Development of detailed transport connectivity datasets1. Using detailed road network data, three spatial variables are constructed to represent different types of transport connectivity. The primary data source is georeferenced road condition data provided by the National Roads Administration (ANE) in Mozambique. The dataset covers a total of 29,800 km of classified roads. For each road link, we can know its physical alignment, length, road class (primary, secondary, tertiary, vicinal or unclassified), surface type (paved or unpaved), and road condition (good, fair, poor or very poor) (Figure
A.1).The dataset is available at two points of time: 2010 and 2015.2. Rural Access Index. The Rural Access Index (RAI) is defined by the proportion of people who have access to an all-season road within an approximate walking distance of 2 km or a walking time of 25 minutes. Based on World Bank (2016), paved primary roads are considered to be passable for all seasons. Other paved roads are considered to be passable if they are in good or fair condition. For unpaved roads, a more stringent threshold is used. If unpaved roads have all-season connectivity if they are in good condition. By overlapping the highly granular population distribution data developed by WorldPop, the share of people who live within 2 km of an all-season road is calculated for each administrative post (postos).3. Port access. This aims at measuring transport accessibility to the global market. There are three major ports in Mozambique: Maputo, Beira and Nacala. To this end, the road condition data are converted to average road user costs in Africa, using standard conversion factors (Figure A.2) that are available in road engineering models, such as the Highway Development and Management Model (HDM4). The estimated road user costs, which mainly comprise fuel costs, vehicle maintenance costs and time costs of drivers and passengers, vary depending on road class, surface type and condition. In general, if a road is better condition, the user cost tends to be lower because the surface is smooth and driving speed can be high