Issue: 2022/Vol.32/No.2, Pages 105-124

ASSET LIABILITY MANAGEMENT FOR THE BANK OF UGANDA DEFINED BENEFITS SCHEME BY STOCHASTIC PROGRAMMING

Herbert Mukalazi , Torbjörn Larsson , Juma Kasozi , Fred Mayambala

Full paper (PDF)    RePEC

Cite as: H. Mukalazi, T. Larsson, J. Kasozi, F. Mayambala. Asset liability management for the Bank of Uganda defined benefits scheme by stochastic programming. Operations Research and Decisions 2022: 32(2), 105-124. DOI 10.37190/ord220207

Abstract
We develop a model for asset liability management of pension funds, which is solved by stochastic programming techniques. Using data provided by the Bank of Uganda Defined Benefits Scheme, which is closed to new members, we obtain the optimal investment policies. Randomly sampled scenario trees using the mean and covariance structure of the return distribution are used for generating the coefficients of the stochastic program. Liabilities are modelled by remaining years of life expectancy and guaranteed period for monthly pension. We obtain the funding situation of the scheme at each stage, and the terminal cash injection by the sponsor required to meet all future benefit payments, in absence of contributing members.

Keywords: closed scheme, finance, asset liability management, scenario generation, stochastic programming

Received: 25 December 2021    Accepted: 19 June 2022
Published online: 15 July 2022