Institutional Portfolio Adjustment to Geopolitical Risk
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Heightened geopolitical risk in recent years has altered investors’ risk calculus. This underscores how geopolitical events can pose challenges to global diversification. Institutional investors such as AP7 must navigate this tension while also ensuring long-term returns. Using Russia’s invasion of Ukraine in 2022 as a geopolitical shock, this paper investigates whether Sweden’s largest public pension fund, the default investment option for approximately six million pension savers, adjusted its country-level equity portfolio exposure to China relative to other emerging markets following this shift. The analysis uses public semi-annual holding disclosures from AP7 Aktiefond, between 2020 and 2025. This fund acts as AP7’s globally diversified equity sleeve. Exploiting the invasion as an exogenous geopolitical shock within a difference-in-differences framework, the study finds that China’s share of AP7’s equity portfolio declined by 3.2 percentage points relative to the control group. The estimates are consistently negative across control-group specifications, though statistical significance varies. Event-study estimates show a steady, sustained pattern, attaining statistical significance from December 2022 through June 2024. This suggests a deliberate reallocation rather than a mechanical index response. The findings provide empirical evidence of how a major state-mandated investor re evaluates country-level exposure in response to geopolitical shocks. Although the paper focuses on AP7, it may highlight a broader trend of geoeconomics considerations increasingly influencing portfolio management. The paper complements existing studies on market-level returns and aggregate capital flows by examining a fund’s country-level portfolio adjustments to geopolitical risk. This has implications for state-mandated funds considering dual challenges of ESG and financial returns during heightened geoeconomic uncertainty.