Constructing Macroeconometric Models: The Case of Kazakhstan

Macroeconometric models designed for forecasting and simulation of the economy have proven to be an invaluable tool for policy makers the world over. The development of such models is an arduous task involving the collection of large amounts of data, estimation, calibration and simulation of the model over many sample sub-periods.

The National Bank of Kazakhstan is moving from a monetary policy which targets the exchange rate to one which targets inflation directly and has requested assistance in developing a macroeconometric model for policy analysis. The project was initially funded by the U.S. Agency for International Development and now is funded by the Asian Development Bank. David Kemme of the Fogelman College of Business & Economics is estimating and simulating a medium sized macroeconometric model of the Kazakhstan economy and guiding a team of researchers in the Research and Statistics Department of the National Bank of Kazakhstan.

The model Kemme is creating is designed for forecasting and policy simulation to assist the Monetary Policy Technical Committee in determination of medium term policy instruments. A preliminary version of the model has already been created, and the first forecasts by the team were presented at an intergovernmental roundtable in Astana, the capital of Kzakhstan.

The research team consists of David Kemme (dmkemme@memphis.edu), in the Economics Department, Warren Coats and Sabit Khakimzhanov, ADB consultants, and Indira Talkhanbayeva, Saida Agambayeva, Nurgaisha Turekhanova, Rufina Shagiakhmetova, Vitaly Tutushkin, Bebit Konurbayeva, and Aliya Algoshina, research economists at the NBK.

FCBE Researchers Examine Empirical Measures to Guide Monetary Policy Makers

The transition from central planning to markets has been a challenge for policy makers in Eastern Europe and the former Soviet Union. As the economies have evolved structurally, policies have evolved as well.

With respect to the evolution of monetary policy, most transition economies initially adopted a fixed exchange rate monetary policy regime. The fixed exchange rate served as a nominal anchor and if the currency is undervalued stimulates exports in the short run. As the macro economy stabilizes the fixed exchange rate cannot be maintained and the monetary authorities then begin targeting a monetary aggregate. This has also proven to be problematic since the economies started with very low levels of monetization and unstable money demand. A more appropriate strategy would be to target inflation directly, and the monetary authorities in several countries, Hungary, Poland and the Czech Republic in particular, have done so and have successfully reduced inflation to meet the Maastricht criteria for EU accession.

In making such policies there are many practical questions. First, in the movement from a fixed exchange rate to a floating rate with a money target, what should be the operational target? In other words, how much exchange rate volatility should be allowed and what changes in the monetary base are necessary to accommodate the resulting changes in foreign exchange reserves, and then what are the changes in domestic inflation resulting from the change in the monetary base?

Kemme and Gennady Lyakir, a Ph.D. student are examining the first policy transition in the Czech Republic by focusing on “exchange market pressure” which is a function of exchange rate change and monetary base changes. Typically policy makers move from a fixed exchange rate to a managed exchange rate to a float to avoid trade and capital flow shocks. However, if exchange rate changes are excessively dampened by increases in the monetary base there will be incipient inflationary pressures. If the exchange rate floats, reducing the inflationary pressures, there are negative effects on trade and capital flows. We estimate exchange market pressure and examine whether it is a useful guide for policy makers in the Czech context.

The research team from the Department of Economics consists of David Kemme (dmkemme@memphis.edu) and doctoral student Gennady Lyakir.

Forecasting Macroeconomic Activity in Countries with Limited Data

Researchers in the Fogelman College of Business & Economics are examining how to forecast macroeconomic activity with limited data. A major problem in the analysis of macroeconomic policy in most economies, and transition economies in particular, is the small sample size and large idiosyncratic variation.

A mixed-estimation method incorporating prior information from OECD (Organization for Economic Co-operation and Development) country data is developed to estimate parameters of a reduced form transition economy macroeconomic model. An exactly identified structural vector autoregressive model is constructed and impulse response functions (for a monetary policy shock) for three transition economies (the Czech Republic, Hungary and Poland) are estimated.

The method provides a systematic way to analyze monetary policy in transition economies where data availability is limited. The methodology is presented in a working paper of the St. Louis Federal Reserve Bank Research Department “Using Extraneous Information to Analyze Monetary Policy in Transition Economies,” http://research.stlouisfed.org/wp/2004/2004-034.pdf .

The research team includes David Kemme (dmkemme@memphis.edu), in the Economics Department, and William Gavin of the St. Louis Federal Reserve Bank.