Table 1. Estimation results, sectoral investment data
with country and industry fixed effects________
|
OLS |
Bayesian model averaging | ||||
|
Coefficient |
p-value |
posterior |
posterior mean |
posterior standard | |
|
inclusion |
conditional on |
error conditional on | |||
|
probability |
__________inclusion______ |
_________inclusion_______ | |||
|
Lagged investment |
0.865 |
0.000 |
1.000 |
0.880 |
0.166 |
|
Real interest rate |
0.009 |
0.130 |
0.654 |
0.006 |
0.003 |
|
Barriers to entry |
-0.046 |
0.008 |
0.998 |
-0.043 |
0.012 |
|
Public ownership |
0.004 |
0.701 |
0.151 |
0.000 |
0.002 |
|
Regulatory independence |
-0.026 |
0.556 |
0.184 |
-0.004 |
0.006 |
|
Incentive regulation |
-0.003 |
0.866 |
0.141 |
0.000 |
0.002 |
|
No price regulation |
-0.016 |
0.133 |
0.975 |
-0.015 |
0.008 |
|
interaction term_______________ |
0.060 |
0.002 |
0.996 |
___________0.053________ |
________0.014__________ |
|
Adj. R-squared Regressions run |
0.741 |
256 | |||
|
Prior inclusion probability |
0.5 | ||||
Note: posterior inclusion probability measures the extent to which any given model contributes to explaining the dependent variable as
compared to the other models. Bold figures for the posterior inclusion probability indicate that it is higher than the prior inclusion
probability of 0.5. The posterior mean conditional on inclusion is the mean of the individual OLS estimates weighted by the posterior
inclusion probability of the individual models including a given variable.
7. Concluding remarks
This paper studied the impact of the regulatory setup on sectoral investment in network industries.
Earlier empirical literature suggested that shifting away from traditional rate-of-return regulation
did not generally cause under-investment in network industries. For instance, overall investment
in the UK railway sector did increase in the aftermath of privatisation and the introduction of
incentive regulation. Empirical studies also highlighted that incentive regulation was very helpful
in promoting the deployment of new technologies in the US telecommunications sectors in the
late 1980s and early 1990s.
The empirical results in this paper based on cross-section OLS regressions and on Bayesian
model averaging suggest that the introduction of incentive price regulation or the establishment of
an independent sector regulator do not have a positive influence on investment by themselves.
However, once these policies are implemented jointly, they are associated with a significant
increase in investment. This result highlights the importance of the overall coherence of the
general regulatory environment in supporting investment incentives. Furthermore, the empirical
results show that lower entry barriers encourage investment in the network industries.
18