Abstract
The econometric results show that it is feasible to estimate robust price and inflation equations for Georgia. The long-term price equation expresses prices as a function of money, the exchange rate, and real income and may be interpreted as portraying equilibrium in the goods market. The paper also represents statistical data of transportation indicators, population and employment, personal income tax, monetary survey, average monthly wages, developments in commercial banking, interest rates, prudential indicators of commercial banks, balance of payments, and so on.
I. Inflation in Georgia1
A. Macroeconomic Developments
1. Georgia experienced one of the highest inflation rates among the BRO countries after the creation of a national currency—the coupon—in April 1993.2 A stabilization program in mid-1994 brought an end to hyperinflation, and introduction in October 1995 of a new currency (the lari) replacing the coupon boosted demand for money. Since then, the National Bank of Georgia (NBG) has conducted a prudent monetary policy, focusing on maintaining price stability. The lari was pegged de facto to the U.S. dollar between October 1995 and December 1998 and price stability helped to remonetize the economy somewhat, although monetization has remained low and dollarization high.
2. In the final months of 1998, the onset of the Russian crisis and widespread public concern regarding domestic budgetary problems led to a sharp decline in the demand for lari (broad money declined by 25 percent in nominal terms from August to November) and growing pressure on the pegged exchange rate. To defend the lari, the NBG intervened heavily in the foreign exchange market (Figure I-1), increased banks’ reserve requirements, withdrew liquidity through interbank auctions, and suspended automatic intra-month budget financing. The attempt was unsuccessful, and after running foreign reserves down to the equivalent of 3 weeks of imports, the NBG allowed the lari to float on December 7, 1998. The lari/dollar exchange rate immediately dropped by 20 percent and monthly inflation soared to 12 percent (Figure I-2).
Lari/U.S. Dollar Exchange Rate and NBG Interventions at the Tbilisi Interbank Currency Exchange
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.Lari/U.S. Dollar Exchange Rate and NBG Interventions at the Tbilisi Interbank Currency Exchange
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.Lari/U.S. Dollar Exchange Rate and NBG Interventions at the Tbilisi Interbank Currency Exchange
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.
Georgia: Logarithmic Changes in Price Level d(p) and Lari/U.S. Dollar Exchange Rate d(e)
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.Georgia: Logarithmic Changes in Price Level d(p) and Lari/U.S. Dollar Exchange Rate d(e)
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.Georgia: Logarithmic Changes in Price Level d(p) and Lari/U.S. Dollar Exchange Rate d(e)
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Georgian authorities.3. After the depreciation, the NBG further tightened monetary policy by limiting credit to the government, but the continued weakness of the fiscal position forced an increase in direct financing in the last months of 1999 and in the first half of 2000. This once again exerted downward pressure on the currency. As the fiscal position improved in the second part of 2000, the NBG was able to restrain the growth in net domestic assets. Moreover, it controlled reserve money growth sufficiently to offset a rebuilding of foreign reserves at the end of the year, which was permitted by favorable balance of payments developments. When the exchange rate began to appreciate at the end of 2002, the NBG intervened by stepping up foreign exchange purchases. Aside from these episodes, the exchange rate has remained largely stable and inflation low throughout the post-crisis period.
B. Model of Inflation
4. Long-term price level behavior is assumed to be governed by the balance between aggregate demand and supply of goods and services. Equation (1), derived under this assumption in Appendix I, links the logarithm of price level (p) with logarithms of the money supply (m), exchange rate (e) and aggregate supply of goods (y):
The equation is suitable for estimation and testing in the cointegration framework (the estimation procedure is discussed in Appendix II).3 After restricting parameters β1 and β2 to sum to one, the following estimates of the parameters have been obtained (with standard errors of unrestricted parameters reported in parentheses):
The exchange rate coefficient is higher than that of money, but they are both close to one-half.
5. As suggested in the theoretical discussion in Appendix I, estimates of equation (1) can be stable, even if there are persistent disequilibria in the money and foreign exchange markets. Persistent pressure on the exchange rate before the Russian crisis may be an example of a disequilibrium of this type. In the theoretical model, disequilibrium in the foreign exchange market in this period would imply that the money market was also out of equilibrium. The shift in real money holdings during the crisis gives some support to this hypothesis. A recursive estimation of equation (1) coefficients (reported in Figure I-3) shows that the parameters are stable even if persistent disequilibria in the money and foreign exchange markets had been present in the sample.
Georgia: Recursive Estimates of Long-Run Coefficients of the Price Equation
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.Georgia: Recursive Estimates of Long-Run Coefficients of the Price Equation
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.Georgia: Recursive Estimates of Long-Run Coefficients of the Price Equation
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.6. Short-run price dynamics are modeled in a single-equation, error-correction mechanism. After testing the assumption that changes in the logs of money and the exchange rate are exogenous to inflation, inflation is determined by these two variables and lagged deviations from the long-run relationship. The exact form of the short-run relationship (lag structure) is determined by application of the general-to-specific methodology as discussed in Appendix II. In addition to these variables which enter the long-run relationship, percentage changes in relative prices of fruits and vegetables and percentage changes in oil import prices also affect short-run price dynamics. The two variables proxy for supply shocks stemming from the agricultural sector and from input prices. In addition, a dummy variable is used for December 1998, the month of a de facto regime change, when the lari started floating against the dollar.
7. Estimates of the final specification of the inflation equation are reported in Table I-1. Figure I-4 shows actual and fitted values, together with residuals. The equation shows that inflation is strongly affected by exchange rate changes and that the pass-through is fast. Changes in money also have a significant impact on inflation, but this effect takes longer to work its way through the economy than exchange rate changes. The adjustment of prices is also affected by the error-correction term, which is highly significant. This suggests that the price level adjusts to its long-run equilibrium, which is a function of the levels of money, exchange rate and output. Lagged inflation terms do not appear in the final specification of the short-run dynamics, indicating that—conditioning on the behavior of exchange rate, money, and relative prices—inflation persistence is very low. Supply shocks originating in agriculture have a high and significant short-term impact on inflation. Changes in oil import prices have a smaller, but also significant impact.
The Error-Correction Equation for Inflation 1/
ECM (error-correction mechanism) denotes deviations from the estimated long-run relationship in equation (1). Δ indicates first difference. D1298 is one for December 1998, zero otherwise. Standard errors reported in parentheses
The Error-Correction Equation for Inflation 1/
| R2 = 0.8583 | Sample: 1996.6–2003.2 |
| Equation standard error: 0.0054 | Number of observations: 81 |
| DW = 2.18 | |
ECM (error-correction mechanism) denotes deviations from the estimated long-run relationship in equation (1). Δ indicates first difference. D1298 is one for December 1998, zero otherwise. Standard errors reported in parentheses
The Error-Correction Equation for Inflation 1/
| R2 = 0.8583 | Sample: 1996.6–2003.2 |
| Equation standard error: 0.0054 | Number of observations: 81 |
| DW = 2.18 | |
ECM (error-correction mechanism) denotes deviations from the estimated long-run relationship in equation (1). Δ indicates first difference. D1298 is one for December 1998, zero otherwise. Standard errors reported in parentheses
Δp: Actual Values, Fitted Values and Residuals from the Error-Correction Model
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.Δp: Actual Values, Fitted Values and Residuals from the Error-Correction Model
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.Δp: Actual Values, Fitted Values and Residuals from the Error-Correction Model
Citation: IMF Staff Country Reports 2003, 347; 10.5089/9781451814545.002.A001
Source: Fund staff estimates.C. Conclusions
8. The econometric results show that it is feasible to estimate robust price and inflation equations for Georgia. The long-run price equation expresses prices as a function of money, the exchange rate, and real income, and may be interpreted as portraying equilibrium in the goods market. Short-run dynamics of inflation are strongly affected by current exchange rate changes, money growth, and changes in relative prices of foodstuffs and oil. The estimated long- and short-run relationships are stable, and may be useful as a tool for policy formulation and evaluation. Estimation of a separate money demand equation turns out to be more difficult, pointing to a longer and more complicated adjustment mechanism governing the behavior of real money balances. Inflation in Georgia exhibits very low persistence, possibly due in part to the use of relatively short-term nominal wage contracts (a legacy of the hyperinflationary period), which may prevent inflation from becoming entrenched after a shock.
9. The results suggest that the NBG faces serious challenges when conducting monetary policy. Public memory of hyperinflation is still fresh, and any external or internal shock quickly exerts strong pressure on the exchange rate. Because the stock of foreign exchange reserves remains small, the NBG has no scope for leaning against downward pressure on the lari, especially when budget financing needs complicate monetary tightening. Yet even when faced with these challenges, the NBG has enjoyed substantial success in keeping inflation low and relatively stable. Looking ahead, further accumulation of foreign reserves and development of indirect monetary control instruments, such as a deeper treasury bill market, would increase the capacity of the NBG to respond to shocks.
APPENDIX I: Derivation of the Long-Run Price Equation
10. The aggregate demand for goods and services is a function of real money supply (M/P) and the real exchange rate (E/P). In log-linear form (denoted by lower-case letters), the aggregate demand is written as:
The aggregate supply is exogenously given and in equilibrium is equal to aggregate demand and real income (Y):
It is assumed that the goods market is always in equilibrium and therefore equation (A2) always holds.
11. Flow demand for foreign exchange (current account deficit) is assumed to be a function of real exchange rate and real income. Real income is fixed at the level of aggregate supply, the available foreign financing is exogenously given, and the real exchange rate tends to equilibrate the foreign exchange market. Money demand is assumed to be a function of real income. Similarly, since real income—the only variable entering the real money demand function—is exogenous, real money balances tend to equilibrate the money market.4
12. If the three markets are on average in equilibrium, it is likely that two unique long-run cointegrating vectors emerge between non-stationary nominal variables in equation (A1) (treating “y” as exogenous). The two cointegrating vectors describe equilibrium at any two of the three markets and equilibrium at the omitted market is described by a linear combination of the two unique cointegrating vectors. It is also possible that the money and foreign exchange markets are persistently out of equilibrium (adjustments towards equilibrium may be very slow or non-linear) and that only the goods market is—by assumption—always in equilibrium. In this case, only one cointegrating vector can be found in the data, corresponding to the equilibrium described by equation (A1). Re-normalizing this equation by expressing price level as a function of money, the exchange rate and income gives equation (1) in the text, with β1 = α1/(α1 + β2), β2 = α2/(α1 + α2) and β3 = 1/(α1 + α2).
APPENDIX II: Estimation and Testing of the Model
Sources, transformations and statistical properties of the data
13. The model is estimated on monthly data for the post-stabilization period (January 1996-December 2002). Domestic CPI and GDP (a measure of income) are available from the Georgian State Department of Statistics (SDS). The available quarterly GDP series has been interpolated under the assumption that a monthly series follows a unit root process. The exchange rate is measured by average lari/U.S. dollar exchange rate, and money is measured by M2, both available from the NBG.5 Fruit and vegetable prices are obtained from disaggregated CPI data and are divided by the total CPI to obtain relative values. Average oil prices are from the IMF’s WEO database. All series are in logs and are seasonally adjusted using a version of the X-12 procedure. Tests for stationarity (ADF) suggest that logs of M2, CPI and the lari/U.S. dollar exchange rate are all order-one integrated. The log of GDP is order-one integrated by construction.
Testing and estimation of cointegrating vectors
14. The Johansen procedure is used to test for the number of cointegrating vectors and estimate their coefficients. The procedure starts by selecting a set of endogenous and exogenous variables and choosing an appropriate lag structure for the VAR system of the endogenous variables. Prices, the exchange rate and money are modeled as endogenous variables. Real GDP is exogenous and restricted to enter only the long-run (cointegrating) relationship. Two other exogenous variables, namely relative food prices and changes in oil import prices—proxies for supply shocks—enter only the short-run dynamics of the system. In addition, two dummy variables are used: for December 1998, the month of a de facto regime change when the lari began to float against the dollar, and for September 1998, the first month after the Russian crisis. The VAR is estimated with six lags of each endogenous variable.
15. Results of the tests suggest that there is only one cointegrating vector between prices, money, the exchange rate and output (Table I-2). After normalizing the parameter of the log of price level to unity, a hypothesis that coefficients of money and exchange rate sum up to one (homogeneity restriction) is tested, together with exogeneity restrictions. The homogeneity restriction is not rejected at the 5 percent significance level. The weak exogeneity of the exchange rate and inflation is not rejected at the 5 percent level, while the weak exogeneity of prices is strongly rejected. The equation with imposed restrictions of homogeneity and weak exogeneity of money and exchange rate is reported in the main text and chosen for further analysis.
Georgia: Tests for the Number of Cointegrating Vectors between p, m, e and y 1/
Georgia: Tests for the Number of Cointegrating Vectors between p, m, e and y 1/
Rank | λmax | λmax using T-nk | λtrace | λtrace using T-nk | ||||
|---|---|---|---|---|---|---|---|---|
| 0 | 39.25 | [0.003]** | 28.38 | [0.003]** | 30.53 | [0.041]* | 22.07 | [0.035]* |
| 1 | 10.87 | [0.223] | 9.09 | [0.285] | 8.46 | [0.425] | 7.07 | [0.489] |
| 2 | 1.78 | [0.182] | 1.78 | [0.182] | 1.38 | [0.239] | 1.38 | [0.239] |
Georgia: Tests for the Number of Cointegrating Vectors between p, m, e and y 1/
Rank | λmax | λmax using T-nk | λtrace | λtrace using T-nk | ||||
|---|---|---|---|---|---|---|---|---|
| 0 | 39.25 | [0.003]** | 28.38 | [0.003]** | 30.53 | [0.041]* | 22.07 | [0.035]* |
| 1 | 10.87 | [0.223] | 9.09 | [0.285] | 8.46 | [0.425] | 7.07 | [0.489] |
| 2 | 1.78 | [0.182] | 1.78 | [0.182] | 1.38 | [0.239] | 1.38 | [0.239] |
Specification of an error-correction equation
16. Since the weak exogeneity of the exchange rate and money is not rejected, it is valid to condition on these two variables in a single-equation inflation model. The “general-to-specific” methodology is followed in searching for the final form of the short-run dynamic inflation equation. The specification search begins from estimation of a relatively unrestricted model. The unrestricted inflation equation includes five lags of inflation; five lagged and current values of changes in the log of money and in the log of the exchange rate; the lagged error-correction term from the long-run price equation; changes in relative prices of fruits and vegetables and of oil prices; and the dummy variables discussed above. In the next steps, restrictions imposed on the model are tested against the unrestricted alternative. Restrictions imposed on the general specification leading to the final equation reported in Table I-2 cannot be statistically rejected, and the final inflation equation easily passes all standard specification and stability tests.
Prepared by Wojciech Maliszewski.
The BRO group includes the Baltics, Russia, and other former Soviet Union countries.
Non-stationary variables are cointegrated if some of their linear combinations are stationary.
A typical formulation of the Cagan-style money demand function expresses the demand for real money balances (M/P) as a function of expected inflation and real income. This formulation is not well suited to modeling the long-run behavior of money balances in Georgia. Tests for stationarity of the series—discussed below—suggest that after 1996, inflation became a stationary series, while real money balances remained non-stationary.
Estimation of the model using M3 instead of M2 yields very similar results.
