Consumption Dynamics Explain Bond-Stock Correlation Shifts
Study reveals how consumption regimes drive asset price relationships across economic cycles.
1Three consumption regimes identified
The research identifies three distinct consumption dynamics regimes through real-time sequential learning: two persistent states (permanent vs. transitory shock dominance) and a transitory disaster state. These regimes fundamentally alter how bonds and stocks correlate, with transitions between them explaining historical market behavior. The framework captures structural breaks in consumption patterns that traditional models miss.
2Late 1990s correlation flip explained
The study attributes the late 1990s shift from positive to negative bond-stock correlation to a regime transition where permanent consumption shocks became dominant. This structural change in consumption dynamics caused real bonds and equities to move inversely, matching empirical observations. The mechanism operates through changing inflation expectations and real interest rate responses to different shock types.
3Disaster regime characteristics
During transitory disaster regimes, consumption shocks exhibit mean-reverting properties distinct from other states. These periods see temporary but severe impacts on consumption growth, creating unique asset pricing implications. The model shows how such episodes generate temporary decoupling of traditional asset relationships until normalcy resumes.
4International evidence
Findings extend beyond US markets, demonstrating similar regime-dependent bond-stock relationships across major economies. The international consistency suggests fundamental consumption dynamics rather than local institutional factors drive these patterns. Cross-country differences emerge in regime persistence and transition frequencies.
5Asset pricing implications
The framework provides new tools for pricing assets during regime transitions when traditional correlations break down. Investors can better anticipate correlation shifts by monitoring consumption shock characteristics. Portfolio hedging strategies require adjustment based on the identified consumption state.
6Methodological innovation
The real-time sequential learning approach captures regime shifts that recursive estimation methods miss. By continuously updating beliefs about shock persistence, the model explains sudden changes in asset comovements. This addresses long-standing puzzles about time-varying bond-stock relationships.
7Practical applications
Asset allocators can use regime signals to adjust portfolio risk exposures preemptively. The findings suggest dynamic hedging ratios that vary with consumption states. Pension funds and insurers particularly benefit from anticipating correlation regime changes in long-duration portfolios.
8Future research directions
The study opens new avenues for investigating how consumption dynamics interact with monetary policy across regimes. Future work could explore finer sectoral differences and develop real-time regime classification indicators for tactical asset allocation. The framework may also apply to other asset class relationships.
Original article: Journal of Finance — Consumption Dynamics Explain Bond-Stock Correlation Shifts