For most of human history the workings of what we casually refer to as ‘the economy’ were pretty much a black box. Indeed, for millennia the concept of an economy hardly existed at all. There were at least two reasons for this. First, before the eighteenth-century Industrial Revolution, there was really no such thing as economic growth. That made the economy an awful lot duller. The output of agricultural societies was pretty much a function of the weather. If rains were good, the harvest was good. If not, it wasn’t. Nor, in this pre-industrial world, were there huge productivity gaps between one region and another. Most people were just scraping by. Thus the size of a region’s economy was largely determined by the size of its population. In AD 1000 China and India accounted for just over half of global economic output, a proportion that remained unchanged for 600 years (and may be heading that way again).1
Second, in an era of monarchs – especially those lucky enough to have been ordained by God – what was going on in the broader economy was of no great concern. For an absolute monarch there was no distinction between his own wealth and that of his realm.2 Given the lack of distinction between the wealth of the monarch and the wealth of the nation, there was little room for anything we might call an economy. Apart from keeping the court in its accustomed luxury, the only thing required of a national economy was to finance war. A nation grew only if it conquered new dominions. If the king could muster armies to grab new territory, the national weal would be increased. But how could you tell whether your nation could bear the cost? Most early attempts to catalogue the size of an economy were driven by the need to work out the monarch’s capacity to wage war.3
So it was in France. In 1781 Jacques Necker, the Swiss finance minister of Louis XVI, presented his famous compte rendu au roi, his ‘report to the king’, the first attempt to take serious stock of France’s finances. Necker, formerly a wildly successful banker – are the alarm bells going off yet? – showed that France’s finances were in rude health. Revenues were said to exceed expenditure by the enormous sum of 10 million livres. The main purpose of the report was to demonstrate that France could easily afford its involvement in the American Revolutionary War, in which, as was customary, it found itself on the opposite side to Britain. Necker, who had made his own fortune through speculation, wanted to prove that France’s finances were so solid it could easily borrow money to finance its war effort. What the compte rendu cleverly omitted, however, was that France had already borrowed heavily under Necker’s own direction. One of the earliest attempts to present a set of national accounts was also a piece of fiction.
Necker’s stab at national accounting was not the first. That distinction is usually given to William Petty, whose publication of the Down Survey in 1652 is considered by many to be the first systematic effort to survey a country’s economy – in this case, that of Ireland.4 With the help of simple instruments and a thousand unemployed soldiers, Petty undertook the comprehensive mapping of land in thirty counties covering 5 million acres. The principal motivation was to carve up Catholic land conquered by Oliver Cromwell and to use it to pay back those who had financed the war as well as the arrears of soldiers’ wages. In addition to mapping the land, Petty conducted a fairly rigorous survey of assets, including ships, houses and personal estates. From this he worked out flows of income that would be generated, a crucial distinction from earlier efforts to catalogue stocks of wealth such as the Domesday Book of 1086.
Later, after the restoration of King Charles II, Petty did the same in England and Wales. This time the objective was to improve the monarch’s capacity to tax his subjects. Petty recommended keeping records on domestic consumption, production, trade and population growth and started to develop methods for assessing the value of labour as well as land.
If early attempts to survey the economy had common themes of war, taxation and subservience to the monarch’s needs, there were other schools of thought pulling in a different direction. In France in the eighteenth century the so-called physiocrats emphasised that the wealth of a nation was rooted in farm production and productive work. Subtly different from Petty, in the physiocrats’ interpretation the ‘productive class’ consisted of mainly agricultural labourers, while the so-called ‘sterile’ class included ‘artisans, professionals, merchants and, lo and behold, the King himself’.5 Viewed from this perspective, the invention of the economy – as something distinct from the monarch – was a profoundly democratic act.
Adam Smith, in his An Inquiry into the Nature and Causes of the Wealth of Nations, first published in 1776, also divided labour into productive and unproductive categories. A man, he wrote, ‘grows rich by employing a multitude of manufacturers: He grows poor by maintaining a multitude of menial servants.’ It wasn’t a very flattering view of the leisured classes. Along with hosts of servants performing useless tasks for do-nothing aristocrats, Smith put the monarch, as well as the army and the navy, into the category of unproductive labour.
What unites these early attempts to catalogue national wealth is an effort to draw what economists today call the production boundary – between activities that should be counted and those that should not. In short, they were trying to answer a question that is still relevant today: precisely what is an economy? In the great economic ledger should the king appear on the plus side, the embodiment in flesh and blood of the national patrimony? Or, as the physiocrats and Adam Smith implied, should he be on the negative side of the ledger, an unproductive spender of the nation’s resources?
The same question of what should be included and what should be excluded has rumbled on ever since. Should we include government spending? How about providers of services, whose contributions to society – healthy minds (psychoanalysts), humour (clowns), education (teachers) – may be harder to count than horseshoes or bushels of wheat? In the twentieth century communist countries largely ignored services altogether. Even today we struggle to measure their economic contribution.
Modern national accounts of the type used by virtually every country in the world today only really began to take shape in the 1930s. Simon Kuznets is usually credited with the invention of GDP, the quintessence of the national accounting system. But Kuznets, rather like Victor Frankenstein, soon saw his creation take on a life – and a direction – of its own.
The man who is said to have invented our way of measuring growth was born in 1901 into a merchant family in the town of Pinsk in what was then part of the Russian empire. Pinsk had a large Jewish population and Kuznets’ parents were Belarusian Jews. As a child he lived under the rule of the tsar, and as an adolescent sympathised with the Mensheviks, whose hopes of reforming tsarist Russia were swept aside by the Bolshevik revolution of October 1917.6 Kuznets then studied at Kharkiv University in Ukraine, where he attended the Institute of Commerce and studied economics, history, statistics and mathematics. He was a young man of great social conscience and ideals.
His tutors at Kharkiv stressed the importance of basing opinions on empirical data, a lesson that stayed with him for life. There was also an emphasis on placing economic theory in a wider historical and social context. Kuznets was a brilliant student and by his early twenties had published his first paper on the wages of factory workers in Kharkiv. His studies at the university were interrupted by the Russian civil war, and in 1922 the family fled, via Turkey, to the US. It was here that the Belarusian émigré was to make a profound and lasting impact on global economics.
Kuznets continued his education at Columbia University, graduating in 1923 and receiving his PhD in 1926. The following year he joined the National Bureau of Economic Research, a think tank founded in 1920. Kuznets would become a distinguished academic economist with something any self-respecting economist aspires to – a curve named after him.7 (Oh, and he also won a Nobel Prize in economics in 1971.) His most lasting achievement, however, came in the intersection between economics and the real world.
Kuznets loved data. He worked closely with the first director of research at the National Bureau of Economic Research, Wesley Mitchell, who was also chairman of President Herbert Hoover’s Committee on Social Trends. That work took Kuznets into the heart of government policy. Hoover’s election campaign had promised Americans ‘a chicken in every pot and a car in every garage’. What they got instead was the Wall Street Crash and the Great Depression. Hoover’s response to the terrible depression that followed, which at its trough saw at least one in every four Americans without work, was slow and inadequate. Essentially, he thought the economy would heal itself. Prosperity, he assured Americans, was just around the corner.
Hoover may not have been entirely to blame. There was no systematic methodology for drawing up an accurate picture of a national economy. A publication in 2000 by the US Department of Commerce, which praised GDP as ‘one of the great inventions of the 20th century’, quotes an economist as saying, ‘One reads with dismay of Presidents Hoover and then Roosevelt designing policies to combat the Great Depression of the 1930s on the basis of such sketchy data as stock price indices, freight-car loadings and incomplete indices of industrial production.’ As hard as it is to believe now in this age of obsession with economic statistics, Hoover had only the crudest notion of what was actually going on.
That was about to change. When Franklin D. Roosevelt became president in 1933, Kuznets was entrusted with the task of creating national accounts. Kuznets outlined his ideas in an article for the Encyclopedia of Social Sciences. His notion was disarmingly simple: to squeeze all human activity into a single number.
Kuznets was the ideal man for the job. He had a near-obsession with measuring things. One writer compares his way of analysing an economy to a doctor on his patient rounds. He based his assessment on observable data and symptoms. But understanding the patient’s underlying condition also required judgement, knowledge and a rigorous inquisition of the facts. For Kuznets, being thorough was more important than being brilliant.8
Kuznets began by categorising American industry into different sectors, such as energy, manufacturing, mining and agriculture. He was given a staff of three assistants and five statistical clerks. ‘Together they hit the road, visiting factories, mines and farms, interviewing owners and managers and writing down figures in notebooks.’9 Although the scale of data collection is vastly bigger these days, survey-based methodology hasn’t changed that much even in the era of big data. To this day sizing up an economy remains primarily an extrapolation of survey data, not a summation of gathered facts.
Kuznets’ team travelled the length and breadth of the USA asking farmers and factory managers what and how much they had produced and what they had purchased in order to make their final product. The team shared data so they could compare results and iron out anomalies. Kuznets knew the data were more or less meaningless in isolation. They had to be interpreted. Though it would take many more years before the first publication, in 1942, of a full set of gross national product statistics, Kuznets’ work bore much earlier fruit.10 In January 1934 he presented his first report to Congress. It ran to 261 pages and, for such a historic document, bore a name that only an economist could dream up: National Income, 1929–32.
The report began with much throat clearing about what the numbers could and could not reveal. His effort was, Kuznets said, ‘an amalgam of . . . estimates’, at best ‘only well-considered guesses’.11 The welfare of a nation, he made clear, could ‘scarcely be inferred’ from such an estimate. Contained within its pages, however, was a bombshell. In the three years following the Wall Street Crash the American economy had almost halved in size.
Kuznets’ findings became the basis for the second, much more ambitious, phase of Roosevelt’s New Deal, in which the government spent massively on public works, farm aid and social security in order to pull the US economy out of its seemingly interminable recession. Kuznets had provided a more rigorous empirical foundation on which to take such radical action than freight-car loadings. Still, he had warned that the estimates of national income were ‘of little value in themselves’. The headline number was not what was important, he said in words that should ring louder than ever today. For example, closer analysis showed that in‑ equality had increased greatly during the Great Depression. Blue-collar wages had fallen much faster than white-collar salaries, and property owners had fared better than most. These findings provided Roosevelt with the evidence he needed to push through his radical employment policies, which included unemployment relief, the banning of child labour and the right of labour unions to organise. Without Kuznets’ report, much of this would have been impossible.
His work was far from finished. In 1936 Kuznets helped organise the first Conference on Research in Income and Wealth with high-level participants from both academia and government. It was during this conference that the term gross national product, or GNP, was first used. The proceedings of the first three annual conferences were published, revealing sharp differences among participants about what should be measured and what should be left out.
Although Kuznets is considered the father of GDP, in several important respects the methodology that evolved by the early 1940s – and which has remained largely in use ever since – went against his most profoundly held beliefs. Kuznets was striving for a measure that would reflect welfare rather than what he considered a crude summation of all activity. He wanted to exclude illegal activities, socially harmful industries and most government spending. On many of these issues he lost. One student of national accounting goes so far as to suggest, ‘Kuznets, far from being the progenitor of GDP, was its biggest opponent.’12
One of the most important consequences of the Second World War was the invention of the atomic bomb. It was developed by scientists, some of whom had fled Nazi Germany, working on the top-secret Manhattan Project in the New Mexico desert. The bomb was not only an outcome of the war; it also helped win it. Less well known is the case of GDP, the invention of which was hastened and moulded by the life-and-death struggle with fascism. As with the atomic bomb, its invention had a material impact on the war. Like some of those who led the Manhattan Project, Kuznets was also a reluctant participant in his own creation.
Kuznets thought that a sensible definition of the economy should exclude defence spending. During the war he bent to pressure to include expenditure on armaments to defeat fascism, but in peacetime, he argued, a country’s ability to wage war did not contribute to people’s welfare. National income statements, he wrote in 1937, should be constructed from the viewpoint of an ‘enlightened social philosophy’ and should discount activities that were detrimental or, in his word, a ‘disservice’. The first item he listed for exclusion was ‘all expenses on armament’. For Kuznets, spending on preparations for war subtracted from a nation’s well-being because it reduced individuals’ capacity to consume and because it was defensive in nature. If such spending was a necessary evil, then it should appear as a minus in the accounts rather than a plus.
But national income was a child of war. Kuznets lost the battle before it had begun. From 1940 the annual conferences that Kuznets had been holding on developing national accounts were held behind closed doors. Discussions of the state of the US economy had become a top-secret part of war planning. Making the link more explicit still, in 1942 Kuznets was transferred to the Planning Committee of the War Production Board. His main task there was to work out whether the economy had enough spare capacity to switch into the manufacturing of munitions. More generally, he needed to assess the economy’s ability to sustain an all-out war in Europe as well as in Asia, where the Americans had been fighting Japan since the attack on Pearl Harbor the previous year.
Kuznets threw himself into the task. He sought to discover how America’s economic capacity could best be employed so that it struck a balance between building a fighting machine and maintaining the domestic consumption necessary to keep the economy ticking over. Within the government and military establishment there was sharp disagreement between those who wanted to commandeer, even nationalise, the means of production so that they could be diverted to the war effort and others, including those working with Kuznets, who concluded that the economy had plenty of spare capacity that could be marshalled without curtailing domestic consumption. These economists may even have influenced the timing of America’s entry into the war in Europe, having concluded that the US would better be able to sustain its effort if it delayed involvement until late 1943 or early 1944.13
Just as Germany had lacked an atomic bomb that could have tilted the war in its favour, so it was missing the statisticians and economists who could also have helped it. Germany ended the war without having made anything like the advances in national accounts secured by the US.14
There was another powerful force at play, namely John Maynard Keynes. In 1940, two years before Kuznets was drafted into the War Production Board, the famous British economist had written an instantly influential pamphlet with the less-than-ambiguous title How to Pay for the War. As Britain struggled to stave off the threat from Nazi Germany, Keynes complained that economic statistics were too fuzzy to work out the amount of resources that could be mobilised for the war effort. He sought, in the words of his pamphlet’s opening sentence, ‘how best to reconcile the demands of war and the claims of private consumption’.
Keynes wanted to work out the fairest way of sharing diminished resources while preserving the government’s ability to raise debt to pay for the war. ‘In order to calculate the size of the cake which will be left by civilian consumption’, he wrote, the government would have to estimate various things, including the economy’s ‘maximum current output’, the sustainability of drawing on foreign reserves to pay for imports and the amount it would need to spend on guns, aircraft and soldiers. According to his rough-and-ready calculations, output could probably be increased by 15–20 per cent by bringing boys and women into the workforce and by lengthening overtime. But, he complained, ‘the statistics from which to build up these estimates are very inadequate. Every government since the last war has been unscientific and obscurantist, and has regarded the collection of essential facts as a waste of money.’ Only the state, he concluded, was in a position to collect and process such statistics. In their absence, the government was stumbling about in the dark.
That was not all. Until Keynes, attempts to define the scope of a national economy had excluded the government. But Keynes thought the government played a vital role in the economy, particularly during business downturns, when he advocated government spending to stimulate demand. If government expenditure were excluded from GDP, then its perceived role in the economy would be diminished. Until that point national income had been considered the sum total of market activity, or the spending of private individuals, including businesses, on investment and consumption.15 In this definition there was no room for government.
Kuznets considered most government spending – including on such things as roads – as a so-called intermediary cost ‘implicit in our economic civilisation’. To Keynes, this was a conceptual error. If government expenditure was excluded, then whatever the state spent on the war effort would count against economic growth in the national accounts. The more the government spent, the less there was available for private consumption and investment. Keynes’ economic views demanded that this definition of the national economy be turned on its head. The government had to be considered part of the economy.
This was an almost revolutionary assertion. It was nothing less than a redefinition of what the economy was. By baking this idea into our national accounts, Keynes continues to show his influence. Without this definitional shift, what we know today as Keynesian fiscal stimulus would be difficult to justify since it would detract rather than add to national income. Only if the government is considered part of the economy can its spending contribute to final output. In this way, ‘a British war-time definition of the economy has become a global consensus’.16
Keynes’ ideas had a huge impact in Britain, where a new system was being put into practice by two young economists, Richard Stone and James Meade, who were appointed by the Treasury to produce Britain’s first modern set of national accounts. These were duly published in 1941, bearing the distinct imprimatur of Keynes’ theoretical input. Keynes’ ideas also quickly took hold on the other side of the Atlantic, where financing the war had become the raison d’être for producing an accurate set of accounts. Kuznets’ objections were swept aside by realpolitik, underpinned by the intellectual contribution of the forceful and influential British economist. That Keynes was the true inventor of GDP, writes one commentator, is ‘one of economic history’s best kept secrets’.17
There was a third area in which Kuznets lost influence over his invention. Kuznets thought anything detrimental to social welfare should also be excluded. This included not only things like armaments, but also advertising, speculation and all illegal activities, such as gambling, extortion and prostitution. What should be added and what should be left out is a bit like making a cake. The type of recipe you select will affect the flavour and texture: you can have the plain sponge of growth or the chocolate cake with extra filling. Kuznets’ tastes were restrained, even prim. He thought national accounts should measure only economic activity that was good for you – definitely on the sponge-cake side of things, then. Kuznets lost that debate and we have been left with a recipe for double chocolate fudge cake with whipped cream and sugar sprinkles. Everything, good and bad, goes into it. Economic growth – like butter and cream – is not always good for your health.
In his slightly dry style, this is what Kuznets had to say on the matter:
It would be of great value to have national income estimates that would remove from the total the elements which, from the standpoint of a more enlightened social philosophy than that of an acquisitive society, represent disservice rather than service. Such estimates would subtract from the present national income totals all expenses on armament, most of the outlays on advertising, a great many of the expenses involved in financial and speculative activities.’18
The way we calculate economic growth today ignores Kuznets’ warnings. The bigger our banks, the more persuasive our advertisers, the worse our crime and the more expensive our healthcare, the better our economies are seen to be performing. That is not what Kuznets wanted. But it is what we got.