Life Earnings and Rural-Urban Migration

Literature › Works

Robert E. Lucas, Jr. (2002). Life Earnings and Rural-Urban Migration.

Type	working paper
Scope	Comparative and methods
Method	theoretical or conceptual
Period studied	1820–2000
Regions	Britain, United States, South Korea, Mexico, Brazil, Philippines, Argentina, India, Thailand, Egypt
Themes	comparative theory, demography urbanisation, education human capital, inequality poverty unemployment
Citation key	robertelucas2002

Argument

Lucas models urbanization as a transfer of labour from a traditional, land-intensive technology to a human-capital-intensive urban technology with unending growth potential. Cities are places where migrants accumulate skills. With an external effect in learning (Eaton-Eckstein), the city becomes ever more attractive, so migration is gradual rather than one-off. The model generates temporary urban 'unemployment' made up of full-time learners, offered as an alternative to the Harris-Todaro lottery explanation. It is a theoretical model with calibrated numerical experiments, and it is compared with an alternative two-skill urban model that implies a permanently widening skill gap.

Evidence and method

Table 1 of World Bank World Development Indicators data on rural population shares and city sizes for eight countries. Historical labour-force shares from Britain, the US and South Korea (Shin 1990). Calibration parameters (Johnson 1948 land share). Numerical shooting-algorithm simulations. Appendix proofs. Nothing on South Africa.

Findings

• Lucas frames the paper's question: why migrants keep arriving despite apparent urban hardship, and why they cannot bid themselves into high-wage city jobs. (p. 4)

  Why can these immigrants not bid themselves into high wage city jobs? Why do they keep coming?

• Migrants come because cities are places to accumulate human capital, and the return to this investment equals the rural wage in equilibrium. (p. 5)

  The immigrants come, as in the model of Eaton and Eckstein (1997), because cities are good places to accumulate human capital, and it is the return to this activity that equals the rural wage in equilibrium.

• Without an external effect, all migration occurs at once, which is unrealistic. (p. 12)

  All the rural-urban migration that ever takes place will take place at date t = 0.

• The learning externality makes the city increasingly attractive, so a decision not to migrate early can be reversed later. (p. 5)

  a decision not to migrate early on can be reversed later on, because the city becomes an ever more attractive destination.

• Farm labour eventually goes to zero for any positive externality, and the asymptotic catch-up lag is 1/(θδ). (p. 23)

  Notice that neither this conclusion nor the asymptotic rate of convergence depends on the size of the externality parameter θ.

• Simulated 'unemployment' of learners peaks at over 15 percent with high θ and declines slowly. (p. 26)

  peaks at more than 15 percent of the urban workforce in about 6 years, and falls to 6 percent 30 years after that.

• Lucas acknowledges that this unemployment is voluntary and that jobs are available to all at skill-appropriate wages. (p. 27)

  jobs are available to everyone who wants one at wages suited to their individual skill levels.