Beliefs About Institutions and Investments in State Capacity

In this study, all participants answer three survey questions about key institutional indicators in Nepal: the proportion of Janajati political leaders, the representation of female deputy mayors/chairpersons, and Nepal’s current political stability score. Participants assigned to one of three treatment groups receive immediate corrective feedback—i.e., the correct answer—on one of these questions, while the control group receives no corrective feedback. The specific question that receives feedback varies by treatment group:

T1 (Janjati Treatment): Correct information on the Janajati question.
T2 (Gender Treatment): Correct information on the Gender question.
T3 (Stability Treatment): Correct information on the Stability question.

All respondents answer three belief-elicitation (or “prior”) questions:
1. Janjati Question: “What do you think is the proportion of Mayors/Chairpersons across Nepal who are Janajatis?”
2. Gender Question: “What do you think is the proportion/number of Deputy Mayors/Vice-Chairpersons across Nepal who are female after the 2022 election?”
3. Stability Question: “The World Bank ranks all countries based on political stability, with scores ranging from 0 (extremely unstable) to 100 (highly stable). For instance, Canada and Norway score above 95, whereas countries experiencing active conflict score below 5. In 2005, during the Maoist conflict, Nepal scored 4 points. On a scale of 0-100, what do you think Nepal scores today?”

Treatment Conditions:
Respondents are randomly assigned to one of four arms, stratified by province and position:
1. Control Group: Answers all three questions without receiving any corrective feedback.
2. T1 (Janjati Treatment): Receives the correct answer immediately after responding to the Janjati Question.
3. T2 (Gender Treatment): Receives the correct answer immediately after responding to the Gender Question.
4. T3 (Stability Treatment): Receives the correct answer immediately after responding to the Stability Question.

2024-12-16

2025-02-10

We are primarily interested in the effect of the intervention on (1) posterior beliefs about electoral inclusivity, institutional protections of rights, and risks of political instability or conflict in the future for Nepal (2) willingness and eventual participation in our state-building task.

Posterior Beliefs
This study examines the impact of an information treatment on politicians’ and bureaucrats’ beliefs about institutional quality and stability in Nepal. We elicit posterior beliefs regarding future institutions after the information treatment.
Beliefs are measured using widely recognized institutional quality indicators (e.g., Freedom in the World, Polity Score, Worldwide Governance Indicators). Responses are collected for present and future time horizons (e.g., 5 and 10 years).

We will examine individual indicators while addressing multiple hypothesis concerns through False Discovery Rate (FDR) corrections. In addition, we will construct summary indices by aggregating standardized outcomes. Each index will be calculated as an equally weighted average of z-scores, with sign adjustments for consistency. We will also explore alternative aggregation methods, such as inverse covariance-weighted indices.

Civil Registry Task
We assess state-building engagement through participation in a Civil Registration and Vital Statistics (CRVS) task, aligned with an ongoing national initiative. Respondents were asked to compile local registration data, with engagement measured across two dimensions:
1. Willingness to engage in state-building task: Self-reported agreement to complete the task.
2. Actual engagement in state-building task: Completion of the task, confirmed via a follow-up survey two weeks later.

This study examines how information about Nepal’s institutions influence the posterior beliefs and engagement of local politicians and bureaucrats. We conduct a randomized information provision experiment with 5,267 local representatives across municipalities and wards in Nepal.

Participants are randomly assigned to one of three treatment groups or a control group. Each treatment group receives corrective information about one of three institutional topics—ethnic representation, gender representation, or political stability—after stating their prior belief. The control group answers the same questions but does not receive any information. Randomization is stratified by province and position, and we test for experimenter demand effects and order effects to ensure robustness.

This study examines how providing factual information about Nepal’s institutions affects the beliefs and engagement of local politicians and bureaucrats. The experiment involves a randomized information treatment conducted across 5,267 local representatives (2,259 municipal-level and 3,008 ward-level) in Nepal.

Participants are randomly assigned to one of three treatment arms or a control group. Each treatment group receives corrective information about a specific institutional feature after stating their prior belief:

Janjati Treatment (T1): Proportion of mayors/chairpersons who are Janajati.
Gender Treatment (T2): Proportion of deputy mayors/vice-chairpersons who are female.
Stability Treatment (T3): Nepal’s political stability score on a 0–100 scale.
The control group answers the same belief-elicitation questions but does not receive corrective information.

Additionally, we test for:

1. Experimenter demand effects by randomly "priming" half of respondents with explicit expectations about the study’s purpose.
2. Order effects by varying the sequence of belief-elicitation questions in the control group to match the different treatment arms.

Randomization is stratified by province and position, ensuring balance across key characteristics.

Randomization was done using the randtreat function in Stata before administering the survey.

Individual

No

NA

5271

1. Control - 1,319
2. Janjati Treatment - 1,320
3. Gender Treatment - 1,323
4. Stability Treatment - 1,305