ii) with y ≥ 0, there might be a non-Ricardian regime, that is, a regime of fiscal
dominance.
Table 3 reports the estimation results of equation (28). The possibility of the fixed
effects model seems to get more statistical validation as one may confirm by the value
of the F statistic. This is a test of the null hypothesis that all effects are the same for
each country, in other words, the hypothesis that all autonomous terms ai for equation
(28) are identical.17
Table 3 - Estimation of equation (28), dependent variable: debt-to-GDP ratio (B )
Variable |
Pooled |
Fixed effects |
Random |
Constant (ai) |
2,456 (7,124) |
2,552 | |
γ (Sit-1) |
-0,618 |
-0,766 |
-0,716 |
φ (Bit-1) |
0,987 |
0,989 |
0,988 |
R |
0,9862 |
0,9874 |
0,9860 |
F a test |
3,904* (14,429) | ||
Hausman b test |
7.930 ** | ||
DW_____ |
1,049 |
1,241 |
1,066 |
The t statistics are in parentheses.
a - The degrees of freedom for the F statistic are in parentheses; the statistic tests the fixed
effects model against the pooled regression model, where the autonomous term is the same for
all countries, which is the null hypothesis.
b - The statistic has a Chi-square distribution (the degrees of freedom are in parentheses); the
Hausman statistic tests the fixed effects model against the random effects, which is here the null
hypothesis.
* - Statistically significant at the 1 percent level, the null hypothesis of the pooled regression
model is rejected.
** - Statistically significant at the 1 per cent level, the null hypothesis is rejected (random
effects model), that is, one rejects the hypothesis that the autonomous terms in each country is
not correlated with the independent explanatory variables (in this case the random effects model
does not produce unbiased and consistent estimators).
17 The F statistic is computed as F (n-1, nT-n-k)=[(Ru2-Rp2)/(1- Ru2)][(nT-n-k)/(n-1)], where u stands
for the model without restrictions, p denotes the pooled regression, that is the model with the
restriction that there is only one autonomous term, n is the number of countries, T is the number of
periods and k is the number of exogenous variables (see for instance, Greene (1997) and Johnston
and DiNardo (1997)).
25