Incentives for Joint Land Titling: Experimental Evidence from Uganda

    1. Study Information

       

      This section discusses the paper presented at the 2018 World Bank Conference On Land And Poverty “Incentives for Joint Land Titling: Experimental Evidence from Uganda,” by Ludovica Cherchi, Markus Goldstein, James Habyarimana, Joao Montalvao, Michael O’Sullivan, and Chris Udry.1

      The experiment was implemented by Associates Research Trust (ART), in collaboration with the Government of Uganda’s Ministry of Lands, Housing, and Urban Development. ART is an applied research and implementation organization in Uganda with extensive experience in land rights and gender. The intervention started in 2017, targeting about 1,090 households from 253 villages across four districts (Mbarara, Sheema, Buhweju, and Isingiro) in the Western Region of Uganda. Within each village, an average of 4 eligible households were randomly selected for the intervention. Those households with a married (or cohabitating) couple owning at least one unregistered parcel of land were deemed eligible for the intervention.

    2. Questions posed
      • What is the effectiveness of price subsidies for registering women on titles and information about registration and the value of registering women, in isolation or jointly, in improving women’s access to formal land?
    3. Description of intervention

      The core intervention offered fully-subsidized land titles to rural households in Uganda. It entailed four door-to-door household visits. During the first visit, households were provided with information about costs and benefits of titling, and offered the opportunity to receive a fully-subsidized title for one parcel of land (randomly selected for households with multiple parcels). At the end of this visit, households were asked whether they accept the offer, and if so which names they want to be listed on the title. During the second visit, parcel boundaries were defined for households that accepted the offer in the presence of neighbors and local government officials, and the households were assisted with filling out the land title application forms. On this visit, households had the option to revise their original decisions made during the first visit. During the third visit, the core land demarcation and surveying activities took place. During the fourth visit, the freehold land titles were delivered to the households after being processed by the Government of Uganda.

      The goal of the experiment was to assess the effectiveness of conditional price subsidies and information, in isolation or jointly, in improving women’s access to formal land. To do so, the 253 study villages were randomized into the following two treatments, fully crossed with each other and stratified by parish.

      Conditional Subsidies vs. Unconditional subsidies

      All households were offered a fully-subsidized title for an eligible parcel of land. They varied the conditionality of the subsidy as follows: (i) half of the households received the subsidy conditional on registering the wife as a co-owner of the land; and (ii) the other half received the subsidy unconditionally. This allowed them to isolate the impact of the gender conditionality over and above the impact of the subsidy.

      Gender Information vs. General Information

      All households were shown a short educational video clip. They varied the content of the video clip as follows:  (i) half of the households were shown general information about titling (such as benefits, legal implications, prices); and (ii) the other half were additionally shown information on the benefits of female co-titling.

    4. Context of findings

      The study took place in Uganda where there is an ongoing land titling intervention. The titling and registration was focused on freehold land held by households and not community held land.

    5. Key findings

      The main findings were as follows:

      • The conditionality and the information did not affect overall household demand for titling, but they did substitute each other in increasing demand for co-titling.
      • Imposing the conditionality significantly raised the co-titling probability by 31 percentage points among the gender uninformed households, and by 14 percentage points among the informed households.
      • Providing information significantly raised the co-titling probability by 16 percentage points among households offered titles unconditionally, and had no impact among the households offered titles conditionally.
      • The intervention, which offered fully-subsidized land titles, successfully generated high overall demand for titling, as well as for co-titling.
      • Imposing the gender conditionality on the subsidy in isolation further raised demand for co-titling, without dampening overall demand for titling.
      • Providing additional gender information in isolation also further raises demand for co-titling, though not as much as the conditionality, and has no impact on demand for titling.
      • There was a reduction in overall demand for titling between visits.
    6. Unanswered questions
      • Why was there a reduction in overall demand for titling as time went on? Did the co-titling promotion have any impact on this?
    NOTES
    1. Cherchi, Ludovica; Goldstein, Markus; Habyarimana, James; Montalvao, Joao; O’Sullivan, Michael; and Udry, Chris, “Incentives for Joint Land Titling: Experimental Evidence from Uganda,” 2018 World Bank Conference On Land And Poverty, The World Bank - Washington DC, March 19-23, 2018.