Skip to main content

Population, population density and technological change


In a model on population and endogenous technological change, Kremer (1993) combines a short-run Malthusian scenario where income determines the population that can be sustained, with the Boserupian insight that greater population spurs technological change and can therefore lift a country out of its Malthusian trap. We show that a more realistic version of the model, which combines population and population density, allows deeper insights into these processes. The incorporation of population density also allows a superior interpretation of the empirical regularities between the level of population, population density, population growth, and economic development at aggregated and disaggregated levels.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3


  1. See Boserup (1981).

  2. One could similarly extend and reinterpret the model of Galor and Weil (2000) by including population density. It would particularly help to explain differences in the rate of technological progress between countries, which is likely to have been driven more by their population density than their absolute population size. As we are interested in reinterpreting theory and empirics, we stick to the Kremer paper, which, unlike Galor and Weil (2000), also contains explicit empirical tests.

  3. According to Galor and Weil (1999), most of human history was characterised by such a “Malthusian Regime”. Only in the last 200 years, humans were able to leave the subsistence level and create and accumulate wealth.

  4. As pointed out by Cigno (1984), this would also be an implication of Arrow's (1962) learning-by-doing growth model. Cigno shows, however, that allowing for substitutability between factors and extending the concept of learning-by-doing to all production, this dependence on initial conditions would not occur in the steady state of a growth model incorporating learning-by-doing, although in the transition path, population size or density can influence per capita income and growth.

  5. Gallup and Sachs (1998) differentiate between the effects of population density in the hinterland and in coastal regions. The beneficial effects of population density only are supposed to appear in coastal regions, while they suggest that high population density in the hinterland hinders economic growth. According to them, this is largely due to the fact that high coastal density facilitates international trade and better access to modern technologies, which are largely imported. Their theoretical and empirical analysis is based on the contemporary world, where trade in goods and ideas may indeed be the most important drivers of technological change. It is at least questionable whether this was true to the same extent in earlier times, where technological change in agriculture was much more important and much less dependent on international trade in goods and ideas. See also Boserup (1981) for further discussion.

  6. Burkett et al. (1999) list a number of additional advantages of higher population density which enable more densely populated countries to improve their growth performance through own or adopted technological progress. They include greater social homogeneity, experience with large-scale politics and skills and work attitudes that are more suitable to industrial societies.

  7. In our model, we only consider technological progress and do not make allowances for technological regress due to either ‘depreciation’ of technical knowledge and/or falling populations. Aiyar and Dalgaard (2001) provide a model, in which imperfect knowledge transfers from one generation to the next may result in technological regress. In particular, the model describes how technological levels might decrease due to a fall in population density, which might explain technological regress in some historical and geographic circumstances. These insights supplement our own analysis here, which we believe is more relevant at the global level examined. For a related discussion, see Kremer (1993).

  8. The global land area has indeed not changed drastically over the past one million years, and in this simple formulation of a global relationship, this assumption may be reasonable. See also discussion below about population and technological change in geographically separate regions, which examines this issue at a more disaggregated level.

  9. If it were slightly less than one, it may also account for the fall in population growth after 1960 in Fig. 1. But see also below.

  10. For the aggregate analysis undertaken here, the assumption of a fixed land area appears reasonable. If one were to examine technological change at a more disaggregated level, settlement patterns that shift the inhabited land areas as well as alter local population densities would be important to account for actual trends in technological change over space and time. See also analysis of geographically separate regions below.

  11. For details, see Kremer (1993), which also includes a careful discussion of the data sources and potential biases.

  12. Using alternative data from Clark (1968) or Durand (1977) supports the contention of vastly different population densities between the Old World and the Americas and Australia up until the earliest times. Clark's and Durand's data have considerably higher numbers for the Old World at ad 1 and consequently lower population growth after that. This conclusion would be strengthened if one excluded Africa South of the Sahara from the Old World, where contact to the New World was rather sparse, and the regions were more or less separated.

  13. When we remove outliers (Italy and India in ad 0 and India and Japan in ad 1,000), the influence of population density becomes much stronger and explains a surprisingly large share of the variation in per capita incomes in 1,500. Arguably, it is useful to remove at least Italy and India from the regressions as they were experiencing a high point of a particular imperial period in ad 0 (Italy and India) and ad 1,000 (India), leading to unusually high population concentrations then.

  14. We exclude Spain from the regressions as the urbanization data at 1,000 are temporarily unusually high, probably related to the warfare between Spaniards and Arabs at the time. Urbanization rates dropped considerably subsequently.

  15. For a related discussion, see Gallup and Sachs (1998). Low population density might have other negative effects such as greater ethnic divisions, which has also been found to reduce economic growth (Easterly and Levine 1997). At the same time, it is not clear whether high population densities are still as essential as they used to be given that modern transport and communication technologies offer, if available, greater technological diffusion even without high population densities. However, also here, Africa seriously lags behind.


  • Acemoglu D, Johnson S, Robinson J (2005) The rise of Europe: Atlantic trade, institutional change and economic growth. Am Econ Rev 95(3):546–579

    Article  Google Scholar 

  • Aghion P, Howitt P (1992) A model of growth through creative destruction. Econometrica 60:323–352

    Article  Google Scholar 

  • Aiyar S, Dalgaard C-J (2001) Why does technology sometimes regress? Working paper. University of Kopenhagen

  • Arrow K (1962) The economic implications of learning by doing. Rev Econ Stud 29:155–173

    Article  Google Scholar 

  • Becker G (1981) A treatise on the family. Harvard University Press, Cambridge

    Google Scholar 

  • Becker G, Glaeser E, Murphy K (1999) Population and economic growth. Am Econ Rev 89(2):145–149

    Article  Google Scholar 

  • Bloom DE, Canning D, Malaney PN (1999) Demographic change and economic growth in Asia. CID Working Paper No. 15, Center for International Development, Harvard University

  • Boserup E (1981) Population and technological change: a study of long term trends. University of Chicago Press, Chicago

    Google Scholar 

  • Burkett JP, Humblet C, Putterman L (1999) Pre-industrial and post-war economic development: is there a link? Econ Dev Cult Change 47(3):471–495

    Article  Google Scholar 

  • Chenery H (1960) Patterns of industrial growth. Am Econ Rev 50(4):624–654

    Google Scholar 

  • Cigno A (1984) Further implications of learning by doing: the effects of population on per-capita income. Bull Econ Res 36(2):97–108

    Article  Google Scholar 

  • Cipolla C (1962) The economic history of world population. Penguin Books, London

    Google Scholar 

  • Clark C (1968) Population growth and land use. Macmillan, London

    Google Scholar 

  • Deevey E (1960) The human population. Sci Am 203(3):195–204

    PubMed  Article  Google Scholar 

  • Durand J (1967) The modern expansion of world population. Proc Am Philos Soc 111:136–159

    Google Scholar 

  • Durand J (1977) Historical estimates of world population: an evaluation. Popul Dev Rev 3:253–296

    Article  Google Scholar 

  • Easterly W, Levine R (1997) Africa's growth tragedy: policies and ethnic divisions. Q J Econ 112:1203–1250

    Article  Google Scholar 

  • Frederiksen P (1981) Further evidence on the relationship between population density and infrastructure: the Philippines and electrification. Econ Dev Cult Change 29:749–758

    Article  Google Scholar 

  • Gallup J, Sachs J (1998) Geography and economic development. In: Stiglitz J (ed) The 1998 annual bank conference on development economics, The World Bank, Washington

    Google Scholar 

  • Galor O, Weil D (1999) From Malthusian stagnation to modern growth. Am Econ Rev 89(2):150–154

    Article  Google Scholar 

  • Galor O, Weil D (2000) Population, technology, and growth: from Malthusian stagnation to the demographic transition and beyond. Am Econ Rev 90(4):806–828

    Article  Google Scholar 

  • Grossman EM, Helpman E (1991) Innovation and growth in the global economy. MIT Press, Cambridge

    Google Scholar 

  • Jones C (1995) R&D based models of economic growth. J Polit Econ 103:759–784

    Article  Google Scholar 

  • Kremer M (1993) Population growth and technological change one million BC to 1990. Q J Econ 108:681–716

    Article  Google Scholar 

  • Kuznets S (1960) Population change and aggregate output. In: Kuznets S (ed) Demographic and economic change in developed countries. Princeton University Press, Princeton

    Google Scholar 

  • Maddison A (2001) The world economy: a millennial perspective. OECD, Paris

    Google Scholar 

  • Malthus T (1798) An essay on the principle of population

  • McEvedy C, Jones R (1978) Atlas of world population history. Penguin, London

    Google Scholar 

  • Nestmann T (2000) Der Einfluss von Bevölkerungsdichte auf das Wirtschaftswachstum: Theorie und Empirie. MA thesis, Department of Economics, University of Munich

  • Putnam P (1954) Energy in the future. Macmillan, London

    Google Scholar 

  • Romer P (1990) Endogenous technological change. J Polit Econ 98:71–102

    Article  Google Scholar 

  • Simon J (1977) The economics of population growth. Princeton University Press, Princeton

    Google Scholar 

  • Simon J, Gobin R (1980) The relationship between population and economic growth in LDCs. In: Simon J, DaVanzo J (eds) Research in population economics 2. JAI Press, New York

    Google Scholar 

  • United Nations (1998) Demographic yearbook 1996. United Nations, New York

    Google Scholar 

  • Willis R (1982) The direction of intergenerational transfers and demographic transition. Popul Dev Rev 8(suppl):207–234

    Article  Google Scholar 

  • World Bank (2002) World development indicators (CD-ROM)

Download references


We would like to thank Michael Grimm, Michael Kremer, participants at research seminars at the University of Munich and Göttingen, two anonymous referees, Sandro Cigno as the responsible editor, as well as participants of a session at the 2001 EEA conference in Lausanne for helpful comments and suggestions.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Stephan Klasen.

Appendix 1

Appendix 1

Table 1

Country Code Surface area (1,000 km2) Population (’000) in ad0 Population (’000) in ad 1,000 Population density in ad 0 Population density in ad 1,000 Urbanization in ad 1,000 (%) GDP per capita in ad 1,500
Austria Au 84 500 700 5.95 8.33 0.00 707
Belgium Be 31 300 400 9.68 12.90 8.25 875
Denmark De 43 180 360 4.19 8.37 0.00 738
Finland Fi 338 20 40 0.06 0.11 0.00 453
France Fr 552 5,000 6,500 9.06 11.77 3.92 727
Germany Ge 357 3,000 3,500 8.40 9.80 5.97 676
Italy It 301 7,000 5,000 23.26 16.61 12.26 1,100
Netherlands Ne 42 200 300 4.76 7.14 0.00 754
Norway No 324 100 200 0.31 0.61 0.00 640
Sweden Swe 450 200 400 0.44 0.88 0.00 695
Switzerland Swi 41 300 300 7.32 7.31 0.33 742
United Kingdom UK 243 800 2,000 3.29 8.23 5.71 714
Portugal P 92 500 600 5.43 6.52 2.5 632
Spain S 506 4,500 4,000 8.89 7.95 24.25 698
Western Europe* WE 3,404 22,600 24,300 6.64 7.14 4.58 774
Eastern Europe ** EE 786 4,750 6,500 6.04 8.27 2.16 462
Former USSR*** FUS 24,971 3,900 7,100 0.16 0.28 3.8 500
United States US 9,629 680 1,300 0.07 0.13 400
Other Western Offshoots**** OWO 17,983 490 660 0.03 0.03 400
Total Western Offshoots TWO 27,612 1,170 1,960 0.04 0.07 400
Mexico Me 1,958 2,200 4,500 1.12 2.29 425
Other Latin America ***** OLA 18,501 3,400 6,900 0.18 0.25 410
Total Latin America TLA 20,459 5,600 11,400 0.27 0.55 416
Japan J 378 3,000 7,500 7.94 19.84 3.9 500
China Ch 9,598 59,600 59,000 6.21 6.14 2.5 600
India I 3,287 75,000 75,000 22.82 22.81 0.8 550
Africa Af 28,821 16,500 33,000 0.57 1.14 400
World W 110,200 230,820 268,273 2.09 2.43 565

Notes: *Austria, Belgium, Denmark, Finland, France, Germany, Italy, Netherlands, Norway, Sweden, Switzerland, United Kingdom, Portugal, and Spain. **Albania, Bulgaria, Czechoslovakia, Hungary, Poland, Romania and former Yugoslavia. *** Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, Lithuania, Moldova, Russian Federation, Tajikistan, Turkmenistan, Ukraine, Uzbekistan. ****Australia, New Zealand, Canada. *****Argentina, Brazil, Chile, Colombia, Peru, Uruguay, Venezuela, Bolivia, Costa Rica, Cuba, Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, Nicaragua, Panama, Paraguay, Puerto Rico, Trinidad & Tobago.

Sources: surface area was taken from World Bank (2002). Population figures as well as GDP data were taken from Maddison (2001). Population density (people per km2) was calculated by dividing total population by surface area. Urbanization figures are taken from Acemoglu et al. (2005).

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Klasen, S., Nestmann, T. Population, population density and technological change. J Popul Econ 19, 611–626 (2006).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Endogenous growth
  • Population growth
  • Population density


  • O3
  • J1
  • N3