Sunday, 6 September 2009

The woes of macro-economics- would a linear audit trail help?

Paul Krugman's essay in the NY Times, already referenced in my earlier posting today, has attracted quite a lot of commentary including an interesting piece by John Lounsbury which is available here .

John Lounsbury cites research work done by
Dirk J. Bezemer at Groningen University which, it is claimed, may contain the seeds of a new paradigm for macro-economists. Bezemer's flow of cash model which is sketched out in Lounsbury's piece just might, it is alleged, enable macro-economists to be less inept at understanding and anticipating large moves in markets, asset valuations and business confidence. Indeed it is the highlighting of their current incompetence which is the essence of Krugman's piece.

Lounsbury summarizes the possible contribution that Bezemer's flow of cash approach could make as follows:

Bezemer pointed out that the critical elements of human behavior and confidence are not reflected in macro economic equilibrium models. In other words, there were no factors for herd mentality, or booms, busts and panics. He did find that they (the behavioral factors) are amenable to modeling in a flow-of-cash model. He did not propose that macro models be discarded; he felt they should be supplemented and expanded to include flow-of-cash factors. Bezemer proposed that principles of accounting, the tool of finance, be folded into the static macro economic models.
It is hard to disagree with the first two sentences but my suspicion is that it is not the equilibrium assumptions of neo-classical economic economic theory per se that are the root cause of that theory's failures to account for the modern economy's proneness to accidents. It is not even the fact that we do not have a complete audit trail to follow as seems to be implied in the flow of cash modeling assumptions and which Lounsbury alludes to with the next observation:

"If one can see all the states of motion from one position to the next, the true function of the subject can be understood. The quality of motion is fundamentally related to effectiveness of function."
The real problem is that even an all encompassing audit trail would suffer from the same intellectual problems which currently haunt the overly mathematical approach that most macro-economists have taken. In my estimation there is no underlying linear logic to the motion or trajectory of asset prices. The same gestalt switches which can allow us to see a glass as half full at one moment and half empty an instant later are at the basis of the indeterminacy of any valuation exercises and that is why, to put it rather simply, we get bubbles and recessions. There are no continuous functions for mathematicians or accountants to understand and/or track. Price gaps on charts and liquidity meltdowns and crashes only highlight and underline the fundamental "discontinuities" in the way that investors perceive the financial world especially during critical episodes.

I would suggest that there are essentially non-linear dynamics at work where tipping points and critical contagion episodes are inherently part of the fabric of economic life and this is what constitutes the "irrationality" of markets. More precisely it constitutes the non-quantifiabilty of some of the key assumptions of macro-economists such as overall market liquidity and the magnitude of the "animal spirits"(surely it was the lack of any measurable quality that Keynes was driving at in using such terminology).

For me Krugman's observation about the lack of liquidity in the baby sitting coop was quite illuminating and, to take as benign a view as one can about current shenanigans with High Frequency Trading, this may well be the real requirement for financial policy-makers i.e. how to promote liquidity in a market place where collective trust and unfettered collateral have been exhausted.

No comments:

Post a Comment