Providing Information and Application Assistance to Improve Social Pension Enrollment in India

Sarika Gupta
Fieldwork by:
Delhi, India
1,257 women
Línea de tiempo:
2015 - 2017
Governance Initiative
Target group:
  • Urban population
  • Women and girls
Outcome of interest:
  • Empowerment
  • Social service delivery
Intervention type:
  • Information
AEA RCT registration number:


Governments often support disadvantaged citizens with pensions, commonly in the form of monthly cash transfers. The Delhi government provides pensions to widows and other disadvantaged women, yet the majority of eligible women are not enrolled in this scheme. The researcher conducted a randomized evaluation to measure the impact of providing eligible women with program information and various levels of application assistance for the Delhi Widow Pension Scheme on program knowledge, application completion, and enrollment. Offering more involved application assistance increased average application rates, particularly among more vulnerable women. Enrollment in the pension program also increased across the levels of application assistance that were offered.

Policy issue

Governments often support disadvantaged citizens with pensions, commonly in the form of monthly cash transfers. Despite the availability of these pensions, many eligible people do not enroll into pension programs. It is possible that these social programs unintentionally exclude certain groups by requiring citizens to navigate bureaucratic processes and prove their eligibility to enroll. Eligible citizens may lack information regarding the program or demonstrate confusion about the application process. One approach to improve access to pension programs is to provide eligible citizens with information and assistance with the application process.  Can information and assistance help particularly disadvantaged citizens overcome barriers to enrollment?

Context of the evaluation

Widows are one of the most vulnerable and disadvantaged groups in India’s social hierarchy. In order to support these women, the Indian government offers state-level pension schemes for widows and other disadvantaged women. One state-level program is the Delhi Widow Pension Scheme, which is available to widowed, divorced, separated, or abandoned women ages 18-59 with a household income below a given threshold (at age 60 they become eligible for old-age pensions). Eligible women receive bank transfers worth INR1,500 every month (about US$25), or 70 percent of monthly median per capita consumption. However, two-thirds of eligible women were not enrolled in the program in 2014. 

To enroll in the pension program, applicants need to obtain an application form online or in-person, fill out the application with personal and financial information, and collect multiple required documents including proof of marital status, personal identification, bank account information, and more. When the application and supporting documents are prepared, applicants have to make two office visits, first to a local politician’s office to obtain a signature on the form and then to a different district office to submit their completed application. 

The application process can be difficult given the complexity in obtaining required documents, the costs associated with visiting multiple offices, and widespread illiteracy among the eligible population. These factors may explain why less than 1 percent of surveyed women were able to complete the pension application themselves.

Many women participating in this study were considered vulnerable across economic, political, and personal empowerment dimensions: about 70 percent were illiterate, 50 percent had no personal income, and women on average had lost a monthly income of INR5,539 (about US$92) that used to come from their spouse. On the political dimension, many participants felt women could not approach government officials as easily as men and almost 20 percent of women had never been to a government office. Lastly, almost 50 percent of women lacked autonomy in the household, as measured by needing permission to do basic activities.

Woman sitting in her beauty parlor
Photo: Paula Bronstein/Getty Images/Images of Empowerment | CC BY-NC 4.0

Details of the intervention

The researcher conducted a randomized evaluation to measure the impact of providing information and various forms of assistance with the application process on enrollment in the Delhi Widow Pension Scheme. The study targeted 1,257 women from 75 slums in Delhi who were eligible for, but not yet enrolled in, the pension program and had all or almost all required documents to apply.  

Participants were randomly assigned to one of three intervention groups or a comparison group in late 2015:

  • Information: Women in this group were offered information regarding pension eligibility, benefits, and the application process in both verbal and written form (322 women).

  • Application assistance: Women were offered the information provided to the information intervention group as well as assistance with completing the application form. The application assistance was provided by field staff trained by the research team (308 women).

  • Combined assistance + accompaniment: Women were offered the same support as the assistance intervention group and were also offered accompaniment to their local politician’s office by a trained field staff member to have the application form signed. Accompaniment was not offered for the final step of submitting the application and supporting documents to the district office (335 women).

  • Comparison group: Women were not offered information or assistance (292 women)

Despite efforts to only include women not already receiving the widow’s pension in the study, 20 percent of intervention group participants were enrolled in the pension program by the time initial surveys and interventions began, making it more difficult to detect significant intervention impacts.

From fall 2015 to fall 2016, participants were surveyed once immediately before the intervention and three times throughout the year after the intervention. Before the intervention, women were surveyed about their personal, political, financial and family background. After the intervention, women were asked about their application experience, including what steps they completed and if they were ultimately accepted into the pension program. As an additional method to verify enrollment, the researcher collected administrative data on the official list of program beneficiaries from the Delhi government’s website one year after the intervention.

Results and policy lessons

Both the application assistance and the assistance plus accompaniment interventions substantially increased application completion, with notable gains for the most vulnerable women. However, providing information on the pension program with no additional assistance only improved the application rates of literate women. Enrollment in the pension program increased at a similar rate across all three interventions.

Pension program knowledge: Offering information on the pension program increased women’s knowledge of the pension program. For example, women who were only offered information and no further assistance were 6 percentage points more likely to know they needed a politician’s signature on their application than those in the comparison group (an 8 percent increase from a base of 73 percent). Offering application assistance or accompaniment in addition to information did not lead to higher increases in pension program knowledge than receiving information alone.

Application completion: Offering any intervention more than doubled the likelihood of women beginning the application process (a 104 percent increase relative to a base of 23 percent). However, only application assistance and assistance plus accompaniment increased the number of women completing the entire application process. Application assistance led to a 6.2 percentage point (41 percent) increase in completed applications, while application assistance plus accompaniment led to a 10.6 percentage point (70 percent) increase relative to the comparison group completion rate of 15 percent. 

Application Completion Varied by Individual Characteristics. While the interventions had important average impacts, some groups of women experienced substantial increases in their application rates.

  • Effects by literacy levels: For women who were offered information alone, application completion rates increased only for literate women, who completed the entire application process at a 14 percentage point higher rate than illiterate women. The application assistance or accompaniment interventions were also unable to improve completion rates for illiterate women despite ensuring they completed more steps in the process, suggesting illiterate women struggle during the final district submission step where no assistance was provided.
  • Effects by political connections: Less politically connected women who were offered application assistance increased their application completion rate by 15 percentage points (99 percent) relative to the comparison group, but politically connected women did not change their application completion rate. This suggests that political connections are useful for obtaining and filling application forms. The accompaniment option was no more beneficial to less-connected women than the application assistance alone, while information had the same effect regardless of political connections. 
  • Effects by household level empowerment: Women lacking autonomy in the household experienced larger increases in their application completion rates from any of the interventions than did more autonomous women. Application assistance with accompaniment to the politician’s office was most effective for women with low autonomy, raising application completion rates by 16.6 percentage points (110 percent), suggesting it is important for women lacking autonomy to receive support in leaving the house if they are to access government resources. 

Pension enrollment: Across the three interventions, the share of women enrolled into the pension program increased by an average of 6.2 percentage points (42 percent) compared to women who received neither information nor any form of assistance, who enrolled at a rate of 15 percent. However, there was no noticeable difference in enrollment rates between the information, application assistance, and assistance plus accompaniment groups, likely due to enrollment prior to the intervention reducing the power of the study to detect meaningful changes. Follow-up surveys suggest overall enrollment was constrained primarily due to problems with obtaining the appropriate supporting documents, an application step where no additional assistance was provided.  

While a lack of information can limit program enrollment, information alone may not be enough to increase enrollment without complementary efforts to ensure women can make use of the information. User-friendly, simple application processes with flexibility around required documents can facilitate greater and more inclusive program enrollment. Increasing outreach while removing barriers to apply is particularly important for including those with the greatest challenges in navigating the application process, such as illiterate women and those who lack autonomy. Further, strengthening support from the bureaucracy to assist eligible citizens in overcoming application barriers can improve enrollment while decreasing dependency on political and social networks.

Gupta, Sarika. "Perils of the Paperwork: The Impact of Information and Application Assistance on Welfare Program Take-Up in India." Working Paper, November 2017.