the IMF
** I have excluded outward short-term flows because they are highly correlated with short-term inflows; the banking sector doesn’t seem to be a consistent source of net financing for the US.
*** The homeland investment act allowed American firms to repatriate profits earned abroad without paying US corporate income taxes. It led to a $200 billion or so fall in US outward FDI, as American firms (temporarily) stopped reinvesting their foreign profits.

That is unfortunate. It is quite possible to believe that the buildup of vulnerabilities that led to the current crisis was a product both of a rise in savings in key emerging markets, a rise that — with more than a bit of help from emerging market governments — produced an unnatural uphill flow of capital from the emerging world to the advanced economies, and policy failures in the U.S. and Europe.

The savings glut argument was initially put forward to suggest that the United States’ external deficit was a natural response to a rise in savings in the emerging world – and thus to defuse concern about the sustainability of the United States’ large external deficit. But it was equally possible to conclude that the rise in savings in the emerging world reflected policy choices* in the emerging world that helped to maintain an uphill flow of capital – and thus that it wasn’t a natural result of fast growth in the emerging world. This, for example, is the perspective that Martin Wolf takes in his book Fixing Global Finance. Wolf consequently believed that borrowers and lenders alike needed to shift toward a more balanced system even before the current crisis.

From this point of view, the savings glut in the emerging world — as there never was much of a global glut, only a glut in some parts of the world — was in large part a result of product of policies that emerging market economies put in place when the global economy — clearly spurred by monetary and fiscal stimulus in the US — started to recover from the 2000-01 recession. China adopted policies that increased Chinese savings and restrained investment to try to keep the renminbi’s large real depreciation after 2002 – a depreciation that reflected the dollar’s depreciation – from leading to an unwanted rise in inflation. The governments of the oil-exporting economies opted to save most oil windfall – at least initially. Those policies intersected with distorted incentives in the US and European financial sector – the incentives that made private banks and shadow banks willing to take on the risk of lending to ever-more indebted households (a risk that most emerging market central banks didn’t want to take) to lay the foundation for trouble.

On one point, though, there really shouldn’t be much doubt: savings rates rose substantially in the emerging world from 2002 to 2007. Consider the following chart – which shows savings and investment in emerging Asia (developing Asia and the Asian NIEs) and the oil exporters (the Middle East and the Commonwealth of independent states) scaled to world GDP.

Investment in both regions was way up. But savings was up even more.

It is unusual for Asia and the oil exporters to show large surpluses at the same time. In 98 the fall in oil prices helped Asia and hurt the oil exporters; in 2000 the rise in oil prices helped the oil exporters and hurt Asia. And way back in 1980, Asia ran a deficit that helped offset the oil exporters’ surplus.

The main reason for the rise in emerging Asia’s savings is simple: China’s GDP rose relative to world GDP, and China’s savings rate rose relative to China’s GDP

The chart is from Stephen Green of StanChart; used with permission

The result was a very large increase in the aggregate savings of the emerging world – especially after 2003. The rise in the combined surplus of Asia and the oil exporters that followed the Asian crisis was around 0.5% of world GDP. The post 2003 “China boom” pushed the combined savings rate of the oil exporters and emerging Asia up another 1% of world GDP.

All my data, incidentally, comes straight from the IMF’s WEO data tables. All I did was to multiply the data on savings rates by regional GDPs and then scale the resulting dollar figure to world GDP in dollars.

That disaggregated data is almost as striking.

It shows, for one, that the “investment drought” argument applies far more to the Asian NIEs (Korea, Taiwan, Singapore, Hong Kong) than to the rest of Asia. Investment in some countries may not have recovered from the 1998 crisis, but the overall data is dominated by the huge rise investment in developing Asia (read China). Plotting the rise in billions of dollars – rather than as a share of global GDP – makes the scale of the rise in investment in developing Asia over the past few years clear.

Savings and investment in India both rose. And China went from a $1 trillion economy investing 30 to 35% of its GDP to a $4 trillion plus economy investing close over 40% of its GDP …

It is also striking that investment in the Middle East was essentially stagnant, in dollar terms, from say 1980 on. That meant that is was falling as a share of world GDP – and certainly falling relative to the Middle East’s population. Comparisons with the “boom” level of 1980 is a bit unfair, but it still isn’t hard to see why the region stagnated when oil prices stagnated.

And it also isn’t hard to see why the region boomed when oil prices soared, as the rise in oil revenue financed a boom in investment. The scale of that boom – in dollar terms – is rather impressive.

The net result: the global economy prior to the crisis was characterized both by high levels of both savings and investment in Asia and the oil exporters and by high levels of consumption and low levels of savings in the US.

In a global economy, a rise in savings relative to investment in one part of the world necessarily implies a fall in savings relative to investment in the rest of the world; sorting out why key macroeconomic variables change is always difficult.

Maybe this equilibrium was a function of excessive demand stimulus by the advanced economies in the aftermath of the last recession – and lax financial regulation that allowed households to over-borrow. High US and European demand allowed the emerging world to save more. Maybe it was a function of policies in the emerging economies, policies sometimes put in place to support undervalued exchange rates. That would explain why the growing US savings deficit didn’t put upward pressure on global interest rates and why the rise in the US external deficit didn’t lead to a rise in US real interest rates — something would have short-circuited the housing boom. Probably it was a mix of both. Emerging market savers (really their governments, as private savers weren’t exactly seeking out depreciating dollars) helped to provide Wall Street and the City the rope they (almost) used to hang themselves.

No matter. We don’t need to assign responsibility for the imbalances that marked the pre-crisis global economy to know that the chain of risk-taking that allowed emerging market savers to finance heavy borrowing by US households didn’t result in a stable system.

* Policies that increased savings in China include a tight fiscal policy and the reforms that increased the profitability of the SOEs, creating a new source of business savings. No comparable reform was put in place to have the SOEs pay dividends (or to use the dividends to support say a social safety net), so the rise in business savings in effect freed up household savings to be lent abroad (with a lot of help from the state banks and the PBoC). Policies that reduced investment include the rise in the banks’ reserve requirement — which meant that Chinese banks had one of the lowest loan to deposit ratios in the emerging world going into the global slump — and more generally the restraints on bank lending. The governments of most oil-exporting economies also saved a large fraction of the oil windfall, especially in 2004 and 2005. Over time discipline waned a bit, but the rise in spending and investment didn’t quite keep pace with the rise in oil prices.

Yet even if the US data doesn’t show quite as much debt as it probably should, it still tells a lot going about what was on in the US – and the global – economy in the run up to the crisis.

It is consequently tempting to try to do a bit of forensic accounting to help understand how vulnerabilities built up. One thing quickly becomes clear. The US was piling up external debts in the run-up to the crisis even if the United States’ net international investment position wasn’t deteriorating.

The data in the NIIP can be disaggregated into debt and equity fairly easily. It is also fairly easy to separate out net official and net private claims. There isn’t a separate breakout for “official” investments in equities – as central bank and sovereign funds’ equity investments are aggregated together with their investments in US corporate bonds. But the US survey data indicates that official holds of equities were over three times official holdings of corporate bonds in the middle of 2008, so I don’t feel too bad considering “other official assets” a proxy for central bank and sovereign funds’ investment in US equities.

But don’t get bogged down in the details. There is no doubt that the US was clearly racking up debts to both official and private creditors in the run up to the crisis. Net US external debt (US borrowing from the world, net of US lending to the world) is now close to 40% of US GDP — a fairly high level for a country with a modest export sector.

The steady buildup of US external debt though was long offset by the rise in the value of US equity investment abroad. In my calculations I valued FDI at cost; valuing it at its market value would have pushed net US equity holdings (US equity investment abroad – foreign equity investment in the US) up even faster and, of course, produced an even bigger fall in 2008.

Looking at the net data alone can be misleading. In some cases gross positions offset; it is useful to know if a stable net position reflects a symmetric rise in gross positions (or a symmetric fall).

Look at the data on “gross” official flows.

In 2008, for example, a rise in US “official” lending to the world – essentially the Fed’s swap lines – offset an ongoing rise in gross official claims on the US.

This data helps to clarify the debate on the dollar’s status as a global reserve currency – a debate that I often feel misses a key point. The dollar was a reserve currency in the 70s, 80s and 90s too. But total central bank claims on the US generally were under 10% of US GDP. They jumped a bit in the mid-1990s, when a surge in capital inflows to the emerging world allowed many countries to rebuild their reserves. They dipped a bit during the Asian crisis, as many countries few on their reserves to finance capital outflows. But they only started to soar in 2003, when the dollar started to depreciate against the euro and Japan and a host of emerging economies resisted pressure on their currencies to appreciate.

Somehow the usual debate on the dollar’s status as a global reserve currency suggests that little has changed over the past few years – the question is only whether countries will continue to want the dollar to be world’s leading reserve currency. But the debate over “market share” ignores the real issue: the size of the market. The dollar’s share of global reserves didn’t rise over the past several years. Rather countries holdings of reserves soared, and a constant (or even slightly falling) share and much larger stock produced a big rise in central bank holdings of US debt.

That shift mattered: It left many emerging economies with a lot of exposure to the dollar, and left the US exposed to the risk that key countries might lose their appetite to continue to hold dollars.

What of private debts? I plotted gross bank claims on the US against US bank claims on the world, and US holdings of debt securities against foreign holdings of US debt securities.

Gross banks soared in the late 1970s. Petrodollar recycling. It actually was more than just recycling though. US bank claims on the world increased faster than foreign bank claims on the US; in aggregate, US banks were using US deposits to provide financing to the rest of the world. That has changed.

Gross flows have generally trended up, but foreign claims on the US and US claims on the world rose together. The pace of increase though did pick up just prior to the crisis – probably because of the expansion of the shadow financial sector. But I am just guessing. Setting 2005 though, US banks weren’t a net source of financing to the rest of the world.

Foreign and US holdings of debt securities have both increased over time. Foreign holdings of US debt soared in the 1990s. That makes sense. A rising dollar made dollar-denominated US debt an attractive asset. And US firms were borrowing heavily to finance a lot of tech related investment; think of it as the US borrowing to build the information super highway (and no doubt to consume a bit too). The big rise in foreign holdings of US debt from 2002 on is a bit harder to understand. It obviously provided a lot of financing to the US household sector – as foreign demand shifted from “straight” corporate bonds to US asset-backed securities. But these inflows came in the face of a declining dollar. They presumably were done by investors with access to dollar financing – so the investors were taking the credit risk but not the currency risk. The US banking system though wasn’t in aggregate providing credit to the rest of the world, so it wasn’t in a position to finance these purchases. Someone else was supplying a lot of dollar financing in the world’s offshore financial centers (London especially). Figure out who, and I suspect that you have figured out a lot.

A disaggregated plot of the different components of the US net debt position shows that the deterioration in the last decade reflects a rise in (net) official claims on the US, and a rise in (net) private holdings of US debt securities.

The mechanics of the rise in official holdings are well known (China, China, China and to a lesser degree Japan, Russia, Saudi Arabia and Brazil). The mechanics of the rise in net private holdings of dollar securities less so. Who wanted to take US dollar risk as well as US credit risk? And how was the dollar risk of say European banks buying US ABS shed?

A small point: some of the rise in private holdings may reflect disguised official flows, or at least official flows intermediated v private intermediaries. The US data shows that “official” investors had $3.5 trillion in “safe” US assets (I am setting aside official holdings of corporate equities, as those could be held largely by sovereign funds). Counting the PBOC’s other foreign assets and SAMA’s non-reserve foreign assets, the global pool of reserves was around $7.3 trillion at the end of 2008. If 60% of all reserves (a low end estimate) were in dollars, total central bank dollar holdings should be around $4.4 trillion; if 70% of all reserves (a high end estimate) are in dollars, the total rises to $5.1 trillion. That is a gap of between $900 billion and $1.6 trillion — a sizeable sum. Net private holdings of US debt are around $2.4 trillion, though gross holdings are obviously much larger.

At least $300 billion of that is in offshore dollar deposits ($300 billion is the gap between the US data and the BIS data in table 5c), and quite possibly more. Some of the $350 billion in official holdings of “risk” assets shown in the US data is in central bank hands. But between $500b and $1000b is missing – perhaps because it is managed by private fund managers. Central banks handing dollars over to private managers thus could be one explanation for persistent private demand for dollar debt even as the dollar fell. But there are clearly others.

What of equities, and specifically portfolio equities?

The story here is simple. The value of US investment in foreign portfolio equities doubled, as a percent of US GDP, from 2003 to 2007.

That isn’t primarily a reflection of large purchases of foreign equities by US residents. Cumulative purchases of foreign equities by US investors from the end of 2003 to the end of 2007 totaled $560 billion, implying valuation gains accounted for about $2.6 trillion of the $3.17 trillion total rise in US holdings of foreign equities. Rather it reflects the dollars’ depreciation, which pushed up the dollar value of US investments abroad – especially in Europe. And it reflects the fact that foreign stock markets dramatically outperformed the US stock market during this period.

Foreign investors bought nearly as many US equities as US investors bought foreign equities ($520b v $560b); the rapid rise in the value of US equity investment abroad relative to foreign investment in the US reflects the underperformance of US equity markets.

That in some sense is the great puzzle of the last six years. The US ran large deficits – and necessarily attracted large financial inflows – during a period when US markets consistently performed worse than foreign markets. Sure, that changed in the crisis. But the out-performance of the US then (US markets fell less than foreign markets) in the crisis hardly explains the persistence of inflows when returns on both safe and risky investments in the US lagged returns on comparable investments abroad. Foreign investors presumably weren’t buying US assets because they anticipated a US financial crisis.

The ability of the US to finance large deficits – and run up a large debt stock – during a period when returns on US investments lagged is the central puzzle of the global flow of funds. And it seems to me that one at least has to consider the possibility that the financing that supported these deficits weren’t entirely a “market” outcome.

* Some discrepancy between the stock and cumulative flows is to be expected — especially as the stocks get bigger. Things like the repayment of principal on Agency MBS and asset-backed securities make proper calculation of the flow difficult. The presence of a gap in the data isn’t a surprise. The surprise is that the revisions consistency work in the United States favor.

Why? The euros’ rise pushed up the value of US investments in Europe. The US external position actually benefits from a falling dollar, as most US liabilities are in dollars while many US assets are not. And foreign equities did better than US equities.

Those who argued that the US was attracting funds because it was the best place in the world to invest generally forgot to note that during the period when the US deficit was rising the US was consistently doing better on its investments abroad than foreigners were doing on their investment in the US.

That changed in 2008. The US borrowed $505 billion from the world.* The dollar’s rise reduced the value of US investment abroad by $685 billion ($583 billion without including direct investment). Foreign portfolio equity investments in the US fell in value by about $1.3 trillion. But US portfolio equity investments abroad fell in value by about $1.9 trillion. Add in changes in the value of US and foreign bonds and changes in market prices reduced the value of US portfolio investment abroad by $720 billion more than the changes in market prices reduced the value of foreign portfolio investment in the US. The size of that total losses rises if changes in the market value of US and foreign direct investment are also factored in. All told, changes in market prices (other than exchange rate changes) led to a $1.7 trillion deterioration in the United States net investment position.

Combine those valuation changes with ongoing US borrowing and the net international investment position deteriorated in a rather dramatic fashion. The US NIIP — the blue line in the following graph — deteriorated by $2.5 trillion if FDI is valued at market prices.

Underlying data.

Even with this deterioration though the US net international investment position is in better shape than would be expected if you just sum up the United States’ current account deficits — or the financial flows that financed those deficits. The red line represents the US net international investment position implied by just summing up those flows. Nor do valuation gains from currency moves — the dollar’s fall pushes up the value of US assets abroad — explain the gap. The combined impact of currency moves and ongoing deficits — the green line in the graph — isn’t that far from the net international investment position expected by just summing up financial flows.

Nor is the gap explained by better relative performance of US equity investment abroad. That had a large impact on the data for several years, but the market moves of 2008 effectively wipted the cumulative “profits” on US investment abroad from market moves. At least in the data going back to 1990.

The main reason why the net international investment position isn’t worse than it is? The cumulative impact of something called “other changes.” William Cline has called these “other” changes – effectively accounting changes to reconcile the data on financial flows with the data on the stock of US and foreign investments – statistical manna from heaven. He is right. Without these changes, the US net international investment position would be much, much worse.

One quick aside: If FDI is valued at cost rather than at market value, the US net international investment position deteriorated by only $1.3 trillion. It would have been close to the $2 trillion Gian-Maria Milesi-Ferretti predicted if not for a mysterious $420 billion in “other” changes.

There are a couple of other ways of looking at the net international investment position data. One is to disaggregate it into a net debt position and a net equity position — i.e. netting US lending against US borrowing, and US equity investment abroad against foreign equity investment in the US.

My rough calculations suggest US portfolio equity investments abroad are worth about $650 billion more than foreign equity investment in the US (US direct investments abroad are worth about $515 billion more than foreign direct investment in the US). That implies that the US has now borrowed about $4.3 trillion more from the rest of the world than it has lent to the rest of the world.

Foreign central banks – setting aside “other foreign assets” a category that includes a lot of equities – hold about $3.5 trillion of US debt. Even after the extension of $530 billion in swap credit, the US government only has about $700 billion in claims on the rest of the world. The net position of central banks on the US therefore mechanically accounts for the majority ($2.8 trillion) of the world’s net holdings of US debt. That is why there is so much focus on the ongoing willingness of central banks to hold US debt.**

Another little tidbit in the US data: Official investors lost $13 billion on their US investments due to market moves. Private investors by contrast lost $1.22 trillion. They are the ones who should be smarting.

The market value of central banks holdings of Treasuries and Agencies rose by $148 billion in 2008 – or just a bit under 5%. The rise in the value of Treasuries alone was about 5% (a bond whiz should be able to back out a rough estimate of the average duration of central banks’ Treasury portfolio from that data).

Conversely, the market value of “other” official assets – a category that includes the equities and corporate bonds that are held by central banks and sovereign funds (at least those that show up in the US data) – fell in value of $161 billion. That is roughly a 30% loss.

No surprise. But it is kind of ironic that all the angst this year has focused on those parts of central banks portfolio that held up well last year.

China clearly accounts for some of the $161 billion in equity losses – as China (SAFE in my view) bought a large number of equities in the second half of 2007 and the first half of 2008. But I would guess that China accounts for a larger share of the mark-to-market gains on central banks’ bond portfolio than of the losses on central banks and sovereign funds’ equity portfolio. Its overall portfolio held up rather well – all things considered – during the crisis.

I feel a bit guilty for pointing out the large losses China (and others) took on their large equity purchases just before the crisis hit. In aggregate, central banks went into the crisis with a portfolio that was heavy on government bonds – and that portfolio did well.

Then again if China was counting on the mark-to-market gains on its bonds to offset the market losses on its equities, it may need to think again. It never realized the gains on its bond portfolio – and now that Treasury yields have moved back up, China’s bond portfolio has experienced some (modest) mark-to-market losses.

One last set of points — on the academic debate over the United States’ exorbitant privilege and its skill at generating “dark matter” to offset its ongoing trade deficits.

Technically the rise in the US net international investment position doesn’t disprove the dark matter thesis.

“Dark matter” is calculated by calculating the net international investment position implied by discounting net income payments at a fixed 5% discount rate, and comparing that sum to the net international investment position.

The income balance (positive, thanks to income on US FDI abroad) implies, according to this methodology, that the US has about $2.4 trillion in (net) assets, while the NIIP (using FDI at market value) shows $4 trillion in net US debt. Presto, $7.4 trillion in dark matter.

Of course, the argument that dark matter represents US skill at investing abroad – skills that lets the US collect risk premia on those investments – is hard to sustain in a year marked by epic losses of US investment abroad. $2.7 trillion in losses without counting direct investment, close to $5 trillion if mark-to-market losses on direct investment are factored in.

Mechanically, a fall in interest rates increases the amount of dark matter, since the Hausmann and Sturzenegger methodology assumes that the US should be paying 5% on its net debt position, and credits the US with dark matter if the interest rate on US net debt falls below 5%.

And I still think that the gap between the returns on US direct investment abroad and foreign direct investment abroad – a gap that stems entirely from exceptionally low reported earnings on foreign investment in the US (especially low reinvested earnings) – is a function of tax laws that allow US firms to avoid taxes on profits abroad that are held abroad. European and US firms both have an incentive to show profits in low tax jurisdictions in Europe, not the US. That is one of the dangers of inferring assets from the income balance!

NOTE: I edited the middle portion of the post significantly on Saturday to clarify the graphs. My initial post was rather cryptic. I hope the edited version works a bit better.

*Technically, the is net financial flows, so equity investments enter in as well. In practice though it was almost all debt. Treasuries actually.
** All these calculations are a bit rough. I rushed a bit – and reserve the right to refine them as I feed more data into my spreadsheets.

Central banks also seem increasingly interested in five year notes. Indirect bids at today’s five year auction were quite high as well.*

Strong central bank demand for Treasuries shouldn’t be a real surprise. Reserve growth picked up in May: look at Korea, Taiwan, Russia and Hong Kong. There are even rumblings - based on the data that the PBoC puts out — that Chinese reserve growth picked up as well. The rise in reserve growth fits a long-standing pattern: emerging markets tend to add more to their reserves — and specifically their dollar reserves — when the euro is rising against the dollar. A fall in the dollar against the euro often indicates general pressure for the dollar to depreciate — pressure that some central banks resist (Supporting charts can be found at the end of a memo on the dollar that I wrote for the Council’s Center for Preventative Action).

And the Fed’s custodial holdings (securities that the New York Fed holds on behalf of foreign central banks) have been growing at a smart clip. Recent talk about a shift away from a dollar reserves by a few key countries actually coincided with a surge in the Fed’s custodial holdings. Over the last 13 weeks of data, central banks added $160 billion to their custodial accounts, with Treasuries accounting for all the increase.

$160 billion a quarter is $640 billion annualized — a pace that if sustained would be a record. Of course, $640 billion in central bank purchases of Treasuries would still fall well short of meeting the US Treasuries financing need. The math only works if Americans also buy a lot of Treasuries. That is a change.

Still, most emerging economies seem to have concluded that the risks associated with holding too few dollar reserves exceed the risks of holding too many dollars. That doesn’t seem to have changed.

China may be in a different position, but it likely will find that scaling back its dollar exposure is hard so long as it wants to maintain a dollar peg …

The rise in the Fed’s custodial holdings isn’t a perfect indicator of dollar reserve growth. If a reserve manager pulls dollars out of a bank and invests the proceeds in Treasuries, that can show up as a rise in the Fed’s custodial holdings. A reserve manager that shifts a Treasury bond from a private custodian to the Fed can produce a similar result. Both no doubt happened last fall and early this winter. Conversely, central banks can — and do — hold Treasuries with private custodians. From mid 2005 to mid 2006 the rise in central banks holdings of Treasuries (according to the survey) exceeded the rise in the Fed’s custodial holdings.

But the Fed’s data is the best high-frequency data we have got. And if central banks reserves are up and the Fed’s custodial holdings are up, Occam’s razor suggests that central banks are adding to their dollar reserves.

That isn’t to say all is well so long as central banks are adding to their dollar reserves.

On one hand, there is a risk that a return to excessive reserve growth will keep the United States trade deficit from continuing to adjust. They could make it harder for exports to spur US growth. A depreciating dollar is once again producing a depreciating RMB.

And on the other, central bank reserve managers are a lot more comfortable holding short-term US notes than longer-term US notes. There consequently is a potentially a gap between what central bank reserve managers want to buy and what the US wants to issue. That is a more subtle version of the argument that central banks won’t finance the US deficit. Central banks might finance the deficit but not by buying the tenors the US really wants to sell.

Central banks could be clustered at the short-end of the curve because they fear that US inflation will rise — and they don’t want to be stuck with longer-term US bonds then. Or they could just worry that they will buy a bond that yields 3.5% only to see yields rise to 4.5%, producing a mark-to-market loss. Or it could just be a mechanical result of central banks aversion to the risk of any (mark-to-market or accounting) loss. More volatility means a high probability that a longer-term Treasury portfolio might lose value. That mechanically might lead some central banks to shorten the maturity of their holdings.

But there is a limit to how far central banks can go. Bills don’t produce any income, and most central banks need some income from their reserve portfolio. When the yield curve is steep, that generates pressure to hold something that has a slightly longer maturity. The two year note seems to be hitting central bank reserve managers’ sweet spot — and today’s auction suggests that central bank demand for somewhat longer tenors could be picking up as well.

UPDATE. From last Wednesday to this Wednesday, the Fed’s custodial holdings were pretty close to flat (looking at the data from the end of the reporting week, not the change in the weekly averages). From May 27 to June 24 (a four week period), the custodial holdings rose by about $30 billion, with an increase in Treasuries of a bit less than $40 billion and a fall in Agencies of a bit less than $10 billion. That is a solid increase, but a much slower increase than in May — when an uptick in capital flows to the emerging world pushed up reserve growth.

* One caveat: the recent rise in indirect bidding may be — in part — a function of changes in auction rules. “Guaranteed bid arrangements” through the primary dealers have been eliminated. See Jansen (and ultimately the reporting by Min Zeng of Dow Jones)

Shipments to China, Japan’s biggest trading partner, fell 29.7 percent, more than April’s 25.9 percent. Exports to Asia slid 35.5 percent from 33.4 percent a month earlier.

hat tip, Yves Smith of Naked Capitalism.

That isn’t good news. US exports to China are also down (15.6% y/y, through in the first four months of 2009, though a bit less in April itself). The eurozone’s exports to China are also down — though the 8% or so fall y/y fall in the eurozone’s exports to China seems a bit more modest than the fall in Japan’s exports to China.

China’s economy may have expanded over the last year, but that expansion clearly hasn’t fed through into more Chinese demand for US, European or Japanese goods.
Not yet at least. Pick your explanation. China’s stimulus may have been directed at domestic producers. The process of substituting Chinese components for Japanese components in China’s exports may be accelerating. Or China’s recovery just may not be quite as robust as some believe.

The best that can be said of Japan’s May trade data is that Japan’s exports to China aren’t down as much as Japan’s exports to the US and Europe.

Shipments to the U.S. fell 45.4 percent in May after dropping 46.3 percent in April, the ministry said. Exports to Europe slid 45.4 percent from 45.3 percent.

The y/y comparison will get more favorable soon. But there is now real way to put all that positive gloss on Japan’s 41% year over year fall in exports. It is an epic fall.

Japan’s May 2009 exports were even a bit lower than its April 2009 exports. There may be some benign explanation for the slight dip in May, but I don’t think there is any way to suggest that the Japan’s May trade data suggests a robust global recovery.

Trade (as we all know) contracted far more rapidly during this cycle than in the past.

But the fall in private financial flows — outflows as well as inflows — has been even sharper than the fall in trade flows. US private investment in the rest of the world rebounded a bit in the first quarter, but private demand for US financial assets remained in the doldrums. Private investors were still pulling funds out of the US in the first quarter.

A close examination of the graph indicates that demand for US financial assets by private investors abroad actually peaked in the second quarter of 2007 — a peak that came after gross private flows (inflows as well as outflows) rose strongly in 2005 and 2006. That surge was — in my view — linked to the chain of risk associated with a world where central banks took the currency risk associated with financing the US external deficit and private intermediaries took the credit risk associated with financing ever more indebted US households.

Any interpretation of what caused the crisis has to explain this surge. But any interpretation of the crisis also needs to explain why US imports and exports continued to rise — and the US trade and current account deficit remained large — even after private inflows collapsed.

I suspect that part of the answer is that a lot of private inflows were linked to private outflows — as special investment vehicles operating in say the US could only buy long-term US mortgage bonds if someone in the US bought their short-term paper. The fact that private outflows collapsed along with private inflows meant that net private flows didn’t fall at the same rate. Indeed, at times - notably in q4 2008 — the fact that US investors pulled funds out of the rest of the world faster than foreign investors pulled funds out of the US provided the US with a significant amount of net financing.

And part of the answer is that private investors never were the only source of financing for the US current account deficit. Strong central bank demand — especially in late 2007 and early 2008 — offset a fall in private flows.

One thing though is sure: the scale of the collapse in private financial flows during this crisis is entirely unprecedented. There were a few instances in the past when private flows (excluding flows into Treasuries) were slightly negative. But outflows of 5% of GDP in a quarter are entirely unprecedented. And now that the US data has been revised to reflect the survey, adding private purchases of Treasuries back in doesn’t change all that much …

Net private demand for long-term US financial assets - that is purchases of US securities and foreign direct investment in the US, net of US purchases of foreign securities and US direct investment abroad — has been weak for some time now.

There is more to say on the details of the balance of payments data, but I’ll live it for another post.

One. China is growing, thanks to China’s government. The World Bank estimates that the government’s policy response will account for about 6 percentage points of China’s 7.2% forecast growth (p. 8). That’s good. There is a big difference between growing as 7% and growing at 1%. This was the right time for China’s government to “unchain” the state banks. Ok, it would have been better if China had allowed its currency to appreciate back in late 2003 and early 2004 to cool an overheated economy instead of imposing administrative curbs on bank credit and curbing domestic demand. Then China might not have ever developed such a huge current account surplus and avoided falling into a dollar trap. But better late than never: this was the right time to lift any policy restraints on domestic demand growth.

China has, in effect, adopted its own version of credit easing. It just works through the balance sheets of the state banks rather than through the balance sheet of the central bank. Andrew Batson:

By some indicators, credit in China is even looser than in the U.S., where the Federal Reserve has extended unprecedented support to private markets. … China’s methods for pumping cash into the economy are quite different from those of other major economies. Its banks, almost all of which are state-owned, made more than three times as many new loans in the first quarter as a year earlier. Central banks in the U.S., Europe and Japan lack such control over lending, and have instead used extremely low interest rates and direct purchases of securities to support credit.

Two. China’s fiscal deficit will be closer to 5 percent of GDP rather than 3 percent of GDP. That’s cause for celebration in my book. Last fall I was worried that the desire to limit the fiscal deficit to three percent of GDP would mean that there was less to China’s stimulus than met the eye (or hit the presses). I was wrong. If the likely future losses on the rapid expansion of bank credit are combined with the direct fiscal stimulus, China almost certainly produced a bigger stimulus program than any other major economy.

Three. China’s current account surplus is now projected to fall in 2009. Exports still haven’t picked up — and we now have data through the first five months of the year. Imports by contrast are starting to pick up. That shows up clearly in a chart of real imports and real exports, a chart that draws on data that that the World Bank’s Beijing office generously supplied me:

Some of that is commodity stockpiling and thus not a reflection of underlining demand. And some stockpiling sounds a lot like simple speculation. But let’s set those debates aside for a bit.

In dollar terms, China’s 2009 current account surplus will be a bit smaller than its 2008 surplus (The World Bank assumes that China’s trade surplus in the last half of 09 to be significantly smaller than its surplus in 08, as the surplus was up in the first five months of 09). And since the fall in commodity prices would be expected to push the surplus up, all other things being equal, that indicates a real shift in net exports. Net exports, according to Dr. Kuijs of the Bank’s Beijing office, will subtract about 2.5% from China’s overall GDP growth in 2009.

The not-so-good news:

One. It isn’t clear that China has put in place policies that will bring about a sustained rise in domestic consumption. The stimulus has worked by pumping up investment, especially state investment. And investment already loomed large in the national accounts. Olivier Blanchard:

“In response to the crisis, China has embarked on a major fiscal expansion, with a focus on investment rather than on consumption. This was the right policy given the need to increase spending quickly, but this increase in investment can only last for a while. The question is whether, as time passes, China will allow an increase in consumption.”

Two. The negative drag from net exports stems from a faster fall in real exports than in real imports, not a rise in real imports that exceeds the rise in real exports. For the year, real exports are forecast to fall by 10% and real imports by almost 5%.

If China’s exports fall faster than global demand, that opens up space that allows others to cut back less. The alternative — fast Chinese export growth amid a shrinking global economy — would be a sure source of trouble. But China still isn’t really acting as a locomotive for overall global demand growth.

Three: Real imports of manufactured goods are still down 16% y/y. The rebound in Chinese imports has been driven entirely by the rise in commodity imports; real imports of primary products were up 17% (y/y) in April.

Back in 2003, when China was going through another lending boom, real imports of all kinds were up way more than they are now. Of course, back in 2003, China’s export sector was also booming and that pulled in imports for the “processsing trade.” The comparison isn’t perfect. But it does highlight how different China’s current lending and investment boom is from past lending and investment booms.

Part of the explanation for the weak rebound in Chinese demand for manufactures is no doubt weak demand for exports, and thus weak demand for imported components.

And part of it is that there is plenty of spare capacity in China to meet a surge in Chinese demand. For say cars. Chinese auto sales may top US auto sales this year – and China seems able to meet that rise in demand without importing a lot of finished cars or auto parts.

But part of it seems to be that Chinese consumers are less interested in Western — or even Korean and Japanese– brands. Maybe Chinese consumers concluded that if foreign banks weren’t better than Chinese banks, they shouldn’t assume that foreign goods were better than Chinese goods.

And China’s government also seems particularly keen on making sure China’s stimulus is spent in China. Jamil Anderlini reports: “Beijing said government procurement must use only Chinese products or services unless they were not available within the country or could not be bought on reasonable commercial or legal terms.”

Kind of risky for a country that still exports way more than it imports.

But it shouldn’t be a total surprise. The usual argument for why China would keep its exchange rate undervalued even though the undervalued exchange rate meant that China was overpaying for foreign assets and thus would eventually take losses was that China needed to keep up Chinese employment, and this was a way to do so. And don’t forget, the undervalued renminbi has encouraged jobs through import substitution – not just the expansion of China’s export sector. China’s hasn’t been interested in undistorted trade; it has been interested in using trade to support domestic activity in China. It isn’t a huge jump then to see why China might want to make sure that its domestic stimulus creates, in the first instance, as many Chinese jobs as possible.

But it does suggest that China’s commitment to say the G-20 is limited. Just giving China a seat at the international negotiating table won’t necessarily change China’s policies.

All in all, I would say the good trumps the bad. But real problems will come if China’s buy China policy is still holding down Chinese demand for the world’s goods when global demand for Chinese goods returns; rising exports and still stagnant manufacturing imports from the world’s biggest surplus country wouldn’t be terribly popular globally.