What caused the Crisis? It is a story in four stages, summarized as follows:
1. Too much credit? An international perspective
2. Too much risk. Reaction to low real interest rate
3. The fatal flaw. The new complex financial instruments
4. The panic. Bank lending dries up.
> Too much Credit? The international Savings Glut.
The story begins with the “international savings glut.”
Various countries at various times have run high current account surpluses, reflecting an excess of savings over domestic investment. For many years, up to 2003, Japan has had the biggest surplus (in US dollar terms). Then from 2004 the oil exporters’ combined surplus became large, as also that of Germany. But there have been many other surplus countries.
Above all, in 2005 the Chinese surplus increased, and by 2007 it was the biggest surplus of all. In 2007, combining all the surplus countries, 21.4% of the total world surplus was accounted for by China, 19.7% by the major oil exporters, 12.6% by Japan and 11% by Germany. (IMF, 2008, p.131).
Naturally this “savings glut”, as it has been called, would reduce world real interest rates, World exports of capital must be equal to total world imports of capital, so that current account surpluses in total will be matched by current account deficits. The decline in world real interest rates will bring this about. The US fiscal deficit became high from 2002 (3.8% of GDP) and in 2003 was about equal to the US current account deficit. It had the opposite effect on world real interest rates, tending to raise them. But on balance, real interest rates fell. The effect of the savings glut coming from Japan, China, Germany, the oil exporters, and some other countries outweighed the effect of the Bush deficit policy.
As borrowing became cheaper and more readily available spending both for consumption and for investment increased in some countries, especially the United States, and, indeed, that was the process by which the world total of current account deficits became equal to total surpluses. In 2007 nearly half of the total world deficits was the US deficit. Other significant deficit countries were Spain, the UK and Australia (but there were also many others).
The story of these “global imbalances” has been told in many places. See, for example Corden (2007). The essential point is that, for various reasons, net savings outside the United States increased from about 1999, and this reduced world real interest rates. This, in turn, induced higher private sector spending in the United States and some other countries, notably Spain, the UK and Australia.
One might distinguish those changes that were exogenous and those that were endogenous, the latter brought about by the decline in real interest rates, and hence equilibrating the world system (Corden, 2007). The most important endogenous effect was to stimulate housing booms in the United States and some other countries, notably the three just listed.
The US fiscal deficit was exogenous. It accounted for approximately the whole of the US current account deficit in 2003 (4.8% of GDP), but after that it declined relative to GDP, and hence relative to the private sector deficit. By 2007 the US current account deficit was 5.7% of GDP while the fiscal deficit was only 2.6%. (This refers to the “general government fiscal deficit”).
Why a US Housing Boom ?
The question arises why so much of the world savings glut was absorbed, at least since 2005, by the US private sector and, indeed by the household rather than the corporate sector. Overwhelmingly it was absorbed by consumption and by housing construction, all resulting from a housing boom. There seem to be three reasons why the world savings glut ended up specifically in a US housing boom, having then the various well-known consequences discussed in this paper.
Firstly the US economy is about 30% of the world economy; that is sufficient to explain why any effect in the US is likely to be relatively large in the world. Even the big housing boom in Spain ( higher relative to GDP than the US boom) could not have such a marked world wide effect even if all the details had been the same as in the US. Secondly, private non-financial corporations in the US and some other countries, did not expand and hence borrow very much during the relevant period because of their caution resulting from the earlier bursting of the ”dot com” bubble. (This is just a hypothesis here). Thirdly, many developing countries, especially in Latin America, were reluctant to borrow and thus generate current account deficits, because of the traumatic effects of their earlier debt crises. Similarly, some Asian countries – notably Korea and China – ran surpluses partly for that reason – to build up foreign exchange reserves, and thus make them independent of the IMF in case of a crisis.
The Monetary Policy Reaction
The main conclusion at this point is that the world savings glut lowered world real interest rates and made credit more readily available, and thus underlies the further effects to be discussed here. But there is something I have left implicit. Where do the monetary policy reactions of central banks, especially the US “Fed”, fit in? During the period under discussion inflation has been low and employment high. It has been a wonderful period – the “Great Moderation”. Thus, I have assumed so far that central banks in general, especially the Fed, successfully pursued policies of “internal balance”. This raises a question: can the ready availability of credit in the United States – which underlies our story here, and which allowed a housing bubble to develop – be attributed to the international savings glut or to the policy of the Fed? I shall come back to this interesting issue in Part II.
Too much Risk? The Search for Yield
Low real interest rates led to the “search for yield”. In general, only investments believed to be risky are likely to offer substantially higher returns. This meant that the various financial intermediaries, notably the commercial and the investment banks, were willing to run more risk for the sake of getting higher returns. One could argue that this was a rational response. After all, capitalism is all about risk taking. “Nothing ventured, nothing gained”. The lower the return from safe investments the more rational it is to go for risky, higher return investments. At a time when the investment decision is made the probability of the bad event happening may be very low. Why not gamble for the sake of the immediate high return and hope for the best?
Only with hindsight does it look like a mistake, or even just bad luck. And so banks and other institutions borrowed short term at low interest rates and made splendid profits by lending at higher rates. There were wonderful private equity deals, and much else. Huge salaries and bonuses were extracted. Leverage went up, and liquidity went down. There is much evidence that many banks became increasingly dependent on short term borrowing from the wholesale market relative both to their retail deposit base and, above all, their own capital. And this was risky.
But was it rational, or just bad management or judgment? As has happened before, the possibility of crises, and that high returns usually involved serious risk, was forgotten because there had been a long period of high growth and macroeconomic stability. This was particularly true because so many of the potential risk takers were young and had never experienced a crisis.
The Principal-Agent Problem
There was another factor, much talked about recently. This was the principal-agent problem applied to banks. The people that took the risks – and so got the business for the firm – received short-term rewards in the form of high salaries and, above all, bonuses, and suffered no losses – indeed may have left the firm – when the crisis came. They had no interest in the long-term success or even survival of the firm. This was not always so; for example, the employees of Lehman Brothers had much (or a substantial part) of their wealth invested in the firm. There was also sometimes a failure of corporate governance, when management failed to supervise and reign in the risk takers within the firm. In any case there was a divergence of interest between the personal interests of the risk-takers who were rewarded for short-term results and the long-term interests of the firm, its shareholders, and its long-term employees.
The Fatal Flaw: The new Financial Instruments
The new financial instruments or “structured finance” were the reason that a crisis which might have been confined to a few US States actually spread all over the USA and indeed the world.
As is well known, it began with the application in the US housing finance sector of the “originate-to-distribute” model. The aim was to spread the risks of mortgage loans, and this was indeed done. Let me quote from IMF (2008), which is the best exposition. “Structured finance normally entails aggregating multiple underlying risks (such as market and credit risks) by pooling instruments subject to those risks (e.g. bonds, loans or mortgage-backed securities) and then dividing resulting cash flows into “tranches” or slices, paid to different holders.”
The US originators, in effect, sold their risks into a market where the buyers were literally everywhere, in the USA and abroad. The principal instruments were MBSs (mortgage backed securities) and CDOs (collateralized debt obligations). Also important were CDSs (credit default swaps), which were a form of insurance against default. These depended on the liquidity and solvency of the insurer, for example the huge insurance company AIG which the government had to take over.
One might note here that, while the originators of the loans were local US banks and mortgage companies of various kinds, the actual construction of the new instruments was done primarily by investment banks (such as Lehman Brothers) for whom this was a very profitable activity, at least unless they retained a significant quantity for themselves to hold.
Three Problems of the new Model
Focusing on US housing loans, where the whole disaster began, there were three problems.
Firstly, the originators of the mortgages had little or no incentive to ensure that the mortgagees could afford to take out the mortgages without default. Only if they retained some of the loans or instruments would they have an incentive, This is very different from the old-fashioned way when local banks retained the risks and therefore made themselves adequately familiar with the mortgagees. It is not surprising that the new procedure is called the “originate-to-distribute model”.
Secondly, it became impossible for mortgagees to renegotiate the loans with the ultimate lenders, since the latter were effectively dispersed. One might contrast this with the negotiations that took place in the eighties between the governments of developing countries and the international banks in resolving the LDC debt crisis.
Thirdly, there was an effect that was fatal, and indeed set off the world credit crisis. Once the US housing market went into decline and a proportion of “sub-prime” mortgagees defaulted there was a critical information problem. Holders of these instruments, which were composites of many different mortgages, did not know – and could not know – what risks they were running. All they knew was that they could make big losses – or they might not. As a result they wanted to get rid of them, and the market value of the instruments fell dramatically. A device which was meant to off-load and spread risk – which indeed it did – spread fear. And this led to the next step, Panic.
Why buy “toxic” Securities?
Before going on one might ask a simple question: why did banks, and other financial entities, such as insurance companies, mutual funds, and investors of various kinds buy these “toxic” securities and put them in their portfolios? The answer was that, in the absence of any default, the returns were expected to be high. The decline of the US housing market was not anticipated. It was another case of “the search for yield” which in old- fashioned language some might describe as “greed”. In effect buyers of these products of structured finance were sold poison – “toxic” was a later description – even though the sellers might not have realized they were selling poison. Assuming that the sellers were not consciously fraudulent one must assume instead that both sellers and buyers were misled by the extraordinary complexity of these structured instruments. As is well-known by now, they were also misled by the credit rating agencies. One should not buy or sell what one cannot understand!
The final stage has been a panic, which has converted a US housing market crisis into a world-wide banking crisis. And, in turn, a banking crisis is likely to create a serious recession – and in the absence of adequate policy reaction – would create a depression.
The process is fairly simple. Banks become unsure about their own balance sheets and, in addition, the balance sheets of other banks – the counterparties – with which they deal regularly through the interbank market. Even if retail deposits are guaranteed by the government, so that a classic run on deposits is avoided, any bank or other financial institution that depends on the interbank market or the wholesale funds market is in trouble. Illiquidity is severely punished.
It all began with the information problem caused by the combination of a downturn in the US housing market and the complexity of structured finance. Banks stop lending, so that ordinary non-financial corporations, small businesses and large, are drained of life-blood. A crisis in Wall Street creates a crisis in “Main Street” It is for that reason, and not for the sake of rescuing Wall Street, that the US government has needed to intervene.
This is where we now are.
Corden, W.M. (2007) ‘Those Current Account Imbalances: A Sceptical View’ The World Economy,30, 363-382.
International Monetary Fund (2008) Global Financial Stability Report April 08.