The simplistic, aggregative economics of our intellectuals

A couple of months ago, I sat next to a leading economist, reputedly of the free-market variety (though our conversation led me to doubt it). I suggested to him that GDP was not a good indicator of the health of the economy. He said he thought it was, and (immediately betraying the weakness of his position) proferred the straw-man challenge that perhaps I preferred Gross Wellbeing.

Though the smarmy economist would not like to acknowledge it, it is possible to be critical of GDP without endorsing the hippy, Cameroony, pseudo-statistic of Gross Wellbeing. Here is one of many reasons why GDP hides a litany of economic and (particularly) political sins.

Consider two economies identical in every way - same size of population, same output of products, same prices - except for one feature. In one economy, the work is done by half the population available for work, working on average 70 hours per week whilst the other half of the working population do not work. In the other economy, the whole population available for work is in employment, working on average 35 hours a week. In aggregate, the same amount of money is paid in wages for the same amount of work to produce the same amount of goods, and the same amount of wages is spent on the same value (though probably not the same types) of goods. The GDP of the two economies would be identical. But are these economies equally healthy? To my mind, clearly not, and for fairly simple economic reasons.

Value is established at the margins. Our marginal valuation of any good will normally vary according to how much we have of that good. To take the Rothbard egg scale, two eggs is better than one egg, but not necessarily twice as good, and the value to us of an extra egg when we already have a dozen is probably less than when we have none (though not necessarily - this is always subjective and dependent on circumstances).

This logic applies as much to leisure and work as it does to any other good. For a given job paying a given wage (and ignoring counter-incentives erected by the state), the value to us of an hour's work is probably (though not necessarily) greater when we have no work than when we have already worked 69 hours that week. Conversely, the value of an hour's leisure time is greater if we have worked so many hours that we have little time for leisure, than if we have no work and every waking hour is leisure time.

In the first economy above, it does not seem enormously contentious to suggest that half the population available for work is getting less leisure than would be ideal, and the other half is getting less work than ideal.

It could be objected that the (labour) market reveals people's preference scales and valuations and is the only valid way to make interpersonal comparisons of these subjective quantities. If a person already working 69 hours is more suited or more willing to do an hour's work than an unemployed person, then in reality the subjective value of that hour's work to the former is higher than to the latter, regardless of what one might expect in the abstract, depersonalised, hypothetical scenario.

And it should be cautioned that, even if one accepts the premise that the first economy is suboptimal and inefficient, this does not justify (as it might seem to) government intervention to "spread the work". In the absence of artificial incentives that inhibit more even distribution of work, if the first economy's labour market chooses that distribution of work, that is the balance that most closely reflects the individual, subjective valuations of the population, and is therefore the optimal balance for that group of people, regardless of what we might think ought to be best for them.

I list the objection and the temptation to intervene in short succession because they, and the value of GDP as a measure of economic health, rely on an assumption of efficiency that is so far from the reality of our modern economies and state-created incentives that it is laughable to base any premise or action on that assumption. The efficiency that is assumed is that people in different circumstances face similar incentives and disincentives to work. In practice, our tax and benefits system has created different disincentives to work depending on one's existing state of employment. For those at the bottom end of the earnings scale, we reward leisure and penalize work (through the withdrawal of means-tested benefits combined with the basic rate of taxation above the personal allowance). For those in the middle of the earnings scale, we are particularly keen to encourage work (for that short gap between the level of earnings where one no longer receives means-tested benefits and the level of earnings where one starts to pay upper-rate tax). And for those at the upper end of the earnings scale, we are increasingly disincentivising work and rewarding leisure, though not (yet) as aggressively as we do to those at the bottom end of the scale. Ironically, it is precisely those egalitarian measures introduced by governments to try to redistribute income to reduce inequalities that are responsible for the disincentives that widen the earnings gap by encouraging those who are out of work not to work, those who are in work to work more, and those who are paid best (and most able to influence their levels of pay) to pay themselves more to compensate for the aggressive taxation. Yet people are surprised that inequality in the UK has widened under the Labour government! It is the direct effect of Brown's interventions, but politicians of all colours are too ignorant (and addicted to appearing to do good with other people's money) to notice.

It is not practical to try to design a different measure of the health of the economy that took this into account. Such a measure could reward economically-inefficient "make-work" schemes, and not adequately reflect the benefit of growth achieved through efficiencies that increased productivity without increasing employment. It could penalize changes in preferences and norms that might be socially-beneficial, such as increased stay-at-home parenting. But no one should be fooled by the difficulty of designing a superior measure of economic health into thinking that our preferred measures (whether GDP or GDP per capita) are adequate, nor that increasing GDP necessarily means a healthier economy. There is a tendency amongst politicians, their advisers (bureaucratic and academic) and the media to look for their keys not where they dropped them, but under the street-light where it is easier to search. We should not kid ourselves that GDP is the key measure of economic health simply because it is relatively easy to compile. Policy-making requires great judgment, awareness of many factors, and consideration of the interactions of the effects of all interventions. It cannot be reduced to maintenance of a stable upward trend in a few basic indices (such as GDP or inflation), as is implied by the decision (assumed by most commentators to be right) to hive-off to the Bank of England responsibility for maintaining a specified level of consumer-price-inflation, or the Government's boasting before the recent downturn about the number of quarters of continuous GDP growth that had been maintained. Most of the damage that resulted in the trouble we are now in was done in that time, partly through the fragmentation of responsibility and poor judgment of those who thought our economic health could be measured by such inadequate and misleading statistics.

Our intellectuals have it exactly back-to-front. They think our economy can be reduced to a few simplistic statistics, and yet immensely complicated measures are needed to "manage" it. In reality, the economy is immensely complicated, irreducable and unpredictable. It is futile and counter-productive to try to micro-manage it.

Those who saw Jim al-Khalili's excellent "The Secret Life of Chaos" on BBC4 last Sunday should have gained an insight into the relationship between chaos, complexity and simplicity (and those familiar with Hayek may have noticed strong similarities between his description of the self-organising nature of chaotic systems and Hayek's concept of spontaneous order). Our micro-managers fail to understand that the economy no more behaves Newtonian, mechanistic rules whose behaviour can be predicted by high-power computers and managed by the pulling of ever-more-intricate levers than does the climate or the universe.