METHODOLOGY AND DATA
Methodology
The five proposed hypotheses are tested with the following simultaneous equations model. The
simultaneous equations model is employed because this study is to test the simultaneous relationships between
internationalization, leverage, and profitability. Moreover, as discussed in the literature review, internationalization
and capital structure could be interrelated and moderate each other’s influence on firm performance. Therefore, a
simultaneous equations model consisted of the following three equations for the purpose of this study.
Equation 1:
1
1
2
2
1
0
ε
α
α
α
+
Α
+
+
+
=
Z
LEV
LEV
INTL
Equation 2:
2
2
2
2
1
0
ε
β
β
β
+
Β
+
+
+
=
Z
INTL
INTL
LEV
Equation 3:
3
3
3
5
2
4
3
2
2
1
0
*
ε
γ
γ
γ
γ
γ
γ
+
Γ
+
+
+
+
+
+
=
Z
LEV
INTL
LEV
LEV
INTL
INTL
ROA
, where INTL: level of internationalization; LEV: leverage; ROA: profitability of international hotels measured by
Return on Assets; Z
1
, Z
2
, and Z
3
are vectors of controlling variables of equation 1, 2, and 3 respectively; A, B, and
Γ
are coefficient vectors for Z
1
, Z
2
, and Z
3
respectively;
ε
1
,
ε
2
, and
ε
3
are error terms for equation 1, 2, and 3
respectively.
Markides and Williamson (1994) treated their system of equations as recursive rather than endogenous because they
believe that feedbacks in the system occur at sufficiently long lags to allow them to pull out individual equations for
separate treatment. In this study, because a firm’s leverage and internationalization are intertwined, the accurate
impact of internationalization can not be presented without accounting for such interlinks. Econometrically, the
remedy for such problem is to use the Seemingly Unrelated Regressions (SUR) procedure, which addresses the
existence of contemporaneous correlation between equations (Singh & Nejadmalayeri, 2004). However, the possible
reciprocal relationship between internationalization and capital structure decisions also means that endogeneity
could exist in the system. Moreover, the inclusion of internationalization and leverage as regressors in the
profitability equation (eq. 3) would cause endogeneity problem. The endogeneity will be checked by the Hausman
test on all pairwise combinations of three dependent variables: internationalization, leverage, and profitability. If the
null hypothesis of none endogeneity is rejected, the Three-Stage Least Squares (3SLS) procedure, instead of the
SUR, will be used to account for the possible endogeneity. The 3SLS procedure takes into account the information
present both within and across the proposed equations (Davidson & MacKinnon, 2004). The detail procedures of
testing hypotheses using the results of 3SLS are outlined as follows: Hypothesis 1 and 2, (
Internationalization and
Do'stlaringiz bilan baham: |