Date created:13/6/03 Last modified:13/6/03 Maintained by: John Quiggin John Quiggin
10 April 2003
Just as with fashions in clothes, fashions in economics tend to spread unevenly, so that terms that are desperately passé in some circles are still chic in others,
'Data mining' is one such term. It refers to the use of clever automatedsearch techniques, such as stepwise regression to discover putatively significant relationships in large data sets. Economists were early and enthusiastic users of data mining techniques, but by the early 1980s they were thoroughly disillusioned.
The problem is that, like methods for picking the winner of the Melbourne Cup, relationships discovered by data mining rarely persist beyond the data set that generated them. These days, the term 'data mining' is likely to occur in economics reading lists right next to the plaintive plea 'Let's take the con out of econometrics?'
By contrast, in other areas, data mining is still viewed with unqualifiedapproaveal. A Google search on "data mining" and "marketing" reveals hundreds of thousands of references nearly all of which seem to be approving.
Similarly, during the bubble years, the terms 'financial engineering' and 'financial innovation' were synonymous with the application of mathematical 'rocket science' techniques to the business of creating weatlh. These days, we read former US SEC Chairman Harvey Pitt, who condemning "financial engineering techniques designed solely to achieve accounting objectives rather than to achieve economic objectives."
The taste for financial innovation has also diminished notably following the downfall of Enron, correctly named by Fortune magazine as America's most innovative company. As Enron shows, most financial innovations over the centuries have proved to be unsound, unworkable or just plain fraudulent. There have been enough successes to allow a steady expansion in the range of financial instruments available to businesses and governments but for every success there have been many failures.
Rigid opposition to financial innovation makes no sense, but it is more important to adhere to sound financial practices that have developed over decades or centuries than it is to be innovative. Over the last few years, stockmarkets have come to recognise this, preferring companies with simple and comprehensible balance sheets to those that embody complex engineering and innovation.
Despite its loss of appeal in the business world, the idea of financial innovation has retained its fashionable cachet in relation to the financing of public infrastructure. A little under a year ago, for example, an editorial in the Sydney Morning Herald (19 April 2002), referring to a railway project that is in jeopardy because of a blowout in estimated costs, said'It is up to the Government to come up with innovative financing arrangements - very likely in partnership with the private sector - to make the line a reality' The idea that financial innovation represents some sort of magic pudding, allowing inconvenient problems like cost blowouts to be wished out of existence, has rarely been put more clearly.
At one level, politicians are aware that this idea is false.Their own Treasuries have repeatedly told them that public-private partnerships are not a means of expanding the overall level of resources available to spend on government-funded social infrastructure. Putting the point succinctly, NSW State Treasury Secretary John Pearce and Victorian Treasury Secretary Ian Little have observed that 'PPPs do not provide governments with an additional bucket of money for use on infrastructure projects'
Unfortunately, the political need to be seen as 'financially innovative' remains as strong as ever, and the associated belief in innovation as a source of buckets of extra money is never far behind. For example, introducing the Victorian PPP guidelines, Treasurer John Brumby said 'Our Partnerships Victoria policy actively seeks to increase the level of investment in Victoria's infrastructure by involving the private sector in the provision of public infrastructure and services.'
In view of steady refinements in standard methods of public procurement, the scope for beneficial use of PPP arrangements is limited. Construction by competitive tender is standard, and governments routinely outsource operations where they can save money by doing so. Only rarely will the bundling of construction and operation that characterises PPP schemes represent an improvement on standard procurement.
But as long as financial innovation is seen to be good in itself, politicians will continue to pursue these schemes, not as a method of optimally allocating a complex set of risk, but as a source of fairy gold, from which valuable public assets can seemingly be spun out of thin air. Of course, just like fairy gold, this illusion will disappear in the light of day, leaving a mountain of debt and poorly-structured projects.Professor John Quiggin is a Senior Research Fellow of the Australian Research Council, based at the University of Queensland and the Australian National University.
Read more articles from John Quiggin's home page
Go to John Quiggin's Weblog