Accounting choices under ifrs and their effect on over-investment in capital expenditures

IFRS
period will exhibit a reduction in over-investment in 
PPE
after 
IFRS
adoption relative to pre-adoption levels because of more timely recognition of 
impairment losses under 
IFRS
strict impairment rules.
5
This argument is consistent with the International Accounting Standards Board’s (IASB) objective:
“ to develop, in the public interest, a single set of high quality, understandable and enforceable global 
accounting standards that require high quality, transparent and comparable information in financial 
statements and other financial reporting to help participants in the world’s capital markets and other users 
make economic decisions (International Accounting Standards Committee (IASC) Foundation, 
Constitution Part A.2).” 

---

5 
My sample is comprised of publicly-listed firms in nine EU countries that 
mandated 
IFRS
adoption in 2005. In order to minimize the self-selection bias related to 
managers’ incentives to voluntarily adopt 
IFRS
before 
IFRS
became mandatory in 2005 
(e.g., Christensen et al. 2008), I include firms only from those nine EU countries where 
voluntary adoption of 
IFRS
was not allowed prior to 2005.
6
I measure over-investment by 
examining whether a firm is more likely to over-invest using two partitioning variables: 
cash levels and free-cash flows. Prior literature (e.g., Jensen 1986; Harford 1999; Lie 
2000; Richardson 2006) shows that firms with high cash levels and high free-cash flows 
are more likely to over-invest. My results suggest that there is more timely loss 
recognition for 
PPE
following 
IFRS
adoption. In particular, I find that the frequency of 
impairment losses for 
PPE
is significantly greater in the post-
IFRS
period relative to the 
pre-
IFRS
period. I also find modest evidence of an increase in the asymmetric timeliness 
of loss recognition in earnings following 
IFRS
adoption. My main results indicate that 
over-investment in 
PPE
(or capital expenditures) is lower following 
IFRS
adoption 
among EU firms that used historical cost accounting with impairment testing in the post-
IFRS
period, consistent with EU firms having more timely loss recognition for 
PPE
under 
IFRS
strict impairment rules.
I further examine UK firms that had the option of using fair-value accounting 
(i.e., revaluation) for 
PPE
under UK GAAP. I find that most UK firms elected to use 
historical cost accounting with impairment testing for 
PPE
after 
IFRS
mandatory 
adoption. I also find that 30% of UK firms switched from fair-value accounting under 
UK GAAP to historical cost accounting with impairment testing under 
IFRS
and only 1% 
switched from historical cost accounting with impairment testing under UK GAAP to 
fair-value accounting under 
IFRS
.
6
See Table C2 (Panel A) for the list of these nine EU countries that did not allow firms to adopt 
IFRS
before it became mandatory in 2005. 

---

6 
Under both UK GAAP and 
IFRS
, increases in fair value for 
PPE
are recognized 
in a revaluation reserve in shareholders’ equity. Because most long-lived assets such as 
PPE
are heterogeneous in nature (i.e., firm-specific) and are not traded in liquid markets, 
fair-value estimates of 
PPE
are likely to exhibit less reliability than historical cost (Ball 
2006; Holthausen and Watts 2001; Kothari et al. 2010; Watts 2006).
7
Hence, I argue that 
the existence of a positive revaluation reserve in shareholders’ equity creates slack that 
self-interested managers can opportunistically use to offset impairment losses among 
assets and delay the recognition of impairment losses in earnings. On the other hand, UK 
firms that use historical cost accounting with impairment testing (under UK GAAP or 
IFRS
) have a zero balance in the revaluation reserve and, hence, self-interested managers 
will have no ability to opportunistically use previous upward revaluations to absorb or 
conceal impairment losses.
8
Thus, relative to fair-value accounting, historical cost 
accounting with impairment testing has greater disciplinary implications because 
impairment losses are recognized in earnings in a more timely manner. In other words, 
self-interested managers are expected to be more disciplined in their investment decisions 
when historical cost accounting with impairment testing is used. Therefore, I predict that 
UK firms that previously used fair-value accounting under UK GAAP and then switched 
to historical cost accounting with strict impairment rules under 
IFRS
will exhibit greater 
reductions in over-investment relative to other EU firms that used historical cost 
accounting with impairment testing prior to 
IFRS
adoption. Using difference-in-
differences and multivariate tests, I find results consistent with this prediction.
7
Barth et al. (2001, p.85-86) also argue that the reliability of fair-value estimates of long-lived assets “…is 
open to question because typically no market for these assets exists…”
8
UK firms that used fair-value accounting for