Abstract
We document a strong positive relationship between natural resource exports and urbanization in a sample of 116 developing nations over the period 1960–2010. In countries that are heavily dependent on resource exports, urbanization appears to be concentrated in “consumption cities” where the economies consist primarily of non-tradable services. These contrast with “production cities” that are more dependent on manufacturing in countries that have industrialized. Consumption cities in resource exporters also appear to perform worse along several measures of welfare. We offer a simple model of structural change that can explain the observed patterns of urbanization and the associated differences in city types. We note that although the development literature often assumes that urbanization is synonymous with industrialization, patterns differ markedly across developing countries. We discuss several possible implications for policy.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
For example, The Economist (2012) writes: “Angola is now one of Africa’s economic successes—thanks almost entirely to oil. [...] Teams of gardeners are putting the finishing touches to manicured lawns, palm trees, tropical shrubs and paved walkways. Soon these will stretch the whole way along the Marginal, the seafront road that is being turned into a grand six-lane highway sweeping around the horseshoe-shaped bay of Luanda, Angola’s buzzing capital. Modern offices, hotels and apartment blocks are sprouting up behind. [...] fancy yachts and speed boats crowd the shores of the Ilha, a once almost deserted strip of sand used mainly by poor fishermen, on which smart restaurants and nightclubs for the new elite are now springing up. [...] Shiny shopping malls are filled with everything the Angolan heart could desire, from gourmet food to the latest fashions and car models. Prices are wildly inflated. Virtually everything [...] has to be imported.”
Weber (1922) also contrasts “consumption cities” and “production cities”. In his description, the former are cities that serve the needs of a political elite, while the latter address the needs of manufacturers and merchants. In his view, consumption cities emerge from rent redistribution rather than rent generation, and they are thus not that different from the “parasitic cities” of Hoselitz (1955) and urban bias theory. While we do not explicitly account for urban bias, this would augment the effects of resource exports on urbanization, not eliminate them. Moreover, our consumption cities are distinct from the “consumer cities” of Glaeser et al. (2001). Using U.S. data, Glaeser et al. (2001)show that high-amenity cities have grown faster than low-amenity cities, and urban rents have gone up faster than urban wages in those locales.
The model is related to a large body of work on structural change and the decline in agriculture; see Herrendorf et al. 2011 for a full survey of the literature. Schultz (1953), Matsuyama (1992), Caselli and Coleman II (2001), Gollin et al. (2002) and Michaels et al. (2012) focus on the “food problem” and how agricultural productivity pushes labor out of agriculture (and possibly into cities). Lewis (1954), Hansen and Prescott (2002) and Lucas (2004) suggest that industrial sector productivity will pull labor out of agriculture. A relatively small literature considers structural change within an open economy (Matsuyama 1992, 2009; Galor and Mountford 2008; Yi and Zhang 2011; Glaeser 2013).
This is not the only possible relationship between urbanization and development, obviously. If the country experiences a Green Revolution, the rise in food production releases labor for the urban sector (Gollin et al. 2007). Rural poverty due to natural disasters can cause rural migrants to flock to cities (Barrios et al. 2006; Henderson et al. 2013a). There are various urban pull factors. If the country experiences an Industrial Revolution, the urban wage increases, which attracts workers from the countryside (Alvarez-Cuadrado and Poschke 2011). If the government adopts urban-biased policies, urban wages increase (Ades and Glaeser 1995; Davis and Henderson 2003; Henderson 2003). Lastly, natural increase in the cities has also become a driver of urbanization in the developing world (Jedwab et al. 2014; Jedwab and Vollrath 2015a). With either a green revolution or an industrial revolution—a productivity increase in agriculture or industry, respectively—urbanization would normally be associated with growth. By contrast, theories in which urbanization is driven by rural poverty, urban bias or urban natural increase will typically imply that urbanization occurs without growth (Fay and Opal 2000; Jedwab and Vollrath 2015b). Our work implies a complementary explanation for urbanization without industrialization.
We could explicitly incorporate a rural non-tradable sector or a rural industrial sector, but this would not change the logic of the model unless individuals had an income elasticity for rural output strictly larger than one. As noted in Sect. 2, though, 91 % of agricultural workers are in rural areas.
Both results follow directly from Eq. (8) and applying the implicit function theorem. If \(R>p_{f}^{*}\overline{c}_{f}\), then an increase in \(\overline{A}\) will actually lower non-tradable labor. In this case, resource earnings are so high that the economy would already have a fraction greater than \(\beta _n\) in non-tradables. Any increase in tradable productivity would substitute workers away from that sector.
The list of countries is available in Web Appendix Table 1.
We recognize that urbanization rates are not in fact collected in each year; the reported figures reflect interpolation and extrapolation from census data and other sources. Nevertheless, we believe that the data from United Nations (2011) offer a generally reliable picture of urbanization, given that population trends tend to be smooth and fairly predictable.
Cash crops include tropical crops that cannot serve as the basis for subsistence and are produced for export: cocoa, coffee, cotton, groundnuts, rubber, sisal, sugar and tea.Footnote 10 We will show that our results are nonetheless robust to including all agricultural exports (i.e., including both cash and subsistence crops).
Although groundnuts are consumed as food in some countries in West Africa, they are more commonly exported as a source of vegetable oil.
The choice of cutoff level is arbitrary, but none of the analysis depends on this precise choice, and in the empirical work of the next section we treat resource exports as a percent of GDP, using the continuous variable instead of relying on a specific cutoff value.
There are arguably good reasons not to identify industrialization with manufacturing alone. Some service sector activities seem to have many of the characteristics of industry, and the GDP share of services rises with development (Herrendorf et al. 2011), even as manufacturing peaks and then declines. Second, industrializing countries may have a comparative advantage in the production of tradable services such as finance and business services (Jensen and Kletzer 2010). These services now account for a large share of GDP in countries such as Japan, which has clearly industrialized. This evolution of world output suggests that a sensible definition of industrialization might include certain parts of the service sector. But it has proven difficult to define the subset of services that might reasonably count as “industry,” or to specify criteria that would usefully characterize this definition.
See Web Appendix Figs. 1–4 for these results.
These relationships are robust to the exclusion of outliers Bahrain (BRN), Kuwait (KWT), and Qatar (QAT), or giving less weight to smaller countries by using the population of each country in 2010 as weights for the estimation of the linear fit. See Web Appendix Figs. 5 and 6.
All data on workers refers to domestic nationals; expatriates are excluded.
See Web Appendix Table 2 for a full breakdown of employment across sectors and locations. We note that in the data, “rural” and “urban” may be defined by administrative boundaries that are imprecise. Some of the rural non-agricultural employment may take place in towns that are misclassified as rural.
While the share of manufacturing and services in GDP in 2010 is a good proxy for the extent of industrialization in 1960–2010, the share of resource exports in GDP in 2010 does not measure well the historical dependence on resource exports. First, many countries have relatively recently started exporting oil, such as Equatorial Guinea and Yemen. For these countries, the historical share of resource exports in GDP does not explain well the contemporaneous urbanization rate. Second, a few countries that were previously resource exporters have become more industrialized over time (e.g., Chile and Malaysia). Lastly, the contribution of resource exports to GDP depends on international commodity prices, which are volatile in the short run. Using the mean share of resource exports in 1960–2010 minimizes these concerns.
Rapid population growth and land pressure could drive urban growth if the rural wage decreases. However, excess urban population growth could also decrease the urban wage. Ultimately, the urbanization rate only increases if urban growth is faster than rural growth, which is not obvious here. For example, we do not find any correlation between demographic growth and urbanization in the data. We nonetheless control for rural density and demographic growth in the regressions.
Urban primacy is defined as the percent of urban population living in the largest city.
Estimated effects of each of the control variables used in Table 1, column (5) are available in Web Appendix Table 4. Alternative regional groupings provide similar results (see Web Appendix Table 5).
Further robustness checks include (i) controlling for or omitting desert countries (see Web Appendix Table 6), (ii) excluding the outliers Bahrain, Kuwait, and Qatar (see Web Appendix Table 7), and (iii) using measures for government spending in 1960, 2010, or both (see Web Appendix Table 8).
Data on resources rents as a share of GDP (%) comes from World Bank (2013). Commodity-specific rents are estimated as the number of units produced times the difference between the world price of the commodity and average unit costs of extraction. Rather unfortunately, data is missing for the 1960s and for many country-year observations post-1970. Besides, production costs are not country-specific in the data, so country-specific rents are imperfectly measured. Our main variable of interest is the contribution of resource exports to GDP. We do not subtract the production costs of natural resources since they directly contribute to the economy by providing an external source of extra income to the domestic factors of production.
These (non-)results on cash crops are in line with Jedwab (2013), who finds a positive effect of cocoa production on urban growth in Ghana and Ivory Coast, and Brückner (2012) and Henderson et al. (2013b) who find a negative effect of agricultural exports on urbanization in Africa when using agricultural price shocks as a source of identification. Positive price shocks should deter urbanization if workers are disproportionately drawn into the agricultural sector. Our empirical results suggest that the specialization effect indeed dominates the income effect for most cash crops.
The results hold when controlling for the share of manufacturing and services in GDP at period t (\(I_{c,t}\)) for the restricted sample of non-missing observations (see Web Appendix Table 9).
In addition to these results, the panel estimates are robust to (i) different definitions of areas and/or regions (Web Appendix Table 10), (ii) controls for or omitting desert countries (Web Appendix Table 11), (iii) excluding outliers Bahrain, Kuwait, and Qatar (Web Appendix Table 12), and (iv) using government expenditures in 1960, 2010, or both as controls (Web Appendix Table 13).
Urbanized countries could have fewer resources (e.g., land or labor) available for resource production. This would lead to a downward bias. It could also be the case that industrial countries do not export the natural resources they produce, using them instead as intermediate goods or consumption goods. For example, China and Mexico are among the world’s top oil producers, but domestic demand accounts for large fractions of production. As these countries are also more urbanized, this leads to a downward bias. From the other side, investments in domestic transportation infrastructure both facilitate the exploitation of natural resources and promote urbanization, leading to an upward bias. Poor institutions that foster an urban bias and restrict industry could also lead to high resource exports as well as high urbanization.
While resource exports fluctuate as a percent of GDP, the binary status of being a resource exporter is highly persistent. In our sample, once a country discovers a resource that is exported,it continues to export that resource in some capacity throughout the period of study. Web Appendix Fig. 7 shows the the fluctuations in resource exports in GDP for 12 countries.
Details of our coding process are available in the Web Appendix, as well as examples of how we coded the discoveries in Botswana (diamonds) and Oman (oil) in Web Appendix Figs. 8 and 9.
We use data for the most recent year. When no data could be found for the period 2000–2010, we use data for the 1990s. Similarly to McMillan and Rodrik (2011), the 11 sectors are: “agriculture”, “mining”, “public utilities”, “manufacturing”, “construction”, “wholesale and retail trade, hotels and restaurants”, “transportation, storage and communications”, “finance, insurance, real estate and business services”, “government services”, “education and health” and “personal and other services”.
FIRE includes real estate services, which are mostly non-tradable. However, census and survey data rarely distinguish them from other FIRE sectors, which leaves us no choice but to include them.
For example, as noted above, Kuwait and Japan have similar per capita incomes and urbanization rates, but Kuwaiti cities have a lower employment share of manufacturing and FIRE workers (11 vs. 33 %). Other pairs of countries for which we corroborate this hypothesis are: Saudi Arabia vs. South Korea and Taiwan; Gabon and Libya vs. Mexico, Chile and Malaysia; and Angola vs. China and The Philippines.
The “commerce and personal services” sector consists of “wholesale and retail trade, hotels and restaurants”, “transportation, storage and communications” and “personal and other services”. We lump them together to capture the whole commerce and personal services (Pers.Serv.) sector.
Web Appendix Table 14 shows that the urban-rural gap in terms of access to improved water, access to improved sanitation facilities, or usage of non-solid fuels as a main source of energy, is indeed more pronounced in resource-exporting countries than in industrial countries. The non-result for inequality within urban areas should then be taken with caution, as urban Ginis are usually measured using household survey data for the urban areas, which may not capture well the top of their income distribution. In contrast, total Ginis may be computed using several sources of data, including tax records.
Web Appendix Table 15 examines urban primacy in our panel specification, and again finds no significant relationship with resource exports.
This variable is created using the same IPUMS and survey data we used to generate the data on urban labor composition.
Schoellman (2012) uses the returns to schooling of foreign-educated immigrants in the United States to measure the education quality of their birth country. The returns are thus based on the returns achieved in a common labor market.
The dependent variable was recreated using existing data on the relative prices of various goods and services for expatriates for the largest city of each country. See Web Appendix for more details. As an example, Luanda in Angola is at least two times more expensive for expatriates than Shanghai in China, although Angola and China have similar income and urbanization levels.
A household is considered to: (i) have access to an improved water source if the household has easy access to drinking water, (ii) have access to improved sanitation if an excreta disposal system is available to the household, (iii) have sufficient-living area if the household lives in dwelling units with less than 2 persons per room, and (iv) live in a durable residence if the residence is built on a non-hazardous location and has a structure permanent and adequate enough to protect its inhabitants from extreme climatic conditions.
Results also hold when controlling for government expenditure in 2010 (Web Appendix Table 19).
As shown by Michaels (2011), resource production can promote long-term development when institutions are strong. Campante and Glaeser (2009) document how human capital and political stability can explain the divergent paths of Chicago and Buenos Aires, which both began as consumption cities in our typology.
References
Acemoglu, D., Johnson, S., & Robinson, J. A. (2002). Reversal of fortune: Geography and institutions in the making of the modern world income distribution. The Quarterly Journal of Economics, 117(4), 1231–1294.
Ades, A. F., & Glaeser, E. L. (1995). Trade and circuses: Explaining urban giants. The Quarterly Journal of Economics, 110(1), 195–227.
Alvarez-Cuadrado, F., & Poschke, M. (2011). Structural change out of agriculture: Labor push versus labor pull. American Economic Journal: Macroeconomics, 3(3), 127–158.
Barrios, S., Bertinelli, L., & Strobl, E. (2006). Climatic change and rural-urban migration: The case of Sub-Saharan Africa. Journal of Urban Economics, 60(3), 357–371.
Brückner, M. (2012). Economic growth, size of the agricultural sector, and urbanization in Africa. Journal of Urban Economics, 71(1), 26–36.
Campante, F. & Glaeser E. L. (2009). Yet another tale of two cities: Buenos Aires and Chicago. NBER Working Papers 15104.
Caselli, F., & Coleman II, W. J. (2001). The U.S. structural transformation and regional convergence: A reinterpretation. Journal of Political Economy, 109(3), 584–616.
Corden, W. M., & Neary, J. P. (1982). Booming sector and de-industrialisation in a small open economy. Economic Journal, 92(368), 825–848.
Davis, J. C., & Henderson, J. V. (2003). Evidence on the political economy of the urbanization process. Journal of Urban Economics, 53(1), 98–125.
De Long, J. B., & Shleifer, A. (1993). Princes and merchants: European city growth before the industrial revolution. Journal of Law and Economics, 36(2), 671–702.
Duarte, M., & Restuccia, D. (2010). The role of the structural transformation in aggregate productivity. The Quarterly Journal of Economics, 125(1), 129–173.
Duranton, G. (2008). Viewpoint: From cities to productivity and growth in developing countries. Canadian Journal of Economics, 41(3), 689–736.
Duranton, G. (2013). Growing through cities in developing countries. Working paper.
FAO. (2013). FAOSTAT. Rome: Food and Agricultural Organization.
Fay, M. & Opal, C. (2000). Urbanization without growth: A not-so-uncommon phenomenon. The World Bank Policy Research Working Paper Series 2412.
Galor, O., & Mountford, A. (2008). Trading population for productivity: Theory and evidence. Review of Economic Studies, 75(4), 1143–1179.
Glaeser, E., Kolko, J., & Saiz, A. (2001). Consumer city. Journal of Economic Geography, 1(1), 27–50.
Glaeser, E. L. (2013). A world of cities: The causes and consequences of urbanization in poorer countries. National Bureau of Economic Research, Inc NBER Working Papers 19745.
Glaeser, E. L., & Gottlieb, J. D. (2009). The wealth of cities: Agglomeration economies and spatial equilibrium in the United States. Journal of Economic Literature, 47(4), 983–1028.
Glaeser, E. L., Kallal, H. D., Scheinkman, J. A., & Shleifer, A. (1992). Growth in cities. Journal of Political Economy, 100(6), 1126–1152.
Gollin, D., Lagakos, D., & Waugh, M. (2014). The agricultural productivity gap. Journal of Economics, 129(2), 939–993.
Gollin, D., Parente, S. L., & Rogerson, R. (2002). The role of agriculture in development. American Economic Review, 92(2), 160–164.
Gollin, D., Parente, S. L., & Rogerson, R. (2007). The food problem and the evolution of international income levels. Journal of Monetary Economics, 54(4), 1230–1255.
Hansen, G. D., & Prescott, E. C. (2002). Malthus to solow. American Economic Review, 92(4), 1205–1217.
Harding, T. & Venables, A. J. (2013). The Implications of Natural Resource Exports for Non-Resource Trade. Discussion Papers CEPR Discussion Papers: C.E.P.R. 9318.
Henderson, J. V. (2003). The urbanization process and economic growth: The so-what question. Journal of Economic Growth, 8(1), 47–71.
Henderson, J. V. (2010). Cities and development. Journal of Regional Science, 50(1), 515–540.
Henderson, V., Storeygard, A., & Deichmann, U. (2013). Has climate change promoted urbanization in Sub-Saharan Africa? Mimeo.
Henderson, V., Roberts, M., & Storeygard, A. (2013). Is urbanization in Sub-Saharan Africa different?. Working paper.
Herrendorf, B., Rogerson, R., & Valentinyi, A. (2011). Growth and structural transformation. Draft prepared for the handbook of economic growth.
Hoselitz, B. F. (1955). Generative and parasitic cities. Economic Development and Cultural Change, 3(3), 278–294.
Jedwab, R. (2013). Urbanization without industrialization: Evidence from consumption cities in Africa. Working paper.
Jedwab, R., & Vollrath, D. (2015a). The mortality transition, malthusian dynamics and the rise of poor mega-cities. Working paper.
Jedwab, R. & Vollrath, D. (2015b). Urbanization without growth in historical perspective. Working paper.
Jedwab, R., & Osei, R. D. (2013). Is structural change in Sub-Saharan Africa different? New historical evidence from Ghana. Working paper.
Jedwab, R., Christiaensen, L., & Gindelsky, M. (2014). Rural push, urban pull and... urban push? New historical evidence from developing countries. The George Washington University, Institute for International Economic Policy. Working Papers 2014-04.
Jensen, J. B., & Kletzer, L. G. (2010). Measuring tradable services and the task content of offshorable services jobs. Labor in the new economy (pp. 309–335)., NBER chapters Chicago: National Bureau of Economic Research, Inc.
Lagakos, D., & Waugh, M. (2013). Selection, agriculture, and cross-country productivity differences. American Economic Review, 103(2), 948–980.
Lewis, A. (1954). Economic development with unlimited supplies of labour. The Manchester School, 22(2), 139–191.
Lucas, R. E. (2004). Life earnings and rural-urban migration. Journal of Political Economy, 112(S1), S29–S59.
Lucas, R. E. (2009). Trade and the diffusion of the industrial revolution. American Economic Journal: Macroeconomics, 1(1), 1–25.
Matsuyama, K. (1992). Agricultural productivity, comparative advantage, and economic growth. Journal of Economic Theory, 58(2), 317–334.
Matsuyama, K. (2009). Structural change in an interdependent World: A global view of manufacturing decline. Journal of the European Economic Association, 7(2–3), 478–486.
McMillan, M. & Rodrik, D. (2011). Globalization, structural change and productivity growth. NBER Working Papers 17143.
Michaels, G. (2011). The long term consequences of resource-based specialisation. The Economic Journal, 121(551), 31–57.
Michaels, G., Rauch, F., & Redding, S. J. (2012). Urbanization and structural transformation. The Quarterly Journal of Economics, 127(2), 535–586.
Potts, D. (2005). Counter-urbanisation on the Zambian copperbelt? Interpretations and implications. Urban Studies, 42(4), 583–609.
Rodrik, D. (2013a). The perils of premature deindustrialization. Prague: Project Syndicate.
Rodrik, D. (2013b). Unconditional convergence in manufacturing. The Quarterly Journal of Economics, 128(1), 165–204.
Sachs, J. D., & Warner, A. M. (2001). The curse of natural resources. European Economic Review, 45(4–6), 827–838.
Schoellman, T. (2012). Education quality and development accounting. Review of Economic Studies, 79(1), 388–417.
Schultz, T. W. (1953). The economic organization of agriculture. New York: McGraw-Hill.
The Economist. (2012). Boom boom: The oil money may start to trickle down. London: The Economist.
UN-Habitat. (2003). Slums of the World: The face of urban poverty in the new millennium?. New York: United Nations.
United Nations. (2011). World urbanization prospects: The 2011 revision. New York: United Nations, Department of Economic and Social Affairs.
United Nations. (2013). Milledium development goals indicators: The official United Nations site for the MDG indicators. New York: United Nations.
USGS. (2013). Mineral industry surveys 1960–2010. Reston: USGS.
Vollrath, D. (2009a). The dual economy in long-run development. Journal of Economic Growth, 14(4), 287–312.
Vollrath, D. (2009a). How important are dual economy effects for aggregate productivity? Journal of Development Economics, 88(2), 325–334.
Weber, M. (1922). Wirtschaft und gesellschaft. Tubingen: Mohr Possible.
World Bank. (2013). World development indicators. Washington D.C.: World Bank.
Yi, K. M. & Zhang, J. (2011). Structural change in an open economy. Report: Federal Reserve Bank of Minneapolis Staff. 456.
Acknowledgments
We would like to thank the editor, three anonymous referees, Nathaniel Baum-Snow, Filipe Campante, Donald Davis, Jeremiah Dittmar, Gilles Duranton, Edward Glaeser, Vernon Henderson, Berthold Herrendorf, Fabian Lange, Margaret McMillan, Markus Poschke, Andres Rodriguez-Clare, Harris Selod, Adam Storeygard, Michael Waugh, David Weil, Adrian Wood, Kei-Mu Yi and seminar audiences at Alicante, American, BREAD-World Bank Conference on Development in sub-Saharan Africa, Brown, Columbia, Durham, EEA, EBRD, Georgetown (GCER), George Washington, John Hopkins, Manchester, McGill, NBER/EFJK Growth Group, NBER SI Urban Workshop, Oxford (CSAE), Pretoria, SFSU (PACDEV), Urban Economic Association Meetings (Atlanta), U.S. Department of State, World Bank, World Bank-GWU Conference on Urbanization and Poverty Reduction, Yonsei (SED) and York. We gratefully acknowledge the support of the Institute for International Economic Policy and the Elliott School of International Affairs (SOAR) at George Washington University.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Gollin, D., Jedwab, R. & Vollrath, D. Urbanization with and without industrialization. J Econ Growth 21, 35–70 (2016). https://doi.org/10.1007/s10887-015-9121-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10887-015-9121-4