Economic results in response in brazlian money


. Econometric results
This section presents the results of the fiscal response function estimations based on the Markovswitching model. Table 1 presents the estimations of the parameters with p-values16 from equation (8)
for a specification of the Markov-switching intercept autoregressive heteroskedasticity (MSIAH) model,17
while annex A1 shows the graphics of the smoothed probabilities or, in order words, the chronology of
regimes.18 After evaluating several competing models, the two-state or two-regime model was found
to be the best fit with the data, based on the different specification tests.19 In effect, application of the
likelihood-ratio (LR) test rejected the null hypothesis of linearity (LR = 189.65,X2(7) = [0.000]**

 e X2(8)
= [0.000]**).20
14 Galí and Perotti (2003) use a forward-looking specification. However, this distinction is not particularly relevant in the case of
single-equation models, as it can be proved that, in this case, the forward-looking specification always has backward-looking
equivalents (Sims, 1999 and 2001).
15 Not least the global economic crisis of 2007–2008, which naturally had repercussions on the Brazilian economy.
16 Standard error statistics were computed numerically using the Hessian matrix of maximum log likelihood function. Unfortunately,
these approximations can be somewhat imprecise.
17 A specification that allows changes in the intercept (I), the parameters of variables (A) and variances (H) in each egime. This
terminology was adopted by Krolzig (1997).
18 Smoothed probability takes into account information from the whole sample and is defined as follows: pr [St
=j|ΨT], where ΨT is
the complete set of information up to moment T. Filtered probability is an optimal inference on the state of the variable at time
t considering the information up to t, while predicted probability considers the information up to t-1.
19 No major specification problem was found upon conducting traditional Durbin-Watson serial correlation, normality and autoregressive
conditional heteroskedasticity (ARCH) tests. Introducing a larger number of regimes led to issues with the numerical optimization
routine, making the transition probability matrix non-ergodic, thereby violating one of the main hypotheses of the model.
20 The likelihood-ratio (LR) test shows a non-standard distribution and cannot be characterized analytically as the transition probabilities
are not identified under the linearity hypothesis. It is possible, however, to show that this distribution can be approximated, as
it is in the interval between two chi-squares. We derive from this that if the distribution rejects the null hypothesis, the LR test
must do so as well. Conversely, if neither rejects the hypothesis of linearity, then the LR test will not do so either. In any other
situation there is nothing to be said (Davies, 1977).

2 CEPAL Review N° 135 • December 2021
Fiscal and monetary policy rules in Brazil: empirical evidence of monetary and fiscal dominance
Table 1
Model MS(2)-AIH(1)
Dependent variable: PRIM
Regime 1 Regime 2
Constant -1.333 (0.183) -1.7741 (0.317)
DLSP(-1) 0.221 (0.284) 0.299 (0.000)
INFLA12(-1) 0.453 (0.093) 0.065 (0.000)
TXPIB12(-1) 0.637 (0.000) 0.341 (0.000)
Standard deviation 0.123 (0.0000) 0.048 (0.000)
Observations 156
Likelihood 197.923
Source: Prepared by the authors, on the basis of data from the Central Bank of Brazil.
Note: p-value in brackets.
The results shown in table 1, as well as the smooth probability graphics in annex A1, support
the affirmation that there is a clear difference in the conduct of fiscal policy between the two regimes.
Analysis of the results in table 1 yields the following remarks. A clear difference is seen in the response
of the primary balance with respect to debt between the two states. Although net public sector debt
has a positive sign in both regimes, it is only significant in regime 

2. This means that the fiscal authority
reacted to the rise in debt in this regime, which supports the interpretation that it pursued a fiscal target
in regime 2. Conversely, in regime 1 fiscal policy did not respond to the rise in public debt, showing an
active fiscal policy.
To further defend this thesis, annex A1 also includes the graphic for the ratio between net debt
and GDP (public sector net debt). Comparing this graphic with the chronology of regimes, it may be
seen that in December 2013 the downtrend in public sector net debt was reversed. In that period,
regime 1 comes into play, where the Treasury ceases to treat net debt as a fiscal policy objective. In
fact, between December 2013 and December 2015 public sector net debt rose by six percentage
points. Between end-2002 and early 2009 regime 2 predominated, in which public sector net debt was
statistically significant. It is interesting to note that public sector net debt trended strongly downward
almost throughout this period.
Coming back to the analysis of table 1, in both regimes public sector net debt responds positively
to output, TXPIB12, which seems to indicate that fiscal policy was countercyclical. However, the fiscal
authority responds less strongly to output growth in regime

 2 than in regime 1. In this case, there are
signs that during regime 1 the fiscal authority may concern itself more with economic performance than
with fiscal targets, which suggests an active fiscal policy in regime 1 and a passive one in regime 2.
Although the computed coefficient of inflation is positive in both regimes, its level of significance
in regime 1 is low, as it is marginally significant at 10%, while in regime 2 it is statistically significant at
1%. However, the response to inflation by the fiscal balance is much lower in regime 2.
Given the foregoing, the results show empirical evidence that in regime 2, the share of the fiscal
surplus in GDP reacts strongly to increases in the ratio between public debt and GDP, while in regime 1
the fiscal indicator shows no response to changes in that ratio. In addition, in regime 2 responses by
the primary surplus/GDP indicator to the rate of inflation and to output growth are much weaker than
in regime 1. It may thus be observed that although fiscal policy adopts a passive stance in regime 2
—insofar as the coefficient between debt and GDP is positive—

 it cannot be said to be 100% passive,
as it responds less to changes in the rate of inflation and output growth than in regime 1. Symmetrically,
regime 1 shows a strongly active stance: this is a necessary but not a sufficient condition to identify a
regime of fiscal dominance. For that, monetary policy would have to be passive during the same period
of regime 1, following fiscal policy.
CEPAL Review N° 135 • December 2021 93
Tito Belchior S. Moreira, Mario Jorge Mendonça and Adolfo Sachsida
Moving to analysis of the transition probability matrix (see table 2), it may be seen that once the
economy is within one of the two regimes, it has a high probability of remaining there. In this case, the
null hypothesis supposes that migration can occur from one regime to another. However, if the fiscal
policy rule is present in regime 1, the computed probability of moving back to regime 2 is very small.
Nothing can be said of the opposite case, because the p-value of the computed probability of transition
from regime 2 to regime 1 is not significant. The fact that the probability of return to regime 2 is small
when the fiscal rule obtains in regime 1 suggests that deterioration of the fiscal framework may become
structural and ingrained, and thus difficult to reverse.

legal consultations and travel advisor in the States and within UK

Media solutions , Media company , online classes , learn german , learn english , perfect language , blood cord , rehab , rehabiliations , rehabilitation center , magazitta

Post a Comment

Previous Post Next Post

Contact Form