The Great Normalisation Following the

The Great Normalisation Following the collapse of the tech bubble, the past 5 years or so have been characterised by two global asset market anomalies: low long-term risk-free real interest rates and very low credit risk spreads (or default risk spreads) of all kinds. High-grade sovereigns paid only slightly lower rates than potential fiscal flunkies, both within the industrialised world (e.g. Bunds vs 10-year Italian or Greek sovereign debt) and comparing advanced industrial countries with emerging markets; corporates paid surprisingly small spreads over their sovereigns; even junk – non-rated instruments – was cheap. Many explanations were offered. Low risk-free real rates were the result of an ex-ante saving glut driven by the high saving propensities of some of the BRICS (Brazil, Russia, India, China, South Africa – the leaders of the new industrialising countries – the acronym invented by Jim O’Neill of Goldman Sachs) and of the commodity exporters that were (and are) the major beneficiaries of the terms of trade windfall that characterises this phase of the globalisation process. In addition, Caballero has argued that in many of the fastest growing emerging markets and other high-saving countries, there has been a relative scarcity of low-risk financial instruments, because of the lack of governance and weakness as regards the rule of law in areas like contract enforcement, creditor protection and minority shareholder protection. This has further depressed the yield on securities issued by those borrowers (especially the governments of the US and the Euroland member states) that were deemed capable of committing to the rule of law and of foreswearing discretionary default. While there may be something to these stories, there remains in my view a large unexplained ‘real risk-free rate gap’ – a long-term interest rate bubble, if you want – up until, say, six months ago. The explanations of the low credit risk spreads were almost all parial equilibrium at best, spurious at worst. Many had a ‘New Paradigm’ flavour. Whenever you hear the words ‘New Paradigm’ as an explanation for why things are really different this time around, it is time to take out the bullshitometer for some careful measurements. The proliferation of new contingent claims, especially CDOs, CLOs and CMOs, and of new financial institutions willing to issue and/or hold these claims, meant that previously illiquid bank loans and household mortgages could now be securitised and sold off. By assigning the receivables from the pool of underlying assets (say, mortgages) to ‘tranches’ of securities with different seniority, a pool of receivables whose average quality was ‘regular junk’ could support some tranches with higher credit ratings as well as, inevitably, an ‘unspeakable junk’ tranche. The theory was that this proliferation of instruments and institutions for repackaging, slicing up and reselling credit risk, would lower the average credit risk spread required for a given amount of fundamental default risk to be held. Nice theory. Just two things wrong with it. First, the total amount of fundamental default risk that needs to be held could well increase when another set of financial intermediaries (issuing new financial instruments) comes on the scene. Second, there is no reason to believe that, whatever the total quantum of undiversifiable risk that needs to be held, it will be held by those best able to bear it. First, the possibility that an increase in the number of institutions and instruments could increase the total amount of risk that needs to be held. When the the chain of financial institutions and instruments interposed between ultimate savers (households, corporations and governments) and ultimate investing entities (non-financial corporations and governments – sometimes households also, although our accounting conventions do not classify this as capital formation), gets longer, not only does the pool of potential risk sharers get larger, but the number of transactors and transactions subject to counterpary risk increases. The likelihood of default occurring somewhere in the chain therefore may well increase. Second, the possibility that risk ends up being held by the wrong parties. Just because today financial engineering permits contingent payment streams to be broken down into what some call ‘their most fundamental pieces’, does not mean that risk now really ends up with those most able to bear it. The same financial engineering techniques permit risk to be bunched and bundled in ever more convoluted ways. The risk still ends up with those most willing to bear it. Whether they are also most able to do so, only time will tell. New contingent claims that can be used to hedge risk can equally well be used to seek out and take on risk that without these new instruments could not have been bought and held

The past couple of decades has witnessed developments that have exacerbated two long-standing sets of problems in the trading of risk. The first is that the pace of product innovation in financial markets, both in exchange-traded instruments but especially in the over- the-counter (OTC) market where designer financial engineering is rife, has outstripped our capacity to understand, let alone price these products. Nor have regulators and supervisors kept up.

Some of these structured finance products are so complex that even their designers probably don’t know what risks are embedded in them. Financial institutions that hold these instruments cannot mark them to market because there is no liquid market for them (this holds even for many CDOs that are, in principle, exchange-traded instruments). Instead they are frequently ‘marked-to-model’, using models whose complex mathematical and computational features neither the institution’s risk managers nor the supervisor/regulator (if there is one) have mastered. Alternatively, they are valued on the basis of quotes from brokers that do not have to deal at the prices they are quoting; such cheap talk is completely meaningless, even when the parties making them have the proper arms-length relationship with the holder of the securities. At times, they are priced using some wet-finger-method based on the credit rating of the instrument, even through these ratings are provided (sold?) by agencies that are too often conflicted and/or far too close to the designers and issuers of the instruments. For illiquid AAA-rated CDOs, it was until recently common practice to value them at par. Of course, even true AAA-rated instruments can be subject to considerable market risk. This market risk can become default risk for the counterparties of the party holding the AAA-rated instrument subject to the market risk.

Not only don’t we know how to price many of these instruments, we often don’t know who ends up holding them because of inadequate reporting obligations for many systemically important institutions. Basel II, with its emphasis on ‘marking to market’ is encoutering the problems of there not being markets for a growing number of newly designed instruments (as well as for the remaining old illiquid instruments), is clearly fighting the wrong war. Marking-to-model is an invitation to deceit, fraud and corruption.

The second problem, highlighted by both is that increasing numbers of financial institutions operate across many different national jurisdictions. Supervision and regulation are organised at the national level. Cooperation and coordination attempts are forever chasing new institutions and new instruments. One can only hope that the development of EU-wide supervision and regulation of financial institutions and markets will permit a significant reduction in the number of ad-hoc arrangements that needs to be concluded.

The problem of global financial markets and global financial operators running circles around national supervision and regulation is aggravated by regulatory competition between nations, which often makes for a race to the bottom, with minimal regulatory and supervisory demands being made on systemically important financial institutions. Many, including most hedgefunds and private equity funds, have at most rudimentary reporting obligations.

When even self-regulation (which really means no regulation at best, and at worst no regulation plus cartelisation of the industry) is considered too onerous, and when voluntary codes of conduct are sniffily rejected, it is clear that we have created and informational black hole and an increasingly under-regulated financial system where no-one knows who owes what and to whom.

All this is a long if not long-winded way of saying that global default risk premia went down, across the board, to ridiculously low levels despite there being no obvious change in the determinants of default risk and in the fundamental price of default risk. We had a default risk premium bubble – an explosion of default risk euphoria. Both global asset market anomalies are being reversed. Long real risk-free rates are almost back to normal. Temporary episodes, like the ‘flight to quality’ we saw last week, may depress real risk-free yields for a bit, but there is no doubt that the UK will not see a real yield of 0.38 percent on 50-year index-linked gilts again any time soon. There is perhaps another 50 bps to go still, at the ten-year maturity, but normalisation has taken place. And this appears to have happened without damaging the remarkable expansion of the global economy. What triggered the real rate normalisation? Some observers point to the reduction in the saving rates of the oil exporting countries. My view is that since the extraordinarily low real rates were not a fundamental but a bubble-driven phenomenon, the bursting of that bubble does not require a fundamental explanation. Credit risk spreads are also being normalised. Starting from the US sub-prime mortgage derivatives markets, corporate spreads have increased throughout the industrialised world, especially in the financial sector where most of the suspect financial claims are likely to be held. Covenant-lite bonds and loans, another manifestation of low spreads in a qualitative dimension, are no longer finding takers at any yield borrowers are willing to offer. There is quite a bit more to go as regards credit risk spread normalisation, however. Emerging market spreads, for instance, still look unrealistically low. It’s true that most emerging markets are far removed from the US sub-prime housing loan markets and the CDOs and CMOs issued against them, but it is an error to believe that a repricing of credit risk was required only in the CDO and CMO markets and for the institutions exposed to them. The blow-up of the US sub-prime housing market, of the financial instruments based on them and of the hedge funds invested in them, was indeed the event that triggered the renormalisation of global credit spreads. But even without the sub-prime housing loan scandal, global credit spreads seriously underpriced global default risk. The US sub-prime debacle is a classic example of a ‘fin-de-credit-boom’ loss of discipline, judgement and control. On the demand side for mortgages issued to customers with impaired credit histories, there were growing populations of crooks and ignoramuses. On the supply side, greed, lack of memory and lack of internal control by banks and other mortgage lenders permitted the proliferation of utterly inappropriate instruments such as interest-only mortgages and negative amortisation mortgages. Asking for income verification was considered superfluous. Finally, the regulators dropped the ball completely, and should pay the price for this failure of regulatory and supervisory oversight. To witness the Fed, having failed to prevent the issuance of ludicrously inappropriate mortgage instruments to known bad credit risks, now engaged in populist ‘leaning on the lenders to go easy on the non-performing borrowers’, is shameful indeed. The normalisation of credit spreads, especially in emerging markets, still has some way to go. However, its impact on the real economy, admittedly an exercise blighted by the need to come up with a convincing counterfactual, appears thus far to be minor at most. This would be what one would expect, if the increase in spreads reflects a repricing, due a bursting bubble, of given fundamental determinants of credit risk, rather than a worsening of the determinants of credit risk. One area of vulnerability remains the US. A history of financial excess, encouraged by regulatory failure, has resulted in greater vulnerability to a further repricing of credit risk than elsewhere in the industrialised world. The housing market is tanking. Growth has slowed to a rate well below potential, and there is considerable downside risk (as well as significant upside risk) to the 2.5 -3.0 percent growth forecasts that represent the Fed’s central scenario for the next year and a half. There is a material risk that the combination of the normalisation of the two global asset market anomalies, together with a significant further slowdown in the US, would trigger a global slowdown, bringing to an end the halcyon semi-decade we have just experienced. The other area of vulnerability is China. Runaway growth (now 11.9 percent at an annual rate) and rising inflation (CPI inflation at 4.4% at an annual rate) point to an unsustainable credit boom. It is also becoming increasingly obvious that China’s growth rate and growth pattern is utterly environmentally destructive and unsustainable. The risk of a combined credit crunch and environmental emergency is real and rising. No doubt the monetary authorities in the US and elsewhere would respond of the global business cycle were to turn down. The Fed, with its triple mandate, would cut rates more aggressively than the ECB and the Bank of England, with their lexicographic or hierarchical inflation targeting mandates. The Bank of Japan, still with one leg in the liquidity trap quagmire, could not do much. China’s monetary authority is not yet a full domestic player, let alone a global player, and would not be able to make a material contribution to the management of the global business cycle. It need not happen. It is not yet my central forecast. But the likelihood is growing of a slowdown in global economic activity, prompted by the normalisation of the two global asset anomalies, a serious slowdown in the US and a growth implosion in China. For the first time in five years, the global growth light is turning from green to amber.

Labels: ,

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website

Maverecon: a guide

Comment: To comment, please register with FT.com, which you can do for free here. Please also read our comments policy here.
Contact: You can write to Willem by using the email addresses shown on his website.
Time: UK time is shown on posts.
Follow: Links to the blog's Twitter and RSS feeds are at the top of the page. You can also read Maverecon on your mobile device, by going to www.ft.com/maverecon