Personal Experiences and Political Views among Nepalese Politicians

Last registered on May 11, 2021

Pre-Trial

Trial Information

General Information

Title
Personal Experiences and Political Views among Nepalese Politicians
RCT ID
AEARCTR-0007654
Initial registration date
May 10, 2021

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 11, 2021, 11:47 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Bocconi University

Other Primary Investigator(s)

Additional Trial Information

Status
On going
Start date
2021-03-01
End date
2022-12-31
Secondary IDs
Abstract
We recruit local politicians in Nepal for a phone survey to test whether their personal experiences, and especially participation in the Nepal People’s War, affect their political views and policy priorities. First, participants are randomized into three groups: i) participants in the control group are not asked to reflect on the link between their personal experiences and policy priorities; ii) participants in the “personal experiences prime” treatment group are asked to reflect on how their personal experiences affect their policy priorities, without any specific mention of the civil war; and finally, iii) participants in the “civil war prime” treatment group are asked to reflect on their personal experiences during the civil war and how these affect their policy priorities. We then measure participants’ policy goals, preferences for redistribution, opinions over targeting strategies related to social protection (avoiding exclusion or inclusion errors, and use of in-kind versus cash transfers), support for federalism, and opinions about political inclusion, and test whether their political views and policy priorities are affected by the control recollection of past personal experiences.
External Link(s)

Registration Citation

Citation
Fiorin, Stefano. 2021. "Personal Experiences and Political Views among Nepalese Politicians." AEA RCT Registry. May 11. https://doi.org/10.1257/rct.7654-1.0
Experimental Details

Interventions

Intervention(s)
Intervention Start Date
2021-03-30
Intervention End Date
2021-05-15

Primary Outcomes

Primary Outcomes (end points)
The main objective of our experimental design is to measure the effect of the controlled recollection of personal experiences, especially during the civil war, on local politicians’:
focus on pro-poor policies (with respect to their own policy goals and the policy goals of donors);
priorities over eliminating errors of inclusion or exclusion from public programs;
preferences for cash versus in-kind social protection payments;
preferences for redistribution;
support for federalism;
beliefs about the current inclusion of marginalized groups into policy decisions, and about which excluded groups need to be included more in policy decisions.
Primary Outcomes (explanation)
Own policy goals are measured using the question:
“Please list the three most important policy goals that you have for your municipality.”
Preferences over the policy goals of donors are measured using these questions:
“Would you recommend to donors that they should focus on supporting these three policy goals? Or, should they focus on other policy objectives?”
“What policy objectives, then, do you think they should focus on?”
We will then categorize the extent to which politicians’ responses are “pro-poor” and are “pro-marginalized”. The Maoists started the People’s War with the explicit goal of empowering historically marginalized caste and ethnic groups, such as the Dalits and especially the Janajatis. Therefore, we are interested both in policies that specifically benefit the poor and that benefit these marginalized groups. We use the Nepal National Governance Survey (NNGS), conducted in 2017-18, which measured citizens’ policy priorities for the new federal government. Specifically, we will use responses to the question: “In your opinion, what should be the three main priorities of provincial and local governments for the next five years? (record three priorities)”. We will focus on policies that are favored by Janajatis and Dalits. We will account for spatial variation in citizens’ policy priorities. We will also have our colleagues at the Nepal Administrative Staff Collect (NASC) confirm these priorities.

Priorities over eliminating errors of inclusion or exclusion from public programs are measured on a 1-4 scale using the following question:
“Would you value more a program that focused on (a) methods to identify uncovered beneficiaries or (b) methods to identify beneficiaries who do not need assistance?”


Preferences for cash versus in-kind social protection payments are measured through an index which aggregates the responses to these two questions, both measured on a 1-5 scale:
“I would like to know if you strongly agree, or agree, with the first or second (or neither). Sometimes social protection payments are given as cash, sometimes they are given in-kind, for example as food). Statement 1: Assistance should be given as cash so that people can decide what they need on their own. Statement 2: Assistance should be given in kind so that scarce social protection resources are not wasted”
“When people receive support from the government in cash, they will use it for things they do not really need. Do you strongly agree, agree, neither agree nor disagree, disagree, or strongly disagree with this statement?”

Preferences for redistribution are measured through an index which aggregates the responses to these two questions, both measured on a 1-5 scale:
“It is a problem if children from poor and rich backgrounds have unequal opportunities in life. Do you strongly agree, agree, neither agree nor disagree, disagree, or strongly disagree with this statement?”
“Do you support more policies to increase the opportunities for children born in poor families and to foster more equality of opportunity, such as education policies? Naturally, to finance an expansion of policies promoting equal opportunity, it would have to be the case that either other policies are scaled down or taxes are raised.”

Support for federalism is measured on a 1-5 scale using the following question:
“The First Directive Principle of the 2015 constitution states: 'It shall be the political objective of the State to strengthen a federal democratic republican system to ensure an atmosphere where democratic rights are exercised by acknowledging sovereignty, independence and integrity of the country to be of utmost importance; by protecting freedom, equality, property and all citizens through rule of law; by embracing the norms and values of fundamental rights and human rights, gender equality, proportional inclusion, participation and social justice'. The federal system has now been in place for four years. Do you strongly agree, agree, neither agree nor disagree, disagree, or strongly disagree that Federalism has achieved these goals?”

Beliefs about the current inclusion of marginalized groups into policy decisions are measured on a 1-3 scale using the following question:
“I would like to know with which of the following statements you agree the most.
Statement 1: In modern day Nepal, historically excluded groups do not have their voices heard when it comes to policy decisions.
Statement 2: In modern day Nepal, historically excluded groups do have their voices heard when it comes to policy decisions, but their preferences are not implemented.
Statement 3: In modern day Nepal, historically excluded groups do have their voices heard when it comes to policy decisions, and their preferences are implemented.”
Beliefs about excluded groups who need to be included in policy decisions are measured using the question:
“When you think of inclusion, along which of the following lines do you think your municipality needs the most attention to include the excluded groups?”

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We ask local politicians and bureaucrats from Nepal to participate in a phone survey. Participants are recruited from a sample of 746 municipalities: in each municipality, we attempt to interview the mayor, the deputy mayor, and the chief administrative officer.

After obtaining participation consent, we ask all respondents a few demographic questions and a set of questions designed to measure their political ideology.

Afterwards, we randomize the controlled recollection of personal experiences during the civil war by randomizing participants into three groups. The first group, comprising a pure control, receives no prime at all. The second group (group 2, or “personal experiences prime” treatment group) receives the following prime: “Over your lifetime, you saw Nepal going through a lot of change. Can you please take a moment to share with us how those experiences shape your policy priorities now that you are a policy maker?”. The third group (group 3, or “personal experiences prime” treatment group) receives the prime: “Over your lifetime, you saw Nepal going through a lot of change. Please reflect on your personal experiences during the civil war. Can you please take a moment to share with us how those experiences shape your policy priorities now that you are a policy maker?” We developed these primes in close consultation with staff at the National Administrative Staff College and through informal discussions with politicians. Discussing experiences during the war, especially for public officials, is common and does not typically constitute a sensitive subject.

Before proceeding with the measurement of our outcomes of interest, all participants are truthfully informed about international donors’ involvement in the research project. We developed our questionnaire in close collaboration with policy partners, including the Foreign and Commonwealth Development Office (FCDO), and made sure they are comfortable with signing up to using the information to align their programming with mayors’ policy priorities. Specifically, we use the language: “In developing some of the following questions, we consulted with international donors who are looking to align their development assistance more closely with your needs. We will communicate the anonymized results of the survey to them and other development actors. Therefore, when responding, we ask that you take into consideration the fact that your responses might have some effect on donor priorities. Of course, all results will be communicated to donors only in anonymous and aggregate form, so your identity and your personal details will remain confidential and will not be linked to your responses.”
The purpose of emphasizing this to respondents is that it alleviates potential concerns of experimenters’ demand effect by increasing the real stakes of participants’ answers.

Next, participants are asked to express their opinion on: i) their own policy goals and their preferences over the policy goals of donors, ii) their priorities over eliminating errors of inclusion or exclusion from public programs, iii) their preferences for cash versus in-kind social protection payments; iv) their preferences for redistribution; v) their support for federalism; and vi) the inclusion of marginalized groups into policy decisions.

Next, all participants are asked a series of questions related to the COVID19 pandemic, that are not directly related to the project described here. Finally, all participants are also asked questions about their political party affiliation and plans for the next election.

Nepal’s people’s war, fundamentally, reflected a social movement to remove a caste-based monarchy and to allow groups who were historically denied political voice greater political representation. The fundamental hypothesis that we would like to test, therefore, is whether and the extent to which the struggle for greater rights during a social movement influences the policy priorities of politicians after they gain power and authority. In particularly, we are interested in whether those who fought for greater rights for historically excluded groups also, then, pursue more pro-poor policies that they have authority over.

A comparison of groups 1 and 2 allows us to test whether the recollection of any type of personal experience during Nepal’s transformation from a monarchy to a federal republic (which spanned the period 1990 - 2017) shapes policy priorities. A comparison of groups 1 and 3 allows a test of whether experiences specifically related to the civil war (which spanned the period 1996 - 2006) shapes these priorities. Finally, a comparison of groups 2 and 3 permits a test of how specific experiences during the civil war impact policy priorities relative to more general changes over the period.
Experimental Design Details
Because we recruit three participants from each municipality (mayor, deputy mayor, and CAO), and because it is not possible to schedule the survey at the same time for all three participants (to ensure high response rate, we need to offer time flexibility to participants), it is possible that participants talk with each other about the content of the survey. We plan to use naturally occurring random variation in the proportion of participants in a given municipality who are treated with the primes in order to estimate possible spillover effects, rather than implementing a randomized saturation design. Indeed, the saturation design would be impractical given our stratification because it would imply too many strata-treatment cells with too few observations in them.
Randomization Method
The randomization is done through Stata.
Randomization Unit
The unit of randomization is the individual participant.

In assigning treatment, we plan to stratify on the following variables. First, position (mayor, deputy mayor, and CAO) to ensure power to test whether experiences are different in different subgroups. Because most mayors are men and most deputy mayors are women, this creates de facto stratification on gender. Second, region (plains/Terai, hill, Himalaya). Politics is highly regional in Nepal, with the Terai having its own struggles, and so we want to stratify by the region of the municipality to guarantee power for subgroup analysis by region. Finally, we stratify mayors and deputy mayors by party (Maoist, United Marxist Leninist, Nepali Congress, and other). We predict that treatment will be localized to respondents who participated directly in the struggle for greater rights, and this group is concentrated among the Maoists. CAOs are bureaucratic, not elected positions, and so we do not stratify them by party.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
The treatment is not clustered.
Sample size: planned number of observations
We will attempt to contact a total of 2224 participants.
Sample size (or number of clusters) by treatment arms
We will attempt to contact 740 participants in group 1, 742 in group 2, and 742 in group 3.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Considering a power of 80% and a significance level of 5%, the minimum detectable size when comparing any two treatment groups on the full sample is of about 0.15 standard deviations.
IRB

Institutional Review Boards (IRBs)

IRB Name
Yale IRB
IRB Approval Date
2021-03-25
IRB Approval Number
2000025903

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials