Our objective is to compare the posterior distribution of the primary surplus response to
debt across different sub-samples. To that end, we combine according to the Bayes rule
the prior information in (2.11) with each of the following three sub-sample periods: 1977-
1991, 1992-1999, 2000-2005. Doing this, we can assess the extent to which the
information in each of these sub-periods is compatible with the estimates for the overall
sample response in Table 2.6. The resulting posterior densities are reported in Table 2.7,
with the corresponding graphical representation in Figure 2.2.
Table 2.7 Posterior Distributions for δ
|
_______________________1977-1991________ |
_________1992-1999________ |
________2000-2005________ | ||||
|
Mean |
Standard Deviation |
Mean |
Standard Deviation |
Mean |
Standard Deviation | |
|
Belgium |
0,05 |
0,022 |
0,02 |
0,019 |
0,04 |
0,027 |
|
Germany |
0,07 |
0,043 |
0,04 |
0,023 |
0,02 |
0,010 |
|
Greece |
-0,02 |
0,035 |
0,07 |
0,049 |
0,00 |
0,032 |
|
Spain |
0,04 |
0,025 |
0,04 |
0,043 |
0,02 |
0,015 |
|
France |
0,07 |
0,028 |
0,07 |
0,032 |
0,04 |
0,042 |
|
Italy |
0,04 |
0,020 |
0,03 |
0,032 |
0,02 |
0,014 |
|
Ireland |
0,06 |
0,027 |
0,00 |
0,020 |
0,10 |
0,041 |
|
Netherlands |
0,04 |
0,020 |
0,01 |
0,033 |
0,05 |
0,057 |
|
Austria |
0,03 |
0,018 |
0,01 |
0,040 |
0,02 |
0,044 |
|
Portugal |
0,15 |
0,058 |
0,11 |
0,008 |
0,10 |
0,070 |
|
Finland |
0,05 |
0,105 |
0,06 |
0,046 |
0,11 |
0,031 |
|
Denmark |
0,08 |
0,038 |
0,01 |
0,015 |
0,05 |
0,075 |
|
Sweden |
0,10 |
0,051 |
0,10 |
0,097 |
0,09 |
0,056 |
|
UK |
0,05 |
0,039 |
0,16 |
0,053 |
0,10 |
0,029 |
|
US |
0,01 |
0,026 |
0,03 |
0,016 |
0,03 |
0,045 |
|
Japan_________ |
0,05 |
0,017 |
-0,02 |
0,028 |
-0,02 |
0,030 |
A general feature of the information in Table 2.7 is that it conveys evidence of time
variation in the response to debt accumulation in most countries. More specifically, we
detect a rather clear shift across sub-periods of the posterior distribution in the cases of
Belgium, Germany, Greece, France, Ireland, Netherlands, Finland, Denmark, UK and
Japan. The rest of the countries in the panel seem to display more stability. On the other
hand, in most euro area countries the shift in the posterior may be associated with the
economic and monetary integration effects modelled as dummy effects in Table 2.6. Only
in the cases of Spain and Austria the integration effects identified in Table 2.6 are not
visible in Table 2.7.
Looking at particular countries, some interesting features are noteworthy. First, the results
for Germany and France nicely reflect the progressive deterioration of fiscal discipline
across sub-periods, especially in the case of Germany. Second, the posterior is centred on
a non-positive mean in the cases of Greece (except in 1992-99) and Japan (except in
1977-91). Third, Finland now displays a progressive and significant improvement in
terms of the sustainability of its fiscal policy. The difficulties visible during the period
1977-91 are overcome in the following two sub-periods, ending with a current sound
fiscal position. This is in sharp contrast with the overall conclusion in Table 2.6.
Overall, Table 2.7 conveys a similar qualitative message of Table 2.6: the response of the
primary surplus to accumulated debt has been generally positive across countries and sub-
sample periods, except in Greece and Japan. This result reinforces the robustness of the
42