Sequence of return risk in South African post-retirement portfolios: the effectiveness of volatility-focused asset allocation strategies to address sequence and associated risks

Schabort, Kilian Petika

Sequence of return risk in South African post-retirement portfolios: the effectiveness of volatility-focused asset allocation strategies to address sequence and associated risks

Master Thesis

2022

Abstract

Sequence of return risk (which is the risk of unfavourable investment outcomes at the most unfavourable time) is an important consideration for efficiently funding retirement portfolio spending goals. This study examines the sensitivity of retirement decumulation portfolios to sequence of return risk (“SOR risk”, also referred to as “sequence risk”) in a South African context and evaluates the effectiveness of five volatility-focused asset allocation strategies in addressing the risk. Using monthly asset index returns separated into bull and bear market regimes for the period 1991 to 2020, correlated non-normal returns were simulated and combined with independently simulated inflation to generate 10 000 independent 30-year simulation trials. The performance of each of the five strategies was measured against a benchmark set of portfolios using simulated data and was validated using partial out-of-sample historical data. Performance was assessed using the sustainable withdrawal rate (SWR) and actuarial coverage ratio (ACR) metrics. The sensitivity results showed that SOR is greatest at the retirement date, declining asymptotically (halving within the first ten years of retirement). The geographic diversification strategy showed clear benefit to reducing SOR whereas the results for the risk parity strategy were not conclusive. The analysis and comparison of the lowrisk, rising equity glidepath, and dynamic cash buffer strategies formed the focus of the study. All three showed considerable SOR potential with the dynamic cash buffer strategy outperforming the others and substantially reducing SOR relative to the benchmark.

Keywords

finance and tax

Reference:

Collections

Masters

Full item page