OH GROSS! The real problem with GDP

Mmmm, what's a house-bound spouse worth?

The first of which is that none of the variations say anything about distribution: of incomes, or work, of consumption. I tend not to think those are important things (or, perhaps, not as important as some others do) and many, many, people disagree with me. For example, there's definitely a movement insisting that we must include inequality in our measure of how well the economy is doing. And I've no problem with that: the problem to me comes when we try to include it into just the one number like GDP that we observe. So, given that we've got Gini's and Lorenz Curves and all sorts of other ways to measure the inequality of incomes etc it doesn't bother me that GDP doesn't directly include this.

The second comes in two parts: it measures things at market prices. But what's the market price of something that doesn't have a market price? Obviously the world is a better place when your wife is smiling at you but that's not exactly an economic transaction (well, not always). More importantly, there's vast amounts of household production that are done which just aren't recorded in our economic figures. Thus those reports that come around every so often with an angry faction insisting that the economy would be 20 per cent, 50 per cent larger if only housewives were paid for housework.

And to be fair to said housewives/househusbands they do have a point. Washing, childcare, cooking dinner, these are things which do have a value. Doing them is also obviously work. And we can make each and every one of them (including more intimate spousely duties) into economic transactions if we want: dry cleaners, day care centres, restaurants and brothels just about covers the four. And the problem with ignoring the value of these is that if people are partaking of professional ladies (for, yes, that is now included in GDP) rather than the wife on a Saturday night then we are recording the nation as being richer. Which, really, given that the same amount of tupping is going on, isn't obviously true. It's just that one version of the transaction is monetised and the other is not.

This was addressed by the Sarkozy Commission, upon which Laureates Joseph Stiglitz and Amartya Sen worked, and their answer was that such household production should be valued at the “undifferentiated labour rate” which really translates into minimum wage.

Yes, I know, that the snotdribblers are looked after by someone who loves them rather than some nursery worker who couldn't get any other job with his or her two GCSEs would appear to have a greater value than that. But this is really Adam Smith on the division and specialisation of labour. Yes, that damn pin factory again. Out in a market economy you're dividing and specialising with, well, really, with the some 3.5 billion other humans who are part of the global economy.

Unless your domestic arrangements are very much more interesting than mine you're going to be dividing and specialising the household labour with no more than one other adult. And if the average wage is around £13 an hour in the UK economy, which it is, then that difference in the efficiency of division and specialisation might well justify around half the pay rate, say £6.50 an hour. This is the solution to the point Keynes made (as a criticism disguised as a joke) in the 1930s as Simon Kuznets was putting together the basic concept of national production accounting. That when a man marries his housekeeper the economy shrinks: because formerly paid for activity is now unpaid for.

So, household production, non-market work, perhaps we should add it to GDP at that minimum wage rate. And I tell you, I would pay very good money indeed to see the members of the Sarkozy Commission telling their spouses who work in the home that the Saturday night legover should be valued at no more than £6.50 an hour.

The second part of this is that of course most things produced by governments don't have market prices. Back in those 1930s this was less of a problem than it is now: governments were a very much smaller portion of the society then than they are now. The way we get around this problem is that we simply declare the value of government to be the amount that we spend upon government.

Which has a very strange implication. If we raise the salaries of bureaucrats then we appear to be richer. And, bizarrely, productivity also rises. Because productivity is the value of production per labour hours, and if we define output value as being wages paid then higher wages raise productivity per hour. Which isn't, I think we all know, how the output of bureaucracies actually works.

All of which brings us back to the original point. Yes, GDP isn't very good but its ease of calculation makes up for most of the not very good bits. Even if Ms Bruni might not quite agree with Nicolas about some of the suggested corrective valuations. ®