PT - JOURNAL ARTICLE AU - Samuel. E. Assabil AU - Don L. Mcleish TI - Assessing the Impact of Longevity Risk for Countries with Limited Data AID - 10.3905/jor.2020.1.080 DP - 2020 Dec 29 TA - The Journal of Retirement PG - jor.2020.1.080 4099 - https://pm-research.com/content/early/2020/12/28/jor.2020.1.080.short 4100 - https://pm-research.com/content/early/2020/12/28/jor.2020.1.080.full AB - The impact of longevity risk has not been well studied in most developing countries due to the lack of suitable mortality data. As a result, most pension companies in these countries (especially those on the African continent) do not account for longevity risk in their annual valuation. This can even lead to their collapse if steps are not taken to address it. In this work, we develop a method of assessing longevity risk where there is a severe lack of mortality data. The method is based on the assumption that there is a nearly linear relationship between annuitant’s hazard function and their mortality at higher ages (post-retirement age), which permits approximating with the Gompertz model. We tried the method on mortality data from Ghana and the results are consistent with those in the standard literature. That is, longevity risk is a treat to pension companies in Ghana even though, in the case of Ghana, this treat has partially been mitigated by the high-interest rate in the country. With this method, pension and life companies that are not able to account for longevity risk as a result of lack of data or newly formed pension companies with even two-year mortality data will be able to assess the longevity risk they face without relying on data or models from other countries.TOPICS: Long-term/retirement investing, pension funds, risk management, frontier marketsKey Findings▪ Longevity risk is present in Ghana and its impact on life companies in the country is potentially high.▪ Unlike most developed countries, longevity risk in Ghana has partially been mitigated by the high interest rate in the country.▪ Ghana’s post retirement mortality could be approximated with the Gompertz Model.