=Paper=
{{Paper
|id=Vol-2018/paper-11
|storemode=property
|title=A Study of Entrepreneurial Activity of the Population in Regions of the Russian Federation
by Means of Panel Data Analysis
|pdfUrl=https://ceur-ws.org/Vol-2018/paper-11.pdf
|volume=Vol-2018
|authors=Ekaterina Liskina,Olga Serova
}}
==A Study of Entrepreneurial Activity of the Population in Regions of the Russian Federation
by Means of Panel Data Analysis==
https://ceur-ws.org/Vol-2018/paper-11.pdf
A Study of Entrepreneurial Activity
of the Population in Regions of the Russian
Federation by Means of Panel Data Analysis
Ekaterina Ju. Liskina1[0000−0002−4169−6062] and
Olga P. Serova2[0000−0002−3550−8838]
1
S. A. Esenin Ryazan State University, Ryazan, Russia
katelis@yandex.ru
2
EPAM, Ryazan, Russia
lyolka92@gmail.com
Abstract. This work aims to construct the best regression model in
which small business density is the indicator of the level of the popula-
tion’s entrepreneurial activity. The regressors are the volume of invest-
ments into the fixed capital of small businesses, the rate on insurance
contributions for individual entrepreneurs to extra-budgetary funds of
the Russian Federation, and the loan rate for small business. The pooled
model, the unrelated model, the fixed effects model, the random effects
model are constructed, and their statistical characteristics are calculated.
We verify statistical hypotheses about the choice of the most preferred
model according to the criteria of Wald test, Breusch – Pagan LM-test,
and Housman test. We select the best model. Based on obtained results
we propose a classification of the regions of the Russian Federation by
the type of the dependence of their entrepreneurial activity on the re-
gressors. We analyze all regions and construct the best regression model
for each of them. The unrelated regression model is found to be the best
for each group of regions. The authors suggest that strong differences
between the regions can be attributed to qualitative factors, such as reg-
istration procedures of small businesses, the rate of regional taxes, etc.
The results of the study can be used in state and municipal programs of
small business development to improve forecasting of economic develop-
ment of each region of the Russian Federation. Also, we point out that
to boost entrepreneurial activity in different groups of regions, different
adjustments are required, such as reduction of the rate on insurance con-
tributions for individual entrepreneurs, improvement of the investment
climate, or loan rates for small businesses.
Keywords: panel data, economy of Russia, entrepreneurial activity,
small business density, pooled model, unrelated model, fixed effects model,
random effects model, decision making for small business support
1 Introduction
Small business is an important factor in the development of the economy as it
is most susceptible to changes in market conditions and state tax policy. The
level of development of small business is an indicator of the well-being of the
economy as a whole. Small and medium entrepreneurship research has been
intensively developing since the late 1980s. An overview of different areas of the
research can be found in the works of A. Yu. Chepurenko [17, 18]. Dynamics
of entrepreneurship development in countries worldwide is evaluated through
two global monitoring programs: “Panel study of dynamics of entrepreneurship”
(PSED) and “Global entrepreneurship monitor” (GEM) [1, 6].
The entrepreneurial activity of the population is one of the small business
development indicators. Entrepreneurial activity is a qualitative dynamic indi-
cator of entrepreneurial capacity in current business climate [10]. The level of
entrepreneurial activity in Russian regions is estimated using monitoring pro-
grams and ratings [2–4,9]. Ratings [3,4,9] evaluate investment climate in regions
including the area of small business. However, ratings do not allow evaluating
the degree of influence of various factors on the entrepreneurial activity of the
population of Russia. It is therefore important to identify the type, degree, and
regional specifics of the individual factor influence on the entrepreneurial activity
level of the population of our country. This knowledge will allow us to scientif-
ically substantiate the ability to control national economic growth by stimulat-
ing small business via the most influential economic factors. For example, the
importance of the research of entrepreneurial performance as the government
participation function is discussed in [16].
At the moment there is a very limited body of works devoted to statistical
analysis of the factors influencing entrepreneurial activity and econometric busi-
ness modeling in Russia. In particular, Pin’koveckaja Ju. S. [13] constructed a
two-factor exponential production function and showed the dependence of small
business turnover on investment in fixed capital and wages. Gorlov A. V. [8]
investigated the degree of influence of macroeconomic factors on the economic
activity of small manufacturing businesses (WFP). The endogenous variable was
the volume of WFP output; the following factors were considered: average num-
ber of employees of WFP, investment in fixed capital, export of goods and ser-
vices, foreign investment in the Russian economy, and others. On the basis of a
selected set of most influential factors, he developed a multi-factor production
Cobb–Douglas function describing the volume of output of small businesses. Us-
ing the constructed model, the author constructed the forecast of small business
production volume dynamics for the planning period of 2016–2018.
Unlike the above works [8,13], we developed an econometric model of the en-
trepreneurial activity of the population of Russia based on a number of macroe-
conomic factors: investment in fixed capital of small businesses, the rate of insur-
ance contributions for individual entrepreneurs, and the average interest rate on
loans for small businesses. The density of small businesses is an endogenous vari-
able and a quantitative characteristic of the population entrepreneurial activity
level (see [9]). The density of small businesses is equal to the ratio of the number
of small enterprises to the number of economically active population. Such defi-
nition eliminates the effects associated with the distribution of the economically
active population in the country.
2 Setting of the problem
We use the following symbols:
SB – Small Business – the number of small businesses in the Russian Feder-
ation;
EAP – Economically Active Population – the number of economically active
population of the Russian Federation, in thousands;
SB
SBD = EAP – Small Business Density;
I – Investments – the volume of investments into fixed capital of small busi-
nesses, thousand rubles;
IC – Insurance Contributions – the tax rate on insurance contributions for
individual entrepreneurs to extra-budgetary funds of the Russian Federation
(before 2010 – the unified social tax);
LR – Loan Rate – the average rate for small business loans;
t – year (t = 2000; 2014);
n – region index number (n = 1; 79).
The regions (Federation subjects) of Russia are the objects of the study. We
collected panel data of listed statistical economic indicators in all regions from
2000 to 2014 [5]. The panel is balanced since the data is present for all regions
and points in time (total 1185 cases). In the selected period the economic system
of Russia had a stable structure. Significant structural changes occurred at the
boundaries of this period for the following reasons:
1. the first and second parts of the existing Tax Code were adopted in 1999
and 2000, which had a significant influence on the structure of the economy
as a whole [12];
2. the current structure of insurance premiums to non-budgetary funds of the
Russian Federation, unified for all types of employers and employees, was
adopted in 2001;
3. the Republic of Crimea was annexed by Russia in 2014.
In [11] and [14] we studied the dependence of SBD on I, IC, and LR, by
methods of regression analysis. We used the cross-data of the constituent entities
of the Russian Federation in each year of the considered period. We showed that
the linear dependence of the density of small businesses on these factors results
in the best statistical characteristics:
SBD = a0 + aI I + aIC IC + aLR LR. (1)
In this paper, we formulate the task to identify regional differences that
influence the level of entrepreneurial activity of the population by panel data
analysis method and carry out the grouping of regions according to the degree
of regional differences.
The calculations are performed using the Microsoft Excel add-in “data Anal-
ysis”. The level of significance is 0.05.
3 Experimental research
In accordance with the theory [7], we constructed the following models of panel
data: the pooled model (OR-model), the unrelated model (UR-model), the fixed
effects model (FE-model), and the random effects model (RE-model).
Statistical characteristics of the constructed models are presented in Table 1.
Table 1. Statistical characteristics of the constructed models
Model R2 The residual sum of squares (RSS)
OR-model 0.25 76773.51
2 2
UR-model Rmin = 0.17, Rmax = 0.95 17797.70
FE-model 0.41 32502.81
RE-model 0.40 39726.79
For the constructed models at a significance level of 0.05 we tested the fol-
lowing base hypotheses:
1. about absence of individual differences (Wald test on the insignificance of
individual coefficients, OR-model preferable to FE-model);
2. about absence of relationship between regions (Wald test on the insignifi-
cance of the coefficients on the regressors in the FE-model, UR-model prefer-
able to FE-model);
3. about absence of relationship between regions (Wald test on the insignifi-
cance of the coefficients on the regressors in the RE-model, UR-model prefer-
able to RE-model);
4. about the absence of random individual differences (Breusch – Pagan LM-
test);
5. about random individual differences preferable to fixed effects (Housman
test, we used the evaluation in the form of the auxiliary regression equation
SBD (λ) = X (λ) α+ZW γ +ε, matrix X (λ) = (I (λ) , IC (λ) , LR (λ))q is con-
σ̃v2
structed according to the adjusted data for the RE-model, λ = 1 − 2 ,
σ̃v2 +T σ̃u
σ̃v2 is the estimation of residual variance of the OR-model, σ̃u2 is the estimate
of the variance of random effects, T = 79; matrix ZW = (IW , ICW , LRW ) is
constructed according to the adjusted data for the FE-model, base hypothe-
ses: γ = 0).
The test results of all hypotheses are shown in Table 2.
We made the following conclusions based on Table 2:
1. Wald test on the insignificance of individual coefficients of the FE-model
showed that individual differences between regions are significant, i.e., the
FE-model was preferable to the OR-model;
Table 2. Verification of statistical hypotheses on the choice of the best model
value Conclusion about
Test
factual critical base hypotheses
1) Wald (OR-model 536.65 2.61 reject
preferable to FE-model)
2) Wald (UR-model – 178.26 2.61 accept
preferable to FE-model)
3) Wald (UR-model –217.49 2.61 accept
preferable to RE-model)
4) Breusch – Pagan 2433.18 3.84 reject
LM-test (OR-model
preferable to RE-model)
5) Housman (RE-model PγIC = 2.6 · 10−29 0.05 reject
preferable to FE-model)
2. Wald test on the insignificance of the coefficients on the regressors in the FE-
model showed that the factor coefficients are insignificant; i.e., the individual
fixed differences of regions such that the regions can not be combined in the
FE-model;
3. Wald test on the insignificance of the coefficients on the regressors in the RE-
model showed that the factor coefficients are insignificant; i.e., the individual
random differences of regions such that the regions can not be combined in
the RE-model;
4. Breusch – Pagan LM-test showed that the RE-model was preferable to the
OR-model;
5. Housman test showed that RE-model was preferable to the FE-model.
Based on 1–5 we can put the studied models in the ascending order of prefer-
ence: OR-model, FE-model, RE-model, UR-model. In other words, with a prob-
ability of 0.95, the individual differences of the regions are random but so sig-
nificant that the UR-model is the best to construct the dependence of SBD on
I, IC, and LR.
We then analyze the significance of the coefficients of the regressors for each
equation of the UR-model, perform the correlation analysis of the factors for
each region’s data, and classify the regions according to the type of dependence
of SBD on I, IC, and LR. The classification results are presented in Table 3.
We drew the following conclusions from Table 3 (significance level of 0.05).
1. There is no Russian region for which all three regressors of equation (1) are
significant.
2. The most numerous group consists of the regions where the small business
density is determined by the volume of investments into fixed capital of
small enterprises (the regression equation has the form SBD = a0 + aI I,
group No 3).
Table 3. Classification of the regions of the Russian Federation by the type of regres-
sion
No Regression equation Regions of the Russian Federation
1 SBD = a0 + aI I + aLR LR Regions: Belgorod, Moscow; Stavropol Krai
2 SBD = a0 + aI I + aIC IC Regions: Rostov, Ivanovo, the Kabardino-Balkar
Republic, Republic of Buryatia, Zabaykalsky Krai,
the Jewish Autonomous region
3 SBD = a0 + aI I Regions: Voronezh, Lipetsk, Orel, Kursk, Pskov,
Tver, Tula, Smolensk, Volgograd, Kirov, Nizhny
Novgorod, Orenburg, Penza, Samara, Saratov,
Ulyanovsk, Kemerovo, Amur, Bryansk, Vladimir,
Kostroma, Kaluga, Tambov, Yaroslavl,
Arkhangelsk, Novgorod, Astrakhan, Novosibirsk,
Omsk, Irkutsk, Kurgan.
Republics: Adygea, Dagestan, Bashkortostan,
Mordovia, Chuvashia, Tyva, Mari El, Udmurt,
Karachay-Cherkess, Karelia, Tatarstan, Altai.
Krai: Krasnodar, Altai, Krasnoyarsk, Kamchatka,
Primorsky,
St-Peterburg-city
4 SBD = aIC IC Region: Ryazan, Tomsk, Sverdlovsk, Magadan,
Sakhalin, Tyumen.
Republics: North Ossetia-Alania, Khakassia, Sakha
(Yakutia), Komi.
Krai: Khabarovsk, Perm
5 SBD = a0 + aLR LR Chukotka
6 No significant coefficients Region: Vologda, Kaliningrad, Murmansk,
of regressors in the Leningrad, Chelyabinsk
equation (1) Republic: Kalmykia, Ingushetia.
Moscow-city
3. There is a group of regions in which the SBD does not depend on any of the
regressors (group No 6).
4. In groups 1–3, the regression equation contains the volume of investments
into fixed capital of small enterprises. Therefore, there is a possibility to
manage small business growth by attracting investments in these regions.
5. Control at the Federal level affects the development of small businesses in
groups 1 and 4, and group 4 – only at the Federal level.
6. Control of small business development can be realized through improvement
of credit conditions in the regions that fall into groups 1 and 5. For a single
region in group 5, such control is possible only via improving credit condi-
tions.
We supplement this research with an analysis of partial panels formed by
groups 1–4. We identified regional differences within each group and analyzed
the nature of these differences (fixed or random). The results of this analysis are
presented in Tables 4 and 5.
Table 4. Statistical characteristics of regression models based on the partial panels of
the groups 1–4
Group The residual sum
Model R2
number of squares (RSS)
1 OR-model 0.34 1102.00
2 2
UR-mode Rmin = 0.79; Rmax = 0.86 205.09
FE-mode 0.37 736.00
RE-mode 0.58 1158.56
2 OR-mode 0.44 2896.02
2 2
UR-mode Rmin = 0.66; Rmax = 0.80 905.31
FE-mode 0.58 2372.19
RE-mode 0.54 2475.35
3 OR-mode 0.16 52733.36
2 2
UR-mode Rmin = 0.42; Rmax = 0.95 9086.06
FE-mode 0.02 172287.67
RE-mode 0.39 139821.51
4 OR-mode 0.30 10969.42
2 2
UR-mode Rmin = 0.28; Rmax = 0.49 7755.74
FE-mode 0.37 8108.54
RE-model 0.56 10382.36
Thus, we found that the UR-model was the best model in each group of
regions. We can explain this fact by the presence of local qualitative factors
influencing the level of regional entrepreneurial activity. Those factors might
include procedures of registration of small businesses, rates of regional taxes,
etc.
4 Conclusion
In this paper, we developed a regression model of entrepreneurial activity level
(i.e., small business density) of Russian population using panel data analysis.
Explanatory variables were the volume of investments into fixed capital of small
businesses, the rate of contributions to extra-budgetary funds of the Russian
Federation for small businesses, and the average interest rate on loans to small
businesses. We classified the regions based on the explanatory variables influence
degree on the small business density. We studied each region group separately
and found that UR-model was the most suitable. It can be explained by the pres-
ence of local qualitative factors influencing the level of regional entrepreneurial
activity, including business regulations.
Previously the authors determined the presence of a direct positive linear
correlation between 1) the density of small businesses and employment (includ-
Table 5. Verification of statistical hypotheses on the choice of the best model for each
partial panel
Group value Conclusion The
Test on base best
number factual critical hypotheses model
1 1) Wald (OR-model preferable 10.69 3.21 reject UR-
to FE-model) model
2) Wald (UR-model preferable –15.51 3.21 accept
to FE-model)
3) Wald (UR-model preferable –17.69 3.21 accept
to RE-model)
4) Breusch – Pagan LM-test 17.30 3.84 reject
(OR-model preferable to
RE-model)
5) Housman (RE-model PγI = 0.42, 0.05 reject
preferable to FE-model) PγLR = 1.4·10−4
2 1) Wald (OR-model preferable 9.72 3.1 reject UR-
to FE-model) model
2) Wald (UR-model preferable –27.21 3.1 accept
to FE-model)
3) Wald (UR-model preferable –27.91 3.1 accept
to RE-model)
4) Breusch – Pagan LM-test 16.58 3.84 reject
(OR-model preferable to
RE-model)
5) Housman (RE-model PγI = 0.20, 0.05 reject
preferable to FE-model) PγIC = 0.01
3 1) Wald (OR-model preferable 1199.40 3.89 reject UR-
to FE-model) model
2) Wald (UR-model preferable –696.29 3.89 accept
to FE-model)
3) Wald (UR-model preferable –686.30 3.89 accept
to RE-model)
4) Breusch – Pagan LM-test 18.96 3.84 reject
(OR-model preferable to
RE-model)
5) Housman (RE-model PγI = 4.2 · 10−84 0.05 reject
preferable to FE-model)
4 1) Wald (OR-model preferable 63.15 3.89 reject UR-
to FE-model) model
2) Wald (UR-model preferable –7.79 3.89 accept
to FE-model)
3) Wald (UR-model preferable –45.28 3.89 accept
to RE-model)
4) Breusch – Pagan LM-test 5452 3.84 reject
(OR-model preferable to
RE-model)
5) Housman (RE-model PγIC = 1 0.05 accept
preferable to FE-model)
ing self-employment) of the population, and 2) between the density of small
businesses and a gross regional product [15]. Increasing the density of small
businesses boosts employment and gross regional product. Therefore, the results
of this study can be used in state and municipal programs of small business
development to forecast economic development of each region of the Russian
Federation. The region classification (Table 3) demonstrates that a change in
economic factors might lead to a larger economic effect. In particular, a signif-
icant increase of entrepreneurial activity can be achieved by reducing the rate
on insurance contributions for individual entrepreneurs at the Federal level. For
most areas, both state and regional measures are important in order to improve
the investment climate. In two groups of regions the entrepreneurial activity can
be boosted by regulating the rate of small businesses loans.
References
1. The panel study of entrepreneurial dynamics. http://www.psed.isr.umich.edu/
psed/home
2. Global monitoring entrepreneurship. http://gsom.spbu.ru/gsom/centers/eship/
projects/gem/ (2013, 2016–2017)
3. Rating of regions by entrepreneurial activity. http://club-rf.ru/theme/388 (2014)
4. Investment ratings of regions of Russia. http://www.raexpert.ru/rankings/#r 1108
(2014–2016)
5. Small and medium entrepreneurship in russia. http://www.gks.ru/
wps/wcm/connect/rosstat main/rosstat/ru/statistics/publications/catalog
/doc 1139841601359 (2015)
6. Global entrepreneurship monitor. http://www.gemconsortium.org/ (2016)
7. Eliseeva, I.I.: Econometrics. Finansy i statistika, Moscow (2007)
8. Gorlov, A.: The factors of economic development of small enterprises in Russia:
the statistical analysis. Biometrika 84(2), 409–417 (2017)
9. for strategic initiatives, T.A.: Improving the investment climate in the Russian
Federation. http://asi.ru/investclimate/#rating (2014–2016)
10. Korneeva, E., Koren’, A.: Analysis of existing approaches to the definition of en-
trepreneurial activity. Naukovedenie 6, 1–8 (2013)
11. Liskina, E., Serova, O.: Econometric analysis of the dependence of entrepreneurial
activity of Russian population on the investmentss volume in fixed capital of small
enterprises. Mathematics and natural Sciences. Theory and practice 11, 224–230
(2016)
12. Nazarov, V.: The tax system of Russia in 1991-2008 years. http://www.ru-
90.ru/node/1170 (2010)
13. Pinkovetskaya, J.: Small entrepreneurship in the regions of Russia: production
functions on the panel data. Proceedings of ISA RAS 2, 80–84 (2014)
14. Serova, O.: Econometric analysis of the entrepreneurial activitys dependence of
Russian Federations population on the rates of insurance contributions to extra-
budgetary funds of the Russian Federation. In: New information technologies in
scientific research: Proceedings of XXI Russian scientific-technical conference of
students, young scientists and specialists. pp. 30–32. Ryazan (2016)
15. Serova, O., Liskina, E.: Econometric analysis of the impact of the level of en-
trepreneurial activity on employment and self-employment of the population and
the gross regional product of the Russian Federation. In: Mathematics: fundamen-
tal and applied research and education: Proceedings of the International scientific-
practical conference. pp. 275–280. Biometrika Trust, UK (2016)
16. Stough, R.: Entrepreneurship and regional economic development: Some reflec-
tions. Investigaciones Regionales Journal of Regional Research 36, 129–150 (2016)
17. Tchepurenko, A.: Entrepreneurship theory: New challenges and future prospects.
Foresight-Russia 9(2), 44–57 (2015)
18. Tchepurenko, A.: Entrepreneurship as a sphere of social studies Russia and inter-
national experience. Sociological Studies 9, 32–42 (2013)