X hits on this document

PDF document

Location Decisions of Foreign Banks and Institutional Competitive Advantage - page 33 / 33

105 views

0 shares

0 downloads

0 comments

33 / 33

Table 4 - First-differences test of institutional competitive advantage in foreign banking - Robustness tests

The table shows the results of several robustness tests. The results presented in this table refer to the estimates for the institutional competitive advantage variable in the different models. The benchmark model is the model whose results are shown in Table 3 first column. See for details of this model the main text and the explanation provided in Table 3. In the first two robustness tests we estimate the model using different estimation techniques (tobit and OLS), in both cases (1 plus) the dependent variable is transformed in logs. In the next two robustness tests we adjust the sample period, i.e. the dependent variable reflects the change in number of foreign banks from source country j in host country i between 1996 and 2000, and 2001 and 2006 respectively. In the following robustness test the dependent variable reflects the change in assets from source country j in host country i between 1996 and 2005. In the sixth robustness test we include regional dummies in the model. Robustness test seven uses the second lag for the explanatory variables instead of the first lag and is estimated for the period 1997-2006. In robustness test eight we exclude the change in host institutions and in trade as regressors. In the next two robustness tests we change the sample by including all countries available in our database except source and host countries that are offshore centers and only developing countries respectively. In the eleventh robustness check we take as denominator of the institutional competitive advantage variable the weighted averaged where the weight is determined by the country size proxied by the dollar GDP. In the final two robustness tests we use alternative variables to capture institutional competitive advantage: International Country Risk Guide composite indicator (ICRG) and the World Bank's Country's Policy and Institutional Assessment (CPIA) index. The robust p-values appear in brackets and ***, ** and * correspond to one, five and ten percent level of significance respectively.

1)

Benchmark

2)

Tobit

3)

OLS

4)

Sample 1996-2000

5)

Sample 2001-2006

6)

Change in assets

7)

Including regional dummies

8)

Second lag explanatory variables

9)

Excluding host institutions and trade

10)

All countries

11)

Only developing countries

12)

Weighted average

13)

ICRG

14)

CPIA

dInstCompAdv

  • -

    1.200 ***

[0.000]

  • -

    0.012 ***

[0.000]

  • -

    0.013 ***

    • [0.001

      ]

      • -

        1.205 ***

[0.000]

  • -

    1.025 ***

[0.000]

  • -

    1.479 ***

[0.000]

  • -

    1.069 ***

[0.000]

  • -

    1.260 ***

[0.000]

  • -

    1.378 ***

[0.000]

  • -

    1.318 ***

[0.000]

  • -

    1.132 ***

[0.000]

  • -

    0.534 ***

[0.000]

  • -

    0.512 ***

[0.000]

  • -

    0.515 ***

[0.000]

32

Document info
Document views105
Page views105
Page last viewedSun Dec 11 01:09:49 UTC 2016
Pages33
Paragraphs642
Words11833

Comments