We have had many bizarre analyses of the credit crunch in recent days, as remote from an understanding of the economic theory as they have been from a knowledge of the facts. I want to sketch for you my current interpretation of events. I will inevitably over-simplify a number of issues and miss out a discussion of others, but the core of my tale will, I hope, be clear and intelligible to lay readers. You will also, I hope, see the political significance. For in my view a deeply flawed analysis of these events is emerging, and risks reaching at-least-temporary consensus, with profoundly negative implications for future policy and future political philosophy.
Here are the factors I currently consider most important in explaining these events:
- Genuinely valuable financial innovation: Over the past decade there have been genuine financial innovations that really did increase the extent to which it was possible to diversify away risk. For example, If I own one one millionth of one million mortgages, I am less exposed to the risk of individual people defaulting on their mortgages than if I own just one mortgage. But how much less exposed am I? I have not eliminated all uncertainty, all volatility in the stream of payments I will receive from mortgage holders, for what if if there is some event that affects the whole mortgage market, so that many more people than is normal default at the same time? Perhaps even the risk of that is reduced - perhaps financial innovation means that the economy as a whole will be less volatile in the future. But how much less? It is often the case with innovations, of all sorts (not just financial), that when they first arrive people aren't sure how best to use them, and sometimes over-estimate how much difference they will make in the short term (even if they indeed will eventually make a huge difference). It appears that in this case, people in financial markets over-estimated how much risk had been reduced by these new products, and consequently paid too much for them. This necessitated a market correction.
- Over-confidence in regulatory badging: The problem of mis-pricing of risk was made worse as a consequence of hubristic regulation. Regulators in many key authorities only permit institutions to function and certain individuals to sell financial products if they are appropriately authorised. This authorisation serves to provide consumers with confidence as to the competence and informedness of those offering them products. I suspect that in this case that confidence was mis-placed. Regulatory authorities did not have a good basis for gainsaying financial institutions in their selling and pricing of these products, or in the purchasing of these products by institutions under regulatory oversight - their very novelty prevented that. Why would regulatory authorities have a better view than market participants about these matters? And ordinary consumers and shareholders trusted that the banks in which they deposited funds or held shares, and the financial companies from which they bought products, must be behaving themselves properly - otherwise the regulatory authorities would be complaining, wouldn't they? The consequence of all this was that if the regulatory authorities said things were okay, then no-one else was going to check. They just bought what was before them, pensions, deposits, shares, and more complicated products, assuming it couldn't be that bad and that the underlying behaviour of the institutions backing these products must be okay. Market monitoring was crowded out by state monitoring, but this was hubris on the part of regulators for they were not well placed to monitor these matters.
- Flaws in the Market's and the regulatory authorities' interaction with ratings agencies: Amongst the agents that over-estimated the reduction in risk because of financial market innovations were ratings agencies. Markets and regulators had probably become complacent in their reliance upon ratings - assuming that if something were AAA rated then it must be okay. Markets will think about ratings agency ratings differently in the future - markets will learn from their mistakes. Regulators will probably change the rules relating to ratings agencies, and will certainly adjust their own degree of reliance upon ratings agencies.
- Use of novel products to by-pass regulatory requirements: We have seen time after time that financial markets respond to restrictions on activities that are frowned upon by devising clever new techniques that fall outside the regulatory net. That seems to me to be perfectly proper behaviour on the part of financial firms - they only need not do what they are forbidden from doing. New financial products of recent years have allowed institutions to by-pass banking prudential requirements and other kinds of regulation. I do not believe that this can be prevented without totally undermining financial innovation and the workings of financial markets. The proper response is to be more modest in one's initial expectations of what regulation can really achieve, and to trust in and value market mechanisms. For a standard sequence of events goes: Transparent market mechanisms create outcome authorities do not like => New regulatory requirements, undermining Market controls => Financial innovation to get around regulatory requirements => Regulatory control undermined as well as market control => Outcome authorities like even less => calls for new regulation (and the cycle starts again).
- Extreme moral hazard in respect of housing: House prices have come to be closely associated with political fortunes in recent decades. Governments have been seen to take political responsibility for house prices, boasting when they rose and suffering when they fell. As a consequence, financial markets anticipated that governments would intervene in the event that house prices fell, to limit defaults and foreclosures. Because of this, lending associated with housing was seen as very much lower risk than would be its natural status in a market in which governments would not intervene. As a consequence, it was appropriate for lenders to take far greater "market" risks in lending on housing - driving up house prices to extreme levels until the point at which the government bailout feasibilities were tested. The answer is not for governments to validate moral hazard expectations of the past by intervening today. Otherwise housing will continue to be subject to wild cycles of this sort. Instead, governments must find ways to eschew responsibility for house prices, and to give this credibility by enacting monetary policy mechanisms that will lean against the wind of house price effects.
- A flaw in the orthodox monetary policy regimes: I have explained before my view that monetary policy targeting short-term inflation is flawed in that when it is most credible it automatically starts to give rise to asset price bubbles. Specifically, in response to the dotcom crash, 9/11, the Enron affair, and then the Iraq War bear market, interest rates were kept very low creating excess liquidity. There was an opportunity to mop up this excess liquidity in 2004 and 2005, but because it was not issuing in inflation at that point, and because there was understood to be a risk that aggressive tightening would threaten already-over-inflated housing markets, the mopping up was not done (indeed, under inflation targeting it was not strictly necessary). This exacerbated the problems above by creating a situation in which much liquidity was available seeking return to be paired with novel products (poorly-understood, but regulatorily-badged and associated with government-back housing) offering decent returns. This drove classic Austrian mal-investment.
- Markets awaiting bailouts: Once matters were underway, market participants in many cases preferred to try to outwait the regulatory authorities so as to secure bailouts or subsidies or other favourable terms (e.g. for takeovers), rather than crystallize losses or enter into market rescues early. This meant that the markets own stabilization mechanisms tended to be undermined by the implicit promise of government intervention if it were needed.
I believe that other factors - e.g. the thought that there was "bad luck" because the above was paired with rising oil prices and inflation - are rather confused. The extent of oil price rises was another manifestation of excess liquidity, and inflation was the consequence of excessive monetary growth.
I do not believe that there was any necessity that the UK suffer equally with the US. In the 1930s, US real GDP fell by around a third. But UK GDP fell only 5% - indeed, in growth terms the 1930s was no worse than the recessions of the early 1980s or early 1990s. Why was that? Well, there are a number of factors, but the single most important was probably that the UK financial system did not suffer from anything like the same degree of crisis that there was in the US. This was for two reasons: firstly, matters in the late 1920s did not get as out of hand on the "boom" side in the UK as in the US. But also, on the "bust" side the UK financial regulation framework kept matters moving.
But don't we live in a much more globalised world now? Could we really avoid suffering almost as much as the US? UK financial services have, over the past few years, become enormously important globally. London has at least come to challenge, if not overtake New York in this area. It is by no means obvious why a US financial crisis should not have been to the benefit of the UK - chasing business here - rather than the detriment. In addition to the problems mentioned above (which were shared, largely, by the UK) I attribute the UK-specific problem to two other key factors:
- The foolish Tripartite system of regulation. The destruction of the Bank of England's role as overseer of financial market stability, empowered with prudential oversight of banks and the untrammelled ability to force fait accompli takeovers of failing institutions will surely rank as one of the most misguided regulatory decisions for decades.
- The use of deposit insurance. Deposit insurance is a deeply flawed concept, which encourages excessive risk-taking by depository institutions (eliminating the need for depositors to monitor the behaviour of banks) and provides political comfort to regulatory authorities (allowing those authorities not to act early so as to restrict excessive risk-taking by depository institutions, because if worst comes to worst the depositors will be ensured and the political damage thereby limited). The only insurance there should be should be for payment system volumes - so that people have confidence to use banks rather than notes and coin to make payments. If you deposit £50,000 you are an investor, and you should accept risk. If you want your money safe, hire a security house to keep it in a vault. If you lend it (which is what depositing at interest is), you risk it.
Once the crisis was underway, there may have been sharp practice by firms covering up their risks - I shall leave that to the FBI. It is also possible that "mark-to-market" requirements introduced in the wake of corporate finance scandals may have destabilised matters once the crisis was in full tilt. There are also other factors that I have not mentioned. But the above represents my current analysis.
Here are a few political implications:
- Some occasional financial instability is the inevitable flip side of the enormous positive benefits we gain from increasing efficiency in financial markets. Regulators must never imagine that they can eliminate all volatility without killing the goose that lays the golden egg.
- Markets learn from their own errors without the need for additional regulation, and would do so on this occasion also.
- Excessive and the wrong sort of regulation and government intervention have been as much to blame as any lack of such regulation. Introducing inappropriate additional regulation may do little more than cause the episode to be repeated in a future generation but just in a slightly different form.
- Macroeconomic management needs to change. The inflation targeting cum fiscal policy rules orthodoxy of the 1990s has failed, either contributing to the problems itself or shown to be impotent in addressing them. We will need to do something different in future.
