CC BY
23of organizational performance. The Academy of Management Journal, 1996; Vol. 3, No. 4,pp. 949-969.
12. Huselid, M.A. The Impact of human resource management practices on turnover, productivity, and corporate financial performance. The Academy of Management Journal, 1995; Vol. 38, No. 3, pp. 635672.
13. Katou, A.A. and Budwar, P.S. The effects of human resource management policies on organizational performance in Greek manufacturing firms. Thunderbird International Business Review, 2007; vol. 49, No. 1, pp. 1-35.
14. Petrescu, A.I. and Simmons, R. Human resource management practices and workers' job satisfaction. International Journal of Manpower, 2008; Vol. 29, No. 7, pp. 651-667.
15. Krishnamurthy, V. The work ethos in Maruti udyog. Productivity. New Delhi, 1985.
16. Schuler, R.S., Dolan, S.L. and Jackson, S. Trends and emerging issues in human resource management: global and trans-cultural perspectives - introduction, International Journal of Manpower, 2001; Vol. 22, No. 3, pp. 195-7.
17. Budhwar, P.S. and Debrah, Y.A. Human resource management in developing countries (ed.). London: Routledge 2001.
18. Singh, K. Impact of HR practices on perceived firm performance in India. Asia Pacific Journal of Human Resources, 2004; Vol. 42, No. 3, pp. 301-317.
19. Yeganeh, H. and Su, Z. An Examination of human resource management practices in Iranian public sector. Personnel Review, 2008; Vol. 37, No. 2, pp. 203-221.
20. Absar, M.M.N., Azim, M.T., Balasundaram, N. and Akhter. S. Impact of human resources practices on job satisfaction: evidence from manufacturing firms in Bangladesh. Petroleum-Gas University of Ploiesti BULLETIN, 2010; Vol. LXII (2), pp.31 - 42.
21. Mahmood, M.H. The institutional context of human resource management: Case studies of multinational subsidiaries in Bangladesh. Unpublished doctoral thesis, University of Manchester, UK. 2004.
22. Nair and Nair. Personnel Management and Industrial Relations. S. Chand and Company Lt, New Delhi. 1999; pp. 24.
23. Mathis, R.L. and Jackson, J.H. Training human resources. Human Resource Management. 10th ed., Thomson Asia Pte Ltd., Singapore. 2005.
24. Ahmed, K. Labour Movement in Bangladesh. Dhaka, Bangladesh. 1978; pp. 27.
25. Filmer, D. and Pritchett, L. Estimating Wealth Effects without Expenditure Data-Or Tears: With an Application to Educational Enrollments in States of India. World Bank Policy Research Working Paper 1994, Development Economics Research Group, World Bank, Washington, DC. 1998.
26. Bhatia, B.S., Verma, H.L. and Garg, M.C. Studies in Human Resource Development. Understanding HRD: Basic Concept (ed.), 1997; Vol. 1, pp. 263280.
27. Ghosh, B. Human Resource Development and Management, Vikas Publishing House Pvt. Ltd., New Delhi. 2000.
28. Uddin, M.A., Habib, M.A. and Hassan, M.R. Human Resource Management Practices in Power Generation Organizations of Bangladesh: A Comparative Study of Public and Private Sector. Journal of Business Studies, Southeast University, 2007; Vol. III. No. 2.
29. Aswathappa, K. Incentive payments. Human Resource Management, 4th ed. New Delhi: Tata Mac-Graw-Hill, 2005; pp. 520.
30. Subramanian, K.N. Wages in India, Tata Mac-Graw-Hill, New Delhi, 1979; pp.194.
31. Pfeffer, J. The Human Equation: Building Profits by Putting People First, Boston, MA: Harward Business School Press. 1998.
ECONOMIC GROWTH AND FOREIGN TRADE: EVIDENCE FROM RUSSIA
Tetin I.
PhD, Assistant Professor I-Shou University, International Finance Dpt.
Kaohsiung, Taiwan Antonenko E.
PhD,
South Ural State University, Research Center for Sport Science
Chelyabinsk, Russia
Abstract
This study builds a VAR model to analyse the dependency of exports, imports and GDP growth through the prism of oil prices for the past 20 years. Causal relationships are estimated using quarterly data from 2000 to 2020. We utilise the Johansen procedure for cointegration testing and Granger causality testing. The results do not confirm the existence of long-run relationships between foreign trade and economic growth in Russia. Moreover, short-run relationships between foreign trade and economic growth in Russia are not verified.
Keywords: economic growth, export, import, VAR, export-led, growth-led.
Introduction
Regardless of size and level of economic development, the economics of any country is interconnected with others through foreign trade. It is known that the size of the country is negatively correlated with its dependency on imports, meaning that smaller countries have a greater degree of openness in their economics. The flow of imported goods and services thus contributes to the development of small economies. Following Hecksher-Ohlin (1) theory, developing countries should import products to fulfil the scarcity of labour and natural resources. On the other hand, exporting the excessive capacity of resources leads to expanding consumption, investment and public spending through the foreign trade multiplier. Foreign trade allows local manufacturers to engage in large-scale production (2) and significantly benefit from trade, especially when domestic markets are crowded with supply.
Two competitive hypotheses between export and economic growth exist. The first one - export-led growth (3,4) raises demand for technological innovation (5,6) and enables more effective and efficient use of resources. The second one prompts the opposite relationship, where growth rates can increase the export (7,8). This article attempts to evaluate the causal relationship between Export, Import and GDP growth in Russia and test whether there is export-led growth or growth-led export in Russia.
Materials and Methods
Quarterly data for exports and imports in billions USD, and real GDP in current prices obtained from the
United Archive of Economic and Sociological Data (www.sophist.hse.ru). Brent oil prices — from the portal (www.bhom.ru). The sample consists of 84 observations, covering the period from 2000q1 to 2020q4.
Methods used in this study include stationarity testing with Augmented Dickey-Fuller (ADF) unit-root test (9), VAR model estimation, Johansen procedure for cointegration testing, and Granger causality testing. The analysis starts with unit-root testing for the null hypothesis: series contains a unit root. If the null hypothesis is rejected, the series is considered stationary. We perform a unit root for the first differenced series, see Table 1. Next, we estimate Vector Autoregressive Model and choose an optimal number of lags with Schwarz Information Criterion, Table 2. After that, the Johansen procedure is performed (10). This procedure tests the null hypothesis of k cointegrating vectors against the alternative of n cointegrating vectors using the following statistic:
n
JTrace = -N ^ ln(1 - Ai)
i=k+1
Here N - sample size, ^ - is the largest i canonical correlation. If the test statistic is greater than critical values, the null hypothesis of k cointegrating vectors is rejected, Table 3. Using the results of the Johansen procedure, VAR or VECM models are estimated (11). In our case, only VAR models can be estimated, and only short-run relationships can be obtained. Following VAR models of order four are obtained:
m
AGDPt =ci+^ aubGDPt-i + ^ ßuAEx— + ^ Y^Alm— + ^ SliAOHt-i + eu ;
i=1 i=1 i=1 i=1 mm mm
AExt = C2+^ a2iAExt-i + ^ ß2iAGDPt_i + ^ y2iAlrnt_i + ^ S2iAOHt-i + £2t\
i=1 i=1 i=1 i=1 mm mm
AImt =c3+^ avMmt-i + ^ ßsiAGDP^ + ^ y^AEx— + ^ 8-iiA0Ht_i + e3t)
i=1 i=1 i=1 i=1 mm mm
AOilt = C4+^ a.4iA0ilt-i + ^ ß4iAGDPt-i + ^ Y^AEx^ + ^ S^AIm^ + £41.
lt = + ^ a4iA^ilt-i + ^ H4i^wl t-i + ^
i = 1 i=1 i=1 i = 1 VAR estimates are given in Table 4, and Table 5 Results
includes VAR Granger causality test results. Granger Results of the ADF test for unit root are presented
causality test tests bilaterally whether the lags of the ex- in Table 1. ADF test shows that the series under concluded variable affect the endogenous variable using sideration are stationary in first differences. Therefore, the null hypothesis: the lagged coefficients are signifi- it is possible to estimate the cointegration between var-cantly different than 0. It also performs the joint test, iables with the Johansen cointegration test. that the lags of all other variables affect the endogenous variable.
Table 1
ADF Unit-root test
m
m
m
Variable t-statistic p-value
AGDP -2.962** 0.0429
AEx -4.667*** 0.0002
Aim -4 710*** 0.0002
AOil -7.367*** 0.0000
Notes: A — first difference operator, *** — denotes significant at 1% level of significance, ** — denotes significant at 5% level of significance
We determine the optimal lag length based on the the VAR lag order selection criteria, we choose the op-VAR model (see Table 2). According to the results of timum lag length of 4, which minimises the Schwarz
information criterion value.
Table 2
VAR lag order selection criteria
Lag LogL LR FPE AIC SC HQ
0 -1436.688 NA 3.43e+11 37.91283 38.03550 37.96186
1 -1379.429 106.9837 1.16e+11 36.82707 37.44042 37.07220
2 -1336.213 76.19575 5.69e+10 36.11088 37.21491 36.55210
3 -1284.421 85.86548 2.24e+10 35.16898 36.76370 35.80631
4 -1236.299 74.71656 9.77e+09 34.32365 36.40905* 35.15708
5 -1204.659 45.79460 6.66e+09 33.91208 36.48815 34.94160*
6 -1184.208 27.44757* 6.18e+09* 33.79494* 36.86170 35.02057
7 -1171.638 15.54651 7.19e+09 33.88522 37.44265 35.30694
Notes: * indicates lag order selected by the criterion; LR: sequentially modified LR test statistic (each te st at 5% level); FPE: Final prediction error; AIC: Akaike information criterion; SC: Schwarz information criterion; HQ: Hannan-Quinn information criterion
Having the results of VAR lag order selection criteria, we imply lag 4 to estimate the cointegrating relationship (see Table 3).
Table 3
Johansen cointegration test
Hypothesized Trace 5%
No. of CE(s) Eigenvalue Statistic Critical Value P-value
None * 0.409408 97.28560 47.85613 0.0000
At most 1 * 0.313627 56.20854 29.79707 0.0000
At most 2 * 0.186119 26.85452 15.49471 0.0007
At most 3 * 0.129204 10.79107 3.841466 0.0010
Note: * denotes rejection of the hypothesis at the 0.05 level
The results indicate no cointegration relationships since we reject the null hypothesis at 5% significance levels for None, at most one, at most two, and most three cointegrating equations. In this case, we only estimate the VAR model to determine short-run relationships between the variables.
Analysing VAR equations, in Table 4, we can see significant short-run relationships between AOil prices
in the previous quarter and AExport volume and Almport volume. AGDP is almost perfectly (adjusted R2 = 0.879) explained by previous autoregressive lags, while other factors do not help in explaining current values of economic growth.
Table 4
VAR models
Variables AGDP AEx Aim AOil
AGDP(-1) -0.125724 -0.000929 -0.001579 -0.002009
[-1.12400] [-0.61418] [-1.84141] [-1.03650]
AGDP (-2) -0.394274 -0.001253 -0.001212 -0.001406
[-3.57870] [-0.84085] [-1.43488] [-0.73613]
AGDP(-3) -0.342233 -0.003230 -0.001005 -0.002657
[-2.85508] [-1.99226] [-1.09403] [-1.27893]
AGDP (-4) 0.768829 0.000369 0.000242 0.001094
[ 6.15276] [ 0.21860] [ 0.25247] [ 0.50538]
AEx(-1) 1.374598 -0.339639 -0.140396 0.424125
[ 0.05806] [-1.06084] [-0.77365] [ 1.03372]
AEx(-2) 35.96478 -0.254441 -0.119703 0.157904
[ 1.33542] [-0.69860] [-0.57983] [ 0.33831]
AEx(-3) -18.88491 0.011829 -0.009099 0.188214
[-0.70738] [ 0.03276] [-0.04446] [ 0.40679]
AEx(-4) -8.333930 0.078480 0.109133 -0.011521
[-0.40493] [ 0.28196] [ 0.69174] [-0.03230]
Alm(-l) -12.14459 -0.075232 -0.107073 -0.026317
[-0.65467] [-0.29988] [-0.75297] [-0.08186]
AIm(-2) -18.72670 0.370311 0.031719 0.474440
[-1.02305] [ 1.49592] [ 0.22605] [ 1.49554]
AIm(-3) 17.44191 0.247234 -0.096609 0.681754
[ 0.89865] [ 0.94191] [-0.64935] [ 2.02677]
AIm(-4) 6.873037 0.200590 0.368429 0.090448
[ 0.34199] [ 0.73803] [ 2.39154] [ 0.25968]
AOil(-l) 23.08089 0.778185 0.497379 0.053104
[ 1.47063] [ 3.66638] [ 4.13428] [ 0.19524]
AOil(-2) -14.43849 0.071705 0.104087 -0.587055
[-0.65790] [ 0.24160] [ 0.61873] [-1.54348]
AOil(-3) 21.24149 0.078930 0.174687 -0.300757
[ 0.99699] [ 0.27394] [ 1.06962] [-0.81452]
AOil(-4) -5.847839 -0.261200 -0.217200 -0.384039
[-0.32640] [-1.07804] [-1.58154] [-1.23684]
Const 338.7853 2.069686 1.598234 0.740859
[ 2.61301] [ 1.18039] [ 1.60812] [ 0.32971]
Adj. R-squared 0.879744 0.523623 0.763307 0.109008
Notes: t-statistics in [ ]
After obtaining VAR coefficient estimates, we causal relations between the variables (Table 5). This perform the VAR Granger Causality test, evaluating test might serve as a complement to VAR estimation
Table 5
VAR Granger causality test
A(GDP) A(EX) A(Im) A(Oil)
A(GDP) - 7.250022 (0.1233) 7.618973 (0.1066) 5.389785 (0.2496)
A(EX) 4.347338 (0.3610) - 1.492795 (0.8279) 1.139698 (0.8879)
A(Im) 2.755698 (0.5995) 3.565626 (0.4680) - 5.976272 (0.2009)
A(Oil) 9.303388 (0.0539) 20.21027 (0.0005) 27.39297 (0.0000) -
All 29.43571 (0.0034) 63.36502 (0.0000) 75.70405 (0.0000) 13.92308 (0.3056)
Conclusion
The results of Granger causality show that we can reject only two out of twelve null hypotheses between the variables in the sample. The hypothesis that has been rejected are: all the lag coefficients of AOil do not have causal effects on AExport, and all the lag coefficients of AOil do not have causal effects on AImports. These results confirm the significance of t-statistics for AOil(-1) coefficients in Table 4.
The null hypothesis that all lags in front of all variables do not have causal effects on AGDP, AExport and Almport are rejected, implying that lags of all other variables affect the conforming endogenous variable. Therefore, estimated VAR equations are statistically significant.
Analysing VAR equations, we can draw the following inference. Oil prices are predeterminants of exports and imports volumes. Exports do not influence the GDP volume. Therefore, the export-led hypothesis of economic development is not confirmed. The opposite relationship from GDP growth to exports is also unconfirmed; thus growth-led hypothesis is inconsistent. The economy of Russia unable to engage benefits of
foreign trade: no evidence of the influence of imports on GDP volume. Moreover, GDP growth does not influence the volumes of imports.
References
1. Feenstra RC. Advanced international trade: theory and evidence. Princeton, N.J: Princeton University Press; 2004. 484 p.
2. Helpman E, Krugman PR. Market structure and foreign trade: increasing returns, imperfect competition, and the international economy. Cambridge, Mass: MIT Press; 1985. 271 p.
3. Wilbur WI, Haque MZ. An investigation of the export expansion hypothesis. J Dev Stud. 1992 Jan;28(2):297-313.
4. Giles JA, Williams CL. Export-led growth: a survey of the empirical literature and some non-causality results. Part 2. J Int Trade Econ Dev. 2000 Jan;9(4):445-70.
5. Balassa B. Exports and economic growth. J Dev Econ. 1978 Jun;5(2):181-9.
6. Grossman GM, Helpman E. Innovation and growth in the global economy. Cambridge, Mass: MIT Press; 1991. 359 p.
7. Vernon R. International Investment and International Trade in the Product Cycle. Q J Econ. 1966 May;80(2):190.
8. Henriques I, Sadorsky P. Export-Led Growth or Growth-Driven Exports? The Canadian Case. Can J Econ. 1996 Aug;29(3):540.
9. Fuller WA. Introduction to Statistical Time Series: Fuller/Introduction [Internet]. Hoboken, NJ,
USA: John Wiley & Sons, Inc.; 1995 [cited 2021 Jul 16]. (Wiley Series in Probability and Statistics). Available from: http://doi.wiley.com/10.1002/9780470316917
10. Johansen S. Statistical analysis of cointegra-tion vectors. J Econ Dyn Control. 1988 Jun;12(2-3):231-54.
11. Stock JH, Watson MW. Vector Autoregressions. J Econ Perspect. 2001 Nov 1;15(4):101-15.
