Heterogeneous firm-level responses to trade liberalization: A test using stock price reactions

This paper presents novel empirical evidence on key predictions of heterogeneous firm models by examining stock market reactions to the Canada–United States Free Trade Agreement of 1989 (CUSFTA). I derive testable predictions for a class of models based on Melitz (2003). Using the uncertainty surrounding CUSFTA's ratification, I show that the pattern of abnormal returns of Canadian manufacturing firms was strongly consistent with predictions related to export (U.S.) tariff reductions, but less so with predictions related to import (Canadian) tariff reductions. Lower Canadian tariffs did have an effect through the implied reduction in intermediate input tariffs, however.     

Semantics and capital market efficiency: Testing the market's reaction to the naming of stock distributions as splits or dividends

The AICPA and NYSE urge corporations to avoid the term “stock dividend” and use the term “stock split” when referring to large common stock distributions. Theoretically these large distributions do not convey any economic value to the stockholders. Therefore a rule was considered necessary so as not to mislead them as to the actual nature of the distributions. The purpose of this study is to examine security price reactions to the differential naming of these distributions. The results strongly suggest that security returns are not differentially affected by naming a large common stock distribution a stock dividend rather than a stock split. That is, the capital market appears to be semantically efficient between terminologies.     

A secure,real-time stock market protocol

Stock markets constitute the largest electronic commerce market in the world. The tremendous growth in trading volume and the need for fast and accurate transaction execution has made the stock market one of the most technology friendly markets. The fastest growing stock exchange, NASDAQ, is a wholly electronic stock exchange with all transactions conducted over computer networks. However, the transaction model used by NASDAQ and other electronic stock markets still borrows heavily from the older traditional models used by non-electronic stock exchanges. Two important requirements of modern day stock market transactions are: (a) customer's ability to place sophisticated transaction orders to buy/sell stock, and (b) customer's ability to detect transaction delays. Modern electronic stock exchanges lack both the ability to place newer, more sophisticated transaction orders and the ability to detect delays in transaction execution. In this paper, we propose a protocol for stock market transaction that can model a new sophisticated model for transaction orders while continuing to support traditional transaction orders. The protocol is augmented with a mechanism to detect delays in transaction execution. It is further shown that the protocol proposed is secure, atomic, anonymous, private, and incurs low overhead costs. This revised version was published online in June 2006 with corrections to the Cover Date.     

A framework for analyzing money stock control under the Monetary Control Act

This paper analyzes the money stock effects of the Monetary Control Act (MCA) under a nonborrowed reserve (NBR) operating procedure. Prior to the passage of the MCA, policy was conducted under an interest rate operating target where reserve requirement reforms such as those introduced by the MCA had little influence on money stock variability. Under an NBR procedure however, the structure of reserve requirements may have a significant impact on monetary control. Our analysis indicates that the relative improvement in monetary control greatly depends on the degree of tightness exercised by the Federal Reserve over total reserves in an MCA regime. The tighter the control, the more significant the estimated monetary control benefits of the MCA under an NBR procedure.     

A quantile-based test of protection for sale model

This paper proposes a new test of the Protection for Sale (PFS) model by Grossman and Helpman (1994). Unlike existing methods in the literature, our approach does not require any data on political organization. We use quantile and IV quantile regressions in our tests, using the data from Gawande and Bandyopadhyay (2000). Surprisingly, the results do not provide any evidence favoring the PFS model. We also explain why previous work may have found support for it.     

A closer look at co-movements among stock returns

Correlation among financial assets is widely recognized; however, the mechanics of the relationship are not well understood. This paper investigates the microstructure of the co-movement of stock returns. The goal is to improve our understanding of correlation among stock returns by examining the conditions under which asset returns co-move on an intra-day basis. The methodology combines a traditional lead–lag model with a modified or pseudo-error correction model. Empirical evidence is presented to suggest the speed of adjustment between paired asset intra-day returns is a function of asymmetric information. Specifically, the wider an asset's spread, the faster the asset will converge to the intra-day returns of other similar assets. This result is consistent with partial adjustment model presented by Chan (Chan, K. (1993). Imperfect information and cross-autocorrelation among stock prices. The Journal of Finance:1211–1230.) which suggests market makers gain from monitoring other market makers in periods of uncertainty.     

The value of celebrity endorsements: A stock market perspective

Are celebrity endorsements worthwhile investments in advertising? To answer this question, we analyze a unique sample of 101 announcements made between 1996 and 2008 by firms listed in the USA. Internet is the main medium of communication for these announcements. We employ event study methodology and document statistically insignificant abnormal returns around the announcement dates. This finding is consistent with the notion that the incremental benefits from celebrity endorsements closely match the incremental costs due to such contracts. Further, we investigate if the announcement date return depends on a number of characteristics that are often used in the endorsement literature. As a result, we find that endorsements of technology industry products coincide with significant positive abnormal returns around the announcement dates. Finally, we find weak support for the match-up hypothesis between celebrities and endorsed products.