The liquidity state dependence of monetary policy transmission
Summary
Focus
Monetary policy shocks often have large effects on long-term interest rates, but the strength of this transmission varies considerably over time. We study how the liquidity of the Treasury market shapes the pass-through from short-term policy surprises to long-term yields. We show that monetary policy is most powerful when markets are liquid and arbitrageurs are well capitalised, and that it is weaker when liquidity is impaired.
Contribution
We combine high-frequency policy shocks with aggregate bond yield data and transaction-level evidence from the US Treasury market. Liquidity conditions are measured by the degree to which bond yields deviate from a smooth fitted yield curve, which reflects the limits of arbitrage capacity. Our framework introduces "liquidity state dependence" as a new dimension of monetary transmission. By linking yield movements to both arbitrageurs' trading activity and investors' reach-for-yield behaviour, we provide a mechanism that reconciles puzzling patterns in the term premium with the structure of preferred-habitat models.
Findings
We find that long-term yields react to policy rate surprises significantly more when market liquidity is high. These effects are driven by changes in real term premia and persist for up to three months after the policy announcement. Micro data confirm that hedge funds and other arbitrageurs trade significantly more at longer durations when liquidity is abundant, while some investors display reach-for-yield behaviour. The results imply that conventional policy is most effective in liquid markets, whereas asset purchases gain traction when markets are illiquid. The results also help to explain why tightening is often more powerful than easing: monetary expansions tend to occur when arbitrage capital is most constrained.
Abstract
We show that monetary policy shocks move long-term government bond yields only when market liquidity is high and arbitrageurs are well capitalized. This liquidity state dependence operates entirely through real term premia, not expectations. Using novel transaction-level data on the US Treasury market, we find that arbitrageurs trade about 40% more duration during FOMC meetings in high-liquidity periods. We propose ways of enriching standard term-structure models to rationalize our evidence that constraints on arbitrage capital suppress transmission. The results introduce new empirical moments for theories of limits to arbitrage, and underscore the role of liquidity conditions in shaping the effectiveness of conventional monetary policy.
JEL classification: E43, E52, E58
Keywords: monetary policy, long-term real rates, limited arbitrage, segmented markets
