CC BY
19процесса интернационализации, транснациональные банки повлияли на развитие процесса глобализации. Этот сравнительно новый научный термин уже получил широкое распространение, хотя еще не имеет устоявшегося и четкого определения.
Подводя итоги, можно сказать следующее, что касаемо традиционной глобализации, сейчас все более или менее ясно: ее субъекты - ТНК и ТНБ - будут также продолжать обеспечивать высокую конкуренцию на мировом валютном рынке и при этом же, сотрудничать, как это уже и происходит сегодня. Такой вопрос как кто в конечном итоге пересилит в конкурентной борьбе: американские или китайские транснациональные банки, то тут все будет зависеть от мощи банков, то есть размеров их активов, инновационных потенциалов и размеров той помощи, которую им в силах будут оказать правительства стран.
Возможности же власти будут во многом определяться уровнем развития страны, интересами, которые будут преследовать правящие элиты, направленностью политики, эффективностью средств экономически, а также готовностью общества этих стран поддерживать усилия власти.
Именно поэтому в наше время определяющее значение несет объективный квалифицированный анализ и прогнозы тенденций развития международных отношений, и практическая реализация выводов в конкретную социально-экономическую, а наряду с ней и оборонную и внешнюю политику.
Список литературы /References
1. Белова И.Н. Международные валютно-кредитные отношения. Учебник. Гриф УМО вузов России, М., 2010.
2. Сироткин В.Б. Обеспечение надежности инвестирования в реальный сектор экономики. М., 2003.
3. iFinance. [Электронный ресурс]. Режим доступа: http://global-finances.ru/krupneyshie-banki-mira-2017/ (дата обращения: 25.06.2018).
THE IMPACT OF ECONOMIC GROWTH ON THE SIZE OF THE FINANCIAL SECTOR OF THE ECONOMY ON CASE
OF INDONESIA Nguyen Thi My Linh (Russian Federation) Email: [email protected]
Nguyen Thi My Linh - Master's Degree Student, INTERNATIONAL FINANCIAL FACULTY, FINANCIAL UNIVERSITY UNDER THE GOVERNMENT OF THE RUSSIAN FEDERATION,
MOSCOW
Abstract: the main attention is paid for influence of economic growth on the size of the financial sector. In work author use quantity methods, statistical models for analysis, including the Darbin-Watson test, the Goldfeld-Kuandt test, and others. The relationship between economic growth and the financial sector in Indonesia is represented through these indicators: domestic credit provided by the financial sector as a percentage of GDP, GDP growth rate, unemployment rate, gross domestic savings on period between 1998 and 2017. Keywords: economy, influence, financial sector, Indonesia, Durbin-Watson test, Goldfeld-Quandt test.
ВЛИЯНИЕ ЭКОНОМИЧЕСКОГО РОСТА НА РАЗМЕР ФИНАНСОВОГО СЕКТОРА ЭКОНОМИКИ НА ПРИМЕРЕ
ИНДОНЕЗИИ Нгуен Тхи Ми Линь (Российская Федерация)
Нгуен Тхи Ми Линь - студент магистратуры, международный финансовый факультет, Финансовый университет при Правительстве Российской Федерации, г. Москва
Аннотация: в статье анализируется влияние экономического роста на масштаб финансового сектора. В работе используются количественные методы, статистические модели для анализа, в том числе тест Дарбина-Уотсона, тест Голдфелда-Куандта и другие. Взаимосвязь между экономическим ростом и финансовым сектором в Индонезии представлена через данные показатели: внутренний кредит, предоставляемый финансовым сектором в процентах от ВВП, темпы роста ВВП, уровень безработицы, валовые внутренние сбережения в период с 1998 по 2017 год.
Ключевые слова: экономика, влияние, финансовый сектор, Индонезия, тест Дарбина-Уотсона, тест Голдфелда-Куандта.
Indonesia, a country which is famous for not only its beauty of an archipelago nation, but also a position as the biggest economy in South East Asia. Since the Asian financial crisis of the late 1990s, Indonesia has had a charted remarkable economic growth. Indeed, for 2006, Indonesia's economic growth accelerated to 5.1% in 2004 and reached 5.6% in 2005 [2]. Real per capita income has reached fiscal year 1996/1997 levels. Growth was contributed primarily by domestic consumption that accounts for around three-fourths of Indonesia's gross domestic product. In 2004, the Jakarta Stock Exchange was the best performing market in Asia with an up by 42%. Constrains that continue to hold back on growth include bureaucratic red tape, low foreign investment levels, and very widespread corruption which is responsible for loss of 51.43 trillion Rupiah or 5.6573 billion US$ or approximately 1.4% of GDP on a yearly basis.
The unemployment rate in 2007 was 9.75%. In spite of a slowing global economy, Indonesia's economic growth reached to a ten-year high of 6.3% in 2007. This rate was enough to pull the rate of poverty from 17.8% to 16.6% based on the Government's poverty line as well as reversing the contemporary trend towards growth in the rate of jobless people, with the proportion dropping to 8.46% in 2008 [3]. Unlike many others of its export-dependent nations, it has found a way to get out of the recession, helped by strong domestic demand (which accounts for nearly two-thirds of the economy) and an enormous government fiscal stimulus package of around 1.4% of GDP which was announced earlier at that year. After China and India, Indonesia is currently the third fastest growing economy in the Group of Twenty industrialized and developing economies (G20). The $512 billion economy grew 4.4% in the first quarter of 2009, followed by an alteration in IMF's 2009 forecast for the country to 3-4% from 2.5%. Indonesia performs stronger fundamentals with the authorities implemented wide-ranging financial and economic reforms, including a fast reduction in external as well as public debt, strengthening of corporate and banking sector balance sheets and reducing bank vulnerabilities through higher capitalization and better supervision. Indonesia's gross national income per capita has risen steadily, from around $560 in 2000 to over $3,370 in 2017, going along with an impressive gain in poverty reduction, pulling the rate to more than a half since 1999, down to around 11% in 2017 [1]. Today, Indonesia is among the 10 largest economies over the world in term of purchasing power parity, and a member of G20.
However, according to ADB, the story is different in this country's financial sector. Despite having an enormous growth in economic development, the Indonesian financial sector remains small, undeveloped and still contains a shallow capital market. I conduct this report, using a strong and effective tool which is econometrics, with an aim to analyze the issue whether having the impact of economic growth on the size of the financial sector or not, and choose Indonesia as the experimental object because of the paradox which I mention above.
The econometrics model in this report is constructed to draw the effect of four factors which are GDP growth rate, unemployment rate, gross domestic saving in billion US$ and total reserve (exclude gold) in billion US$, on the domestic credit provided by financial sector in percentage of GDP.
All data using in this article is collected from Bloomberg. The analyzed period of time is 21 years, from 1998 to 2017. The independent variables in this model are GDP growth rate, unemployment rate, gross domestic saving and total reserve, which seem to affect economic growth in the long-term, while the dependent variable one is domestic credit provided by financial sector, which is one of the main index describing the depth of financial sector.
Model specification
Mathematically, the model was constructed as follow: Yt = C0 + C:*X1t+ C2*X2t+ C3*X3t+ C4*X4t + e
E(et) = 0
a(st) = const
In which:
Y: domestic credit provided by finanacial sector
X1t - X4t: GDP growth rate, unemployment rate, gross domestic saving and total reserve, respectively.
C0-4: parameters whose value need to be estimated. These estimations will help point out the effect of X1t, X2t, X3t and X4t on Yt.
E: expectation of disturbance term
et: disturbance term, which denote random factors which affect to the dependent variable but not exist in the model
g: the standard deviation
The estimated econometric model can be written down as: Yt = 67.48 - 1.27*X1t - 2.47*X2t + 0.43*X3t - 1.33*X4t + et R2= 0.79 F= 15.51
Dependent Variable: Y
Method: Least Squares
□ ate: 01/26/17 Time: 23:38
Sample: 1995 2015
Included observations: 21
Y=C{1 J+C{2)*X1 +C{3)*X2+C{4)*X3+C{5fX4
Coefficient Std. Error t-Statistic Prob.
C{1) 67.47530 6.423544 10.50437 0.0000
C{2) -1.265753 0.440409 -2.874050 0.0110
C{3) -2472732 0.908908 -2.720551 0.0151
C{4) 0 431652 0.095593 4.515505 0.0004
C{5) -1.326719 0.278957 -4.755993 0.0002
R-squared 0 794940 Mean dependent var 3273934
Adjusted R-squared 0.743676 S O. dependent var 12.63673
S.E. of regression 6.423127 Akaike info criterion 6.761944
Sum squared resid 660.1050 Schwarz criterion 7.010640
Log livelihood -66.00041 Hannan-Quinn enter. 6.315917
F-statistic 15.50653 □urbin-Watson stat 1 725272
ProbfF-statistic) 0.000023
Fig. 1. Results of variables
To see, whether there is any correlation between all chosen variables, a correlation matrix should be constructed in excel. Correlation matrix represents whether there occur a linear relationship between each exogenous variable and the explained variable. The bigger the correlation coefficient, the more linearly dependent in Y on a specific X. Correlation coefficient lies somewhere in between 0 and 1. So if it is equal to 0, the variable are independent, while if it is equal to 1, the variable are perfectly dependent.
Table 1. Table of variables
Y X1 X2 X3 X4
Y 1
X1 -0,18699 1
X2 -0,69265 0,160766 1
X3 -0,01666 0,3078 -0,1986 1
X4 -0,13857 0,224296 -0,13097 0,982148 1
If the sign is positive, it indicates that there is a positive linear dependence. If there is a "-'' in front of the correlation coefficient, it shows that there is a negative linear relationship between the variable and the Y.
It is useful to create scatter diagrams to graphically show, how the statistics for each variable are scattered through a trend line, representing the dependence of effect variable.
Here given 4 diagrams, representing 4 explanatory variables and Y response to the change of each of them.
-15,00 -10,00 -5,00 0,00
GDP growth
y = -0,5526x + 35,193
5,00
10,00
Fig. 2. The dependence of domestic credit
Diagram 1 represents the dependence of domestic credit provided by financial sector (% GDP) from the GDP growth rate.
y = -4,5334x + 65,984..
0,00 2,00
4,00 6,00 8,00 Unemployment (%)
10,00 12,00
Fig. 3. The relationship between domestic credit
Diagram 2 shows the relationship between domestic credit provided by financial sector (% GDP) and unemployment rate.
_ 80,00
it e (%60,00 "C -c
er yt tro40,00 U £ u
£ -a 8 20,00
g 2 .E 0,00 ^ o n
oD por ann
0,00 50,00 100,00 150,00 200,00 250,00 300,00 350,00 Gross domestic saving (billion US$)
Fig. 4. Domestic credit and GDS
Diagram 3 shows how domestic credit provided by financial sector (% GDP) depends on gross domestic saving.
R2 = 0,0192 7,611
0,00 10,00
20,00 30,00 40,00 50,00
Total reserve (billion US$)
60,00 70,00
Fig. 5. Relationship between domestic credit and the total reserve
Diagram 4 represents the relationship between domestic credit provided by financial sector (% GDP) and the total reserve.
In all graphs a trend line is used to graphically show a precise dependence between each of exogenous variables and the dependent one. The equation of each trend line and determination coefficients also represent the strength and character of dependence between variables. It is useful to create scatter diagrams to graphically show, how the statistics variable are scatted
In order to clarify whether this model is plausible or not, the following tests are performed.
1. R2 test
R2 is the Coefficient of Determination. It indicates how many points fall on the regression line. The value of R2 in my case is 0.79, which shows that 79% of total variation of Y is explained by the variations of factors Xi, and only 21% among them is the result of other factors which are not included in this model.
2. F-test
F-test is conducted in order to check the correctness of the R2 value and the quality of the specification of the model. We must use data analysis to assess the needed variables. If f-crit is lower that F, than the quality of specification of our model is high. In order to find f-crit, we use the function (=FINV) using excel, while value of F can be derived from the AVOVA table.
In my case, I have F= 15.51, while f-crit has a smaller value, equals to 3.01. Therefore, it can be concluded that the value of R2 is not random as well as the model has good quality of specification.
3. T-test
This test checks the significant of coefficients. We use the criterion |t| > t-crit, in which t, or t-stat, is the coefficients devided by its standard error, and t-crit is a value which can be calculated by using the function (=TINV) in excel. If the criterion above is fulfilled, then the variable has a significant effect in the model In my case, I have those value as follow: t-crit= 2.12
t-stat for X1 - ti= -2.87 ^ |t11 > t-crit ^ the variable is significant t-stat for X2 - t2= -2.72 ^ |t2| > t-crit ^ the variable is significant t-stat for X3 - t3= 4.52 ^ |t3| > t-crit ^ the variable is significant t-stat for X4 - t4= 4.76 ^ |t4| > t-crit ^ the variable is significant
4. Goldfeld - Quandt test
This test is based on the hypothesis that the error variance is related to a regressor X. The Gauss-Markov 2nd condition states that the disturbance tern in a regression model is homoscedastic; it has the same potential distribution in all observations. The GQ test is done in order to check the residuals of a model and verify the condition of Gauss-Markov theorem.
To conduct the G-Q test, first I will choose an independent variable which has the highest t-stat, sort it in term of ascending number, split the data into 2 equal part, and make regression analysis for both of them. After that, we calculate the GQ index by dividing residual sum squares of the first part for one of the second part, or GQ= RSS1/RSS2, and compare to these conditions:
If GQ-1 < f-crit GQ: residuals of the model are homoscedastic If GQ-1 > f-crit GQ: residual of the model are heteroscedastic In which f-crit GQ can be calculated by using formula (=FINV)
The results from my data are GQ= 1.89, so GQ-1= 0.53, while f-crit GQ= 5.05. We can see that GQ-1 < f-crit GQ, hence the residuals of the model are homoscedastic, the second Gaus-Markov theorem is confirmed and we may use the OLS (ordinary least squares) in order to estimate model coefficients.
5. Durbin - Watson test
The aim of this test is to detect the presence of autocorrelation in the residuals from a regression analysis. To carry out this test, first, we need to find out the lower level of D-W statistics and the upper one, which are denoted by dl and du.
In my model the significant level is 5%, the number of my observations is 21, the number of estimated parameters is 4, so we have di= 0.927, du= 1.812
The DW statistic, which is used to compare to dl and du, can be derived from the result of using Eviews application. And in my case, I have DW= 1.73. To check our model is plausible or not, we make a string with 4 intervals: 0 to dl, 4 - dl to 4, dl to du and 4 - du to 4 - dl and du to 4 -du. Then we watch the position of the value of DW statistic. According to the law, if the observed number lays between 0 and dl, there is positive autocorrelation in residuals, if it is between 4-dl and 4, there is negative autocorrelation in residuals. In both cases we cannot use OLS in order to estimate the coefficients. If the above number lays between dl and du, or between 4 - du and 4 -dl, there is no information about autocorrelation. If it is between du and 4 - du, there is no autocorrelation in residuals, the third Gauss-Markov condition is confirmed and we may use OLS for estimating the coefficients.
We can see that dl < DW stat < du, hence there is no information about autocorrelation.
The last step of our model-testing is checking confidence intervals in order to make sure that our model is adequate. We have to make sure that the estimate value of Y lies between the lower and the upper boundaries of Y. That is where we finally use the last interval of 2017.
1) First, we must calculate the estimated domestic credit provided by financial sector (% of GDP) as of 2017: Yt = C(0) + C(1)*X1t + C(2)*X2t + C(3)*X3t + C(4)*X4
2) Then, we calculate the boundaries:
Lower Boundary (Yt-) = Estimated Yt - t-crit * standard error
Upper level (Yt+)=Estimated Yt + t-crit * standard error
We have obtained:
Table 2. Results of calculations
Real Yt 39.07 Yt- 21.68
Estimated Yt 35.29 Yt+ 48.90
It follows that the real value of Yt in 2017 belongs to the confidence interval, so the model is adequate and may be used for forecasting.
The last step before forecasting is estimating of a model error. It is calculating in the following way:
O = |Ytheoretical - Yreal|/ Yreal*100%
o = |35.29 - 39.07|/ 39.07*100% = 9.67%
this means that in 90.33% of cases this model would give an exact right result.
Conclusion
To sum up, I have analyzed the relationship between economic growth and financial sector on Indonesia, by using domestic credit provided by financial sector in percentage of GDP to represent for the financial sector, as well as using GDP growth rate, unemployment rate, gross domestic saving in billion US$ and total reserve (exclude gold) in billion US$ to stand for economic growth, in the period between 1998-2017.
After examination of the econometric model and passing a number of tests, I come to a conclusion that my model is plausible in chosen condition, as well as other conclusions which will be listed as follow:
• There is a negative relationship between domestic credit provided by financial sector and GDP growth rate as well as unemployment rate and total reserve. Meanwhile, that kind of connection is positive toward remain variable.
• If there is an increase of 1% in GDP, the domestic credit provided by financial sector would decrease by 1.27% as a percentage of GDP.
• If there is an increase of 1% in unemployment rate, the domestic credit provided by financial sector would decrease by 2.47% as a percentage of GDP.
• If there is an increase of 1 billion US$ in gross domestic saving, there would be a further increase of 0.43% in domestic credit provided by financial sector as a percentage of GDP.
• If there is an increase of 1 billion US$ in total reserve, the domestic credit provided by financial sector would decrease by 1.33% as a percentage of GDP.
• If all independent variables equal to zero, the domestic credit provided by financial sector would be 67.48% of GDP, thank to other factors which cause effect but not exist in the model.
References / Список литературы
1. Bloomberg's database. [Electronic resource]. URL: https://www.bloomberg.com/europe/ (date of acces: 10 April 2018).
2. The Growth and Development of the Indonesian Economy - Stephen Elias and Clare Noone. [Electronic resource]. URL: https://www.rba.gov.au/publications/bulletin/2011/dec/pdf/bu-1211-4.pdf/ (date of acces: 13 April 2018).
3. Constrains to Indonesia's economic growth - Steven R. Tabor. [Electronic resource]. URL:https://www.adb.org/sites/default/files/publication/178041/ino-paper-10-2017.pdf/ (date of acces: 10 April 2018).
4. Accounting for Indonesia's economic growth: recent past and near future - Pierre van der Eng. [Electronic resource]. URL: http://www.uq.edu.au/economics/cepa/docs/seminar/papers-nov2006/Van-der-Eng-Paper.pdf/ (date of acces: 13 April 2018).
HOW DIGITAL PLATFORMS INFLUENCE THE LABOR MARKET Voronina E.I. (Russian Federation) Email: [email protected]
Voronina Ekaterina Igorevna - Student, DEPARTMENT OF FINANCE AND MANAGEMENT, INSTITUTE OF LAW AND MANAGEMENT, TULA STATE UNIVERSITY, TULA
Abstract: the concept of digital platforms in a foreign economy is considered and analyzed Internet platforms of talents, ascertaining the reasons for their appearance, and the positive influence they have on the economy as a whole are called upon. The results of a survey on the topic of digital platforms among students of the Tula State University are presented, the corresponding conclusions are drawn. The topic is relevant, as the development of such Internet platforms is gaining rapid pace in all countries of the world.
Keywords: digital economy and platforms, index of readiness for transition to digital platforms, online talent platforms.
КАК ЦИФРОВЫЕ ПЛАТФОРМЫ ВЛИЯЮТ НА РЫНОК ТРУДА Воронина Е.И. (Российская Федерация)
Воронина Екатерина Игоревна - студент, кафедра финансов и менеджмента, институт права и управления, Тульский государственный университет, г. Тула
Аннотация: в данной статье рассмотрено понятие цифровых платформ в международной экономике, проанализированы интернет-платформы талантов,
117
