International Capital Flows, Land Conversion and Wage Inequality in Poor Countries

Abstract

Conversion of land from agricultural to industrial use has met with strong opposition in many developing countries in recent times. A number of relevant papers study welfare effects associated with willful conversion, no conversion or politically forced and manipulated conversion of land into other activities. No study, however, deals with the evolving skilled-unskilled wage gap consequent upon full or partial convertibility of land in the short and the long run. This paper shows that in a multiple-commodity world with land as a crucial input both for agriculture and industry, an influx of capital for supporting industrial production must widen the skilled-to-unskilled wage gap. If the rate of conversion of land exceeds a critical value in the short-run, the wage gap rises. Even in the long run, possibility of full conversion can raise the wage gap if agriculture is—as is indeed the case—labor-intensive in developing countries. Showing the dependence of wage inequality on the degree to which land is convertible into other uses is the paper’s unique contribution to the literature.

Appendices

Appendix 1

$$ \hat{w}{\theta}_{L1}+{\hat{R}}_1{\theta}_{T1}=0 $$

(19)

$$ \hat{w}{\theta}_{L2}+{\hat{R}}_2{\theta}_{T2}=-\hat{r}{\theta}_{K2} $$

(20)

$$ {\lambda}_{L1}{\hat{a}}_{L1}+{\lambda}_{L2}{\hat{a}}_{L2}-{\lambda}_{L1}{\hat{a}}_{T1}=0 $$

(21)

(21) follows from differentiation of eq. (6)

Note: ${\hat{T}}_1={\hat{T}}_2=0$, also ${\hat{a}}_{T2}=0$ (by prior assumption).

Therefore, ${\lambda}_{L1}\left({\hat{a}}_{L1}-{\hat{a}}_{T1}\right)+{\lambda}_{L2}{\hat{a}}_{L2}=0.$

or, $-{\sigma}_1{\lambda}_{L1}\left(\hat{w}-{\hat{R}}_1\right)-{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\left(\hat{w}-\hat{r}\right)=0$

$$ -\hat{w}\left({\sigma}_1{\lambda}_{L1}+{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\right)+{\lambda}_{L1}{\sigma}_1{\hat{R}}_1=-{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\hat{r} $$

Substituting from (19) and solving

$$ \hat{w}=\frac{\lambda_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}}{\left({\sigma}_1{\lambda}_{L1}+{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\right)}\hat{r} $$

(22)

Note that, as

$$ \hat{r}<0,\hat{w}<0 $$

(23)

Appendix 2

With μ > 0 we can redo the calculations and get to (24).

$$ -\hat{w}\left(\frac{\sigma_1{\lambda}_{L1}}{\theta_{T1}}+{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\right)-\left(\mu |\lambda |\frac{\theta_{L1}}{\theta_{T1}}-\mu |\lambda |\frac{\theta_{L2}}{\theta_{T2}}\right)=\hat{r}\left({\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}+\mu |\lambda |\frac{\theta_{L2}}{\theta_{T2}}\right) $$

where, |λ| = (λ_L1λ_T2 − λ_T1λ_L2)

or,

$$ \hat{w}=\frac{\hat{r}\left[\mu \left|\lambda \right|\frac{\theta_{K2}}{\theta_{T2}}+{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}\right]}{\frac{\lambda_{L1}{\sigma}_1}{\theta_{T1}}+{\lambda}_{L2}{\sigma}_2\frac{\theta_{K2}}{1-{\theta}_{T2}}+\frac{\mu \left|\lambda \right|}{\theta_{T1}}\frac{\left|\theta \right|}{\theta_{T2}}} $$

(24)

Divide both numerator and denominator by μ.

Thus, when

$$ \mu \to \infty, \hat{w}=\hat{r}{\theta}_{K2}\frac{\theta_{T1}}{\mid \theta \mid }>\hat{r}{\theta}_{K2,\kern0.75em \mathrm{since}}\kern0.2em \hat{r}<0. $$

(25)

Fig. 2

Graph Showing Agricultural Land as percentage of Land Area

Notations for Figures 2 and 3:

East Asia & Pacific (excluding high income) – EAP-XHY

Europe & Central Asia (excluding high income) – ECA-XHY

Fragile and conflict affected situations - Fragile

Heavily indebted poor countries - (HIPC)

Latin America & Caribbean (excluding high income) –LAC-XHY

Least developed countries: UN classification - LDC

Low income - LY

Lower middle income - LMY

Middle East & North Africa (excluding high income) MENA-XHY

Sub-Saharan Africa (excluding high income) - SSA-XHY

Fig. 3

Country Group-wise Agricultural Value added as percentage of GDP