Working Paper: NBER ID: w20968
Authors: Denis Gromb; Dimitri Vayanos
Abstract: We develop a model of financially constrained arbitrage, and use it to study the dynamics of arbitrage capital, liquidity, and asset prices. Arbitrageurs exploit price discrepancies between assets traded in segmented markets, and in doing so provide liquidity to investors. A collateral constraint limits their positions as a function of capital. We show that the dynamics of arbitrage activity are self-correcting: following a shock that depletes arbitrage capital, profitability increases, and this allows capital to be gradually replenished. Spreads increase more and recover faster for more volatile trades, although arbitrageurs cut their positions in these trades the least. When arbitrage capital is more mobile across markets, liquidity in each market generally becomes less volatile, but the reverse may hold for aggregate liquidity because of mobility-induced contagion.
Keywords: arbitrage; liquidity; capital constraints; asset prices
JEL Codes: D52; D53; G01; G11; G12; G14; G23
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| arbitrage capital (G19) | investment capacity (G31) |
| investment capacity (G31) | asset prices (G19) |
| asset prices (G19) | profitability of arbitrageurs (G19) |
| profitability of arbitrageurs (G19) | capital replenishment (G31) |
| shock depleting arbitrage capital (G19) | profitability of arbitrageurs (G19) |
| capital mobility (F20) | liquidity responses to shocks (E44) |
| liquidity responses to shocks (E44) | volatility in individual markets (G19) |
| capital mobility (F20) | aggregate liquidity volatility (E41) |