Pension liabilities and conservative financial reporting*

ABSTRACT

We investigate the relationship between a firm’s conditional financial reporting policy and the tendency to understate pension liabilities in the statement of financial position. We focus on two primary actuarial assumptions for pension liabilities, the discount rate and the salary growth rate, because a small bias in these two actuarial assumptions has a big effect on the estimate of pension liabilities and firms have some latitude in choosing these rates. We predict and find that firms are less likely to choose an upwardly biased discount rate and/or a downwardly biased salary growth rate when their earnings reflect bad news on a timelier basis than good news. Even though the salary growth result is somewhat weak, overall results suggest that firms are less likely to understate their pension liabilities by choosing pension obligation increasing actuarial assumptions when their financial reporting is conditionally more conservative.

KEYWORDS:

• Pension accounting
• defined benefit obligation
• accounting conservatism
• actuarial assumptions

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The effect of pension assumptions on pension liability is large. For example, Samsung Electronics reports in the notes to its 2015 financial statements that a one percent increase (decrease) in the pension discount rate would give rise to a decrease (increase) of approximate 11% (13%) in pension obligations.

2. To ease concerns about the pervasive pension deficits in Korea and to protect employees from creditors in the event of corporate bankruptcy, ERBSA requires Korean firms to increase their external funding gradually. The minimum external funding ratio is set to 80% in 2016 and will increase to 90% in 2018 and eventually to 100%.

3. George Coats (2010). ‘Korea’s pension explosion,’ on 4 October 2010, from https://www.professionalpensions.com/global-pensions/feature/1740346/korea-s-pension-explosion.

4. Gopalakrishnan and Sugrue (1995) report that a 1% increase of the discount rate would decrease the pension liability by about 20%. Similarly, in Korea, a 1% increase of the discount rate decreases pension obligations by approximately 13.4% (Yoon et al. 2014). For instance, Amorepacific Company disclosed in the notes to their 2014 financial statements that a 1% increase of the discount rate would decrease pension obligations by 7.9% and a 1% increase of the salary growth rate would result in an 8.9% increase in pension obligations.

5. Our regression results (Panel A of Table 2) show that corporate governance is not associated with the discount rate, but the level of conditional conservatism is. It appears that the effect of accounting conservatism on actuarial assumptions is distinct from that of corporate governance.

6. Recently, several studies raise the validity issue of the Basu (1997) model in capturing conditional conservatism. Some opponents argue that the Basu (1997) coefficient is biased (e.g. Patatoukas and Thomas 2011). However, proponents of the Basu model do not agree with those arguments based on analytical and empirical evidence (e.g. Basu 1997; Ryan 2006). Ball, Kothari, and Nikolaev (2013) claim that the inclusion of firm-fixed effects can remove the biases documented by Patatoukas and Thomas (2011).

Additional information

Funding

This work was supported by the Korea University Business School [Research Grant].