The name is absent

I start by focusing on the core of the theoretical study which is expressed by two hy-
potheses: We find a significantly negative effect of investments on the investor’s
termination rate. Dependent on the employed estimation model increasing invest-
ments by ten percentage points reduces the investor’s termination rate for incoming
calls by about 0.25 to 0.47 percentage points. Rearranging equation (4) the theoretical
model predicts that a ten percentage points investment increase results in an increase
in cost efficiency of about 0.5 to 0.94 percentage points.

Moreover, concerning the direct impact on incoming traffic no significant coefficients
could be identified what corresponds to the results of the theoretical model. Never-
theless, investments indirectly affect traffic as at least the coefficients of the combined
estimation approach in column (6b) provide evidence for a negative termination rate
coefficient. I come back to this finding when calculating the effect of investments on the
investor’s short-run profit. So far, the estimation results confirm the first hypothesis
concerning own investment effects.

With regard to the second hypothesis we observe a decrease in termination rates and
an increase in incoming traffic to competitors due to investments. While the pos-
itive investment effect on traffic is in line with the outcome of the theoretical model,
following theory a significantly negative effect on competitors’ termination rates was
only identified in line with LRIC regulation.

A more detailed analysis of alternative investment effects due to regulation schemes re-
quires the consideration of interaction effects between investments and regulation. Table
5 provides the extension of the estimations above where I have replaced the regulation
dummies by their interaction terms with investments employing the 3SLS estimation
method. Columns (1) and (2) are the results where either cost-based or incentive regu-
lation are compared to "no regulation", i.e. I exclude the other regulation scheme from
the observations. In column (3) I keep both regulation schemes in the data.

We again find support for the first hypothesis on the investor’s own termination rates
and incoming traffic. Moreover, also the effect on incoming traffic to competitors’ net-
works is found to be positive. Nevertheless, no evidence could be found concerning the
expected outcome on competitors’ termination rates. While the direct common invest-
ment effect is in the range of the previous estimations, particularly the interaction term
is found to be positive and even (weakly) significant for the first approach.

Thus, the estimation results support the second hypothesis (investment externalities)
with regard to the quantity of incoming traffic effect. Nevertheless, concerning the ef-

23

<<   26   27   28   29   30   31   32   >>

More intriguing information