Sector Results and Learning:
Land

MCC’s rural development projects are often built around interdependent multisector investments; however, in practice, separate implementation of these projects often occurs as they move at different speeds with distinct stakeholders. MCC should remove dependencies except when absolutely needed or consider how to structure these investments to ensure alignment. In Ghana, the Agriculture Project expected that complementary investments in farmer training, land access, access to credit, and rural roads would support agricultural productivity and farmer access to markets. However, in practice, the implementation of roads, agriculture and land programs were distinct and occurred in different locations. The agriculture and Land Tenue Facilitation (LTF) efforts were not advancing at the same rate, and the LTF selected a district in a peri-urban rather than a rural agricultural area. This led LTF to spin off from the Agriculture Project and act as a standalone project. In the end LTF did not lead to the project goal of improving agricultural productivity though it had standalone benefits of increasing off-farm incomes.

Similar occurrences have happened in other land interventions. In MCC’s Participatory Land Use Planning under the Green Prosperity (GP) Project in Indonesia, the land activities moved forward at a quicker pace than the rest of GP and areas selected were distinct from those of GP investments. At other times, the interventions have occurred in tandem. In Senegal, the land and irrigated infrastructure activities implemented in tandem in the same irrigated perimeter.

The key is understanding the critical issues, including whether an interdependency in fact exists or whether there are really separate but related interventions that should stand on their own. The best strategy may be to reduce interdependencies where possible and structure these as standalone projects with their own evaluation assessment and economic rate of return. In cases where interventions are dependent on one another and key to obtaining results, MCC should consider how to best structure these multisector investment to maintain alignment of all project activities.

The significant investment by the Compact to ensure the achievement of the Participatory Land Use Planning Activity (PLUP) targets may result in interventions that are too expensive for the government to take to scale. MCA-I and MCC are considering ways to improve the efficiency of PLUP activities, based on the differing implementation models employed by the various PLUP implementers. The experience so far indicates that targeting local, rather than international, firms to manage PLUP implementation would have been a more cost-effective option.

Qualitative data can help clarify and provide insights on unexpected findings. In MCC’s early evaluations, such as the evaluation of the Peri-Urban Rangeland Project (PURP), MCC employed largely household surveys with little qualitative analysis. As such, there was no way to obtain an understanding of some unexpected findings. More recently, MCC has used mixed method data collection methodologies, including impact and performance evaluation elements with triangulation of data from qualitative, quantitative and administration data sources. Allowing time for analysis of the household data sets to inform the qualitative instruments and add perspective on some of the results finding in household surveys could be beneficial as well.

Severe weather can create difficulties in evaluating aspects such as mortality rates and control of animal numbers. In Mongolia, the year of the baseline data collection (Phase 1-late 2010 and Phase 2-early 2012) followed a dzud (2009-2010). This left lower animal numbers and stocking rates on the land than there would have otherwise been and higher than normal mortality rates when measuring animal deaths over the few years prior to the baseline. Although control herders faced the same severe weather issues, it still was difficult to interpret some of the data findings and related causes. For example, animal numbers increased in all areas for both treatment and control herders between baseline and endline, but animal stocking rates of program herder groups remained largely within the carrying capacity of the land that was determined by the PURP implementers in 2009 during parcel mapping activities.

Capturing tenure status requires triangulation of data between administrative, survey, and project data. Due to outdated official records and lack of awareness by households on differences in land rights documentation, these different data sets often provide conflicting information on a parcel’s tenure status. Tenure status drives key outcomes like perceptions of tenure security and conflict. When administrative data and survey data do not show the same tenure status, land program implementation data can also provide missing tenure information, including what may be driving the inaccuracies between statutory and perceived tenure status in program areas. An early review of land office databases, along with sharing of project documentation from collections of land records in the field for program areas, can help clarify tenure status and which tenure status is best to use in the analysis. Perception of tenure status drives tenure security and related outcomes while land administration data provides official figures and can provide key information on land transactions rates over time.

The administrative data and self-reported tenure status often do not match due to gaps in administrative data/outdated or non-matching records combined with misunderstanding of tenure documentation by households. These differences in tenure status can lead to variances in interpretation of findings. In Mongolia Special Hashaa Plot Survey, the evaluator used the General Authority of State Registration (GASR) official tenure status as the main data source and noted the large discrepancies between perceived tenure status and official tenure status. GASR administrative records and self-reporting of registration only matched with 73% of Mongolia survey respondents. This could have been due to respondents confusing the certificate of governor’s decision with a registered title or outdated GASR files due to someone previously holding a title and transferring it informally to a new owner. Those who perceived they had title had higher land values. By comparison, in Ghana the evaluator used the survey data to determine tenure status. Later MCC evaluations try to triangulate tenure data where possible to understand outcomes. In future evaluations, it will be key to coordinate with the land project implementation team to obtain and use both land administrative data and land implementation data in addition to survey data.

When establishing a comparison group in land, care should be taken to avoid splitting communities which are under the same land administration unit. During Land Tenure Facilitation, a road was used to form program and comparison groups. The road split left the community split in half. Although the split was good from a statistical point of view as the households and land were comparable, it created problems at a practical level as these households were all under the same land administrative unit. Beyond the practical issues, when looking at effects and longer-term sustainability of land governance systems, it is helpful to have your whole administrative area mapped and registered. Two divergent systems of land rights are difficult to manage and sustain.