Road improvements in Kampala, Uganda can be challenging to implement due to cost of land as shown by this road used by motorists and pedestrians flanked by amenities which could benefit from upgrades. Photo by emretopdemir / iStock Editorial / Getty Images Plus via Getty Images.
Land costs and benefits of road improvements: Evidence from urban Uganda
Infrastructure is crucial for economic development; however, high land acquisition costs may threaten the implementation of the most beneficial projects. An analysis of novel survey data on ongoing road improvement efforts in Kampala reveals how road upgrades can boost citywide welfare, but need to account for barriers to raising and using funds.
Large additional investments in transportation infrastructure are needed to accommodate the rapid population growth of African cities. Only one-third of roads in African cities are paved, making road improvements a potentially high-return and scalable intervention to improve urban mobility. Yet, studying the impacts of road improvements in African cities is challenging.
Quantifying the multifaceted costs and benefits of road improvements
Modern economic models of transportation networks are rarely applied to developing country cities, partly because of the absence of quality data (notable exceptions include Tsivanidis (2023)). In addition, there are important factors in determining the valuation of road improvements beyond construction costs. For example, road improvements commonly require additional land that must be acquired from private landowners, inflating the total price of projects.
In a new IGC study on road improvements, I quantify the net returns of about 140 km of road improvements implemented between 2017 and 2024 in Kampala, Uganda. This study examines the trade-offs of the citywide benefits of traffic speed improvements with government-incurred land costs.
To do so, I collect new data – including 18 months of Google Maps data to measure transport efficiency improvements, a survey with 548 owners of land on the side of upgraded roads to understand private costs, and a survey with 377 real estate brokers to recover property values all over the city – and estimate a spatial general equilibrium model capturing citywide impacts of road improvements.
Road improvements had substantial local and citywide benefits
This study finds that road improvements in Kampala implemented during this period had both local and citywide impacts.
Local traffic speed increased by up to 5 km/h, or about 25% of the average travel speed in the city. Landowners on the side of upgraded roads reported improved road quality, decreased flooding and dust levels, and improved overall security (Figure 1). Due to the improvements in local amenities and connectedness, the value of properties on the side of the upgraded roads also increased by an average of 26%.
Figure 1: Change in road quality and local amenities following road upgrades
Notes: Landowners were asked to rank the quality of various road-level amenities before and after the road was upgraded on a scale of 1 (the worst road in Kampala) to 10 (the best road in Kampala); they reported an improvement in road quality in Kampala following road upgrades. Source: Figure generated by the author.
Road improvements also had citywide benefits through rerouting traffic patterns and relocating firms and residents. Through the lens of the model, I estimate that the implemented road improvements decreased average commuting time by 6.6% and increased total property values in the city by 1.36%. Importantly, the total net welfare gains at the city level were five times the local welfare gains, indicating far-reaching impacts of road improvements beyond property holders.
High land acquisition costs – but not all owners are fairly compensated
While road improvements had large benefits, they also faced considerable costs. The legal environment in Uganda entitles landowners to market-value compensation for land acquired for public projects, such as road improvements. This substantially increases the total costs faced by the government. I estimate that, if all land had been acquired at market value, land costs would have represented up to 45% of total costs.
However, the study uncovers that many owners – 80% of those surveyed – consented to give up an average of 73 square metres of land without claiming any financial compensation. As a result, land acquisition costs decreased by an estimated 72%.
Owners with well-established property rights are more likely to claim compensation
As shown in Figure 2, interviewed owners consented to forgo compensation not only because of private benefits, but also because of the prohibitively high costs of obtaining copies of administrative documents needed for compensation.
Figure 2: Owners’ stated reasons for not claiming compensation
Note: These reasons were given by respondents whose property was affected by road improvements and who consented not to claim the monetary compensation they were legally entitled to. Source: Figure generated by the author.
I further compare these reported private costs across Kampala’s three property right regimes (Figure 3). I find that leasehold owners report the lowest cost of getting a copy of these required administrative documents and are most likely to claim the compensation they are legally entitled to.
Figure 3: Property right regimes in Kampala
Notes: This map displays the three dominant land tenure systems in Kampala, with the parish administrative boundaries. Leasehold is the clearest property right regime, as local land boards maintain a registry of active leases. Property right clarity is weaker under freehold and mailo regimes. Source: Figure generated using Kampala Capital City Authority GIS data.
Why do high land payments threaten net welfare gains?
High cost of raising domestic funds
In practice, these compensations are often funded by raising property taxes and other fiscal costs on city residents, acting like a transfer to affected landowners. However, this transfer is not costless: research shows that for every dollar due in property taxes, the Kampala city government recovers only 39 cents, consistent with the high costs of raising domestic funds.
As a result, compensating owners at market value may cover their losses, but may also threaten the very implementation of the project. This study shows that compensating all owners at market value would have led to negative net welfare gains from road improvements in Kampala because of the high cost of raising funds.
Road improvements that would yield large benefits also yield large land costs
Road improvements at the centre of the city would yield some of the largest benefits. However, this is also where the land is most expensive and landowners have the most established property rights, which implies high effective land acquisition costs for the government. As a result, the government may avoid these areas, lowering the returns from the investment.
Land costs must be weighed against citywide benefits
Road improvements have the potential to generate large citywide benefits. However, like many other infrastructure projects, the underlying land must be acquired from private landowners. Failure to account for land acquisition costs during the design, funding, or evaluation stages of these projects may result in biased cost-benefit analyses and a failure to enact the most effective improvements.
Currently, road improvements in Kampala are mostly funded by the World Bank and the African Development Bank. However, these funds cannot be used for land acquisition, likely due to concerns about corruption.
Despite these concerns, this study shows that relaxing these restrictions could lead to welfare gains of more than 50% in the context of road improvements. While fewer roads may be improved as funds are redirected towards land acquisition, the fiscal loss from raising domestic taxes would be mitigated, making high-benefit improvements in high-land cost areas relatively more attractive.