THE PATTERN OF ECONOMIC CHANGE,

Janusz Szyrmer and Alex Vishnevsky
August, 1994

Introduction

A well known feature of a transition economy is its low predictability. The process of transition from one economic system to another, by its very nature, must disrupt an old order and an old structure, before a new order and a new structure emerge. The old Soviet-type centrally planned economy was "...a coherent whole--a true system--with its own inherent logic, necessary components, and natural interaction of those components " /1, p.11/. Despite a variety of "built-in" instabilities, such as five-year investment cycles, pervasive excess demand in both factor and product markets, hidden unemployment and hidden inflation /2/, this economy managed to secure a relatively stable--smooth and predictable--growth pattern for several decades. This economy remained in the neighborhood of a specific social and economic "equilibrium". Since the late 1980s Eastern Europe has been undergoing fundamental systemic changes. It has had all the features of a "badly behaving" far-from-equilibrium system. In all post-communist economies there occur a lack of political stability, abrupt structural changes, and a high inflation.

The Pattern of economic growth

This report presents preliminary results of a research project on the pattern of economic growth/decline in Eastern Europe. Its objective is an attempt to identify a common trend and country specific deviations from this trend in order to better explain the transition process, evaluate policies and predict. We attempt to explain the observed "economic outcomes" by means of a number of political, social, geographic and economic variables, specific for each country or region. The predicted outcomes include GNP, consumption levels and structure, inflation, and unemployment. The predictor variables include:
- Initial conditions: statehood, political environment, initial level of economic development, stability, centralization, etc.;
- External impact (foreign sector): debt, balance of payments, trade, foreign assistance, direct foreign investment;
- Internal impact (policy): soft/hard budget constraint, budget deficit, collective consumption, fiscal policy, tariffs, currency exchangeability.

Project stages (tasks):

Stage 1. Identify the pattern of the changes in GNP, 1985-94:

Stage 2. Build a statistical model to explain the outcomes by means of initial conditions, external and internal impacts (policy)

Stage 3. Model applications - policy simulations and forecasts

This report presents the initial results from stage 1: The general pattern of change.
Model specification: 

    GNP = b0  + b1 * t + b2 * sin(b3 *t)

Data for six Eastern European countries (Table 1):

   GNP = Gross National Product, first year of each time series = 100;
   t   = 1984-93 for Romania,
         1985-94 for Bulgaria, Hungary and Poland,
         1986-95 for Russia and the Czech Rep.;

   b0  = 100;
   b1  = long-term/medium-term trend;
   b2  = cycle depth (amplitude);
   b3  = cycle speed;

   b1, b2, b3 = OLS estimates of the regression coefficients.
Results:

1. The six countries display a similar trend (with the shape of a sine curve) (Fig.1).
2. The "transition cycle" is assumed to begin around 1985. It started slightly earlier for Romania, and has been delayed for the Czech Rep. and Russia (Fig.2).
3. During the first six years of this cycle the GNP trend for all the countries was quite similar (Fig.2). From the seventh year on major differences in the pattern have emerged.
4. The pattern of changes in the Bulgarian GNP is the closest to the average pattern. Hungary and the Czech Rep. are performing sightly better than the average. Poland displays the best performance -- significantly above the average. Russia and Romania performed below the average.
5. All the regression coefficients are statistically significant (except one) - possess a high t statistic. The long-term trend slope (b1) is positive only for Poland (+0.278). For Romania and Russia this coefficient has the lowest values (highest negative values), -3.176 and -2.601, respectively. The cycle depth coefficient (b2) is the highest for Russia (18.68) and Bulgaria, the lowest for Hungary and the Czech Rep., which are the most stable countries of the region. The speed coefficient (b3) is the highest for Czech Rep. and Poland and the lowest for Romania. The former are believed to transform faster than others, the latter has been slow in introducing market reforms.

References