On bubbles

Alan Wood writing in the Oz last week made the point that "Regulators should be neither bubble poppers nor blowers" (at least that was the headline).

In so far as the comment is actually about central bankers rather than all regulators the comment is probably valid. Wood quotes from a monograph by Stephen Kirchner of the Centre for Independent Studies. Kirchner makes many valid arguments about the limitations of monetary policy as a means to control bubbles. He cites Bernanke's claim that the Great Depression was caused by a move to high interest rates, but Galbraith in his analysis I think also agrees that monetary tightening was the final trigger the problem however was that an asset bubble had run for so long before the tightening.

Kirchner does less well in arguing that bubbles don't exist, or if they do are a benign part of the system. He attacks the usual claim (Plagrave no less) that a bubble can be defined as "asset prices that exceed an asset's fundamental value because current owners believe they can resell at an even higher price". His basis is the simple statement that "few people buy or hold an asset in the expectation of making a loss" hence invalidating the degfinition. But in this he ignores the real benefit of an asset which is either meant to be the utility of its use (it is a house and you live in it), or it is the cashflow you expect to earn from it (it is a house and you earn rent from it). In this he has probably been easily misled by the distortion of the modern equity market that measures "shareholder value" in terms of asset price increase rather than divident flow.

Not only does he fail to accurately identify speculation, he tries to defend speculation by attacking its attackers, saying "much of the popular hostility to financial speculation reflects this concern that it results in undeserved gain." Well that might be the poluar hostility, my hostility is entirely based on the fact that it all ends in tears.

When Kirchner allows himself to admit that bubbles may exist he claims that of fifteen stock market crashes over the course of a century "only" eight resulted in financial system distress. He then seeks solace in a neo-Darwinian (with overtones of Schumpeter) defence that these moments of stress result in innovation. The argument is that "bubbles breed variety, and variety feeds economic evolution". This is a particularly callous view and ignores the real human pain and suffering that can be experienced. And, as Hayek stated in The Road to Surfdom there is also a randomness about who this affects. Hayek warned about the arrogant presumpton that those who are doing well are doing so because of their talents rather than just luck. ("And this is none the less true because in competition chance and good luck are often as important as skill and foresight in determining the fate of different people." Chapter 8 PP 105-106)

Wood in his piece referred to a speech by RBA Assistant Governor Guy Debelle, the concept of "macro-prudential stabilisers". The concept can be understood better in prudential regulation based on Basel I, with a simple capital adequacy ratio based on risk-weighting of certain asset classes and amount of capital required to fund the portfolio. In such a system the risk assessment of an asset class could be increased automatically if the prices of the assets are increasing "too quickly" (as assessed against the overall price level).

This becomes more problematic in a Basel II type world, where ultimately the "risk" of assets was determined by ratings agencies. One of the many problems we've encountered was the way ratings agencies dealt with increasing house prices - they assumed this made lending more secure as a potential defaulter woul have an asset worth more than the original purchase. This, of course, is a great theory so long as the price path continues.

However, it works disasterously when, as it did with sub-prime mortgages, it feeds the machine that offers increasingly risky loans.

Apart from building rules into risk assessment that should discount the value of assets undergoing excessive aset-specific inflation, another very simple policy instrument is consumer credit laws as apply in Australia. In Australia a lender cannot make a loan merely on the basis that the loan is "secure", the lender also needs to be satisfied that the borrower can adequately service the loan (and not by refinancing, which was the other under-pinning assumption with some sub-prime loans).

Monetary policy is definitely not a good mechanism for popping bubbles - as bubbles are by definition increases in the price of one asset class out of sink with the price level in general. Monetary policy can only work on the overall price level.

That does not mean that we don't need to take regulatory action globally to ensure improvement in risk assessment of assets undergoing excessive price growth, and some "duty of care" legislation on lenders.

Finally, the other lesson we learnt in the process is that the assumption that corporate finance markets need less direct supervision than consumer markets on the assumption that the lenders and borrowers are better able to "protect and inform" themselves is complete and utter nonsense. For some insights into the ways of the finance markets one can read this item in the New Yorker.

There are other great reads in two books House of Cards and Fool's Gold. The latter makes particularly interesting reading in the sections dealing with the processes in the large investment banks who basically got remunerated for increasing profit but not managing risk. What gets missed is that this only could occur because these banks had moved to be listed rather than partnerships. They represent another failure of what I refer to as managerial capitalism.

In this context it is interesting to note the moves by APRA on remuneration of executives of DTIs, and the simultaneous inquiry by the Productivity Commission. The PC specifically trucks out the following;

The alignment of board and executive interests with those of shareholders has been a fundamental issue of corporate governance dating back to the first joint stock companies in the 17th Century. What is known as a ‘principal–agent problem’ arises when the interests of boards and executives do not accord with those of shareholders and the wider community. Shareholders (the ‘principal’) have an interest in their company maximising returns over time, and the board and executives (the ‘agent’) are tasked with running the company to achieve this goal. However, executives — like any worker — have their own interests to consider, which can potentially conflict with those of the company’s shareholders.

I'd be prepared to dispute this and actually argue that it became an identified issue only with the publication of Berle and Means study in the 30s. I'm also prepared to have a fight over the underlying premise that the shareholders interest is in "maximising returns over time". If what I sought was a "utility" stock promising a good divident stream that's what I want management focussed on. More generally, investors are interested in both the level of the return and the volitility of that return.

Either way all attempts to tie executive remuneration to short term performance targets, or long term outcomes via "options" are doomed to fail. I've written previously on this topic. The disaster that befell Wall Street is just another example.

Note: Next instalment sometime will be on risk management and risk amplification.