CC BY
0DOI 10.24412/1829-0450-2024-1-60-66 Поступила: 18.03.2024г.
УДК 2964 Сдана на рецензию: 19.03.2024г.
Подписана к печати: 16.04.2024г.
ESTIMATION OF THE LINEAR RELATIONSHIP BETWEEN THE INVESTMENT ENVIRONMENT AND THE FACTORS AFFECTING IT
G. Petrosyan
Russian-Armenian (Slavonic) University gevorg.petrosyan307@gmail.com ORCID: 0009-0001 -5561 - 761X
ABSTRACT
From the theoretical point of view, the study of the factors affecting the social order, legal framework, business activity and investment environment has an important empirical significance. Identifying the factors affecting the investment environment and their coordination in external and internal spheres will enable the strategists to obtain greater information for the creation or improvement of investment opportunities in various spheres. It is also of great importance for reducing and managing negative factors affecting investments. This will lead to a reduction in the level of risks, which will certainly make the country even more attractive for investors, who can more easily evaluate and forecast their investments. In this article, an attempt was made to evaluate the endogenous and exogenous factors affecting the investment environment with the help of a linear regression model. Researchand Development expenses in GDP has the highest affection level in the model. This evaluation can be an initial point for researchers and investors for their investment strategies.
Keywords: investment, endogenous and exogenous factors, linear model, research, investment environment.
Introduction
There are many works that have attempted to evaluate the factors affecting the investment environment at the level of countries and regions. Linear regression models were mostly used, which presented the linear relationship between investments as a dependent variable and factors affecting it as independent variables. An attempt will be made to carry out a similar assessment for the Republic of Armenia, but at the same time, other, more innovative methods will be used, which will allow for a more accurate assessment of the degree of importance of the factors affecting the investment environment, which in turn will make it possible to make forecasts more easily.
In this article, an attempt will be made to answer the following questions through the evaluation of models: “What are the external and internal factors that affect the investment environment of the Republic of Armenia?”, “What weight
and significance do the influencing factors have?”, “How will the volume of investments change?”, as a result of a change in any factor” and “Will investments in the Republic of Armenia increase or decrease in the following years?”.
Body Work Factor collection
For the evaluation of the model data for the several factors were collected. The main database which had been used before the evaluation was the World Bank database [1]. Overall, 37 factors were collected which covered data for the timeline of 1990-2023 for the Republic of Armenia. The factors have been differentiated as exogenous and endogenous.
The exogenous factors are:
• Military expenditure (in GDP %)
• Economic International Trade Taxes (% in Profit)
• Import of goods and services (% of GDP)
• Annual growth of imports of goods and services
• Export of goods and services (% of GDP)
• Annual growth of exports of goods and services
• Financial Export Tax (% of tax revenue)
• Trade (% of GDP)
The endogenous factors are:
• Social Health Expenditure (% of GDP)
• People who use the Internet (% of Population)
• An assessment of transparency and corruption in the social sector
• Research and Development Spending (% of GDP)
• Government spending on education (% of GDP)
• Population
• Emission of carbon dioxide
• Mortality rate
• Volume of economic GDP
• Annual GDP growth
• GDP per capita
• Annual growth in GDP per capita
• Gini index
• Inflation rate
• Unemployment rate (WB modeling)
• Unemployment rate (national estimate)
• Trade rating
• Assessment of logistics activities
• Government debt (% of GDP)
• Financial Organizations that use banks to finance investments (% of all comp.)
• Domestic Financial Consumer Expenditures (% of GDP)
• Assessment of business environment regulation
• Assessment of ease of doing business
• Density of new businesses (per 1,000 people)
• Time to start a business
• Profit tax (% of all commercial profits)
• Tariff rate
• Revenue from taxes (% of GDP)
• Goods and Services Tax (% in Profit)
Empirical studies of the determinants of FDI in CEE countries (e.g., Breton et al. 1998; Dunning 1994; Holland et al. 2000) conclude that the primary factors stimulating investment in transitional economies are market size, openess,growth rates, trade barriers, budget deficits, taxes, labor costs, political stability, cultural ties and geographical location, among others. Recording to that for the Republic of Armenia author tried to gather as much data as possible from the available sources to achive the goal to understand the affection of those determinants on FDI.
In order to improve the quality of further analysis, artificial data augmentation was carried out by means of exponential interpolation. Its purpose is to convert annual data into quarterly data and thus turn 33 rows of data into 132 rows. Exponential interpolation was preferred over linear interpolation in order to reduce the degree of autoregression in later models.
Statistical analysis
An evaluation of the correlation matrix between the factors was carried out, which will enable selection of evaluation factors for future models. The scores of the correlation matrix fall in the range [-1,1] and the closer the scores are to the edges of the range, the better the given factor can be included in the analysis. However, it is necessary to pay attention to the sign as well. It should be logical. There are cases when the degree of correlation is close to one of the edges, but the sign may not correspond to reality. This can be due to a number of reasons, but one of the most common ones is the presence of problems and other factors in the given factor series, which may in turn affect both the dependent and the selected independent variable simultaneously. In that case, multicollinearity occurs, which can ultimately affect the accuracy of the model being evaluated.
Table 1. Correlation between dependent and independent variables.
Factor Investment volume Increase in investment volume
Organizations that use banks for investment financing (% of all comp.) 0.772367 0.442955
GDP volume 0.633228 0.040039
Annual GDP growth 0.246811 0.438664
GDP per capita 0.614595 0.024352
Annual growth of GDP per capita 0.231278 0.43334
Healthcare expenditure (% of GDP) 0.417077 -0.09224
Consumer spending (% of GDP) -0.43393 -0.00883
Military expenditure (% of GDP) 0.32403 0.068328
Internet usage (% of Population) 0.240883 -0.26717
Gini index -0.50226 0.026759
Density of new businesses (per 1000 people) 0.123673 -0.30184
Time to start a business -0.27326 0.251119
Inflation rate -0.32863 -0.39906
Profit tax (% of all commercial profits) -0.09048 -0.41633
Tariff rate 0.309219 -0.1581
Income from taxes (% of GDP) 0.335734 -0.2138
Goods and services tax (% of profit) 0.573183 -0.01846
Unemployment rate (WB modeling) 0.388134 0.34508
Unemployment rate (national estimate) 0.686395 0.462244
International trade taxes (% of profit) 0.277382 -0.23078
Trade (% of GDP) -0.58517 -0.76683
Import of goods and services (% of GDP) -0.61194 -0.54482
Annual growth of import of goods and services 0.304798 0.307486
Export of goods and services (% of GDP) -0.39813 -0.74864
Annual growth of exports of goods and services 0.273873 0.308025
Research and development costs (% of GDP) 0.489802 0.639519
Government spending on education (% of gDp) 0.717465 0.592256
Score of transparency and corruption in social sphere 0.367615 -0.65547
Factor Investment volume Increase in investment volume
Population 0.601917 -0.28242
Carbon dioxide emission 0.290805 -0.39285
Mortality rate 0.165489 -0.06514
Looking at the results of table 1, it can be concluded that 23 of the selected factors are more correlated with the volume of investments, and 13 with the growth of the volume of investments. Also, the signs of correlation with the volume of investments are more logical than with the growth of the volume of investments. It means that it is more appropriate to select the investment volume as a dependent variable with the selected factors than the increase in the investment volume. However, this can be accepted as a hypothesis.
After the evaluation of the correlation matrix 8 factors were chosen and the OLS model was evaluated.
Table 2. Evaluation of factors affecting the volume of investments using the OLS method.
OLS Regression Results
Dep. Variable: Model:
Method:
Date:
Time:
No. Observations Df Residuals:
Df Model: Covariance Type:
fdi_net
OLS
Least Squares Thu, 22 Feb 2024 15:23:40 129 120 8
nonrobust
R-squared:
Adj. R-squared: F-statistic:
Prob (F-statistic) Log-Likelihood:
AIC:
BIC:
0.930
0.925
198.3
2.12e-65
-131.24
280.5
306.2
coef std err t P>|t| [0.025 0.975]
const 10.1324 1.493 6.785 0.000 7.176 13.089
gdppercap 0.2657 0.021 12.440 0.000 0.223 0.308
health_exp -0.7448 0.091 -8.202 0.000 -0.925 -0.565
fin cons exp -0.0265 0.008 -3.243 0.002 -0.043 -0.010
tax_rev -0.8909 0.081 -11.007 0.000 -1.051 -0.731
tax good serv 0.1464 0.034 4.285 0.000 0.079 0.214
unempl_nat 0.2622 0.026 10.183 0.000 0.211 0.313
res dev exp 12.1777 3.000 4.059 0.000 6.233 18.117
gov_exp_edu -0.8320 0.198 -4.194 0.000 -1.225 -0.439
Omnibus: 3.023 Durbin-Watson: 0.308
Prob(Omnibus): 0.221 Darque-Bera (DВ) : 2.857
Skew: 0.364 Prob(JB): 0.240
Kurtosis: 2.960 Cond. No. 5.57e+03
The dependent variable is FDI inflows to Armenia. The chosen 8 factors are: GDP per capita, Healthcare expenditure (% of GDP), Consumer spending (% of GDP), Income from taxes (% of GDP), Goods and services tax (% of profit), Unemployment rate (national estimate), Research and development costs (% of GDP), Government spending on education (% of GDP).
In case of all factors being unchanged, the model predicts that after 2023 December if the trend of the FDI inflows into the Republic of Armenia remains and also nothing in the non-researched variables will be changed which can alternate the results of the estimated model above, an investment of 1 billion 13 million US dollars will be made in Armenia. By the end of 2023, that figure was 990 million US dollars. 4 factors have an inverse effect on the growth of the volume of investments. The factor of the share of analysis and development costs in GDP also provides the biggest increase here, in case of a 1 percent increase in GDP share, the Republic of Armenia will be invested 121.7 million USD more. This means that if it is raised from 0.2 percent to 1.2 percent in the GDP share, 121.7 million USD will be invested. In some countries, it can reach up to 5.5 percent (for example, in Israel) [2].
The signs of the impact of two factors on the investment climate are quite interesting: signs of the unemployment rate and government spending on education. From the estimated model, it can be observed that these two factors would logically have opposite signs. At first glance it may seem illogical, but it raises important questions. The positive effect of the unemployment rate can be attributed to reducing the costs of business activities. In countries with high unemployment rates, there is a greater demand for job seekers and lower salary conditions can be discussed in that case. For several developed countries, Armenia is seen as a market for not-so-high-salary workers, particularly in the field of information technology, which enables investors to look for Armenian specialists for their open positions.
And the negative impact of the government's spending on education can be seen in the fact that the education sector in Armenia is seen by investors as not so competitive, as well as the quality problems of professionals entering the labor market. In Armenia, there is a low supply of qualified labor in several professions, and there are professions where specialists exceeding the demand enter. This can also have a negative effect on investors' search for qualified labor. The negative impact of spending in certain areas of the government can also be explained by the problems of the tax field, which is explained by underspending in some areas and overspending in others, which can affect interest rates, and high interest rates directly affect investment decisions, and of course in a negative way.
Thus, signs of two factors that seem illogical at first glance can be quite logical and it depends on the internal and external situation of a specific country.
Conclusion
One of the integral parts of each analysis is the evaluation of the model using the PC method, which enables to understand the degree of significance of the selected factors, as well as to understand and study the linear relationship between independent and dependent variables. Estimating the model using this method is not particularly complicated and can be a good way to get primary information for many investors. However, doing a more in-depth study is even more important.
ЛИТЕРАТУРА
1. База данных Всемирного банка развития. URL: https://databank.worldbank.org/sour-ce/world-development-indicators.
2. Ктоян А., Хачикян С., МовсисянМ., Акопян К. Статистический анализ экономических задач с использованием пакета SPSS, Ехегнадзор, 2015. С. 15.
REFERENCES
1. World Bank Development Database. URL: https://databank.worldbank.org/source/world-de-velopment-indicators
2. Ktoyan A., Khachikyan S., MovsisyanM., Hakobyan K. Statistical analysis of economic problems using SPSS package, Yeghegnadzor, 2015. P. 15.
ОЦЕНКА ЛИНЕЙНОЙ СВЯЗИ МЕЖДУ ИНВЕСТИЦИОННОЙ СРЕДОЙ И ФАКТОРАМИ, ВЛИЯЮЩИМИ НА НЕЕ
Г.С. Петросян
Российско-Армянский (Славянский) университет
АННОТАЦИЯ
С теоретической точки зрения изучение факторов, влияющих на социальный порядок, правовую базу, деловую активность и инвестиционную среду, имеет важное эмпирическое значение. Выявление факторов, влияющих на инвестиционную среду, и их координация во внешней и внутренней сферах позволит стратегам получить больше информации для создания или улучшения инвестиционных возможностей в различных сферах. Это также имеет большое значение для сокращения и управления негативными факторами, влияющими на инвестиции. Это приведет к снижению уровня рисков, что, безусловно, сделает страну еще более привлекательной для инвесторов, которым будет легче оценивать и прогнозировать свои инвестиции. В данной статье предпринята попытка оценить эндогенные и экзогенные факторы, влияющие на инвестиционную среду, с помощью модели линейной регрессии. Расходы на исследования и разработки в ВВП имеют самый высокий уровень влияния в модели. Эта оценка может стать отправной точкой для исследователей и инвесторов при разработке их инвестиционных стратегий.
Ключевые слова: инвестиции, эндогенные и экзогенные факторы, линейная модель, исследования, инвестиционная среда.
