16
year effects. The results are not reported; however, we find year effects but not industry effects.
The year effects on firm disclosure levels are pronounced; they are significant at a 5 percent level
for 2001–2002 and 2004–2005; the year effects on firm performance levels appear in 2002 and
2004–2005. We believe these year effects may be associated with Chile’s adopting the
International Financial Reporting Standards (IFRS), which may positively influence firm
performance and disclosure levels in recent years.
Second, previous studies suggest that when using time series cross-sectional regressions,
if the relationship between insider ownership levels and firm performance is endogenously
determined, we cannot make inferences about the causality of the relationship. To address this
concern, we test for the existence of an endogenous relationship by evaluating whether changes in
insider ownership and institutional ownership levels are related to changes in firm performance.
According to Lemmon and Lins (2003), this is one of the best ways to address concerns of an
endogenous relationship in time series data. The revised model specification for our robustness
test is as follows:
ΔQ
it
or ΔDIS
it
= β
0
(Constant) + β
1
SIZE
it
+ β
2
PPE
it
+ β
3
DEBT
it
+ β
4
ΔOWN
it
+ β
5
ΔOWN
2
it
+
β
6
ΔAFP
it
+ β
7
ΔNon-AFP
it
+ β
8
PEB
it
+ e
it
Where,
ΔQ
= change in Tobin’s Q from (t-1) to t
ΔDIS
= change in disclosure index from (t-1) to t
SIZE
= natural logarithm of total assets
PPE
= property, plant and equipment as a percentage of total assets
DEBT
= total debt
ΔOWN
= change in insider ownership from (t-1) to t
ΔAFP
= change in pension fund or AFP ownership percentage from (t-1) to t
ΔNon-AFP
= change in other (Non-AFP) institutional ownership percentage
Convention