This article develops a model of the upstairs market where order size, beliefs, and prices are determined endogenously. We test the model’s predictions using unique data for 5,625 equity trades during the period 1985 to 1992 that are known to be upstairs transactions and are identified as either buyer or seller initiated. We find that price movements prior to the trade date are significantly positively related to trade size, consistent with information leakage as the block is “shopped” upstairs. Further, the temporary price impact or liquidity effect is a concave function of order size, which may result from upstairs intermediation.

We reevaluate the IPO underpricing phenomenon using the stochastic frontier methodology. The advantage of the stochastic frontier is that it can be used to measure the level of deliberate underpricing in the premarket without using after-market information. This is accomplished through the estimation of a systematic one-sided error term that measures “inefficiency” or the difference between the maximum predicted offer price and the actual offer price. Data for the analysis are comprised of 1,035 IPOs of common stock issued by firm commitment between 1975 and 1984. IPOs appear to be deliberately underpriced in the premarket in both hot-market and nonhot-market periods. Moreover, the determinants of the maximum IPO price have different effects in the two time periods.

[New evidence is provided on the determinants of stock-return variances. First, when the Tokyo Stock Exchange is open on Saturday, the weekend variance increases; weekly variance is unaffected, however, despite an increase in weekly volume. Second, the listing of U.S. stocks in Tokyo substantially increases the number of trading hours, but Tokyo volume is negligible for these U.S. stocks and their 24-hour variance is unaffected. The overall results are consistent with the predictions of private-information-based rational trading models, but inconsistent with both the irrational trading noise and public-information hypotheses.]

[In a model of takeovers under asymmetric information, we identify a separating equilibrium in which the value of the bidder firm is revealed by the mix of cash and securities used as payment for the target. The model predicts that the revealed bidder value is monotonically increasing and convex in the fraction of the total offer that consists of cash. We examine the model restrictions using data from Canada, where mixed offers are both relatively frequent and free of the confounding tax-related options characterizing mixed offers in the United States. We find that the average announcement-month bidder abnormal return in mixed offers is large and significant. However, maximum likelihood estimates of parameters in both linear and nonlinear cross-sectional regressions fail to support the model predictions.]

New evidence is provided on the determinants of stock-return variances. First, when the Tokyo Stock Exchange is open on Saturday, the weekend variance increases; weekly variance is unaffected, however, despite an increase in weekly volume. Second, the listing of U.S. stocks in Tokyo substantially increases the number of trading hours, but Tokyo volume is negligible for these U.S. stocks and their 24-hour variance is unaffected. The overall results are consistent with the predictions of private-information-based rational trading models, but inconsistent with both the irrational trading noise and public-information hypotheses.

[This article reexamines the value of tax trading when the tax rate on long-term realizations is less than that on short-term realizations. In particular, the value of the option to realize long-term capital gains and repurchase stock in order to increase one's tax basis and restart the option to realize future losses short term is examined empirically. Our estimate of the incremental value of restarting, which is based on the results of simulations of several alternative tax trading policies over a large number of independent return sequences, is generally much smaller than that reported by Constantinides (1984). The incremental value of restarting is shown to depend critically on the particular pattern of realized returns and the assumed tax treatment of unrealized capital gains at the end of the simulation period. The effects of stock price volatility, transaction costs, portfolio offset rules, and realization cutoff levels on the value of tax trading are also investigated.]

This article reexamines the value of tax trading when the tax rate on long-term realizations is less than that on short-term realizations. In particular, the value of the option to realize long-term capital gains and repurchase stock in order to increase one’s tax basis and restart the option to realize future losses short term is examined empirically. Our estimate of the incremental value of restarting, which is based on the results of simulations of several alternative tax trading policies over a large number of independent return sequences, is generally much smaller than that reported by Constantinides (1984). The incremental value of restarting is shown to depend critically on the particular pattern of realized returns and the assumed tax treatment of unrealized capital gains at the end of the simulation period. The effects of stock price volatility, transaction costs, portfolio offset rules, and realization cutoff levels on the value of tax trading are also investigated.

We infer conditional swap rate moments model independently from swaption cubes. Conditional volatility and skewness exhibit systematic variation across swap maturities and option expiries (conditional kurtosis less so), with conditional skewness sometimes changing sign. Conditional skewness displays some relation to the level and volatility of swap rates but is most consistently related to the conditional correlation between swap rates and swap rate variances. From realized excess returns on synthetic variance and skewness swap contracts, we infer that variance and (to a lesser extent) skewness risk premia are negative and time varying. For the most part, results hold true in both the USD and EUR markets and in both precrisis and crisis subsamples. We design and estimate a dynamic term structure model that captures much of the dynamics of conditional swap rate moments.

We develop a tractable and flexible stochastic volatility multifactor model of the term structure of interest rates. It features unspanned stochastic volatility factors, correlation between innovations to forward rates and their volatilities, quasi-analytical prices of zero-coupon bond options, and dynamics of the forward rate curve, under both the actual and risk-neutral measures, in terms of a finite-dimensional affine state vector. The model has a very good fit to an extensive panel dataset of interest rates, swaptions, and caps. In particular, the model matches the implied cap skews and the dynamics of implied volatilities.

[We develop a two-sided asymmetric information model of asset sales that incorporates the key differences from mergers and allows the information held by each party to be impounded in the transaction. The buyer's information is conveyed through a first-stage competitive auction. A seller with unfavorable information about the asset accepts the cash offer of the highest bidder. A seller with favorable information proposes a take-it-or-leave-it counteroffer that entails buyer equity. Thus, the cash-equity decision reflects the seller's but not the buyer's information in contrast to the theoretical and empirical findings for mergers. The central prediction of our model is that there are large gains in wealth for both buyers and sellers in equity-based asset sales, whereas cash sales generate significantly smaller gains that typically accrue only to sellers. Our empirical results are consistent with the predictions of our theoretical model.]