English / ქართული / русский /

Journal number 2 ∘ Giorgi Tetrauli

Expanded Summary

In the cyclical theory of economic development Juglar's medium-term cycles, known as business cycles, play the most important role. In the second part of 19th century, when economics was developing into a science, Juglar was the economist who made an attempt to introduce new statistical methods in economic analysis. Cycles that he discovered are strongly correlated with different financial and economic cycles, including Kondratiev's Long Waves ad Kitchin's short-term cycles. According to empirical evidence, Juglar's cycles have the strongest impact on economic activity and because of this they are defined as basic cycles.

According to numerous surveys, medium-term (business) cycles had the strongest impact on the dynamics of economic development, and because of that they are treated as basic cycles.

The goal of this article is to discover medium-term cycles in economy of Georgia, as well as make corresponding deductions about the future economic development of our country.

Juglar cycle, also called a business cycle, is one of the main researched periodical phenomenon in the economic theory. Its average length is 7-11 years and is related to the periodical changes in business activities.

It is rather problematic to reveal medium economic cycles in a dynamics of a macroeconomic indicator. Different widespread approaches used to study these cycles are listed below:

ü  Rigidity of prices and salaries;

ü  Irrational behaviour of economic subjects (first of all, in investment-related decisions);

ü  Irrelevant government policy (excessive stimulation of economy) or a too much intervention from the government side;

ü  Impact of exogenous factors, which require from the market a long time for adaptation;

ü  Stochastic element in the market mechanism itself.

According to the approach used in the survey described in this article, Juglar cycles are caused by time lags in the process of making investment decisions and can be described by fluctuations in investment amount (fixed capital formation). The mechanism of the business cycles can be explained as follows: the upswing of the cycle takes place if there are enough resources and necessary circumstances. Besides, resources of future periods take part in making decision about investment and production expansion. Future resources come in action in case of successful business strategies (that is, credit resources), which stimulate investment. However, because of the time lags in dissemination of market information, business strategies become incorrect, just as decisions about investment. This describes one of the main market imperfections. As a result, investments are not compatible with the current state of economy and cause excessive investing. This in turn causes fluctuations in the economic activity and the business cycle phenomenon itself.

Hodrick-Prescott (HP) filter was developed in 1997 and is nowadays used for analysis of time series to reveal trend, cyclical and stochastic elements within the dynamics of a concrete indicator. This method is mostly used in business-cycle theories.

The use of HP-filter is based on the assumption that a time series can be represented as a sum of cyclical and growth components

Where:  -is a growth component, -cyclical component of a time series.

Thus, a cyclical component can be calculated as follows:


This means, that a cyclical component using HP-filter is derived by eliminating the growth component.

The analysis was divided into following steps:

  • Collection of the data about a concrete macroeconomic indicator;
  • Choice of a target length of the cycles;
  • Use HP-filter to reveal cycles in the time series;
  • Make conclusions based on the final sinusoid.

HP-filter analysis can be performed used Eviews software, as it was done in this survey. As initial time series, the real Gross Domestic Product per capita in 1965-2014, in constant 2005 USD was chosen.

The goal of the survey was to find Juglar cycles, thus the target length of the cycles was set equal to [7;11] interval. HP-filter enables to gradually smoothen the dynamics of the initial time series, along with eliminating the growth component (trend). First cycles shorter than 7 years were smoothened, on the next step-less than 11 year long. Subtracting the results from each other gave values, which reflect cyclical movements of 7-11 year length.

Finally, use of HP-filter on the economy of Georgia enabled to find several cycles, whose lengths was the same as the Juglar classical cycle period:

  1. 1967-1973-first cycle;
  2. 1973-1982–second cycle;
  3. 1982-1993-third cycle;
  4. 1993-2003-forth cycle;
  5. 2003-2010-fifth cycle.

Based on the periods of the found cycles, following conclusion about the future economic development of Georgia was made: taking into account the average period of the cycles (9 years), 2010 year corresponds to the depression phase of the last Juglar cycle, while current period can be treated as a downturn phase which will continue until 2019-2020 years. This corresponds to a decrease in investments and total economic activity.