Impacts of Improved Infrastructure on Labor Allocation and Livelihoods: The Case of the Jamuna Multipurpose Bridge, Bangladesh

Abstract

The present study econometrically examines the impacts of an improved road transportation system on labor allocation, input use, and livelihood diversification in an economically lagging part of Bangladesh. Before 1998, the northwest part of Bangladesh, home to 22% of nearly 160 million people, was de-linked from the economically advanced eastern part of Bangladesh. On June 23, 1998, the Jamuna Multipurpose Bridge (JMB) on the river Jamuna was opened for use. The bridge has established direct road and rail transportation systems between the northwestern part and the eastern part of Bangladesh. Using these phenomena, and linking the Household Income and Expenditure Survey (HIES) data, HIES2000 and HIES2010, this study elucidates the impacts of improved infrastructure on labor allocation and the input application behavior of the sampled households. Applying the difference-in-difference estimation approach, this study demonstrates that the JMB that has linked the northwest and eastern regions has encouraged labor out-migration, resulting in the most vulnerable day-laborers in both farm and non-farm sectors moving to other jobs with relatively high marginal returns. The improved transportation infrastructure had also increased the daily wage rates, which were lower than the comparison group, and the cropping intensity in the Rajshahi and Rangpur divisions. The study, therefore, concludes that investment in infrastructure can be instrumental in unleashing economic growth and alleviating poverty in the economically struggling parts of developing countries.

Résumé

La présente étude examine économétriquement les effets d'un système de transport routier amélioré sur la répartition de la main-d'œuvre, l'utilisation des intrants et la diversification des moyens de subsistance dans une région en Bangladesh en retard sur le plan économique. Avant 1998, la partie nord-ouest du Bangladesh, qui abritait 22% des quelque 160 millions de personnes, était dissociée de la partie orientale économiquement avancée du Bangladesh. Le 23 juin 1998, le Pont Multifonctionnel Jamuna (JMB) sur la rivière Jamuna a été ouvert. Le pont a établi des systèmes de transport routier et ferroviaire directs entre le nord-ouest et l'est du Bangladesh. En utilisant ces phénomènes et en reliant les données HIES (HIES) et HIES2010 de l'Enquête sur les revenus et les dépenses des ménages (HIES), cette étude élucide les impacts de l'amélioration des infrastructures sur la répartition de la main-d'œuvre et le comportement des ménages échantillonnés en matière d'application des intrants. En appliquant l’approche d’estimation des écarts dans les différences (DID), cette étude démontre que le JMB qui a relié les régions du nord-ouest et de l’est a encouragé la migration de main-d’œuvre, ce qui a permis aux travailleurs les plus vulnérables des secteurs agricole et non agricole de s’orienter à d'autres emplois avec des rendements marginaux relativement élevés. L'amélioration de l'infrastructure de transport avait également augmenté les taux de salaire journaliers qui étaient inférieurs à ceux du groupe de comparaison et l'intensité du système de culture des ménages des divisions de Rajshahi et de Rangpur. L’étude conclut donc que l’investissement dans les infrastructures peut jouer un rôle déterminant dans le décollage de la croissance économique et la réduction de la pauvreté dans les régions économiquement moins développées dans les pays en développement.

Appendix

See Table 9.

The first segment of Table 9, in which we considered only the rural respondents (n = 5194), shows that, in general in 2010, the probability that a sampled respondent would be self-employed in the farm sector and a day-laborer in both the farm and non-farm sectors was higher, where the base occupation category is the self-employed operator in the non-farm sector. It shows that, compared to the control group, in general, the likelihoods that respondents of the Rajshahi and Rangpur Divisions were self-employed in the farm sector and day-laborers in the farm and the non-farm sectors are positive and significant. In the Rajshahi Division, the respondents, in general, were less likely to work as salaried workers compared to the control group. In 2010, however, the respondents in the Rajshahi and Rangpur Divisions were less likely to work as self-employed in the farm sector (p < 0.00) and a day-laborer in the farm and the non-farm sectors (p < 0.00) compared to the control group. The findings lend support to our conjecture that the improved infrastructure that established a direct route to the eastern region encouraged labor out-migration and reallocation from low-paid day-labor jobs to other jobs. It demonstrates the importance of the improved infrastructure on livelihood diversification. Interestingly, in 2010, the likelihood of working as a salaried worker in the Rajshahi Division compared to the control group was not statistically significantly different. This indicates that the proportion of the salaried workers in the Rajshahi Division increased in 2010 compared to 2000 than the control group

In the second second segment of Table 9, we have considered all respondents and added multiplicative dummies for rural respondents consisting of multiplications of the year 2010 dummy, the Rajshahi (RAJD) and Rangpur (RANGD) dummy and the rural dummy (RD). The results show that in both the Rajshahi and Rangpur Divisions, the likelihood of being self-employed in the farm was lower in 2010 compared to the control group, but the difference was not significant in the rural areas. This indicates that the importance of the farming sector in the rural areas both in the control and experiment regions is the same. Importantly, the likelihood of working as a day-laborer in both the farm and non-farm sectors in the rural areas of the Rajshahi Division, and the likelihood of working as a day-laborer in the farm sector in the rural area of the Rangpur Division has been reduced significantly by 2010, compared to the control group.