Aggregate Trading Research Articles

Drivers of demand and supply in the Euro interbank market: the role of “Key Players” during the recent turmoil

We study frictions in trading patterns in the Euro money market. We characterize the structure of lending relations during the period of recent financial turmoil. We use a network-topology method on data from overnight transactions in the Electronic Market for Interbank Deposits (e-MID) to investigate two main issues. First, we characterize the roles of borrowers and lenders in long-run relationships by providing evidence on network formation at a 3-month frequency. Second, we identify the “key players” in the marketplace and study their behavior. In our formalization, key players are “locally-central banks” within a network that lend (or borrow) large volumes to (from) several counterparties, while borrowing (or lending) small volumes from (to) a small number of institutions. Our results are twofold. We show that the aggregate trading patterns in e-MID are characterized by largely asymmetric relations. This implies a clear difference in the roles of lenders and borrowers, with market positions changing only gradually over time. We also find that the large net lenders exploit their positions as network leaders by imposing aggressive pricing policies on their counterparties.

Trading, Profits, and Volatility in a Dynamic Information Network Model

We introduce a dynamic noisy rational expectations model, in which information diffuses through a general network of agents. In equilibrium, agents’ trading behavior and profits are determined by their position in the network. Agents who are more closely connected have more similar period-by-period trades, and an agent’s profitability is determined by a centrality measure that is closely related to eigenvector centrality. In line with the Mixture of Distributions Hypothesis, the market’s network structure influences aggregate trading volume and price volatility. Volatility after an information shock is more persistent in less central networks, and in markets with a higher degree of private information. Similar results hold for trading volume. The shape of the autocorrelation functions of volatility and volume are related to the degree of asymmetry of the information network. Altogether, our results suggest that these dynamics contain important information about the underlying information diffusion process in the market.

The Vulture Spread Premium: Distressed Investorss Impact on Recovery Rates and Bond Prices

Debt pricing models typically ignore the bankruptcy process by specifying recovery rates as an exogenous function of the state space. I develop a parsimonious model in which corporate default induces a transfer of bond ownership away from traditional diversified holders toward risk-averse activist investors (vultures). Vulture funds improve emergence recovery values but demand a premium which increases with the amount of risk they take. The ratio of activist wealth to defaulted debt emerges as the key state variable that drives prices and returns for defaulted debt and expected recovery rates for pre-default bondholders. In empirical tests, this ratio is a significant determinant of risk-adjusted returns and explains 82% of the time series variation in aggregate post-default trading prices. Exactly as the model predicts, the relationship between the activists wealth ratio and returns is strongest firms with assets that are difficult to monetize and for fulcrum classes where the creditors are likely to emerge from bankruptcy holding the newly issued equity. Through its determination of recovery rates, my renegotiation framework can be easily incorporated into a partial equilibrium asset pricing model where it helps to reduce pricing errors relative to models with exogenously-specified recovery rates. The improvement spans the rating spectrum and for 10 year debt in particular, is able to simultaneously match AAA-BBB and AAA-B spreads.

The capital asset pricing model in economic perspective

The capital asset pricing model (CAPM) is theoretically incomplete in its demand-side focus, risk-averse investors and internally inconsistent homogeneous beliefs; is not conclusively supported empirically; and yet it legitimizes a notion that investors can earn higher returns by bearing undiversifiable risk. Our article does not merely extend the CAPM with more realistic assumptions, it completes its original framework by including (1) risk-taking investors in the investor population, (2) investors who can have heterogeneous expectations or beliefs – an overlooked but required condition for the CAPM to be an internally consistent and meaningful model of competitive financial asset pricing under uncertainty and (3) a positive-sloped short-run supply curve based on a reasonable interpretation of the nature of financial asset trade. Upon a complete economic interpretation, it is shown that the equilibrium (systematic) risk-rate of return relationship depends on whose aggregate trading activity dominates, risk-averse or risk-taking investors’. There is no universal, or even general, positive relationship between systematic risk and rate of return. This has far-reaching implications for investors and investment advisors who serve them.

Retail Traders and the Competitive Allocation of Attention

I consider a rational expectations framework in which attention-constrained individuals compete against each other and institutions. The model reconciles a set of empirical facts that cannot be simultaneously explained by standard theories: retail trader portfolios are highly correlated, retail traders tend to follow some stocks and ignore others, and aggregate trades from retail traders are profitable over short-horizons and predict future returns (even controlling for past returns or past volume). Additionally, the equilibrium herding behavior of individuals suggests that their actions are not subsumed by institutions. Instead, the herd competes directly against institutions. The analysis also yields novel predictions regarding portfolio composition and turnover.

Aggregate Market Attention Around Earnings Announcements

This study examines the relation between both number and news content of earnings disclosures by firms and aggregate stock market trading activity. Consistent with the Hirshleifer, Lim, and Teoh (2009a) distraction hypothesis, among announcing firms the number of contemporaneous announcers reduces trading responses. However, among non-announcers the opposite holds. Trading in non-announcing stocks increases with number of announcers. Hence, while high numbers of announcing firms distract attention from each other they actually increase the level of attention given to the aggregate market by investors. This effect is also stronger when announcing firms are large in size, which is consistent with the idea of larger firms providing more macro-relevant information to the market (e.g., Anilowski, Feng, and Skinner 2007; Bonisall, Bozanic, and Fischer 2013). The news content of announced earnings also has several distinct marketwide trading impacts. Negative aggregate news is positively associated with marketwide trading, consistent with the idea that bad news is generally more macro-relevant than good news (Williams 2014). Earnings surprise magnitudes, both overall and individual, also seem to matter much more for marketwide trading when the number of announcers is large, consistent with the idea that the market is more sensitive to earnings news in periods when earnings disclosures are clustered (e.g., “earnings seasons”).

Mutual Fund Herding and Fund Performance

I develop fund-level measures of the similarity in trading of mutual fund managers, resulting in the identification of leaders, contemporaneously herding managers, and followers. I find evidence of a persistent group of funds whose trades lead the aggregate trades of the mutual fund industry; these leader funds exhibit strong subsequent performance, consistent with informed trading. In contrast, there is no evidence that managers that trade together, either contemporaneously or with a lag, outperform. These findings suggest that managers of leader funds receive in advance private signals regarding the information upon which other funds focus.

An Economic Analysis of the Competitive Risk-Return Paradigm

The Capital Asset Pricing Model (CAPM) is theoretically incomplete in its demand-side focus, risk-averse investors, and internally inconsistent homogeneous beliefs; is not conclusively supported empirically; and yet it legitimizes a notion that investors can earn higher returns by bearing un-diversifiable risk. Our paper does not merely extend the CAPM with more realistic assumptions, it completes its original framework by including (1) risk-taking investors in the investor population, (2) investors who can have heterogeneous expectations or beliefs — an overlooked but required condition for the CAPM to be an internally consistent and meaningful model of competitive financial asset pricing under uncertainty, and (3) a positive-sloped short-run supply curve based on a reasonable interpretation of the nature of financial asset trade. Upon a complete economic interpretation, it is shown the equilibrium (systematic) risk-rate of return relationship depends on whose aggregate trading activity dominates, risk-averse or risk-taking investors’. There is no universal, or even general, positive relationship between systematic risk and rate of return. This has far-reaching implications for investors and investment advisors who serve them.

Trading directions and the pricing of Euro interbank deposits in the long run

We investigate the relation between aggregate trading imbalances and interest rates in the Euro money market. We use data for OTC contracts as well as information from the major electronic trading platform in Europe to study the presence of cointegration between trading pressures and money market rates. We report strong evidence of a long-term linear relation between trading imbalances and liquidity prices for Euro interbank deposits.

Why Allow Internalization?

Over the last decade, an increasing portion of U.S. equity trading volume has been executed away from public exchanges. A common form of this off-exchange trading is internalization, a practice whereby a brokerage fills a client's order from its own inventory. I develop a model to study the impact of internalization on traders’ behavior and market outcomes. In equilibrium, trades in the limit order book, relative to internalized orders, stem from the more informed traders. The bifurcation of the order flow increases transaction costs for the more informed traders. Consequently, it reduces these traders’ demands and the aggregate trading volume. When all orders are submitted via possibly internalizing broker-dealers, the volume reduction lowers price efficiency. When traders can access the limit order book directly, the concentration of informed flow on the exchange may dominate the impact of the reduced volume: there exists an equilibrium where internalization improves price efficiency.

Structural Distortions in the Euro Interbank Market: The Role of ‘Key Players’ During the Recent Market Turmoil

We study the frictions in the patterns of trades in the Euro money market. We characterize the structure of lending relations during the period of recent financial turmoil. We use network-topology method on data from overnight transactions in the Electronic Market for Interbank Deposits (e-Mid) to investigate on two main issues. First, we characterize the division of roles between borrowers and lenders in long-run relations by providing evidence on network formation at a yearly frequency. Second, we identify the ‘key players’ in the marketplace and study their behavior. Key players are ‘locally-central banks’ within a network that lend (or borrow) large volumes to (from) several counterparties, while borrowing (or lending) small volumes from (to) a small number of institutions. Our results are twofold. We show that the aggregate trading patterns in e-Mid are characterized by largely asymmetric relations. This implies a clear division of roles between lenders and borrowers. Second, the key players do not exploit their position of network leaders by imposing opportunistic pricing policies. We find that only a fraction of the networks composed by big players are characterized by interest rates that are statistically different from the average market rate throughout the turmoil period.

Political Instability & its Impact on Karachi Stock Exchange (Impact on Trading Volume and Trading Value, 2011)

The objective of this paper is to examine the relationship between aggregate stock market trading volume and daily stock returns during 2/20/2009-3/11/2011 during which the events happened in Pakistan. This paper is evaluating the instability in the stock market due to the political events and the dependency of fluctuations in stock returns and trading volume. The hypothesis of this study states, “The political events have a great impact on trading volume and stock return.” Six political events (attack on Sri Lankan cricket team, restoration of C.J. Ashura attack, Pakistan temporarily suspends NATO supply, assassination of Governor Salman Taseer, assassination of Minorities Minister Shahbaz Bhatti) have been considered. The pre and post events data has been collected, in order to determine the impact of these events on KSE. The results show that stock returns moved too much due to change in the trading volume, the same results were found, changes occur in the value of correlation between the trading volume and stock return. The results show that due the nature of event the relation between the trading volume and the stock return fluctuate, from my study I have concluded that the event effect the value of Pearson correlation and due to event the value is decrease from their pre event value.

Learning and Complementarities in Speculative Attacks

We study a model where the aggregate trading of currency speculators reveals new information to the central bank and affects its policy decision. We show that the learning process gives rise to coordination motives among speculators leading to large currency attacks and introducing non-fundamental volatility into exchange rates and policy decisions. We show that the central bank can improve the ex ante effectiveness of its policy by committing to put a lower weight ex post on the information from the market, and that transparency may either increase or decrease the effectiveness of learning from the market, depending on how it is implemented.

Taking Stock: Historians and Canada’s Securities Markets and Exchanges

Most Canadians normally pay little attention to the daily gyrations of stock markets; however, markets rudely intruded into our consciousness in mid2008 when index values plummeted by over 40%. As many of us painfully witnessed the rapid contraction of our mutual funds and pensions, and saw planned retirements quickly recede from view, we learned a forceful lesson: our financial prospects are intimately enmeshed with market fortunes. Statistics Canada reported in 2007 that “Pensions, investment funds and equities account for 76% of household financial assets of $3.7 trillion” (Statistics Canada 2007). Evidently, market fortunes affect the lives of huge swaths of our population. It is also clear that stock markets are a critical part of Canada’s economy. Before the global credit crisis that began in 2007, the nation’s two stock exchanges—the Toronto and the Venture exchanges1—recorded an aggregate trading volume of over 149 billion shares. The value of these shares was over $1.7 trillion. About 3,951 corporations were listed on the exchanges with a combined capitalization value of $2.2 trillion, fully 157% of Canada’s annual gross domestic product (GDP) (TSX Group 2007). Stock markets matter, and they matter in ways we never really understood. The financial pain and panic of the credit crash elicited important questions: How had stock markets become so vast and so powerful that, within months, they could vaporize, or at least seriously damage, our personal wealth, fortunes, and retirement prospects? Where and when did these powerful market institutions originate? Does this history help us to understand the recent market crash and its lethal fallout for Canadians? Naturally, I expected that historians would provide the answers to these questions. To my surprise I discovered that they have shown scant interest in the history of our stock markets.2 Indeed, over two decades ago, in his massive history of Canadian business entitled Northern Enterprise: Five Centuries of Canadian Business, Michael Bliss concluded, “The history on all non-bank financial institutions in Canada is very weak [and that] on stock exchanges and brokerage and investment houses is even worse” (1987, 601). Little has changed since then.3 This investigation summarizes what we know about the history of our stock market institutions, reveals the gaping holes that still need to be filled in order to explain their role in our lives today, and identifies the dangers of a short collective memory about our financial history.

Do Individual Investors Trade Stocks as Gambling? Evidence from Repeated Natural Experiments

Multiple natural experiments of large jackpot lotteries in Taiwan are used to document that some individual investors trade stocks as a form of gambling. Those investors substitute lottery gambling for stock trading. This substitution effect between lottery and stock is substantiated by the following five key findings. First, when the jackpot exceeds 500 million Taiwan dollars (about 15 million U.S. dollars), the number of shares traded by individual investors decreases by about 7% among stocks with high individual trading fraction, low market capitalization, high past returns, and high past turnover. Second, the reduction in individual trading is about 7% among stocks with lottery features, i.e. high return volatility and skewness. Third, the magnitude of the decline increases monotonically with the jackpot. Fourth, firm-level trading activity reacts negatively to large jackpots, and is statistically significant for a sizable number of firms. Finally, the aggregate trading activity by individual investors declines by about 5% on large jackpot days.

Does the January Anomaly Survive in Equity Exchange-Traded Funds?

We document that the well-known January effect largely disappears in U.S. equity Exchange-Traded Funds (ETFs). We also find at the aggregate level that plan sponsors, who enjoy tax-deferred/exempt status, buy past losers at year-end, whereas investment managers, who are taxable on average, tend to sell these losers. The differential aggregate trading behaviors extend to the end of October, the tax year-end for mutual funds, a major class of investment managers. We interpret the trading by plan sponsors as tax arbitraging and the trading by investment managers as tax-loss selling. Further analysis supports our interpretation and suggests that these trading behaviors are driven by the high liquidity of the ETFs.

Aggregate Trading Behaviour of Technical Models and the Yen/Dollar Exchange Rate 1976-2007

The study analyses the interaction between the trading behaviour of 1,024 moving average and momentum models and the fluctuations of the yen-dollar exchange rate. I show first that these models would have exploited exchange rate trends quite profitably between 1976 and 2007. I then show that the aggregate transactions and positions of technical models exert an excess demand pressure on currency markets since they are mostly on the same side of the market. When technical models produce trading signals almost all of them are either buying or selling, when they maintain open positions they are either long or short. A strong interaction prevails between exchange rate movements and the transactions triggered by technical models. An initial rise of the exchange rate due to news, e.g., is systematically lengthened through a sequence of technical buy signals.

The Interaction between the Aggregate Behaviour of Technical Trading Systems and Stock Price Dynamics

This study analyzes the interaction between the aggregate trading behavior of technical models and stock price fluctuations in the S&P 500 futures market. It examines 2580 widely used trading systems based on 30-minutes-prices. The sample comprises trend-following as well as contrarian models. I show that technical trading exerts an excess demand pressure on the stock market. This is because technical models produce clusters of trading signals that are on the same side of the market, either buying or selling. Initial stock price changes triggered by news are strengthened by a sequence of trading signals produced by trend-following models. Once 90% of the models have signaled a particular position, stock prices tend to move in the direction congruent with the position-holding of the models. This phenomenon has to be attributed to the transactions of non-technical traders, perhaps amateurs. Once price movements lose their momentum, contrarian technical models contribute to reversals of the trend.

Short sale constraints and the information content of trading

We examine changes in the information content of trading when short sale constraints between prohibition and restriction exist on a stock exchange. This is made possible by a unique institutional arrangement at the Stock Exchange of Hong Kong. It maintains a list of stocks which can be sold short under regulations. Stocks not on the list are prohibited from short selling. The list is revised on a quarterly basis based on predetermined criteria. We find that the probability of information-based trading (PIN) significantly increases when a stock is added to the list. Further analysis shows that this is mainly because uninformed traders are driven out of the market. Elimination of uninformed traders also causes the aggregate trading volume to decrease rather than increase. In comparison, the PIN does not change when a stock is dropped from the list. We also find that market liquidity, measured by volatility and bid–ask spreads, slightly decreases when a stock is added to the list and significantly increases when a stock is dropped from the list. Possible explanations are discussed.

Learning and Complementarities: Implications for Speculative Attacks

We study a model where the aggregate trading of currency speculators reveals new information to the central bank and afiects its policy decision. We show that the learning process gives rise to informational complementarities among traders, which generate coordination motives in their trading and reduce the informational content of the market signal. We analyze the tradeofi between information and coordination, derive comparative statics regarding the e‐ciency of the central bank’s decision, and suggest policy tools to improve it. Overall, our results apply to a wide range of problems, where decision makers learn from the actions of heterogeneously informed agents.