Purpose

In this paper, we aim to answer two questions. First, whether firms manipulate reported earnings via pension assumptions when facing mandatory contributions. Second, whether firms alter their earnings management behavior when the Financial Accounting Standard Board (FASB) mandates disclosure of pension asset composition and a description of investment strategy under SFAS 132R.

Design/methodology/approach

Our basic approach is to run linear regressions of firm-year assumed returns on the log of pension sensitivity measures, controlling for current and lagged actual returns from pension assets, fiscal year dummies and industry dummies. The larger the pension sensitivity ratios, the stronger the effects from inflated ERRs on reported earnings. We confirm the early results that the regression slopes are positive and highly significant. We construct an indicator variable DMC to capture the mandatory contributions firms face and another indicator variable D132R to capture the effect of SFAS 132R. DMC takes the value of one for fiscal years during which an acquisition takes place and zero otherwise. D132R takes the value of one for fiscal years after December 15, 2003 and zero otherwise.

Findings

Our sample covers the period from June 1992 to December 2017. Our key results are as follows. The estimated coefficient (t-statistic) on DMC is 0.308 (6.87). Firms facing mandatory contributions tend to set ERRs at an average 0.308% higher. The estimated coefficient (t-statistic) on D132R is −2.190 (−13.70). The new disclosure requirement under SFAS 132R constrains all firms to set ERRs at an average 2.190% lower. The estimate (t-statistic) on the interactive term DMA×D132R is −0.237 (−3.29). When mandatory contributions happen during the post-SFAS 132R period, firms tend to set ERRs at 0.237% lower than they would do otherwise in the pre-SFAS 132R period.

Originality/value

When firms face mandatory contributions, typically firm experience negative stock market returns. We examine whether managers manage earnings to mitigate such negative impact. We find that firms inflate assumed returns on pension assets to boost their reported earnings when facing mandatory contributions. We also find that managers alter earnings management behavior, in the case of mandatory contributions, following the introduction of new pension disclosure standards under SFAS 132R that become effective on December 15, 2003. Under the new SFAS 132R requirement, firms need to disclose asset allocation and describe investment strategies. This imposes restrictions on managers' discretion in making ERR assumptions, since now the composition of pension assets is a key determinant of the assumed expected rate of return on pension assets. Firms need to justify their ERRs with their asset allocations.

中文翻译：

---

企业面临强制性供款时的盈余管理

目的

在本文中，我们旨在回答两个问题。首先，当面临强制性供款时，公司是否通过养老金假设操纵报告的收益。其次，当财务会计准则委员会 (FASB) 要求披露养老金资产构成和 SFAS 132R 下的投资策略描述时，公司是否会改变其盈余管理行为。

设计/方法/方法

我们的基本方法是对养老金敏感度指标的对数进行公司年度假设回报的线性回归，控制养老金资产、财政年度虚拟变量和行业虚拟变量的当前和滞后实际回报。养老金敏感度比率越大，过高的ERR对报告收益的影响就越大。我们确认了回归斜率是正的并且非常显着的早期结果。我们构建了一个指标变量DMC来捕捉企业面临的强制性缴款和另一个指标变量D132R来捕捉 SFAS 132R 的影响。DMC在发生收购的会计年度取值为 1，否则取值为 0。D132R 2003 年 12 月 15 日之后的会计年度取值为 1，否则取值为 0。

发现

我们的样本涵盖了 1992 年 6 月至 2017 年 12 月期间。我们的主要结果如下。DMC上的估计系数 ( t -statistic)为 0.308 (6.87)。面临强制性供款的公司往往将ERR平均设置为高 0.308%。D132R上的估计系数 ( t -statistic)为 -2.190 (-13.70)。SFAS 132R 下的新披露要求限制所有公司将ERR设置为平均低 2.190%。交互项DMA × D132R的估计值 ( t -statistic) 为-0.237 (-3.29)。在后 SFAS 132R 期间发生强制性供款时，公司倾向于设置ERR 比 SFAS 132R 之前的时期低 0.237%。

原创性/价值

当公司面临强制性供款时，通常公司会经历负的股票市场回报。我们检查经理是否管理收益以减轻这种负面影响。我们发现，当面临强制性供款时，公司会夸大养老金资产的假设回报以提高其报告的收益。我们还发现，在 SFAS 132R 下的新养老金披露标准于 2003 年 12 月 15 日生效后，在强制性供款的情况下，管理者改变了盈余管理行为。 根据新的 SFAS 132R 要求，公司需要披露资产配置并描述投资策略。这限制了管理者在制定ERR 时的自由裁量权假设，因为现在养老金资产的构成是养老金资产假设预期回报率的关键决定因素。公司需要通过资产配置来证明其ERR 的合理性。