Table 4 - Estimation of equation (29), dependent variable: first difference of the
|
____________________debt-to-G |
)P ratio ( b )_______________________ | ||
|
Variable |
Pooled |
Fixed effects _____model |
Random |
|
Constant (xi) |
0,396 (2,145) |
0,398 | |
|
k (Sit-ι) |
-0,339 |
-0,360 (-3,708) |
-0,346 |
|
w (bit-1) |
0,594 |
0,567 |
0,586 |
|
R'' |
0,4077 |
0,3963 |
0,4077 |
|
F α test |
0,460 (14,413) | ||
|
Hausman b test |
2,557 | ||
|
DW_____ |
2,083 |
2,068 |
2,068 |
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.
Also, it is possible to think of another test to assess empirically the FTPL. As already
mentioned, the FTPL stresses the point that the price level could be determined by
equation (25). For instance, if the government raises taxes then, according to that
equation, there should be a price level decrease, resulting from the fact that the fiscal
surplus is also higher. That is, if it is possible to observe a sustained negative correlation
between prices and fiscal revenues one could conclude that there is in place a non-
Ricardian fiscal policy, supporting the idea of the FTPL. However, if there is instead a
regime of monetary dominance, the price level should be independent from the
evolution of the government revenues.
A possible specification to test the hypothesis mentioned above could be the following
equation
Pit = λi + ωpPit_1 + ωRRit_1 + εit,
(30)
where P is the average annual change of the price deflator of private final consumption
expenditure and R is the first difference of total public receipts as a percentage of GDP.
28