This paper tests the influential hypothesis that irrational traders will be driven out of financial markets by trading losses. The paper’s main finding is that overconfident currency dealers are not driven out of the market. Dealers with extensive experience are neither more nor less overconfident than their junior colleagues. We set the stage for this investigation by providing evidence that currency dealers display two forms of overconfidence: They underestimate uncertainty, and they overestimate their professional success. This is notable because one might have expected the opposite: currency dealers face strong incentives for accuracy, they have access to comprehensive information, and they have extensive experience.

This paper documents clustering in currency stop‐loss and take‐profit orders, and uses that clustering to provide an explanation for two familiar predictions from technical analysis: (1) trends tend to reverse course at predictable support and resistance levels, and (2) trends tend to be unusually rapid after rates cross such levels. The data are the first available on individual currency stop‐loss and take‐profit orders. Take‐profit orders cluster particularly strongly at round numbers, which could explain the first prediction. Stop‐loss orders cluster strongly just beyond round numbers, which could explain the second prediction.

This paper identifies a new approach to shrouding, the practice of hiding supra-competitive product prices, and sheds light on an extraordinarily opaque and hitherto unexplored segment of the world's largest OTC market. The product is liquidity for FX transactions between global custody banks and their client funds, transactions in which the clients are not actively involved and cannot identify execution costs ex post. We develop a stylized model of shrouding that predicts FX dealers will shift among three pricing strategies over time within individual client relationships. We conduct an extensive econometric analysis of the complete FX trading record of a mid-sized global custody bank over one calendar year. The data provide support for all ten of the shrouding model's testable hypotheses.

Financial market crashes can occur even in the absence of news. This paper highlights four properties of price-contingent trading that increase the frequency of such events. Price-contingent trading is common across financial market, since it includes algorithmic trading, technical trading, and dynamic option hedging. The four properties we consider are: (1) high kurtosis in the distribution of order sizes; (2) clustering of trades within the day; (3) clustering of trades at certain prices; and (4) feedback between trading and returns. The paper estimates the relative importance of these factors using data from the foreign exchange market. Calibrated simulations indicate that interactions among these factors are at least as important as any single one. Among individual factors, the orders’ size distribution and feedback effects have the strongest influence. Overall, price-contingent trading could account for half of realized excess kurtosis. The paper suggests that extreme returns unaccompanied by news are statistically inevitable in the presence of price-contingent trading.

This study analyzes the wide dispersion in bid-ask spreads across clients in over-the-counter (OTC) markets. Our data detailed data comprise a dealing bank's complete trading record in a major OTC contract and include client IDs, seven client types, and precise markups. Average spreads are lowest for hedge funds (<1 basis point, bp) and highest for individuals and for small and medium enterprises (>50 bps). Regression results suggest that the primary source of variation is clients’ execution efficiency, meaning their ability to minimize execution costs. Efficient trading, which can require investments in knowledge and technology, has three dimensions: reliance on low-cost platforms; familiarity with market technologies, conventions, and negotiating strategies; and breadth of dealing relationships. Relations between proxies for client execution efficiency and client incentives to invest, such as trade frequency, are consistent with rational inattention.

This paper provides evidence that “illusory correlations”—a well-documented source of cognitive bias—lead some agents to be imperfectly rational noise traders. We focus on the head-and-shoulders chart pattern, considered by technical analysts to provide one of the most reliable trading signals. Our findings indicate that the pattern is associated with a substantial rise in trading volume even though it does not profitably predict directional movements. We further substantiate the connection between head-and-shoulders trading and imperfectly rational noise trading by showing that the pattern is associated with lower bid-ask spreads.

This paper provides evidence that informed traders dominate the response of limit-order submissions to shocks in a pure limit-order market. In the market we study, informed traders are highly sensitive to spreads, volatility, momentum and depth. By contrast, uninformed traders are relatively insensitive to all these market conditions. The dominance of the informed over limit-order submissions is magnified by contrasts between them and the uninformed in the use of aggressively-priced limit orders.