Home Business & Finance
In this section we first show that our concerns about inference as a result of dynamic heterogeneity bias and the presence of cross-sectional dependence have been justified. We then present the results from our panel VAR model and explore a number of robustness checks. First we investigate to which extent our concerns about dynamic heterogeneity bias and cross-sectional dependence were justified. Figure 8.4 shows the draws of X, the estimated degree of dynamic heterogeneity. The mean of X is .0039 which is similar to the corresponding statistic those reported in Jarocinski (2010). In his Bayesian panel VAR application the assumption of pool-
Fig. 8.4 Impulse response functions to log-level net output shock—financial regulation
ing led to a substantial increase in the persistence of impulse responses. This suggests that our concern about dynamic heterogeneity bias and its potential impact on our inference in this case has been warranted.
Evaluating the FRc , t term at high and low values of financial liberalization permits us to obtain average VAR coefficients under regimes of high and low financial liberalization. The remaining interaction terms, KAc , t, FXRc , t and TRc , t are evaluated at the median values of their distribution. We can then obtain impulse responses under both regimes and compare them to assess whether the reaction of the current account varies with financial liberalization as predicted by the theory. The impact impulse response for the log-level of net output was normalized to 1 in order to ensure that we are comparing current account responses to log
Fig. 8.5 Impulse response functions to log-difference net output shock— financial regulation
level net output shocks of identical size across both regimes. This allows us to assess whether there is a statistically significant difference between the impulse responses due to financial liberalization, rather than to the size of net output shocks. Figure 8.5 shows impulse responses to a log-level net output shock with the financial liberalization index evaluated at both the 100 percent percentile (‘High Financial Liberalization’ column) and 0 percent percentile (‘Low Financial Liberalization’ column) of the distribution of this variable. These percentiles correspond to 0.1528 and 1.0, respectively. The third column shows impulse responses obtained from a distribution of the difference in the impulse responses obtained under the high and low financial regulatory regime. The red and green dashed lines indicate the 90 percent and 68 percent
Fig. 8.6 Impulse response functions to log-level net output shock—capital account openness
confidence bands, respectively. Figure 8.6 repeats this exercise but for a log-difference net output shock.
The median impact response of a log-level net output shock on the current account is 0.71 under and 1.29 under high financial liberalization, which represents an 81 percent increase in the impact of the shock if a country switches from high to low financial regulation (Fig. 8.5). As one can see in column three of Fig. 8.5, this difference is statistically significant. Furthermore looking at column three in Fig. 8.5, one can see that the median of the log net output response is not statistically significantly difference from zero throughout. This suggests that the difference in the current account to net output response cannot be attributed to changes in the nature of the net output shock. For a log-difference net output shock (Fig. 8.6), the median impact response is -0.64 under low and -1.14 under high liberalization (79 percent decline). Again, the third column suggests that the difference in impulse responses is statistically significant and that the net output impulse responses are statistically not very different from each other. In both cases, the absolute value of the current account response is larger and more persistent under in countries with high financial liberalization. Our theory predicts that, all else equal, the change in financial repression/liberalization should affect the impulse response of the current account balance to either type of shock in a similar way. Indeed, the change in the impact response and the persistence profile is similar across both shocks, which provides additional verification for the theory.
We repeat the same exercise for the other economic structure variables, namely capital account openness (KAc , t), the exchange rate regime (FXRc , t) and trade openness (TRc , t) in Figs. 8.7 and 8.8, 8.9 and 8.10 and 8.11, respectively. It is easy to see that the effect of the other inter-
Fig. 8.7 Impulse response functions to log-difference net output shock— capital account openness
Fig. 8.8 Impulse response functions to log-level net output shock—financial regulation controlling for FX regime
action terms is either not statistically significantly different from zero or only slightly significant for only one, but not both, of the shocks.
The empirical results from the panel VAR estimations provide support for the theoretical results derived in Sect. 8.2. Financial liberalization affects the size and persistence of response of the current account to net output shocks in a statistically significant way. This is true both for net output shocks in log-differences and log-levels. In simple words, we find that the reaction of the current account to net output shocks is larger and more persistent under higher financial liberalization (less financial repression).
Fig. 8.9 Impulse response functions to log-difference net output shock— financial regulation controlling for FX regime