Broker-dealers serve customers either by trading out of inventory or by matching trades with counterparties in over-the-counter markets. In a dynamic model featuring broker-dealers rebalancing customer trades at an inter-dealer market, we derive endogenous inventory holdings for broker-dealers, an aspect overlooked by existing over-the-counter market literature. By comparing equilibrium outcomes, we find that allowing the broker-dealer to hold positive inventory improves market liquidity and customer well-being. This highlights unintended consequences of post-crisis regulations restricting banks' balance-sheet capacity. We also explain the non-monotonicity of spreads in a model of varying competition among central dealers in the inter-dealer market.

This paper examines whether U.S. technology firms receive different pricing and non-price terms in the syndicated loan market compared to non-technology firms. The analysis reveals that technology borrowers face significantly less favorable terms, including 12.41 basis points higher loan spreads, approximately 4.9% shorter maturities, and loan sizes reduced by 3.3% of total assets. Evidence suggests these differences are driven by higher information asymmetry, as technology firms benefit more from reductions in opacity than non-technology peers. The relationship between firm quality and loan maturity appears non-monotonic, with higher-quality firms potentially opting for shorter loan terms. The findings are robust to entropy balancing, endogeneity concerns, and credit rating subsamples. Overall, the results indicate that technology firms face systematically worse loan terms, with important implications for borrowers, lenders, and market participants amid the sector’s growing presence in syndicated lending.

The rise of social media and retail trading platforms has potentially changed the importance of retail investors in the stock market. Using data from Reddit WallStreetBets, we examine the market consequences of abnormal social media attention and its relationship with retail investor trading activities. We find that high social media attention on Reddit is associated with higher retail trading activities, cumulative returns and price informativeness. When social media attention occurs in conjunction with earnings announcements, social media attention increases the sensitivity of retail trading responses to earnings news, as well as the discussion tones on Reddit and price informativeness measure. While Reddit helps the information diffusion to the general retail population, investors on Robinhood exhibit opposite trading directions, consistent with previous studies that Robinhood investors are different from other retail investors. 

In a frictionless model with fanatic retail investors flocking in to invest in "meme" stocks, we find that a risk-neutral hedge fund with market power offsets all demand from the fanatics, plus a tilt that bets on the expected changes in the fanatics' demand. When there are market frictions associated with shorting the asset or holding inventory, the hedge fund's demand is attenuated. In equilibrium, mispricing is determined by the directional change in the fanatics' demand. The fanatics as a group always lose money in the economy. In particular, "diamond hands," who buy the asset high and never sell, have a negative present value of investing because they pay too much for the cash flows they receive. The expected loss of the fanatics is divided up equally by the rational long-term investors who trade on mispricing and the hedge fund who front-runs the fanatics.