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

PPE
after 
IFRS
mandatory adoption. Separation of ownership 
and control gives rise to agency conflicts between managers (insiders) and outside 
shareholders (e.g., Berle and Means 1932; Jensen and Meckling 1976; Ang et al. 2000). 
Hence, as managerial ownership declines, the interests of managers and outside 
shareholders are less aligned and, as a result, the severity of the agency problem 
increases.
37
Prior literature (e.g. LaFond and Watts 2008; LaFond and Roychowdhury 
2008; Ahmed and Duellman 2010) shows that timely loss recognition is a governance 
mechanism that mitigates agency conflicts between managers and outside shareholders. 
Therefore, if EU firms are indeed using timely loss recognition (i.e., historical cost 
accounting with strict impairment rules) after 
IFRS
adoption to mitigate agency conflicts, 
then I expect that as managerial ownership declines, the greater will be the effect of 
timely loss recognition on reducing over-investment, ceteris paribus. To test this 
conjecture, I introduce into equation (4) two indicator variables: 
HIGH
and 
LOW
. 
HIGH
equals one for firms that are in the highest quartile of closely-held shares 
(
CLOSELY_HELD
), and zero otherwise. 
LOW
equals one for firms that are in the lowest 
quartile of 
CLOSELY_HELD
, and zero otherwise. I then interact each of these two 
indicator variables with 
IFRS*OVER_INV
to capture the effect on over-investment after 
IFRS
adoption among firms with high and low managerial ownership. Specifically, I
37
Following LaFond and Roychowdhury (2008), I measure separation of ownership and control by the 
percentage of the firm owned by managers (insiders). This is consistent with long-standing arguments in 
the agency theory (see Jensen and Meckling 1976 and Demsetz 1983) 

---

35 
estimate the following regression: 
CAPEX
it
 = β
0
 + β
1
IFRS
it
 + β
2
OVER_INV
it
 + β
3
IFRS
it
*OVER_INV
it
 + β
4
HIGH
it

+ β
5
LOW
it
 + β
6
IFRS
it
*HIGH
it
 + β
7
IFRS
it
*LOW
it
+ β
8
HIGH
it
*OVER_INV
it
 
+

β
9
LOW
it
*OVER_INV
it
 
+
 
β
10
IFRS
it
*HIGH
it
*OVER_INV
it

+

β
11
IFRS
it
*LOW
it
*OVER_INV
it
 + β
12-21
CONTROLS
it
+ β
22-31
CONTROLS
it
*OVER_INV
it
 
+ 
ε
it
 
(8)
 
Consistent with my conjecture, I predict that the estimated 
β
10
coefficient on the 
interaction between 
IFRS
, 
HIGH
, and 
OVER_INV
to be negative but insignificant and the 
estimated 
β
11
coefficient on the interaction between 
IFRS
, 
LOW
, and 
OVER_INV
to be 
significantly negative. 
Table C9 presents the multivariate results based on equation (8) of the effect on 
over-investment after 
IFRS
mandatory adoption. As predicted, the estimated 
β
3
coefficient on the interaction between 
IFRS
and 
OVER_INV
is significantly negative. 
However, I find that the estimated 
β
10
coefficient on the interaction between 
IFRS
, 
HIGH
, 
and 
OVER_INV
is not significant. This finding suggests that EU firms with high 
managerial ownership, where there is greater alignment of managers’ and shareholders’ 
interests (less agency conflicts), have no incremental reductions in over-investment in the 
post-
IFRS
period relative to other EU firms. Consistent with my prediction, I find that the 
estimated 
β
11
coefficient on the interaction between 
IFRS
, 
LOW
, and 
OVER_INV
is 
significantly negative. This finding suggests that EU firms with low managerial 
ownership have greater improvement in investment efficiency (i.e., greater reductions in 
over-investment) in the post-
IFRS
period relative to other EU firms. Further, the 
estimated 
β
10
coefficient for high managerial ownership firms is significantly different 
from the estimated 
β
11
coefficient for low managerial ownership firms (F = 2.62; p < 0.1). 
This result indicates that EU firms with low managerial ownership exhibit greater 
reductions in over-investment in the post-
IFRS
period relative to EU firms with high 

---

36 
managerial ownership.
38
Overall, these findings are consistent with my conjecture that 
the effect of using historical cost accounting with strict impairment rules (i.e., more 
timely loss recognition) on reducing over-investment after 
IFRS
adoption is greater as 
managerial ownership declines. Further, these findings imply that outside shareholders of 
EU firms appear to be demanding timely loss recognition as a means of addressing 
agency conflicts with managers after the mandatory adoption of 
IFRS
. Therefore, these 
findings corroborate those of LaFond and Roychowdhury (2008) who find that outside 
shareholders of U.S. firms demand greater conservative financial reporting (i.e., more 
timely loss recognition) to mitigate agency conflicts arising from greater separation 
between ownership and control.