Poverty and access to roads in Papua New Guinea

In this article, our overall goal is to understand how effective access to infrastructure is in reducing poverty in PNG. To meet this goal, we pursue three objectives. First, we examine poverty in PNG and demonstrate the relationship between poverty and access to infrastructure. Next, we identify the determinants of poverty, most importantly testing whether or not access to infrastructure, ceteris paribus, is a significant factor in a household's poverty status. Finally, we want to understand what infrastructure-related policies will be effective in overcoming poverty in PNG. To narrow the scope of our analysis, we focus on access to roads in rural areas for two reasons. First, as we will show, most of the poor in PNG live and work in rural areas, a characteristic common to most Asian countries. Second, the main infrastructure problems in PNG are in the rural areas. Access to services, markets, and transportation, measured in travel time, is much better in cities. Finally, we believe access to roads and the distance between the point at which villagers can gain access to the road and the nearest government station (which is usually the center of economic activity in many districts) can best proxy for the level of infrastructure. In our analysis, however, we allow for the possibility that access to roads is endogenous and correct for the potential endogeneity by using an instrumental variable approach to estimating the model. To meet our goals and objectives, the rest of the article is organized as follows. In Section II, we first describe our study's data set and explain how we created our measures of poverty. In the following section, we examine the contours of poverty and its relationship to the access that PNG residents have to infrastructure. Section IV creates a model of the determinants of poverty in PNG and presents results about the effect that access to infrastructure has on the poor. We simulate the impact of various investment strategies on poverty. Our modeling and simulation sections draw heavily on the work of G. Datt and D. Jolliffe. The final section concludes. © 2003 by The University of Chicago. All rights reserved.