The economics of greenfield urban planning

Where, and when, do we begin?

The 20,000 Plots project (20k) was implemented by the Tanzanian government from 2000 to 2005. It was an example of "de-novo" planning, where greenfield agricultural land on urban fringes is bought and converted into titled plots with roads and occasionally some basic services. People can then buy plots and build on them. The project ultimately delivered more than 36,000 plots in 12 areas in Dar es Salaam, the country's largest city.

And how well does de-novo planning work?

Urban planning shapes land use in developed country cities but in developing countries, and especially in Africa's large and growing cities (including Dar es Salaam), planning is often ineffective and instead informal settlements arise. A common approach to overcoming this problem is de-novo urban planning. But there is no economic framework for evaluating how planners' choices within de-novo areas, for example the size of plots, affect outcomes such as: the distribution of owners with different incomes between and within areas, the speed of neighbourhood development, land values and consumer welfare. Our study starts to fill this gap by providing a methodology for analysing how consumers value different planning choices.

Sounds like a lot of data - what was collected?

Data sources included: project maps showing plot sizes and planned uses. Satellite images of the project areas and the area around them which provided information on how much of a plot was built on. Interviews with local experts and with leaders of 34 local administrative areas asking what they thought determined the difference in price between 20k and non-20k areas. Questionnaires for local real estate agents, who provided sales dates and prices for individual plots in 20k areas and nearby non-20k areas that were sold in market transactions. Questionnaires for residents in over 3,200 plots within 20k areas covering issues such as their education, employment, the main building and its facilities, and the neighbourhood amenities. Information on how non-residential plots were being used (or not) was also collected.

Our evidence suggests that offering relatively more smaller plots would make de-novo projects more valuable and more inclusive

And, was the 20,000 plots project a success?

Earlier World Bank de-novo projects were halted in the 1980s due to criticism that costs were not recouped and the poor were excluded. We find the government's cost of about $33mn (2021 prices) was recouped, although the poor were still largely excluded. A plot of bare land of a given size in 20k areas is worth roughly twice as much as in nearby informal areas. There are two key reasons for this: the protection of owners' property rights and preservation of access through local unpaved roads which connect to main paved ones. A key limitation of 20k was the oversupply of large plots, which command lower land values and have lower population density. Our evidence suggests that offering relatively more smaller plots would make de-novo projects more valuable and more inclusive.

---

21 October 2025     Paper Number CEPCP720

Download PDF - Allocating land for housing in Tanzania

Download Press Release

This CentrePiece article is published under the centre's Urban programme.

This publication comes under the following theme: What determines urban growth and urban decline and what should be the role of policy?