Examining non-parallel interest rate risk premia with a focus on South African markets

Abstract

In this thesis we analysed sovereign yield curves for emerging and developed markets to identify the proportion of parallel and non-parallel shifts over time. We found that non-parallel shifts were more prevalent in emerging markets due to relatively higher political and economic risks. Key drivers included systemic risk events like wars, debt distress, and pandemics. Geopolitical risks were seen to increase in frequency in the modern era and developing a strategy to exploit non-parallel risk premia is an attractive proposition.

We determined which policy regimes are appropriate to extract non-parallel risk premia. Our research suggests that countries with opposing monetary and fiscal policies possess greater non-parallel return opportunities whilst countries with complementing policies require tactical non-parallel strategies to optimise returns. We identified the South African (SA) swap and bond interest rate markets as suitable candidate for maximising non-parallel risk premia via a long butterfly strategy (which is immune to parallel curve shifts but exposed to non-parallel curve shifts). Various butterfly weighting methodologies were analysed with a cash and duration neutral methodology being the preferred option due to its conservatism and practicality.

Ten swap and bond butterfly risk factors were defined using monthly SA data from 2001-2024. We back-tested and analysed these factors during interest rate cycles, curve scenarios and risk-off periods. Most of the top ranked swap and bond butterfly factors displayed strong and persistent outperformance over their corresponding bottom ranked factors, resulting in improved risk-adjusted and absolute returns especially during positive and steep twists, and bull and bear flattening curve scenarios. The All-Factor Rank which combined all the factors demonstrated improved diversification by balancing upside and downside risks. By demonstrating this outperformance our results contradict the strong efficient market hypothesis of not being able to consistently outperform the market on a risk-adjusted basis.

Trade costs were an important factor that required pragmatic management, we compared monthly and quarterly rebalancing frequencies which resulted in monthly rebalancing producing greater gross returns but due to high trade costs, less frequent rebalancing as in the quarterly frequency reduced trade costs and improved net returns. Trade costs can be significant so having a pragmatic rebalancing strategy with efficient market makers that limit trade costs to one basis point of spread duration make these butterfly risk factors an effective and successful portable alpha strategy.