:: Article

GDP

By Richard Marshall.

GDP, Diane Coyle, Princeton, 2014.

Does GDP measure anything? Coyle’s taut, explosive little book is haunted by this question. Either it does or it doesn’t, and either way its flaws become important not necessarily because they make it a less accurate measure but rather because as an invention helping us to answer certain questions it isn’t as efficient as it needs to be. Coyle’s sharp narrative links GDP with war and crisis. Coyle plunges into her ‘brief but affectionate history’ of an economic entity with Keynes howling against the treaty of Versailles after WWI and then shrieking for better data to fight the war caused by that treaty’s idiocy. If you need a way to get to grips with the North Atlantic economic crisis of 2008 and its terrible consequences then this is a great book to start working out some of the key issues.

‘GDP is the way we measure and compare how well or badly countries are doing.’ It is a fiction by which economies are steered and have been since Keynes’ shriek. It became fully formed in the 1940s; before then there were different concepts used. It’s a reified abstraction that adds together disparate things and then adjusts them and standardises in complicated ways so that countries are roughly comparable ‘ as long as they are adjusted again for some hypothetical exchange rates.’ It is complicated, abstract and important. It has been challenged – by environmentalists who think it overemphasises growth, ‘happiness’ advocates think it should be replaced by well-being factors, and those who think it disguises inequality and social disharmony. It indicates human possibility and innovation. Coyle acknowledges its flaws, knows there’s a need to find out how happy countries are but that doing that is a different job to the one assigned to GDP. Whether or not this will erode the dominance of GDP in the future is open to debate: if well being data becomes less flawed and gathering data for it more sophisticated then perhaps GDP will diminish in importance. But for Coyle this is not imminent.

The book describes its history, sets out its limits and defends it. But it also argues that it isn’t a good enough measure of economic performance – it isn’t agile enough to cope with rapid innovation and intangible digital services. Chapter one gives its historical background from the eighteenth century to the 1930s. WW2 invented the GDP. It built on a frame perhaps established back in the day when in 1665 William Petty estimates expenditure, land and assets of England and Wales. Even then, war was the catalyst. Petty wanted to know if they could afford to fight the Dutch and the French. This and a new method of double entry was significant. The rival French were left behind – not until 1791 did France have a similar statistical calculation in play. By then the French revolution was in swing.

The British innovation of statistics was developed through the eighteenth century. A key issue was what to include in any calculation. Adam Smith in 1776 made a distinction between productive and unproductive labour which proved important, long lasting and incapable of coping with our modern economic reality of super-fast innovation and variety. For Smith and those who followed him only those making physical commodities as in agriculture and industry would count in the calculation. Services were a cost in Smith’s view, and the servant to his master model exemplified this approach: ‘A man grows rich by employing a multitude of manufacturers: He grows poor, by maintaining a multitude of menial servants.’ This idea stayed fixed throughout the nineteenth century. It’s in the Marxist fix. The Soviets ignored unproductive labour, which was 2/3s of the Western GDP.

Then it was thrown aside by Alfred Marshall in the 1890s : ‘Wealth consists of material wealth and personal and non-material wealth.’ Definitions of GDP now in use date from the Great Depression of the 1930s and WW2. Colin Clark in the 20s and 30s introduced quarterly calculations for the British Government. Simon Kuznets did the same in the USA for Roosevelt, applying Clark’s methods to the US economy. He showed in January 1934 that the US economy had halved between 1929 and 1932. His report was a bestseller. Roosevelt’s new recovery programme used the stats. Kuznets aimed to measure welfare and not just outputs. But welfare was seen as a peacetime luxury. He failed to persuade people of mixing welfare into GDP, something that Coyle thinks was wise.

The definition of ‘economy’ was subsequently radically altered. Rather than being the ‘private sector’, with government being a rather small player in economic life, government’s role expanded over two centuries. Victorians saw government building roads and providing costly services. By the time of the WW2 it was a large presence. Government in this new formula was seen as adding to the National income rather than subtracting from it. It was at this point that Keyne’s shrieked. Keynes wanted stats that were not obscurantist and were scientific to use in wartime planning. As ever with economic innovation, being at war or in crisis motivated many of the developments. Calculating the GDP was no exception. When British Treasury officials saw Keyne’s rant they jumped to and started to do what he wanted them to do, although they were reluctant to acknowledge him. Marshall Aid depended on good financial tracking of economies too. The UN picked up on this and developed the System of national Accounts (SNA).

Keynes had a theory in his ‘The General Theory of Employment, Interest and Money’ that linked the tools of government with the size of the economy. As Coyle says, ‘ … the development of GDP, and specifically its inclusion of government expenditure, winning out over Kuznet’s welfare-based approach made Keynesian macro-economic theory the fundamental basis of how governments ran their economies in the postwar era.’ Hence the story of GDP is intimately entwined with macro-economics. Demand management seemed scientific. The idea of control was enhanced by the development, begun by Jan Tinbergen, of ways of using accounts stats to make models of the economy. Keynes was sceptical about models but they became a key tool after 1940 through to the 1970s and beyond.

It has taken another crisis, that of 2008, for the usefulness of models to be challenged. After all, none of the models predicted the crisis! Macro economists continue to argue wildly about how to calculate the required aggregated individual behaviours and aggregated measures – such as whether spending or cuts will boost GDP growth – for their models. A key issue is about the size of the multiplier – if this is greater than one then you need a stimulus package, and an austerity measure will hurt. Which way you go tends to rely on political alignment. In 2013 the IMF’s chief economist concluded that austerity had done more harm than good to short term GDP growth. But the GDP is merely a construct and no mechanical calculation will ensure it corresponds to its independent reality out there in the world because, as Coyle says, ‘ There is no such entity as GDP out there in the real world waiting to be measured by economists.’

Is there a definition of GDP? Coyle engaging says that it’s hard to pin down and is rather like a video game ‘… with increasing levels of difficulty.’ And it’s getting more complicated as time goes on. Statistical methods are getting more complex and economies too. Perhaps the relationship is analytical. If we note that services are more difficult to measure than the number of tractors then why this is so becomes kind of common sensical. There are three basic ways of calculating the GNP: add up all output, add up all expenditure, or add up incomes. It’s a ‘gross’ measure – it doesn’t factor in wear and tear of assets. (That would make it a net measure). But what counts as each of these ? – there’s the rub.

The mystical circular flow is about how national accounts have to balance the books. News and tv talk about balancing expenditure and the flow calculation that does this is consumer spending plus investment spending plus government spending plus exports less imports! This, says Coyle, is simple to state but tricky to unpack because of the attention to detail it requires for each element to be meaningful. There are fuzzy boundaries. For example, Coyle asks whether my ten years of using a car counts as consumption if my company’s purchase of a computer counts as investment? This illustrates the main point that the practicality of calculation is messier than it might seem when setting out what needs to be done. Double counting has to be avoided, seasonal fluctuations accounted for, factor cost adjustments made, alignments between different sources so figures match made, price indexing and inflation considered and all these to ensure that there won’t be too great a variation in the ‘reality’ being presented. Of course, because of these factors, different ‘… choice of technique can lead to strikingly different “real” conclusions’, as Coyle dryly notes.

To illustrate the issues, Coyle asks; ‘Is Africa poor?’ In particular, she asks, ‘Is Ghana poor?’ The ‘horrible technicalities’ in calculating the GDP makes it a difficult question to answer even though it looks stupidly easy. Until November 2010 Ghana was ‘low-income’ and so poor. But between the 5th and the 6th November 2010 its GDP rose by 60%. It had overnight become a ‘middle income’ country because of a change in the measurement. Nigeria, Uganda, Tanzania, Kenya, Malawi and Zambia are changing right now. Nigeria is expected to grow by 40% overnight and will be reckoned to be suddenly as well off as South Africa. So Africa may not be poor after all.

The new weights being used include globalisation and mobile phones and there are big moves afoot for measurement to keep up with the new realities. Old weights miscalculate reality. For the last twenty years sub-Saharan economies have been growing three times faster than they appeared to have been on the old weightings. The divergence from actual patterns of reality are important because policy decisions are buckled to assessments about a country’s economic situation, and GDP (and the closely linked GNP) is a dominant factor. Current calculations for GDP now use ‘chain-weighting’ where ‘… weights used to combine the separate prices into one index change steadily year by year.’ This prevents weights from diverging each year away from the actual pattern of the economy. If this method had been used earlier then the picture of economic growth patterns would look very different from the accepted figure. So, for example, Angus Maddison of the OECD says that ‘Acceptance of the new measure for this period [pre 1950] would involve a major reinterpretation of American history.’ US productivity would be shown to be lower than the UK in 1914, and growth of GDP was lower than the UK by 1929. In the 1970s Thatcher came to power on the back of a calculation of the GDP that showed the UK economy in crisis. But later recalculation showed that things weren’t as bad as had been originally thought.

Practical issues around collecting the stats also add to the complications in calculating accurately the GDP. The GDP is made of this vast patchwork of stats and processes fitting within its conceptual framework. The concern about ‘production boundaries’, which is about having to work out and agree on what counts as economic output is a real challenge. What to do with financial services is another headache. Exchange rates have also to be worked out and factored in to make comparisons work. A solution was Purchasing Power Parity (PPP) developed in the 1950s and 1960s. This uses ‘…data on all prices in the economy to adjust the actual exchange rate to one that reflects living standards more realistically.’ This is controversial. Low income countries will have cheap non-traded goods and services and this ends up overstating the income of these countries and distorts the measurement of living standards among different countries. They make things look overall more encouraging than they really are. We’re always wanting to know whether poor countries are getting wealthier nor not. China, for example, has surely a richer urban population than before 2000, say, but elsewhere across the world the situation is not so easy to assess. So with every way of calculating the GDP there are both technical and practical problems, as well as ideological biases that are sometimes unconscious and other times not so much. PPP calculations are suspect.

Before the development of the GDP theories of how economies grew were elegant and simple. The empirical evidence wasn’t available up to the 1950s and so theoretical and practical challenges of calculating the GDP were hidden. Theories by Roy Harrod, Evsey Domar, Robert Solow and Paul Rosenstein-Rodan were largely free of empirical evidence. Complaints against macro-economics and the simplicity of their models are often recognitions of this state of affairs. Economists are right to be annoyed by having their more recent efforts tarred with the same brush. Many are more responsive to empirical data and have greater theoretical agility than such critics would imagine, and as Coyle’s book nimbly shows, are well aware of the challenges of making something useful for contemporary modernity.
In the aftermath of the 1939-45 war Marshall Aid was an enlightened response compared to what had happened after WWI. GNP was devised to answer the question: was the plan working? It was. 1945 to 1975 was a golden age. As Coyle pithily explains: ‘ … GDP does not measure the nation’s assets or balance sheet, only its flow of income, expenditure, and production from year to year. Wipe out a portion of the assets, whether through natural or manmade disaster, and the activity or repairing and replacing will increase the growth of GDP.’ This is what happened.

What was the explanation for the growth? It could be that at this time it caught up with an already existing long term trend, or it could be that technology was revolutionising domestic scenes, or that we became mass consumers at this time. Whatever the cause, something happened. MacMillan was right when he said we’d never had it so good in 1957. The USA was battling with the USSR for supremacy in the Cold War. Who was ahead? GNP helped answer the question. The planned economies of the Soviets and China were unable to deliver the living standards of the US and Europe and only seemed for a long time to compete by refusing to add services to their calculations. But the golden age ended in the 1970s, round about the time of punk.

There were four factors for the failure of the golden age to continue. The first was that economies moved to disappointing growth or even recession combined with high and accelerating inflation. Secondly, the Cold War just kept going and this was demoralising. False statistics from the Soviet Union covered up the disasters of central planning for decades (as I noted above – they hid the fact that they didn’t add services to their figures). Thirdly, the environmental movement predicted that we were running out of natural resources very quickly and although this hasn’t happened the link between environmentalism and economics is a lasting legacy. Finally the de-colonialisation processes misunderstood the developmental needs of former colonies as well as focused on welfare needs rather than GDP. These four factors were entangled and caused a crisis.

After the crisis in the 1970s neo-liberal policies reacted to the perceived crisis caused by these four factors. Organised labour was destroyed, economies were deregularised , and state owned assets were privatised. Supply side economics needed to be streamlined and made more efficient. Activist fiscal policies were downplayed, central bank mechanisms were emphasised. Keynesianism was overthrown by monetarism. There was little evidence at the time on which to base these judgements. Understanding the causes of growth was therefore limited. Technology became increasingly understood as a factor of growth. ‘”Technology” takes the form of ideas in people’s minds, or education and skills, or ideas embodied in equipment and products’ says Coyle. Empirical data began to be slowly available but it seems for much of the time there was little understanding that would have justified anyone claiming to know how technology related to growth for much of the time economists put forward their macro-models in response to the 70’s crisis. There is still not enough due caution when economists use data around this. Clever guesswork and bold guesses is a mark of much economics. Computers and the internet, mobile phones , smartphones and fibre-optics are somehow connected to growth but there’s no settled view of details. And as Robert Solow wrote in 1987: ‘You can see the computer age everywhere but in the productivity statistics.’ It took a decade before the computer age was visible there. In the 1990’s people started talking about the New Economy or the New Paradigm. Alan Greenspan was an enthusiast. ‘I’ve been looking at business cycles since the late 1940s. There has been nothing like this’ he said in 1995. ‘The depth and persistence of such technological changes appear only once every fifty or one hundred years.’

Coyle thinks he was right and wrong. Today makes what he says look delusional. Growth has been pathetic for the last five years. But up to the mid-noughties it was growing pretty fast. Measuring services posed a challenge that the focus of technology doesn’t address. Services raise questions the calculation of GDP can’t answer, such as; ‘ Is the hairdresser’s productivity just the number of haircuts, ort the premium he can charge because the quality of his cuts or ambience of the salon?’ GDP finds innovation a problem as well, and gets left behind by rapid changes where over and under estimations in the wake of rapid innovation make calculations grievously wrong. Hedonism is a problem too alongside economic variety and innovation – how to calculate the hedonic index remains a problem. GDP lags and misses stuff. It wasn’t designed to measure such things. Perhaps it has reached its limit of usefulness.

Coyle likens the 2008 crash to a Greek tragedy. It involves arrogance, foolishness and destruction. She writes well on each: ‘The arrogance was the triumphalism about the prevailing model of economic growth. It was based on technological innovation, of course, but also on financial market deregularisation and the broader ideology of “free markets”, and the globalisation of finance and trade.’ Coyle isn’t anti-globalisation, saying that it was a force for good generally, opening up cross-border investment and being the main reason for declining levels of poverty in China and India, for example. But there was irrational exuberance and widespread fraud, deception and market manipulation. The financial and corporate universes were morally debased and systematically corrupt. Arrogance continues to dominate the attitudes in these sectors where rather than guilt a sense of being aggrieved is a more common response to criticisms about the way the financial industries and big businesses were being run. This is a hugely important point in the book. Financial services were corrupt and there’s no need to look further. The crisis was one of systemic immorality. Were not their mates in power many of these bad guys would have been sent down for a long time.

Follies include the creation of ‘toxic financial instruments that multiplied and focused risks,’ the self delusions of useless regulatory bodies and the general loss of perspective about the purpose of business which ought to have understood that ‘… profit and share price increases are a side-effect, not a goal.’

The tragic downfall of the world endorsed and believed by Thomas Friedman who supposed that global capitalism was sweeping the globe in bestsellers such as ‘The Lexus and the Olive Tree’ and ‘ The World is Flat’ shows Friedman to be nothing but an inept prophet. The North Atlantic crisis is reflected in the GDP being less than 1% from 2008 until now in the OECD countries, compared to China’s 9% figure over that time (though China is going to be in trouble in the next five years or so). The aftermath of the crisis has raised questions about the purpose of finance and how it counts in GDP. Having had to subsidise the financial institutions in this aftermath to the tune of billions has raised the question about just what these institutions actually contribute to GDP. ‘The estimated cost of the crisis, including economic output forgone because of the resulting recession, is between one and five times the whole world’s GDP’ says Coyle. Andrew Haldane at the Bank of England says, ‘ The scars of the current crisis seem likely to be felt for a generation.’

Coyle denies that finance is properly accounted for in economic stats. Finance seems to have grown twice as fast the whole economy since 1850. ‘Real GNP doubled between 1980 and 2008,’ notes Coyle of the UK, ‘ … but the measured real value added of the financial sector trebled.’ Similar trends are noted in the US. This is more mirage than miracle, notes Haldane and caused by the way it is measured. It’s a statistical mirage that affects all countries’ GDP. A study on the US states that , ‘making conservative assumptions, we show that the current official method overestimates the service output of the commercial banking industry by at least 21% (amounting to $116.8 billion in 2007: Q4 for example) and GDP by 0.3% ($52.9 billion in 2007: Q4 for example) between 1997 and 2007.’ Coyle concludes that ‘ … the size of the financial sector in recent years has been overstated by at least one-fifth, maybe even by as much as one-half.’ This matters because political leaders shape economic policy around key sectors. The Protection of the financial sector has been a key area of politics recently: if up to half of what it is supposed to contribute is a dummy figure, a mirage, then politics is being distorted in reacting to it. London toff mayor Johnson has made much of how we need this valuable sector. He’s feeding us a mirage. Questions are now being raised as to whether the financial sector should be included in the GDP at all.

Questions about the ‘production boundary’ are again raised by these worries about whether financial services should be included as part of the GDP. This is the ‘imaginary line dividing productive from unproductive activity.’ This joins other issues that raise the issue, such as the inclusion of the shadow or informal economy and the inclusion of negative welfare activities (such as arms production) and issues stemming from ‘no growth’ advocacy. Off the books activity is a large proportion of economic activity in poor countries – 20% in Italy, 25% in Greece, up to 44% in very poor countries. How to resolve the production boundary issue is a live issue but not one that ought to make us doubt the use of GDP.

And the crisis has also led to questions being raised as to whether GDP dominance of economic thinking should be replaced by a measure looking at well-being or social welfare. Coyle argues vigorously that there is no need for substituting an indicator of wellbeing or happiness for GDP. Growth of GDP is good enough to track welfare because GDP is closely linked with social welfare. There are no convincing reasons for thinking that happiness is not linked to economic growth and anyway there are already good indicators measuring happiness. Coyle thinks that when the measurement of welfare or happiness becomes more sophisticated then there will be a need to discuss which of the two measures should dominate economic thinking, but they are two different concepts and need to be kept apart. There should be no diluting of the collection of GDP stats mixing the two together. Coyle also thinks there is a need for a sustainability indicator. There is nothing at present to do this. This limits the ability of governments now to secure growth for future generations. She thinks there is also a need to improve the collection of the stats required for GDP. The new technologies need to be used.

But alongside these issues is a deeper question about whether the new kind of economies, where change and variety are exponentially fast, have brought GDP to its limit. It is an invention – according to the US Commerce department ‘one of the greatest inventions of the twentieth century’, – and can’t actually be an accurate measure of anything because it doesn’t refer to anything beyond itself. There is no clear definition of economy. The economy is an intangible. The crisis has flushed all this out into the open. The usefulness of macro-economics is also raised obliquely by all this.

Does GDP measure anything? Well, certainly nothing as tangible and real as it seems to. But Coyle is cautiously optimistic. She summarises the situation regarding GDP thus: ‘ At present we are in a statistical fog, without the information needed about either the negative aspects of growth when it is unsustainable and depletes the natural and other assets available for the future, or the positive aspects, when it delivers innovations and creativity. GDP, for all its flaws, is still a bright light shining through the mist.’

GDP works rather like a fairy story. It serves a very important function, as all fairy stories do. Sicilian fairy stories all end ‘happy ever after’ and expect audiences to know that it isn’t true.


ABOUT THE AUTHOR
Richard Marshall is still biding his time.

First published in 3:AM Magazine: Sunday, March 16th, 2014.