Comparison of the Efficiency of Pure and of Hybrid Inflation Targeting from the Point of View of Inflation Control

Abstract—

The article compares the effectiveness of various monetary policy regimes in terms of controlling inflation. It is shown that the hybrid inflation targeting regime, which combines the management of inflation as the main goal and the exchange rate as an additional one, makes it possible to simultaneously reduce the volatility of both inflation and the exchange rate. Thus, it is preferable to the pure inflation targeting regime in order to ensure the stability of the financial conditions for the development of the economy. Hypothesis testing was carried out by modeling the joint dynamics of inflation, exchange rate, and oil price volatilities in the VAR model. Volatility estimates were obtained using EGARCH models on monthly data for the period 1980–2021, for oil prices, 1992–2021, for the consumer price index, and 1995–2021 for the exchange rate.

Appendix A

The results of estimation of the VAR-model of the joint volatility dynamics, full specification

Index	Dependent variable
	$Vol\_in{{f}_{t}}$	$Vol\_e{{r}_{t}}$	$Vol\_oi{{l}_{t}}$
$Vol\_in{{f}_{{t - 1}}}$	0.93***
	(0.0233)
$Vol\_e{{r}_{{t - 1}}}$	0.090***	0.901***
	(0.027)	(0.033)
$Vol\_oi{{l}_{{t - 1}}}$	0.032	0.075	0.949***
	(0.044)	(0.069)	(0.013)
$inf\_cycl{{e}_{t}}$	0.304
	(0.689)
$\log \left( {e{{r}_{t}}} \right)$	0.002	–0.007
	(0.004)	(0.006)
$\log \left( {oi{{l}_{t}}} \right)$	0.0014	–0.003**	–0.003***
	(0.0012)	(0.002)	(0.0008)
$d\_09\_13$	–0.273***	0.294**	0.008
	(0.091)	(0.135)	(0.039)
$d\_14\_16$	–0.309**	0.444*	–0.022
	(0.171)	(0.234)	(0.044)
$d\_17\_21$	–0.326**	0.360*	–0.034
	(0.182)	(0.278)	(0.034)
R2	0.90	0.88	0.89
Residual sum of squares	12.9	40.3	7.4
Mean value of dependent variable	–10.9	–6.6	–4.3
Standard deviation of dependent variable	0.79	1.28	0.55
Covariance determinant of random error matrices		0.000412
Logarithmic maximum likelihood function		–62.5
AIC		0.788
SIC		1.120
Number of coefficients		21
Number of constraints		6
***, **, and * are the significance levels of 1, 5, and 10%, respectively. In parentheses under the estimated coefficients are their standard errors.

Appendix B

Analysis of the Impulse-Responses Functions

The analysis of the impulse-responses functions was carried out according to the most complete specification of the model in order to avoid the exclusion of important relationships during their construction. Function graphs are shown in Fig. 2. A significant response of inflation volatility to a shock (single innovation) of exchange rate volatility is delayed, begins to operate from the second month and continues for 20 months. A similar response of inflation volatility is also observed to the shock of oil price volatility, however, it is somewhat lower in level. The response of exchange rate volatility to an oil price volatility shock is immediate and remains significant for 8 months. A significant inertial response of inflation volatility to its own shock persists for 14 months, for the exchange rate for 12, and for oil prices for 20 months.

Fig. 2.

Inflation Volatility Impulse Response Functions, exchange rate and oil prices obtained using the Cholesky decomposition (responses to single innovations ± 2 standard errors).

The vertical axis represents the responses of the variables listed at the top of the table to single innovations in each of these variables. The boundaries of the confidence interval are determined by twice the standard error. The horizontal axis indicates the number of the period corresponding to the number of months that have passed since the innovation.