Working Paper Abstracts – 2000
Liquidity Provision during Circuit Breakers and Extreme Market Movements
Michael A. Goldstein, Kenneth A. Kavajecz
On October 27, 1997, circuit breakers caused the New York Stock Exchange (NYSE) to halt trading for the first time in history as the Dow Jones Industrial Average (DJIA) lost 554 points. The next day, the NYSE traded a record 1.2 billion shares as the DJIA increased by 337 points, the largest single-day point gain to date. Using data on the limit order books and specialists’ quotes, we examine liquidity provision by limit order traders and floor members during these extreme market movements. We find evidence that specialists fulfilled their obligations to ‘lean against the wind’ even though liquidity provision by limit order traders declined precipitously. An analysis of activity during the circuit breaker reveals that limit order traders generally either remained inactive or withdrew liquidity during the market-wide trading halt.
Bookbuilding and Strategic Allocation
Francesca Cornelli, David Goldreich
Under the bookbuilding procedure, an investment banker solicits bids for shares from institutional investors prior to pricing an equity issue. After collecting this demand information, the investment banker prices the issue and allocates shares at his discretion to the investors. We examine the books from 39 international equity issues. For each issue we consider all institutional bids and the corresponding allocations. We infer some of the criteria the investment banker uses to allocate shares. We find that the investment banker awards more shares to bidders that provide information (such as a limit prices) in their bids. In addition, regular investors receive favorable allocations – especially when the issue is heavily oversubscribed. The results support the theories that justify the use of bookbuilding.
A Theory of Takeover Bidding
In this paper, we present a model of takeover bidding. Grossman and Hart (1980a) argue that small shareholders in a diffusely held firm would hold on to their shares rather than tendering and earn the higher post takeover return. This “free rider” problem precludes profitable takeovers from ever occurring. However, this problem does not exist in our model. In equilibrium every type of the raider successfully hides her private information in the offer stage, thus earning positive profits. We also show that the higher the fraction of the shares the incumbent initially holds, the higher the equilibrium tender offer price. Contrary to Grossman and hart (1980b), voluntary full disclosure of private information may not be the outcome even when misrepresenting is not feasible. Moreover, enforcement of disclosure laws may lead to a sub-optimal outcome. Finally, we discuss the design of the optimal corporate charter and show that the optimal choice of the supermajority rule depends on the incumbent’s initial holdings.
Two-Class Voting: A Mechanism for Conflict Resolution?
Ernst Maug, Bilge Yilmaz
This paper discusses the merits of two-class voting procedures where voters are separated into classes that vote separately. A prominent example are Chapter 11 bankruptcy proceedings, where claim-holders who decide on a workout proposal are divided into classes, and the approval of the proposal needs a majority of the votes in each class. Some political mechanisms employ a similar process. We analyze two-class voting in a context where voters have conflicts of interest as well as differences of opinion regarding a proposal. We investigate how voting mechanisms aggregate information dispersed among voters. We find that two-class voting provides a significant improvement over single-class voting in all those situations where voters have significant conflicts of interests, and where the electorate is relatively evenly divided between different interest groups. Then voting in homogeneous groups provides voters with protection against expropriation and allows them to reveal their information through voting. however, two-class voting is inefficient for relatively homogenous electorates.
Strategic Voting and Proxy Contests
We analyze proxy fights where privately informed shareholders are uncertain about the management ability of the raider. We show that the shareholders vote towards compensating for the initial bias formed by supermajority amendments and the shares that the incumbent controls. Consequently, the amount of support given to the raider is an increasing function of the initial bias favoring the incumbent. This compensating behavior may reverse the effects of the incumbent’s defensive strategies. More specifically, we show that anti-takeover measures may increase the likelihood of a takeover by an inferior raider. In contrast to earlier sincere voting models we also show that simultaneous (strategic) voting can serve as a Pareto dominant voting mechanism for aggregating dispersed information.
Nested Information and Manipulation in Financial Markets
Archishman Chakraborty, Bilge Yilmaz
We construct a model of trading in a financial market with an insider who may or may not be informed about the fundamentals. Rational traders called followers possess part of the private information of the insider (which the market makers do not possess) and trade on this information. This increases the competitive pressure on the informed insider and, when his private information is sufficiently long-lived, leads him to manipulate in every equilibrium. The presence of the followers also enables the uninformed trader to profitably manipulate by using the followers to generate “momentum” in the price process. We show that when there are many followers and a large number of trades before all private information is revealed, each type of the insider will manipulate in every equilibrium. The results are related to the literature on disclosure of insider trades and to dual trading.
Archishman Chakraborty, Bilge Yilmaz
In asymmetric information models of financial markets prices (imperfectly) reveal private information held by traders. Informed insiders thus have an incentive not only to trade less aggressively but also to trade in the “wrong” direction so as to “confuse” the market and increase the noise in the trading process. They thus manipulate the information content of the market prices for private profit. The result holds when the value of the insider’s information does not decay immediately and when the market is uncertain about the presence of an informed insider. We prove the results for a Glosten-Milgrom type Bid-Ask model as well as for a Kyle type of market order model where the insider faces price uncertainty.
Predictable Changes in NAV: The wildcard option in transacting mutual fund shares
John M.R. Chalmers, Roger M. Edelen, Gregory B. Kadlec
This study documents positive autocorrelation in daily mutual fund returns and examines its causes and consequences. We provide evidence that nonsynchronous trading in the underlying assets held by funds is a primary source of the autocorrelation. The autocorrelation in fund returns presents investors with an option to (indirectly) trade fund assets at stale prices. We refer to this option as the mutual-fund wildcard option. We show that just four roundtrip trades in fund shares yields an average abnormal returns of 1.8% in domestic equity funds, 3.8% in high-beta small-cap domestic equity funds, and 4.7% in foreign equity funds. Approximately 45% of the equity fund universe allow this frequency of transacting without load or transaction fees.
With fixed costs of participating in the stock market, consumers with high income will participate in the stock market, but consumers with lower income will not participate. In a fully-funded defined-contribution social security system tries to exploit the equity premium by selling a dollar of bonds per capita and buying a dollar of equity per capita, consumers who save but do not participate in the stock market will increase their consumption, thereby reducing saving and capital accumulation. Calibration of a general equilibrium model indicates that this policy could reduce the aggregate capital stock substantially, by about 50 cents per capita.
Evaluating and Investing in Equity Mutual Funds
Lubos Pastor and Robert F. Stambaugh
Our Framework for evaluating and investing in mutual funds combines observed returns on funds and passive assets with prior beliefs that distinguish pricing-model inaccuracy from managerial skill. A fund’s “alpha” is defined using passive benchmarks. We show that returns on non-benchmark passive assets help estimate that alpha more precisely for most funds. The resulting estimates generally vary less than standard estimates across alternative benchmark specifications. Optimal portfolios constructed from a large universe of equity funds can include actively managed funds even when managerial skill is precluded. The fund universe offers no close substitutes for the Fama-French and momentum benchmarks.
The Equity Premium and Structural Breaks
Lubos Pastor and Robert F. Stambaugh
A long return history is useful in estimating the current equity premium even if the historical distribution has experienced structural breaks. The long series helps not only if the timing of breaks is uncertain but also if one believes that large shifts in the premium are unlikely or that the premium is associated, in part, with volatility. Our framework incorporates these features along with a belief that prices are likely to move opposite to contemporaneous shifts in the premium. The estimated premium since 1834 fluctuates between four and six percent and exhibits its sharpest drop in the last decade.
Stock-Return Predictability and Model Uncertainty
We investigate the implications of uncertainty about the return-forecasting model for the investment opportunity set. Asset allocations are computed through various approaches that differ in their treatment of model uncertainty. The optimal portfolio choices can differ to economically significant degrees, especially for short-horizon high risk-tolerance investors. We decompose the variance of predicted stock returns into several components, including model uncertainty and parameter uncertainty. The model-uncertainty component can be significantly higher than the parameter-uncertainty component, especially when predictive variables, such as dividend yield and book-to-market, are at their recently observed levels, and there is substantial prior uncertainty about whether returns are predictable.
Risk Aversion and Optimal Portfolio Policies in Partial and General Equilibrium Economies
Leonid Kogan and Raman Uppal
In this article, we examine analytically the optimal consumption and portfolio policies in an economy with incomplete financial markets where agents have power utility over intermediate consumption and bequest, and face portfolio constraints and a stochastic investment opportunity set. The source of changes in the investment opportunity set could be a stochastic instantaneous interest rate, stochastic volatility, and/or a stochastic risk premium. We find analytically the conditions under which investment in the risky asset can increase with risk aversion. We then nest this portfolio problem in a general equilibrium setting (for a production economy and also for an exchange economy) with multiple agents who differ in their degree of risk aversion. We derive the optimal portfolio policies when the evolution of the investment opportunity set is determined endogenously and also characterize explicitly the interest rate, stock price and risk premium in general equilibrium. The exact local comparative statics and approximate but analytical expressions for the optimal policies are obtained by developing a method based on perturbation analysis to expand around the solution for an investor with log utility.
Catching Up with the Joneses: Heterogeneous Preferences and the Dynamics of Asset Prices
Yeung Lewis Chan and Leonid Kogan
We examine how cross-sectional heterogeneity in preferences affects equilibrium behavior of asset prices. We obtain explicit characterization of the competitive equilibrium in an exchange economy in which individual agents have catching up with the Joneses preferences and differ only with respect to the curvature of their utility functions. We show that heterogeneity can have a drastic effect on the behavior of asset prices, in particular, on their conditional moments. Dynamic re-distribution of wealth among the agents in heterogeneous economies leads to time-variation in aggregate risk aversion and market price of risk, generating empirically observed negative relation between conditional return volatility and expected returns on one hand and the level of stock prices on the other hand. This stands in contrast with the behavior of homogeneous economies with the same preferences, in which such relation is positive. Quantitatively, the heterogeneous model is capable of replicating various empirical properties of asset prices. Keywords: Equilibrium Asset Pricing, Heterogeneity, Catching Up with the Joneses, Time Varying Risk Aversion, Stochastic Volatility, Predictability
Does the Internet Increase Trading? Evidence from Investor Behavior in 401(K) Plans
James J. Choi, David Laibson and Andrew Metrick
Does the Internet Increase Trading? Evidence from Investor Behavior in 401(k) Plans ABSTRACT: We analyze the impact of a Web-based trading channel on the trading activity in two corporate 401(k) plans. Using detailed data on about 100,000 participants, we compare trading growth in these firms to growth for a sample of firms without a Web channel. After 18 months of access, the inferred Web effect is very large: trading frequency doubles, and portfolio turnover rises by over 50 percent. We also document several patterns of Web-trading behavior. Young, male, and wealthy participants are more likely to try the Web channel. Frequent traders (before Web introduction) are less likely to try the Web. Participants who try the Web tend to stick with it. Web trades tend to be smaller than phone trades both in dollars and as a fraction of portfolio. “Short-term” trades make up a higher proportion of phone trades than of Web trades.
Intraday Analysis of Market Integration: Dutch Blue Chips traded in Amsterdam and New York
Erik Hupperets and Bert Menkveld
Market integration is studied for Dutch stocks cross-listed at the NYSE. Trading starts in Amsterdam and ends in New York with a one-hour overlap. Both markets are not perfectly integrated in that they can be viewed as one market with the well-documented U-shape in volatility, volume and spread. Increased values for the hour of overlap suggest informed trading. Zooming in on this hour, markets are integrated in that price discovery on both sides of the Atlantic reflects the same underlying, new information. Not consistent across all stocks is the origin of this information, Amsterdam, New York or both.
The Structure of the US Equity Markets;
Marshall E. Blume
In 1975, Congress directed the SEC to develop a national market system in which all orders to buy or sell equities would interact. A national market system abhors fragmentation and assumes that one market will best serve the needs of all investors. Such an assumption does not capture the realities of modern markets. Investors have different needs and different markets will develop to serve these needs. Fragmented markets are a natural result of competition. Within the US, the sharing of trade and quote information among markets helps to mitigate any deleterious effects of fragmentation. The markets of tomorrow will be global. In a global market, the SEC will have to give up its goal of a national market system and focus on other issues. For example, it will be a challenge to provide just the sharing of trade information across borders. Further, technology will allow a market center or order gathering function to be located anywhere in the world. This threat of relocation will place constraints of US regulators, and global trading will make it more difficult for US authorities to regulate investment practices and to protect US investors.
Dynamic Asset Allocation under Inflation;
Michael J. Brennan and Yihong Xia
This paper develops a simple framework for analyzing the asset allocation problem of a long-horizon investor when there is inflation and only nominal assets are available for trade. The investor’s optimal investment strategy is given in simple closed form using the equivalent martingale method. The investor’s hedge demands depend on both the investment horizon and the maturities of the bonds in which he invests. The optimal strategy can be decomposed into three components: first, a portfolio that mimics a hypothetical indexed bond with maturity equal to the investment horizon; secondly, the mean-variance tangency portfolio; thirdly, an additional investment in the hypothetical indexed bond to hedge against changes in the investment opportunity set. When short positions are precluded, the investor’s optimal strategy consists of investments in cash, equity and a single nominal bond. When the model is calibrated to recent data on US interest rates and inflation, only high frequency movements in real interest rates are detected so that the optimal allocation between stock and bond is found to be relatively insensitive to the horizon. A longer calibration period reveals low frequency variation in real interest rates that induces more pronounced horizon effects. Reasons for the differences in the two calibration exercises are suggested.
Jim Poterba finds that consumers do not spend all of their assets during retirement, and he projects that the demand for assets will remain high when the baby boomers retire. Based on his forecast of continued high demand for capital, Poterba rejects the asset market meltdown hypothesis, which predicts a fall in stock prices when the baby boomers retire. I develop a rational expectations general equilibrium model with a bequest motive and an aggegate supply curve for capital. In this model, a baby boom generates an increase in stock prices, and stock prices are rationally anticipated to fall when the baby boomers retire, even though, as emphasized by Poterba, consumers do not spend all of their assets during retirement. This finding contradicts Poterba’s con-clusion that continued high demand for assets by retired baby boomers will prevent a fall in the price of capital.
An Exploration of the Effects of Pessimism and Doubt on Asset Returns
Andrew B. Abel
The subjective distribution of growth rates of aggregate consumption is characterized by pessimism if it is first-order stochastically dominated by the objective distribution. Uniform pessimism is a leftward translation of the objective distribution of the logarithm of the growth rate. The subjective distribution is characterized by doubt if it is mean-preserving spread of the objective distribution. Pessimism and doubt both reduce the riskfree rate and thus can help resolve the riskfree rate puzzle. Uniform pessimism and doubt both increase the average equity premium and thus can help resolve the equity premium puzzle.
Variable Selection for Portfolio Choice
Yacine Ait-Sahalia and Michael W. Brandt
We study asset allocation when the conditional moments of returns are partly predictable. Rather than first model the return distribution and subsequently characterize the portfolio choice, we determine directly the dependence of the optimal portfolio weights on the predictive variables. We combine the predictors into a single index that best captures time-variations in investment opportunities. This index helps investors determine which economic variables they should track and, more importantly, in what combination. We consider investors with both expected utility (mean-variance and CRRA) and non-expected utility (ambiguity aversion and prospect theory) objectives and characterize their market-timing, horizon effects, and hedging demands.
Trading and Voting Choice
David K. Musto and Bilge Yilmaz
The political choice between candidates with different redistribution policies plays out very differently in a complete financial market. When voters have the opportunity to trade election-contingent securities, we find that 1) wealth considerations have no effect on voting, so the interaction between candidates’ redistribution policies and the distribution of wealth has no effect on who wins, 2) an election in which a candidate promises wealth redistribution results in redistribution of wealth, and the redistribution is the same regardless of who wins, and 3) if one candidate prefers some amount of redistribution and the other does not, the candidate who prefers redistribution will propose more redistribution than the amount he prefers.
The Role of Trading Halts in Monitoring a Specialist Market
Roger Edelen and Simon Gervais
We model an exchange as a collection of specialists, each a monopolist market maker in a subset of the stocks listed on the exchange. Specialists can obtain net private benefits at the expense of the exchange (the collection of all specialists) by quoting a privately optimal pricing schedule. Conversely, a coordinated pricing schedule makes all specialists and customers better off. However, coordination requires a system of monitoring and punishment, which can break down when information asymmetries between the exchange and a specialist are high. This breakdown can cause the specialist to seek a temporary halt to trading to alleviate unjustified punishment, or the exchange to halt trading to prevent the quoting of damaging privately optimal pricing schedules. We use a sample of NYSE halts and a proxy for the exchange-specialist information asymmetry to test this theory. As predicted, we find a significant increase in estimated information asymmetry immediately preceding trading halts.
Predictable Changes in NAV: The Wildcard Option in Transacting Mutual-fund Shares
John M.R. Chalmers, Roger Edelen and Gregory B. Kadlec
Economic distortions can arise when financial claims trade at prices set by an intermediary rather than by direct negotiation between principals. We demonstrate the problem in a specific context, the exchange of open-end mutual fund shares. Mutual funds typically set the price at which fund shares are exchanged (NAV) using an algorithm that fails to account for nonsynchronous trading in the fund’s underlying securities. This results in predictable changes in fund share prices, which lead to exploitable trading opportunities of 0.8% per trade at international and small-cap domestic equity funds. A simple modification to the pricing algorithm suggested by nonsynchronous trading theory eliminates much of this predictability. However, one can never rule out the possibility of distortions that arise from other unknown sources when intermediaries set prices.
High-and Low-Frequency Exchange Rate Volatility Dynamics: Range-Based Estimation of Stochastic Volatility Models
Sassan Alizadeh, Michael W. Brandt, and Francis X. Diebold
We propose using the price range in the estimation of stochastic volatility models. We show theoretically, numerically, and empirically that the range is not only a highly efficient volatility proxy, but also that it is approximately Gaussian and robust to microcstructure noise. The good properties of the range imply that range-based Gaussian quasi-maximum likelihood estimation produces simple and highly efficient estimates of stochastic volatility models and extractions of latent volatility series. We use our method to examine the dynamics of daily exchange rate volatility and discover that traditional one-factor models are inadequate for describing simultaneously the high- and low-frequency dynamics of volatility. Instead, the evidence points strongly toward two-factor models with one highly persistent factor and one quickly mean-reverting factor.