Global imbalances and the accumulation of risk

Daniel Gros 11 June 2009



It is often argued that a key factor behind the current financial crisis has been the large US current account deficit. However, the raison d’être of a financial system is dealing with imbalances (between savers and investors). Hence the question is why should the existence of current account “imbalances”, even if they persist for some time, provoke the biggest financial crisis in living history?

The answer must come from the huge, structural build up of a mismatch between asset supply and demand that arose from what are commonly called “global imbalances”. As is well known, the current account deficit of the US arose from an unsustainable increase in consumption (and residential construction). This excess of domestic spending was financed mainly through an increase in the mortgage debt of US households. One key characteristic of mortgages is that they are long-term (often for 30 years). The consumption spree of US households thus led to a large additional supply of long-term (private) assets.

However, this supply of longer-term assets was not matched by a corresponding demand for this type of assets. The excess savings from China (and other emerging economies and oil producers) were mostly intermediated by their central bank, which accumulated huge foreign exchange reserves. These reserves were (and still are) almost exclusively invested in short- to medium-term, safe (i.e. government) and liquid securities (mostly in the US).1 There was thus a need for maturity (and risk) transformation on a very large scale to meet a persistent excess demand for safe and liquid assets.

Figure 1 shows the relevant data. There is a close correlation between the US current account deficit and reserve accumulation, but it is not perfect since the US deficit had already been very large some time before the ‘search for yield’ started. But before 2003 reserve accumulation had been much lower than the US deficit (which had thus been financed largely by private capital transfers). By contrast, after this date reserve accumulation increased relative to the (increasing) US deficit until, by 2006, reserve accumulation actually surpassed by far the US deficit. There is thus certainly a link between the US current account deficit and the build up of the crisis, but this not as straightforward as sometimes believed.

Part of the build up of reserves went also into Euros. IMF data suggest that this part was relatively minor (20-30 %), but it might still have had an impact on government debt in the euro area, contributing to lower interest rates and a compression of yield differentials in Europe as well. Securitisation started in the euro area around this date, although it never acquired the same scale as in the US.

Figure 1. Reserve accumulation by emerging economies and the US current account deficit (USD billion)

Source: IMF, World Economic Outlook database April 2009, ”Change in reserves”

Another way to look at the same phenomenon is to note that the increased demand for US government debt by emerging economy central banks led to lower yields, thus forcing those savers in the OECD countries which would normally have held government assets to frantically “search for returns”. But this was a search for yield on safe (and liquid) assets. The AAA tranches on securitised US mortgages (and other debt) seemed to provide the safety plus a “yield pick up” without any risk, at least in the sense that the securities were rated AAA.

As long as US house prices kept on increasing and unemployment remained low, actual delinquencies remained low and there seemed to be no reason for market participants to question the high ratings of these securities, even though the incentive for the ratings agencies to provide favourable ratings were well known. AAA-rated residential mortgage-backed securities thus provided an important source of liquidity by their widespread use as collateral.

From flows to stocks

Most analysis of global imbalances has focused on the size of the flows, namely the current account deficit of the US relative to US GDP or world savings. Accordingly, most concerns about global imbalances emphasised the magnitude of the exchange rate adjustment that would be required to rebalance US spending and absorption. However, this aspect turned out not to have been crucial. Instead the severity of the present crisis is due to the unprecedented magnitude of the cumulated imbalances in the stocks of assets and liabilities.

The magnitudes of the imbalances between asset supply and demand that cumulated over time are gigantic. Over the period 2000-07, the cumulated US current account deficit amounted to almost $5 thousand billion and US household debt increased by almost $7 thousand billion, of which approximately $5 thousand billion was in the form of mortgages. Meanwhile the foreign exchange reserves of emerging markets increased by about $4 thousand billion (of which the Chinese central bank accounted for about a third). The financial system thus had to transform thousands of billions of dollars of US household mortgages into the type of assets in excess demand from those investors who had been crowded out of the government debt market due to the reserve accumulation by emerging market central banks. In doing so, it took an enormous macro risk (Brender and Pisani 2009).

The key technology that permitted the transformation of US mortgages into safe liquid assets was securitisation. Until 2007, it was widely believed that securitisation should lead to a better distribution of risk since the “originate to distribute model – in its pure form – implies a full risk transfer to the buyers of the various forms of asset-backed securities (ABS) and residential mortgage-backed securities (RMBS). However, in the context of global imbalances this could not have happened on a large scale since the massive buying of US government paper by emerging market central banks had displaced other investors whose preference previously had been for safe, short-term, liquid assets. ABS, especially RMBS do not, a priori, have these qualities. A piece of a pool of mortgages represents a longer-term asset; it is only as safe as the underlying mortgages and is only liquid if there is a demand for this specific asset. Government paper of a given maturity is highly substitutable, whereas every asset-backed security constitutes a special case and thus by its nature much less liquid. Ultimately an RMBS more closely resembles an equity investment in a regional mortgage lender than a government bond.

The excess demand for short-term, safe, liquid assets created by emerging economies’ accumulation of reserves could not have been satisfied by the securitisation of US mortgages (and consumer credit) without massive credit and liquidity “enhancements” by the banking system. A clean securitisation with full risk transfer to the investor was thus not possible from a general equilibrium point of view.

How residential mortgage-backed securities were made safe, short-term, and liquid? The exact way in which this was achieved varies enormously from case to case, but the general rules of the game were the following:


As already mentioned above, the appearance of safety was created by the slicing of tranches coupled with high (AAA) ratings for the most senior tranches (in reality most often about 85% of the total as experience suggested that a total loss of over 15% was extremely unlikely to occur). This service was provided by the ratings agencies for which it represented a major source of income.2

Short- to medium-term

Banks or shadow banking institutions like special investment vehicles used RMBS (and similar assets) as collateral to borrow more funds, e.g. by issuing asset-backed commercial paper, which is short-term and thus the kind of assets that were in excess demand. Issuance of asset-backed commercial paper, which started surging around 2003 (around the same time as reserve accumulation by emerging economies also increased, as shown in Figure 1), constitutes a classic maturity transformation, which was very profitable (given the absence of capital requirements) as long as central banks kept short-term interest rates low and promised (as did the Federal Reserve) to increase them only at a “measured pace”.


Asset-backed commercial paper was already more liquid than the assets with which it was backed. However, such programs were usually possible only if a bank provided a back-up line of credit. Only the banking system could provide the back the stop liquidity that was required by the ultimate investors.

All these elements were necessary to recycle excess emerging market savings to dis-saving US households. Banks had to provide the maturity transformation and the credit enhancement that later proved so costly to them. This transformation required, of course, a huge increase in the balance sheet of the banking (and shadow banking) system and thus a huge increase in leverage.3 This increase in leverage, in turn, acted as a powerful amplifier once risk returned.


When one looks at the risk that persistent global current account imbalances may create for finance stability, one has to take into account the way that current account deficits are financed and how flow imbalances accumulate into large stock disequilibria.

Looking forward, this analysis implies that the current (smaller but still sizeable) US current account deficit should not lead to similar asset supply and demand mismatches since US households are now starting to save and it is the US government which is running the deficit, thus supplying exactly the kind of assets needed by emerging economy central banks.


Brender, Anton and Florence Pisani (2009) “Globalised finance and its collapse” Dexia, Brussels.
Steil, Benn (2009) “Lessons of the Financial Crisis”, Council on Foreign Relations, Centre for Geoeconomic Studies, Special Report No. 45, March.

1 Brender and Pisani (2009) reports that about onethird of all foreign exchange reserves are in the form of bank deposits. Little is known about the maturity composition of the remainder, most of which is invested in interest-bearing securities. The scarce available data on the composition of USD foreign exchange reserves that can be gleaned from the US Treasury International Capital data suggests that over half of foreign official holdings of US securities had a maturity of less than three years.
2 Benn Steil (2009) shows that the correlation between profits of the major ratings agencies and the number of securitised assets rated by them is almost perfect.
3 An increase in capital commensurate with the risk taken by the financial sector would of course have limited the damage, but it would probably have made this transformation too expensive.



Topics:  International finance

Tags:  global imbalances, US current account deficit, financial crisis

Director of the Centre for European Policy Studies, Brussels


CEPR Policy Research