Adjustment dynamics between broker–dealer leverage and stock market: a threshold cointegration analysis

Abstract

We empirically investigate the short- and long-run dynamic relationship between security broker–dealer leverage and stock prices. To explore various potential adjustment dynamics, we use the error correction term to construct several threshold indicators that distinguish the deviation from long-run equilibrium. Using these indicators, we estimate different types of threshold cointegration and error correction models and analyze the results in the context of boom and bust cycles in financial markets. Our results show that, in the long run, a 1% increase in the stock price leads to approximately a 0.43% increase in leverage. We show that a momentum autoregressive model with an endogenous threshold does a better job of fitting the financial data. The asymmetric error correction structure shows that changes in leverage experience significant corrective measures when the stock prices plunge during recessions.