Maddison counting

ANGUS MADDISON, who died on April 24th at the age of 83, described himself as a chiffrephile—a lover of figures. Like many men, he had his first serious crush at the age of 13. He read “How to Pay for the War”, by John Maynard Keynes; it was the annex on national income that most tickled his fancy. For the next 70 years he pursued ever more elusive numbers, estimating GDP for a growing range of countries over a lengthening span of time. In 1995 he published GDP estimates for 56 countries as far back as 1820. In 2001 his romantic adventures culminated in an estimate for world output in the year 1AD: $105.4 billion at 1990 prices.

GDP is a modern term, but the urge to count the nation's produce and compare countries' standards of living predates Adam Smith. Maddison saw himself as heir to a tradition that began with William Petty, the pioneer of “political arithmetick”, who in 1665 estimated the income of England and Wales at £40m. That calculation was of pressing concern to Petty, who wanted to show the king how to pay for the war against the Dutch. But why did Maddison care about the GDP of the distant past?

He believed that the “pace and pattern” of economic activity had deep historical roots. Economies, he thought, do not “take off”, as if from nowhere. Even the industrial revolution was too gradual to warrant the term revolution and too broad to be considered merely industrial. Take, for example, the progress of maritime technology. By 1773, John Harrison was claiming a £20,000 prize from the British Parliament for inventing a seaworthy chronometer. Captain James Cook had reached Australia's east coast, and thanks to sauerkraut and citrus juice, he had lost none of his crew to scurvy.

Even scholars who believed there was a lot of economic progress to measure before the 19th century doubted there was enough data to measure it. Maddison made the most of whatever was available. He drew on one scholar's work on probate inventories in 17th and 18th century England, which showed that each generation passed on more property, furniture and houselinen to its descendants than the last. His economic portrait of Mughal India was influenced by a 16th-century survey by Abu Fazl, vizier to Emperor Akbar. His estimates of Japan's population relied on the annual register of religious affiliation, brought in after the Portuguese were expelled and Christianity outlawed in 1587. One of his students, Bart van Ark, now chief economist of the Conference Board, says Maddison urged him to venture beyond libraries and statistical offices. Even a painting in a museum might provide some clue to a country's standard of living centuries before.

“There is room for two or three economic theorists in each generation, not more,” wrote Colin Clark, one of Maddison's heroes. Every other economist, he added, should be content to build knowledge by steadily laying “stone on stone”. Maddison laid the foundations for many big thoughts. Ten days before his death he was cited in a speech by Robert Zoellick, president of the World Bank, declaring the end of the “third world”. Maddison's figures show that Asia accounted for more than half of world output for 18 of the last 20 centuries. Its growing clout in the world economy is, therefore, a “restoration” not a revolution.

Even as they foreshadow the rise of Asia, his numbers also help explain the historical rise of Europe. His estimates of per head GDP provide a useful empirical crosscheck for a grand thesis proposed by Daron Acemoglu, Simon Johnson and James Robinson in 2005. They argued that European countries prospered after 1500 in so far as they imposed checks on monarchical power and enjoyed access to the Atlantic Ocean, with its lucrative trade in commodities and slaves. Maddison's estimates also appear in their work explaining why poor colonies became rich, and rich colonies became poor. They conclude that sparsely populated colonies benefited over the long run from the property rights that European settlers brought with them. Richer, well-populated colonies suffered from efforts to suck them dry.

Messrs Acemoglu, Johnson and Robinson caution that Maddison's figures for the years before 1820 are “no more than educated guesses”. Maddison freely conceded that the further back he went, the more he had to rely on “clues and conjecture”. In an intemperate article last year, Gregory Clark of the University of California, Davis, described these numbers as “fictions, as real as the relics peddled around Europe in the Middle Ages”. Credulous economists demanded numbers, “however dubious their provenance”, and Maddison supplied them.

Go figure

Quantification can create the illusion of precision. For example, Maddison assumes that African GDP before 1820 remained more or less at subsistence levels. If that is all that can be said, does it add anything to put a number on it ($400-$425 per head)? But he was not selling comforts to the credulous. He believed that numbers sharpened debate. Quantification, he wrote, “is more readily contestable and likely to be contested.” In disputing his figures, scholars would be inspired to provide their own. Even those who disagreed with his work would be influenced by it.

Given the length and depth of his career, it is tempting to say that this intellectual influence is impossible to measure. But that would be contrary to his faith in quantification. His curriculum vitae counts 20 books and 130 articles, plus another 19 volumes that he edited or co-authored. His work has been translated into 12 languages and two books have racked up more than 2,000 citations, according to Google Scholar. He supervised 13 doctoral students, as well as co-founding the Groningen Growth and Development Centre at the University of Groningen, which he joined in 1978, and the Club des Chiffrephiles in 1990. But as even Maddison admitted, “no sensible person would claim that [quantification] can tell the whole story.” He was deeply fond of numbers. And a large number were deeply fond of him.