Data sources, methodology and citation guidance

Pension Watch is a platform focused on social pensions that brings together data from multiple sources and presents it in a clear and accessible way. While the primary focus is on social pensions, these are situated within the broader social protection landscape to provide context on coverage, adequacy and system design.

The platform is based on a downloadable dataset that contains all data presented on Pension Watch. This dataset is accompanied by a detailed data dictionary, which includes variable definitions and metadata. This enables users to explore the data in greater depth and conduct their own analysis.

Pension Watch is designed for a range of users, including national-level organisations both within and beyond the HelpAge global network, government stakeholders, and researchers and academics. For these audiences, the platform provides credible and comparable data that can support advocacy for social pensions, inform the design and strengthening of national systems, and contribute to further research on the role of social pensions in ensuring income security across the life course.

This document outlines the methodology used to develop the Pension Watch platform and its underlying dataset. It presents the key variables used, explains the main calculations and assumptions applied, and presents the sources of data included in the dataset.

How to cite Pension Watch

Please cite Pension Watch when using data from the platform, country profiles, charts or downloadable dataset. Include the platform name, the relevant country, scheme or indicator where applicable, the dataset version or update date, the URL used, and the date accessed.
Suggested citation: HelpAge International. Pension Watch dataset. Version [month year]. Available at: [URL]. Accessed [day month year].

Future updates

Pension Watch was developed between September 2025 and March 2026. The platform and underlying dataset will be regularly updated to reflect new data releases and emerging research. Updates will include revisions to existing data where changes occur at the national level and the addition of new schemes where relevant. Where significant changes are made, these will be documented and shared.

Pension Watch is intended as a living resource, and feedback from users will help inform future updates and improvements.

Feedback and contact

If you notice any errors in the Pension Watch platform or dataset, or if you have comments, questions or suggestions, please contact: pensionwatch@helpage.org
Your feedback is valuable and helps us maintain and improve the accuracy and relevance of Pension Watch. Thank you for helping us make Pension Watch the best possible resource on social pensions.

Overview

Pension Watch presents data for 228 countries and territories, based on the UN M49 list with the addition of Kosovo and Zanzibar, a semi-autonomous region within Tanzania, which is presented separately in Pension Watch due to having a distinct universal social pension system. Dependent territories which are originally featured in the UN M49 list have been removed as their coverage data would be captured under the main country. Countries are classified by income level according to the World Bank income classification. Each country in the dataset is assigned an identifier code (iso_alpha3).

Below are outlined key indicators, sources, and calculations and assumptions used throughout the database and platform.

Country profiles

Country information
Summary	Basic country information including population data
Data descriptions	Country_name: Official country name in English per UNSTATS M49
	iso_alpha3: Three-letter country code per ISO 3166-1 alpha-3 standard
	Region_name and region_code: Geographic region name and code per UN M49 classification
	Subregion_name and subregion_code: Geographic sub-region name and code per UN M49 classification
	Intermediate_region_name and intermediate_region_code: Intermediate geographic region name and code per UN M49 (where applicable)
	Income_group and income_group_code. World Bank income classification based on Gross National Income (GNI) per capita using the Atlas method
	Pop_60plus Total number of persons aged 60 years and older in the country/territory
	Pct_60plus Percentage of the total population aged 60 years and older
	Pop_60plus_2050 Projected total number of persons aged 60 years and older in the country/territory in 2050
	Pct_60plus_2050 Projected proportion of the total population aged 60 years and older in the country/territory in 2050, expressed as a percentage
Sources	Standard country or area codes for statistical use (M49), UNSTATS M49 codes, retrieved November 2025: https://unstats.un.org/unsd/methodology/m49/
	Data on country income classifications were sourced from the World Bank. World Bank Country and Lending Groups, Income classifications set on July 1, 2025, remain in effect until July 1, 2026: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups
	Population data and projections were obtained from the United Nations Population Fund (UNFPA) website; the latest available UN population data is 2023. UN Population Division, World Population Prospects, 2024 revision: https://population.un.org/wpp/downloads?folder=Standard%20Projections&group=Most%20used
	Zanzibar population data, 2022 census: https://www.nbs.go.tz/nbs/takwimu/Census2022/Age_and_Sex_Distribution_Report_Tanzania_volume2a.pdf
Coverage data
Summary	Old-age social protection coverage rates by country, disaggregated by contributory schemes and tax-financed schemes where data is available.
Data descriptions	Oasp_coverage_all_60plus: Percentage of people aged 60 years and older who receive old-age social protection benefits (all types).
	Calculated using the ILO data on the percentage of people above the national statutory retirement age (SRA) receiving old-age social protection benefits. Calculated as Coverage rate (60+) = (Coverage rate at SRA × Population at SRA+) / Population 60+
	Where countries have a range of retirement ages depending on occupation, then the midpoint is used. Where countries have a retirement age expressed in years and months, these are rounded to the nearest year.
	Coverage_year_updated: Most recent year that the ILO old age social protection coverage data based on Statutory Retirement Ages was updated.
	Has_social_pension: Indicates whether the country currently operates at least one social pension scheme. Used as filter variable for displaying scheme-level data. All scheme-level variables are conditional on this field = 1
	Oasp_coverage_contributory_60plus: Indicates percentage of people aged 60 years and older who receive contributory old-age social protection benefits
	Oasp_coverage_taxfinanced_60plus: Percentage of people aged 60 years and older who receive tax-financed old-age social protection benefits
Source and version date	Data on social protection coverage, including the proportion of older people receiving contributory and tax-financed pensions, as well as broader social protection coverage, were obtained from the International Labour Organization’s (ILO) World Social Protection Data Dashboard: https://www.social-protection.org/gimi/WSPDB.action?id=32.
	A series of calculations were undertaken to standardise these data across countries (discussed above). Coverage data is included in the dataset where the reference year is in 2016 or later. Data predating 2016 is excluded and displayed as ‘no data’. Reference years for each country are displayed in the database
	Missing statutory retirement ages obtained from the World Bank: https://data360files.worldbank.org/data360-data/metadata/WB_GS/WB_GS_SG_AGE_RTRE_BEN.pdf
Additional notes	Countries with a statutory retirement age lower than aged 60 may display coverage for the 60plus group at over 100%.
Basic scheme information
Summary	Basic scheme information: year of introduction (current and first scheme year), legal basis, and overseeing policy and administrative organisations.
Social pension scheme inclusion criteria	Inclusion: tax-financed, regular, non-contributory, cash payments provided to older people. Schemes funded from other non-tax public financing mechanisms such as natural resource revenues (e.g. Brunei, Timor-Leste) are also included. All schemes operate at national level, except for Zanzibar, a semi-autonomous region within Tanzania, given its significance to HelpAge International and HelpAge Tanzania, and the absence of a nation-wide social pension in Tanzania.
	Exclusion: Sub-national schemes; Schemes which provide lump sum payments; Schemes which require a contribution period, unless: (a) contributions function as an eligibility gateway (i.e. a threshold to access the scheme rather than a basis for calculating its amount); (b) the scheme is predominantly funded from tax or other non-tax public financing mechanisms; and (c) transfers are not tied to contribution amounts (i.e. they are functionally non-contributory – see notes below for examples); Schemes which provide vouchers or subsidies on goods or services.
	Only social protection programmes specifically targeting older people were included in the dataset and analysis. Schemes such as veterans’ pensions, carer allowances, and medical expense cards, and other social assistance programmes not specifically designed for older people were excluded from the analysis.
Data descriptions	PW_scheme_ID: Assigned scheme ID for the Pension Watch database
	Scheme_year_updated: Most recent year that the data was updated.
	Scheme_info_source: Source of the data.
	Scheme_name: Official scheme name in English
	Scheme_name_local: Name of the social protection scheme in the original language
	Current_scheme_year: Year the current scheme was introduced. Where the year of implementation is available, this has been used, otherwise the year of the relevant legislation is shown.
	Current_scheme_year_source: Source of information about the year the current scheme was introduced.
	Has_legislative_basis: Indicates whether the scheme has been established through a formal legal instrument (e.g. a law, decree, or regulation).
	Admin_org: The government ministry, department, agency, or institution responsible for administering and managing the social pension scheme.
Source	The majority of data on the social pension schemes are from the International Social Security Association (ISSA), dated from January 2023 to July 2024: https://www.issa.int/databases/country-profiles
	Where gaps were identified, additional desk research was conducted, and relevant reports were consulted to validate and complete the information. More information on specific sources is available in the dataset under ‘Scheme_info_source’
	Data on the year of introduction of each scheme was sourced through desk research.
	The full list of sources is available in the dataset under the column “current_scheme_year_source”.
Additional notes	Aruba’s social insurance pension is included because no contributions (or credited contributions) are required to qualify. The scheme operates on a residency-based model rather than a contribution-based model.
	China’s Basic Pension Insurance for Urban and Rural Residents is included on the basis that it is state-financed, with the 15-year contribution requirement functioning as an eligibility criterion.
	Macao, China’s Old-age pension is included on the basis that it is state-financed, with the 5-year contribution requirement functioning as an eligibility criterion.
	Netherlands’ social insurance pension is included because no contributions (or credited contributions) are required to qualify. The scheme operates on a residency-based model rather than a contribution-based model.


Targeting and eligibility
Summary	Targeting and eligibility information includes age of eligibility, targeting approach and qualifying conditions.
Data descriptions	Eligibility_age_male: The minimum age at which men become eligible to receive social pension benefits under the scheme, as specified in national legislation or policy.
	Eligibility_age_female: The minimum age at which women become eligible to receive social pension benefits under the scheme, as specified in national legislation or policy.
	Targeting_approach: The targeting approach can be selected from one of two categories, either ‘social assistance (poverty targeted)’ or ‘universal/near universal’
	Targeting_approach_notes: Additional notes about the targeting approach, where appropriate.
	Qualifying_conditions: Additional qualifying conditions selecting from the following categories: means-tested, no contributory pension, work not allowed, no other benefits, other.
	Qualifying_conditions_notes: Additional notes about the qualifying conditions, where appropriate.
Source	The majority of data on social pension schemes is obtained from the International Social Security Association, dated from January 2023 to July 2024: https://www.issa.int/databases/country-profiles
	Where gaps were identified, additional desk research was conducted, and relevant reports were consulted to validate and complete the information.
Transfer amount and scheme expenditure
Summary	Information on the social pension transfer amount, expressed in local currency and converted to PPP, % of GDP per capita, and as a proportion of national poverty lines. Also includes scheme expenditure data.
Definitions	Currency: The official name of the national currency used in the country, as specified in ISO 4217.
	Transfer_LCU The standard monthly cash benefit provided to eligible beneficiaries under the social pension scheme, expressed in the national currency. For schemes with tiered transfer amounts that vary by age, the minimum amount is used as it reflects the amount received by the majority of recipients. For schemes where amounts vary by household composition and/or income, the base rate for a single person is used. These are the base figures from which all derived comparators are calculated.
	Calculation_method Calculation method used to determine the transfer amount, selected from the following categories: age-tiered, scaled, household-adjusted, flat rate, pct poverty line, pct min pension, pct minimum wage.
	Transfer_notes Additional notes about the calculation method, where appropriate.
	Transfer_uprating_basis: Basis for uprating the benefit amount, either legally mandated or ad hoc.
	Transfer_source: Source of the transfer amount information.
	Transfer_source_date: Date of the source of the transfer amount information.
	Transfer_usd_international: The monthly transfer amount converted from local currency to international dollars using 2021 Purchasing Power Parity (PPP) conversion rates by the World Bank. Purchasing Power Parity (PPP) adjustment accounts for differences in the cost of living across countries, enabling meaningful comparison of transfer amounts in real terms. For countries that have experienced high inflation since 2021, this figure should be interpreted with caution as it may overstate the current purchasing power of the transfer; the Local Currency Unit, (LCU) amount and GDP per capita comparator are recommended for these cases.
	Transfer_pct_gdp_capita: Monthly transfer amount as a percentage of GDP per capita (current Local Currency Units, LCU). The monthly transfer amount expressed as a percentage of annual GDP per capita divided by twelve, both in local currency units. This measures the generosity of the transfer relative to the economic capacity of the country, allowing comparisons of fiscal effort across countries at different levels of development.
	Transfer_pct_poverty_line: The monthly transfer amount in international dollars expressed as a percentage of the relevant World Bank monthly poverty line. This measures the adequacy of the transfer relative to a standardised poverty threshold – that is, the extent to which the pension alone would be sufficient to keep a recipient above the poverty line. The World Bank defines separate poverty lines for each income group (low income: $3.00/day; lower middle income: $4.20/day; upper middle income: $8.29/day; high income: $29.43/day), expressed in 2021 international dollars, with the monthly equivalent calculated at 30.4167 days. For high income countries this comparator is less analytically meaningful, as social pensions in these contexts are typically designed to provide income replacement rather than a basic safety net and will often exceed the poverty line by a significant margin; GDP per capita is the preferred comparator in these cases.
	Scheme_cost_pct_gdp: The total annual scheme expenditure, expressed as a percentage of the country’s gross domestic product, indicating the fiscal scale of the program relative to the national economy.
Source	Monthly transfer amounts (Local Currency Unit, LCU) are sourced entirely through desk research to ensure the most up to date possible. Where no data could be found, the amounts in ISSA Country Profiles were used. The full list of sources is available in the dataset under the column ‘transfer_source’.
	International dollars conversion rate, World Bank Purchasing Power Parity (PPP) conversion factor, GDP (local currency per international $), 2021.
	GDP per capita (current Local Currency Units), World Bank. World Development Indicators, 2024: https://data.worldbank.org/indicator/NY.GDP.PCAP.CN
	International poverty lines, World Bank (2025) Global Poverty Revisited Using 2021 Purchasing Power Parity (PPP) and New Data on Consumption: https://documents1.worldbank.org/curated/en/099503206032533226/pdf/IDU-e2e09dcf-0af2-481a-a60a-64adf28171d0.pdf
	Data on social pension expenditure was compiled from a range of sources. These include two reports by Development Pathways:
	· Kidd Stephen, Athias Diloa, Raj Dhanisha, Beyond the unaffordability myth – A pragmatic roadmap to universal social security, 2026: https://www.developmentpathways.co.uk/wp-content/uploads/2025/11/Beyond-unaffordability-paper_Final.pdf
	· Kidd Stephen, Raj Dhanisha, Athias Diloa, Adding life to years – A review of tax-financed pensions in the Asia-Pacific region, 2026: https://www.developmentpathways.co.uk/wp-content/uploads/2026/01/Adding-life-to-years-FINAL.pdf
	For Latin America and the Caribbean, data on social protection expenditure were extracted from the Economic Commission for Latin America and the Caribbean (CEPAL) database: https://dds.cepal.org/bpsnc/ps.
	In several cases, particularly for high-income countries, expenditure data were identified through additional desk research.
Additional notes	2021 Purchasing Power Parity (PPP) conversion rates for the Local Currency Units (LCU) transfer amounts are used so that it aligns with the WB 2025 poverty lines, which are benchmarked to 2021 PPP conversion rates.
	Argentina, Türkiye and Venezuela have experienced hyperinflation / severe currency instability in the period between 2021 (year of the PPP conversion rates) and 2026. For these countries, the LCU amounts and % of GDP per capita are more reliable measures of adequacy than international dollars or poverty lines.

Social pension cost data

Social pension cost calculator
Summary	This calculator provides a simple estimate of the annual cost of introducing a social pension in a single country or territory. The estimated cost is presented in local currency, as well as a percentage of GDP and total government expenditure. It also estimates the number of people who would be eligible to receive the social pension.
Calculation	The calculation is based on the selected eligibility age and monthly pension amount. It multiplies the estimated number of eligible people within the selected age bracket by the chosen monthly transfer amount and by 12 months. An additional 5% is then added to the total to account for estimated administrative costs.
Sources	Population data was obtained from the United Nations Population Fund (UNFPA) website; the latest available UN population data is 2023. UN Population Division, World Population Prospects, 2024 revision: https://population.un.org/wpp/downloads?folder=Standard%20Projections&group=Most%20used
	GDP and government expenditure data was obtained from the International Monetary Fund’s (IMF) World Economic Outlook (WEO), October 2025: https://www.imf.org/external/datamapper/datasets/WEO

Social pension evidence pages

Diffusion of social pensions
Summary	Countries where social pension schemes are in place, alongside their World Bank income classification.
Sources	Data on existence of social pensions across countries was obtained from:
	· The International Labour Organization’s (ILO) World Social Protection Data Dashboard: https://www.social-protection.org/gimi/WSPDB.action?id=32;
	· The International Social Security Association’s (ISSA) Country Profiles: https://www.issa.int/databases/country-profiles (dated from January 2023 to July 2024)
	Data on country income classifications were sourced from the World Bank. World Bank Country and Lending Groups, Income classifications set on July 1, 2025, remain in effect until July 1, 2026: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups
Introduction of social pensions
Summary	Timeline displaying the year countries introduced social pensions alongside their income classification at the time of introduction.
Sources	Data on country income classifications were sourced from the World Bank. World Bank Country and Lending Groups, Income classifications set on July 1, 2025, remain in effect until July 1, 2026: https://datahelpdesk.worldbank.org/knowledgebase/articles/906519-world-bank-country-and-lending-groups
	Data on the year of introduction of each scheme were sourced through desk research. The full list of sources is available in the dataset under the column “current_scheme_year_source”.
Additional notes	The timeline begins in 1990, as this is the period from which World Bank income classifications are consistently available and comparable across countries.
	Nauru, Tuvalu and Kosovo are not classified by the World Bank. Their income classifications are therefore estimates based on Gross National Income (GNI).
Benefit adequacy (universal vs targeted)
Summary	Comparison of the transfer value (% of GDP per capita) of universal and poverty-targeted social pensions
Source	Targeting approaches were obtained by country profiles developed by the International Social Security Association (ISSA), dated from January 2023 to July 2024: https://www.issa.int/databases/country-profiles
	GDP per capita (current LCU), World Bank, World Development Indicators, 2024: https://data.worldbank.org/indicator/NY.GDP.PCAP.CN
Poverty reduction impact
Summary	Impact of social pensions on poverty reduction among the poorest 20% of the population.
Source	Data was obtained from the World Bank, ASPIRE database – The Atlas of Social Protection: Indicators of Resilience and Equity: ASPIRE – The Atlas of Social Protection: Indicators of Resilience and Equity \| DataBank
	https://www.worldbank.org/en/data/datatopics/aspire
Impact of social pensions
Summary	Evidence on how social pensions affect income, health, nutrition, and wellbeing outcomes.
Sources	Evidence draws on research conducted by HelpAge International, its global network, and partner organisations. Key research and evidence are summarised in this document:
	· HelpAge International (2020). Why social pensions? Achieving income security for all in older age. http://pension-watch.net/silo/files/why-social-pensions-achieving-income-security-for-all-older-people-in-myanmar.pdf