Saturday, August 27, 2011

Impact of infrastructure spending on household welfare in rural Nepal

Did expenditure on access to rural roads irrigation infrastructure, and extension services have significant impact on household welfare over the period of Ninth Five-Year Plan (1997-2002) in Nepal?

A new research monograph by IFPRI shows that the effect of rural roads is robust across two different econometric strategies, while the effect of irrigation and extension services on household welfare is less robust. Access to rural roads improved households’ welfare as measured by land values, consumption growth, poverty reduction, and agricultural income growth. The research also shows statistically significant impacts of irrigation using a hedonic model, while an alternative panel data approach did not yield significant  estimates of the impact of access to irrigation or extension services. What explains this inconsistency? The authors point their fingers to measurement inaccuracies of the irrigation variable at the household level, which is aggregated from the plot level, rather than to true ineffectiveness.

The authors (Dillon, Sharma and Zhang) argue agriculture growth expectations was below the target set in the Ninth Five-Year Plan and large gains in poverty reduction have been largely driven by rural-to-urban migration and remittances. This is also somewhat corroborated by the latest findings of NLSS III and migration survey.

They argue that growth in rural areas can be attained by harnessing its comparative advantages, particularly those hinged on agro-ecological environment. Realizing this potential would require increasing connectivity of rural areas with roads, integrating farmers with markets and increasing their productive capacity by improving access to irrigation.


In our hedonic estimates of the effect of extension on land values in 1995/96, we find that access to extension had a positive yet insignificant effect, while our 2003/04 estimates suggest a larger, statistically significant effect. However, in the panel household analysis, we find that access to extension in 1995/96 did not have a significant impact on growth in household welfare. Due to civil strife, the initial frequency of extension visits might not have been related to agricultural productivity and other welfare indicators seven years later. The inconsistency of the estimated impact of extension service between the two methods calls for more in-depth research in the future.