Note: For each bank, the first line contains the dates of the different privatization phases.
As for merger experiences, only one took place in Portugal in the last decade: the Banco BPI is the bank resulting from the merger of Banco Borges & Irmão (BBI), Banco Fonsecas & Burnay (BFB) and Banco de Fomento e Exterior (BFE) which took place in the first half of 1998. Our data set is for the period 1990-97; hence we do not have any information for Banco BPI. Therefore, we used 1997 data for the 3 above-mentioned banks to simulate this merger operation. This operation will be referred to as case 1. We also use 1997 information to simulate 3 other possible merger operations that one way or another the media has been talking about in the last two months. Hereafter, they will be referred to as cases 2, 3 and 4, and involve the following simulations:
Case 2: CGD+BNU
Case 3: BCP+BPSM
Case 4: BTA+CPP+BSN.
We will assume two different scenarios, respectively A and B, in order to test the sensibility of our results. Under scenario A, the resulting bank produces output levels equal to the arithmetic sums of the individual banks involved in the simulated merger. Under scenario B, the resulting bank uses 3 inputs in order to produce outputs which are lower than the sum of individual banks’ input and output levels7. However, we also assume that there will not be any laid-off employees (labor legislation prevents that) and that there will not be any branch closures8. We also assume that there are no further cost synergies from the (restructured) product mix. The two scenarios are summarized in table 5.
Table 5: Merger scenarios.
Variable | Scenario A | Scenario B |
Y1
|
Sum of individual banks
|
Sum of individual banks
|
Y2
|
Sum of individual banks
|
50% of the sum
|
Y3
|
Sum of individual banks
|
50% of the sum
|
X1
|
Sum of individual banks
|
50% of the sum
|
X2
|
Sum of individual banks
|
Sum of individual banks
|
X3
|
Sum of individual banks
|
75% of the sum
|
W1
|
Average of individual banks
|
Average of individual banks
|
W2
|
Average of individual banks
|
Average of individual banks
|
W3
|
Average of individual banks
|
Average of individual banks
|
We have used all three models to simulate these merger operations. Results are summarized in table 6. In general, our simulations show that any of the mergers can generate substantial cost savings, which are lower for the BCP+BPSM case. The most successful cases seem to be the first two, that is, the BFB+BBI+BFB and CGD+BNU.
Table 6: Merger simulation – Inefficiency indices
| BFB+BBI+BFE |
CGD+BNU
|
BCP+BPSM
|
BTA+CPP+BSN
| DEA* |
0.256
|
0.119
|
0.059
|
0.116
|
Scenario A
|
0.055
|
0.000
|
0.000
|
0.000
|
Scenario B
|
0.075
|
0.000
|
0.040
|
0.000
| SFA-T* |
0.164
|
0.126
|
0.225
|
0.172
|
Scenario A
|
0.000
|
0.000
|
0.000
|
0.000
|
Scenario B
|
0.000
|
0.000
|
0.000
|
0.000
| SFA-E* |
0.116
|
0.134
|
0.077
|
0.079
|
Scenario A
|
0.003
|
0.003
|
0.003
|
0.003
|
Scenario B
|
0.000
|
0.000
|
0.000
|
0.000
|
* Effective 1997 unweighted average of the banks involved.
Share with your friends: |
