An Assessment of Macroeconomic Performance Indicators in Kazakhstan Via Factor Models

Abstract

We use factor models to assess accuracy of macroeconomic performance indicators in Kazakhstan over 2011-2018. Specifically, the annual GDP growth rate and unemployment rate are analyzed based on the fundamental factors underlying the economy. The model is a version of dynamic factor models with exact factor structure. We exploit a two-step estimator that provides consistent estimates of the factors when n ! 1 and T ! 1. The findings indicate that there are two fundamental factors responsible for driving real and nominal variables in the economy. These two factors are just vectors whose entries were estimated via principal components and Kalman Filter. Although the growth rate of GDP from public sources indicate that there has not been a recession in the country, we find that the alternative GDP growth rate implied by the fundamental factors points to the recession in 2016. In addition, the finding is supported by the backcasts of unemployment rate which shows us that the unemployment rate started rising after the severe decline in oil prices that negatively affected the Kazakhstani economy. We also find that the official unemployment rate was below the backcasts of unemployment rate by 2 percentage points on average.