ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

World History in Statistics

World History in Statistics

The World Economy by Angus Maddison; Development Centre Studies, Organisation of Economic Cooperation and Development, Indian edition published by Academic Foundation, New Delhi, 2007


World History in Statistics

Tirthankar Roy

European settler countries in the Americas and the Pacific, the distance between poorer and richer regions of the world increased to very high and almost certainly unprecedented levels. In the same time

conomic historians of south Asia have long been familiar with Angus Maddison’s work. One of his books, Class Structure and Economic Growth: India and Pakistan since the Moghuls, was published in 1971. The book appeared three years after a famous symposium in the Indian Economic and Social History Review around an article published by Morris D Morris. This symposium, where the big picture of Indian economic growth in the 19th century was debated by eminent historians, was remarkable for its sparse use of statistics, especially statistics of total and average incomes. Rereading that debate today, one cannot help being struck by its speculative tone, while elementary benchmarks for settling growth questions remained missing. Class Structure dealt with many of the same issues that were discussed in the symposium but approached the subject of economic growth by a different methodological route. The book approached south Asian economic history with a set of carefully constructed statistical tables. The part of the narrative that the book’s many critics found hard to argue with was the quantitative one. Coming in the wake of one debate, the book opened up another, on the quality and construction of national income statistics for India. In this latter field contemporary and later research, mainly by S Sivasubramonian, overtook the methods and results of Class Structure.1

In the last 40 odd years, Maddison has been engaged in a project that is much larger in scale but similar in scope to his work on India: telling the story of the world economy through a set of statistical tables. The book under review is a two-inone combo that brings two different works together. The first one, A Millenial Perspective, is a review of what it calls “the contours of world development” in the last 1,000 years, based on both descriptive analysis and long-range statistics on a number of indices, including GDP, population, life expectancy, trade and labour productivity. The density and quality of data

Economic & Political Weekly

april 19, 2008

The World Economy by Angus Maddison;

Development Centre Studies, Organisation of Economic Cooperation and Development, Indian edition published by Academic Foundation, New Delhi, 2007; pp 653, Rs 1,295.

change depending on time and region. The second work is called Historical Statistics, which presents total and per capita GDP at purchasing power parity for most countries of the world, covering mainly the period 1820-2001. Valuable source notes accompany these tables.

The core set of tables on GDP is the work that the author is most well known for. It has evolved in the level of detail, coverage, precision and method of estimation, partly in response to specific criticism from area experts. One major subset of the data that has seen development and refinement is the historical GDP of China. While the quantitative approximation of growth became more detailed and inclusive, the output of this vast project has spawned important interpretive pieces, including several by Maddison himself, on inequality, productivity and sources of economic growth.

Three Questions

This dataset now forms the raw material using which economic historians can answer three questions about the fundamental patterns of international economic inequality in the last 200 years. First, how has the pattern of inequality changed? Second, has the distribution of quality of life followed the same path as the distribution of income? And third, what are the reasons for divergence and convergence in these indices?

On the first question, we now know based on this dataset that world income inequality was already quite high in the early 19th century, when the industrial revolution was under way in Britain and beginning in France.2 Thereafter, with the spread of the industrial revolution to western Europe and the beginning of economic growth in the regions that Maddison calls “European offshoots”, that is, span, there was closing of inequality within western Europe and the offshoots. Further research with this data also confirms that inequality peaked in about 1950 but did not either worsen or fall significantly in the subsequent 50 years. This story is one of increasing inequality but not one of rich-getting-richer-and-poorgetting-poorer kind of inequality. Throughout the 19th century, 1820-1910 to be precise, income levels in all regions increased and life expectancy improved.

The dramatic rise in world inequality derived largely from the fact that most world regions grew much more slowly than those located on two sides of the north Atlantic. In particular, the unusually slow growth of India and China, throughout the most populous regions of the world, was clearly among the major sources of divergence in the modern world economy. The increase in income per capita in India and China between 1820 and 1950 was in the range of 10-20 per cent, whereas the increase in western Europe was fourfold. In the early 19th century, GDP per head in the United Kingdom was about three times greater than that in China. By 1910 the differential had risen to 6:1 and by 1950 to

10:1. Among other major world regions, Latin America’s economic growth roughly coincided with the world average, whereas Africa’s fell below the average in the last third of the 20th century. The absence of change in world income distribution since 1950 reflects a slowing of economic growth among western Europe, north America and the offshoots, a catching up by Japan and east Asia, the take-off of China and India beginning in the 1980s and the poor post-1970 performance of Africa, which only now shows signs of weakening. Together, this diverse pattern makes the application of any straightforward “core” and “periphery” approach to world economic inequality impossible for the last two centuries.

These GDP data form one component necessary for an answer to the second question – have well-being and income moved together or diverged? The data here for the 19th century are patchier. It is


plausibly believed that the inequality in life expectancy worsened, as that in income did, in the 19th century. But unlike income, world inequality in life expectancy fell after 1930. In other words, if catching up in income was slow or absent, in social indicators it was accelerated from the interwar period. Measures of literacy and nutrition also suggest an earlier and faster catch up, although most measures are reliable only for the second half of the 20th century [Kenny 2005]. Why welfare indices show a quicker and earlier closing of the gap is an interesting problem. To the extent GDP trends are relatively more market-determined and welfare trends are relatively more policy-determined, the divergence hints at the manner in which modern states, especially in the developing world, defined their priorities over a considerable length of time in the 20th century. That being said, in part the trend is a spurious one, since most welfare indices are subject to an absolute ceiling (100 per cent for literacy, for example), whereas GDP is not.

It is with respect to the third question, namely, sources of long-run economic growth and of world inequality, that these numbers, along with the text supplied with the numbers, fall short of expectation. The textual explanations in Maddison’s work generally display a bias towards what can be measured. Many historians would consider that the really interesting answers to the two big analytical questions

– why there was divergence in the world and why there was convergence among the north Atlantic economies – defy quantification.3 Their answers would range from colonialism, to environment and resource endowments, to culture and know ledge, to institutions, to the relationship between states and societies. The great paradox of interpretive economic history is that any one of these explanatory models seems unable to capture the experience of the whole world. For example, the colonial exploitation might in principle work to explain Indian stagnation but fail for China, which was never colonised. British India saw greater development of growthpromoting institutions such as modern banking or a stable currency system than Japan but still fell behind Japan in the 20th century.

Faced with this paradox, comparative economic history of the world seems to be moving in three directions. First, several recent works have taken the explanatory models further back in time, an enterprise that has produced significant results for China and western Europe. Second, there is an implicit awareness that monocausal models would not work and that historians need to take a more serious account of the interactions between major explanatory variables. For example, resource endowments, institutions and technology may well follow an interactive and connected evolutionary path. We should then be asking multi-layered questions like the following. Did land scarcity or water scarcity in India give rise to particular kinds of communal property rights in natural resources that affected modes of usage, technological change and longterm land yield? And third, much detailed applied research on subjects that would have addressed only the specialist until recently, now addresses the great inequality problematic. The availability of a huge stylised fact has given meaning and purpose to comparative history and turned much economic history comparative.

Maddison’s work has given concrete quantitative shape to that stylised fact and thus enabled asking many of the “why” questions about long-term patterns of growth. The World Economy is not just a book. It is an enterprise, and a primary reference for empirical studies in the history of economic growth.



1 National Income of India in the Twentieth Century, Oxford University Press, Delhi, 2000.

2 I base the discussion on Francois Bourguignon and Christian Morrisson (2002), which uses an earlier set of data constructed by Maddison.

3 A recent work claims that the theory of factor-price equalisation through trade explains the convergence among the Atlantic economies, [O’Rourke and Jeffrey Williamson 2001].


Bourguignon, Francois and Christian Morrisson (2002): ‘Inequality among World Citizens: 1820-1992’, American Economic Review, Vol 92, No 4, pp 727-44

Kenny, Charles (2005): ‘Why Are We Worried About Income? Nearly Everything That Matters is Converging’, World Development, Vol 33, No 1, pp 1-19.

O’Rourke, Kevin and Jeffrey Williamson (2001): Globalisation and History: The Evolution of a Nineteenth-Century Atlantic Economy, MIT Press, Cambridge, Massachusetts.

april 19, 2008

Economic & Political Weekly

Dear reader,

To continue reading, become a subscriber.

Explore our attractive subscription offers.

Click here


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top