MONETARY POLICY OPTIMIZATION BASED ON THE DSGE MODEL OF KAZAKHSTAN'S ECONOMY

Authors

  • D. Shults Russian Foreign Trade Academy
  • A. Kyssykov Center of applied research «Talap», Kazakhstan

DOI:

https://doi.org/10.26577/be-2019-4-e1
        37 70

Abstract

The model describes the economy in the short term (excluding investments), in the case
of inflation targeting policy and represents a system of 15 linearized equations for key macroeconomic
indicators of the main economy sectors: households, enterprises, the National Bank, and the external
sector. The parameters were estimated by Bayesian methods for the period 2010-2018 and in the subperiod
2015-2018. The advantage of the approach is the possibility of estimating parameters in short
time series due to the use of prior information. From the estimates obtained, it follows that the
National Bank pays attention not only to inflation, but also to business activity and changes in the
exchange rate. As is known from theory, the optimal policy for the monetary regulator may differ from
the optimal one for society. To determine the parameters of the optimal monetary policy, the function
of social losses was derived and it was shown that, in addition to the traditional variables of the output
gap and inflation, the fluctuations in the interest rate and exchange rate should be its components. The
work takes into account some sources of welfare losses. These average annual losses of society are
estimated at 3.2% of the equilibrium level of consumption. The optimization carried out according to
the current version of the DSGE-model allows us to draw the following conclusions. A “double
mandate” policy and the inclusion of an exchange rate in Taylor’s equation can increase public
welfare. The sensitivity coefficients of the current interest rate policy can be revised upward, due to
which society losses can be reduced. When pursuing a monetary policy, one should focus not on the
CPI, but on indicators of internal inflation, perhaps an indicator of core inflation and/or PPI.

Downloads

Published

2019-12-23