Nearly all of this work requires quality data. This means data that is accurate, detailed, up to date, complete and reliable. However, since February 2021 the project has been unable to directly contact stakeholders, particularly the military regime to secure any information . The project was faced with the challenge of producing good quality analysis - on which decisions could be based now and in the future - without easily accessible data.

However, the experience of the project showed that despite the challenges, an adapted and flexible approach ensured that valuable analysis could still be produced. The experience of the project may provide useful lessons for similar situations where data is difficult or impossible to obtain.

So how did we adjust our approach and obtain relevant data and what were the approaches that we undertook to perform analysis?  How did these different from a standard approach to data collection and analysis? The following lessons were learnt:

1. Use existing data and make relevant adjustments. The project could call on data and information secured at the end of 2020 for a proposed actuarial valuation of existing social security schemes in Myanmar. We used this data and adapted it to take into account known developments in the labour market, the economy and the health situation since collection date so that it could be used for analysis in 2021.
2. Run sensitivity analysis, what-if scenarios and stress testing to assess what would happen if experience is not in line with base assumptions. Typically, such approaches are used to run projections on different scenarios. However, we also used these methods to reflect that the data was subject to uncertainty.
3. Use previous historical experience in other countries which had suffered similar abrupt shocks to assess what might have happened in Myanmar.  Whilst the nature of shocks vary, some of the impacts may be similar and better documented than is the case in Myanmar. This historical experience was useful as an input into decision making and analysis
4. Use published official information but check it. Some military regime Offices continued to publish information which could be used (e.g., membership figures for social security schemes). Such figures were cross checked with other (independent) sources and analysis to assess the reasonableness of the figures
5. Rely on the expertise of local colleagues. Their knowledge, judgement and expertise contributed to a better understanding of the reality on the ground and factors impacting the labour market, the economy and finance.
6. Reflect data uncertainty in a cautious set of assumptions. Missing and unreliable data leads to being more cautious than normal in both the base and pessimistic / optimistic scenarios.
7. Future proof the outputs under the project so the when the situation improves, up to date information can be used. For example, the simulation model provided allows users to easily update data. As part of the project, substantial training and knowledge transfer resources were provided to support data collection, management and analysis in the future.
8. As actuaries, be driven by professional standards regarding data and assumptions used in the work. The ILO ISSA Actuarial guidelines also cover issues to consider including missing and uncertain data, sensitivity analysis, risk management and reporting.
9. When unable to meet with stakeholders, cost and provide a wider range of different design and financing options for future discussions. As stakeholders were unable to identify and discuss options with us, we included a wider range of alternative designs than normal. These met both ILO standards and conventions and reflected good international practice. In this way, stakeholders will have a basis for discussing alternative options in the future.
10. Adopt a flexible but cautious approach by always respecting standards and guidelines to ensure professional work is delivered.

By adapting for a potential future situation when better data may be available, the project ensured that the outcomes could be relied on for discussions on policy, financing, and management decisions when possible. In addition, the knowledge transfer legacy of the project to provide training and resource material has allowed an increased focus on data management in the material provided and assessing a wider range of alternative options facilitates future discussions.

Good data is essential to ensure actuarial work can be relied upon. Under the project, we learnt the reality of this in crisis times and the approaches that were undertaken will provide useful input for future projects where data challenges exist.

[1] The UNCT Myanmar principle of engagement post military coup d’état disallows certain forms of engagement, while the issue of UN credentials has not yet been resolved. It was however impossible to undertake any independent data survey.

Disclaimer: This article is produced as part of the programme “Improving synergies between Social Protection and Public Financial Management (SP&PFM)” financed by the European Union, and the project “Building resilience for the future of work and the post COVID-19: Bolstering Unemployment Insurance and Labour Market Policy Development in Myanmar (BUILD)” funded by the Ministry of Health, Labour and Welfare of Japan. Its contents are the sole responsibility of the authors and do not necessarily reflect the views of the European Union and the government of Japan.