Speeding up the exchange does not necessarily improve liquidity. On the one hand, more speed enables a high-frequency market maker (HFM) to update his quotes faster on incoming news. This reduces his payoff risk and thus lowers the competitive bid-ask spread. On the other hand, HFM price quotes...

Liquidity providers (LPs) on decentralized exchanges pay a fixed transaction cost (gas price) whenever they update their positions. Different economies of scale across LPs lead in equilibrium to the fragmentation of liquidity supply between low- and high-fee pools. Using data on liquidity...

We document low cross-sectional correlations between high-frequency market maker (MM) inventory positions, suggesting poor risk sharing. Using a unique data set on Canadian futures markets, a simple inventory cost estimate is 300% above the optimal benchmark. Our model explains how heterogeneity...

Exchanges implement intentional trade delays to limit the harmful impact of low-latency trading. Do such "speed bumps" curb investment in fast trading technology? Data is scarce since trading technologies are proprietary. We build an experimental trading platform where participants face speed...

We document low cross-sectional correlations between market maker (MM) inventory positions in Canadian futures markets, suggesting imperfect risk sharing. We build a model to understand how cross-sectional heterogeneity in inventories shapes liquidity provision in time-priority markets. The...

We provide empirical evidence that CAPM-betas positively predict asset returns when market returns are predicted to be high, which occurs about every other month. Consequently, the product of beta and the predicted market return (CAPM) predicts asset returns by combining the out-of-sample...