Measurement error and panel data1

The underidentification of linear models with measurement error does not necessarily extend to panel data models, as has been shown by GAiliches and Hausman (1986). We discuss and extend some of their results for a simple case and address particular issues concerning identification and asymptotic variances.     

A note on securities market response to pension fund termination*

Abstract. A recent article in this journal (Hsieh, Ferris, and Chen, 1990) presents evidence demonstrating that financially distressed firms that terminate overfunded pension plans experience significantly positive abnormal returns, while nondistressed terminating firms experience returns not different from zero. The article concludes from this evidence that the market regards the property rights to excess assets as residing fully with the sponsoring firm prior to termination, particularly when the sponsoring firm is not financially distressed. Accordingly, the reporting requirements of Statement of Financial Accounting Standards (SFAS) No. 88, which require immediate recognition of gains resulting from termination, are inconsistent with the market's view of the ownership rights to excess plan assets. Similarly, the article suggests that the excise tax imposed on asset reversions is “punitive” because property rights to fund assets resided with the company prior to termination. This paper questions three aspects of the Hsieh, Ferris, and Chen article: (1) the plausibility of the theory explaining differential reaction to the announcement to terminate for distressed and nondistressed firms; (2) the effect of the distressed/nondistressed classification criteria, event date specification, and event window length on the results; and (3) the appropriateness of using the empirical results, if reliable, to evaluate accounting treatment or federal tax policy. Résumé. Dans un récent article paru dans cette même publication, Hsieh, Ferris et Chen (1990) soumettaient des faits démontrant que les entreprises en difficulté financière qui mettent fin à un régime de retraite surprovisionné enregistrent des rendements anormaux positifs importants, tandis que les entreprises qui mettent fin à un régime de retraite sans être en difficulté financière enregistrent des rendements qui ne sont pas différents de zéro. De là, les auteurs concluaient que le marché estime que les droits à l'excédent d'actif appartiennent exclusivement à l'entreprise responsable du régime avant la cessation, en particulier lorsque cette dernière n'est pas en difficulté financière. Par conséquent, les exigences de la norme SFAS No. 88 en matière d'information à fournir, qui prévoient la constatation immédiate du produit de la cessation du régime, ne sont pas conformes à la façon de voir du marché relativement aux droits à l'excédent d'actif du régime. De la même façon, selon les auteurs, la taxe d'accise imposée en cas de retour de l'actif est “punitive”, étant donné que les droits à l'actif du régime appartenaient à l'entreprise avant la cessation du régime. Les auteurs s'interrogent ici sur trois aspects de l'article de Hsieh, Ferris, et Chen: 1) la pertinence de la théorie expliquant la réaction marginale à l'annonce de la cessation d'un régime pour les entreprise qui sont en difficulté financière et celles qui ne le sont pas; 2) l'incidence sur les résultats des critères permettant de classer les entreprises comme étant en difficulté financière ou ne l‘étant pas, de la précision de la date de l’événement et de la longueur de la période pré-post relative à l‘événement; et 3) l'opportunité de l'utilisation des résultats empiriques, s'ils sont fiables, pour évaluer le traitement comptable ou la politique fiscale fédérale.     

VARIANCE BOUND TESTS OF BOND MARKET EFFICIENCY

In this paper we examine variance bound tests of the joint hypothesis that (1) bond markets are efficient and (2) the term structure is determined by the expectations hypothesis. Both the Singleton and Shiller tests are shown to be seriously biased toward rejecting the joint hypothesis in finite samples. Flavin's test is unbiased but has a very high variance leading to many false rejections of the joint hypothesis. When corrected as suggested by Flavin, Shiller's test is unbiased and has a relatively low variance. Unfortunately, it is also sensitive to measurement error.     

THE EFFECTS OF THE INSIDER TRADING SANCTIONS ACT OF 1984: THE CASE OF SEASONED EQUITY OFFERINGS

Previous empirical research indicates that corporate insiders tend to increase (decrease) their shareholdings before events that increase (decrease) firm value. More recent evidence suggests, however, that passage of the Insider Trading Sanctions Act of 1984 (ITSA) may have deterred this behavior. Our results indicate that before passage of the ITSA, insiders exploited their access to nonpublic information by selling shares before the announcement of equity issues. However, after passage of the ITSA insiders no longer displayed this behavior. We conclude the ITSA has a deterrent effect, which is more heavily concentrated on insiders at the highest level of the firm who are most visible to regulators and other market participants.     

Sustaining Collusion in Growing Markets

The impact of demand growth on the collusion possibilities is investigated in a Cournot supergame where market growth may trigger future entry and the collusive agreement is enforced by the most profitable ‘grim trigger strategies’ available. It is shown that even in situations where perfect collusion can be sustained after entry, coping with a potential entrant in a market which is growing over time may completely undermine any pre‐entry collusive plans of the incumbent firms. This is because, before entry, a deviation and the following punishment phase may become more attractive thanks to their additional effect in terms of delaying entry.     

Constructivist Negotiation Ethics

The success of Discourse Ethics is premised on the discovery and use of shared values. If this is true what type of negotiation style, especially when used in an intercultural setting, is best suited to make use of shared values. Research focusing on moral arguments between Germans and Americans uncovered an array of shared values. But the existence of shared values, by itself, was not an adequate predictor of a negotiation's success. What did prove to be a predictor of success was the use of a Constructivist style of negotiation by both parties.     

Cross-Listings and Home Market Trading Volume: The Case of Malaysia and Singapore

Cross‐listings of equities internationally are becoming more common. Using data for Europe and North America, previous studies reject the order flow diversion hypothesis, which states that international cross‐listings reduce home‐country trading volume. We test this hypothesis using data for equities cross‐listed in Singapore and Malaysia. We find that trading volume in Malaysia fell 42.9% when Singapore markets were closed for holidays. Furthermore, we show that trading volume in Malaysia did not increase following the implementation of regulations that ended the trading of Malaysian equities in Singapore in 1998. Hence, we reject the order flow diversion hypothesis.     

Real Investment Implications of Employee Stock Option Exercises

This paper examines a real cost of awarding employee stock options. Based on the observation that managers are extremely concerned about earnings-per-share dilution in equity related compensation, we predict and find that firms experiencing significant employee stock option (ESO) exercises shift resources away from real investments towards the repurchase of their own stocks. We further find weak evidence of a decline in subsequent firm performance (as measured by return on assets) for several years following the cut in discretionary investments as a result of stock option exercises, though this result is sensitive to the metric used to measure performance. Collectively, our findings indicate that ESO exercises potentially impose a real cost on the firm in terms of foregone investment opportunities.