Abstract

We claim that the greater the extent to which a performance measure matches delivered performance, the simpler and more robust are the compensation plans based on it. Stock price changes poorly match delivered performance when they anticipate performance. This causes three problems: (1) Price-based plans become complex as they require knowing the extent to which prices lead performance; (2) Price-based plans are less robust to unforeseen events; (3) Price-based plans require period-by-period changes in pay-for-performance relationships even when the underlying production function remains unchanged. The relative use of earnings- and price-based compensation depends on the extent to which each matches delivered performance.

top

Abstract

This study investigates the market response to the requirement that the principal executive and financial officer of an SEC registrant each state under oath that the firm's annual and quarterly financial reports are materially accurate and complete pursuant to the Securities Exchange Act of 1934. We hypothesise that investors should recognise the importance of these changes in financial reporting and, thus, respond at or around those events that should reveal the most information about those changes, specifically, the SEC order to certify (27 June 2002), the passage of the Sarbanes-Oxley Act (25 July 2002), and the first certification filing by a registrant. We find that investors did respond on the identified dates, and in the ways hypothesised. We conclude that investors responded to certification and / or Sarbanes-Oxley.

top

Abstract

This paper examines how audit tenure affects earnings quality by investigating the effect of audit-firm and audit-partner tenure on the level of discretionary accruals. We find that familiarity helps to produce higher earnings quality, but excessive familiarity results in lower earnings quality. Besides, Big 5 auditors are superior in obtaining learning experience in the initial period of engagement, which implies that the negative effect on earnings quality is more serious for clients of non-Big 5 auditors if audit-firm rotation is mandated. These empirical results are valuable to the regulator even though they are analysed on a non-mandatory auditor rotation regime.

top

Abstract

This paper revisits the familiar problem of specifying empirical regressions of returns on pricenormalised earnings variables. The focus is on relevant organising concepts related to the righthand-side variables / regressors. Issues raised are: (i) the crucial role of assuming market efficiency; (ii) why the notion of "unexpected earnings explain unexpected returns" obscures rather than highlights the nature of the regression specification; (iii) why a regression specification benefits from the inclusion of the prior period's earnings in addition to contemporaneous earnings; (iv) why the startof-period book value-to-price ratio should be included as a regressor; (v) why an hypothesis relating accounting data, such as eps and bvps, to price per share precedes a regression on returns specification; (vi) how the analysis extends to include analysts' eps expectations, and changes in eps expectations, as (price-normalised) regressors.

top

Abstract

The effects of trading Level I ADRs in the US OTC market were investigated for 119 firms from Hong Kong, the United Kingdom (UK), Australia, Japan, South Africa, Germany and Brazil during the period February 1992 to April 2001. Even though US GAAP accounting and disclosure requirements are not met, trading Level I ADRs in the less-regulated OTC market has benefits for some firms: surprisingly, we found that after commencement of a Level I ADR program 25% of sample firms showed at least one favourable change in accounting variables and market measures.

top