Andy Haldane, the BoE’s chief economist, always produces speeches which are worth a read and his latest, entitled Whose Recovery? is no exception. Although the UK has been one of the fastest growing G7 economies over the past couple of years, Haldane makes the point that not everyone has felt the benefit of these apparent gains and that “aggregate activity measures are sometimes a poor proxy for the average person’s income.“ Indeed, since GDP measures the sum of all activity in the economy, its performance is boosted by a rise in labour input. Depending how we measure it, GDP per head in 2015 was either 0.1% higher than its 2007 peak (based on total population) or 3.5% higher (based on economically active people aged 16-64). Either way, both illustrate a sharp slowdown in income (output) per head relative to pre-crisis growth rates. The simple GDP per head of population measure posted average annual growth of 2.4% over the period 1980-2007 whilst the age-adjusted activity measure grew at a 2.0% annual rate.
This clearly illustrates that productivity growth must have slowed, and we will leave a discussion of this for another time, but it demonstrates why productivity is so important in generating a rise in living standards and why GDP figures can be misleading. Moreover, the aggregate picture glosses over significant regional differences in output per head, and as Haldane notes “only in London and the South-East is GDP per head in 2015 estimated to be above its pre-crisis peak.” You will need no reminding that London voted overwhelmingly in favour of remaining in the EU on 23 June (although the rest of the south east did not). Perhaps this is one reason why the narrative of a strong UK recovery which would be jeopardised by Brexit did not wash in large parts of the country, as many people asked themselves the same question posed by Andy Haldane: Whose recovery?
It is thus evident that another in the long list of problems with GDP as a measure of wellbeing is that it is an aggregative measure which takes no account of distributive factors. Ironically, much of the evidence on income distribution suggests that the greatest widening of inequality took place in the second half of the 1980s, with a further leg up over 1997-2002, but it has since remained stable and actually come down a little in the past five years. But obviously not enough for a large section of the population to feel that they are getting a big enough slice of the pie.
As I noted in my previous post, GDP is a much used and much abused statistic. There is definitely something wrong with GDP as a measure of welfare if it assigns a positive value to pollution-emitting activities today whilst efforts to clear it up tomorrow will similarly be assigned a positive value. Surely a better measure of net social benefit would be to offset the two. It is for reasons such as this that statistical authorities are placing more emphasis on measuring things such as national wellbeing, which is an area where the ONS in the UK has done a lot of work lately. There is clearly a strong argument for economists to raise their eyes to a wider horizon. But the fact remains that it is still so much easier to measure the physical output of things, and to pretend that we have a handle on the output of services, which is why we continue to focus our attention on GDP statistics.
When discussing the use and abuse of economic statistics I am reminded of the old story of the drunken man who is found by a policeman scrabbling under a streetlight. “What are you doing down there?” asks the policeman. “I am looking for my keys,” says the drunk. After five minutes fruitless searching the puzzled policeman asks, “Are you sure you lost your keys here?” “No,” replies the drunk pointing to the darkness on the other side of the street, “I lost them over there, but the light is so much better here.” I suppose the moral of the story for economists is always carry a decent torch capable of shining a light on the statistics.
No comments:
Post a Comment