CC BY-NC-ND
6
DOI: 10.17323/2587-814X.2021.4.61.75
Modeling the links between institutional and actual globalization in the countries of the world
Elena D. Kopnova
E-mail: ekopnova@hse.ru
Lilia A. Rodionova
E-mail: lrodionova@hse.ru
National Research University Higher School of Economics Address: 20, Myasnitskaya Street, Moscow 101000, Russia
Abstract
The paper is devoted to modeling the links between the institutional and actual level of globalization in the countries of the world. Vector models of error correction, quantile regression, and a stochastic frontier model are considered. As a measure of globalization and its components, the KOF-index of globalization system is used, which allows us to analyze individual globalization processes in the economy, social sphere and politics. According to 2020 data, we determine the dynamic relations between the actual and institutional components of globalization, and the priority of the institutional component for informational and financial globalization is revealed. The example of financial globalization shows the uneven degree of influence of the institutional component on the actual globalization, in particular, its prevailing importance for less globalized countries, indicating the alignment of the degree of internationalization in the global financial system. The degree of effectiveness of the impact of institutional measures, together with the overall level of well-being on the actual financial globalization is analyzed. It is shown that the spread across the countries of the world in the efficiency indicator is almost 70%. Almost 10% of countries have a low efficiency of up to 50%. One third of the countries has average efficiency (50-75%). The share of countries with high efficiency over 75% is about 60%.
Key words: KOF index of globalization; cointegration; vector error correction model; forecast errors variance decomposition; quantile regression; stochastic frontier model.
Citation: Kopnova E.D., Rodionova L.A. (2021) Modeling the links between institutional and actual globalization in the countries of the world. Business Informatics, vol. 15, no 4, pp. 61—75. DOI: 10.17323/2587-814X.2021.4.61.75
Introduction
Globalization is the most important factor of social progress. It is determined by the strengthening of economic, social and political interactions of countries and peoples, regardless of national borders [1]. According to some data [2], globalization is capable of raising world GDP per capita by almost a third. The integration of trade and investment flows, the convergence of markets and the development of multinational corporations characterize economic globalization. Social globalization is expressed in the development of communication technologies, international cultural centers and personal contacts. Political globalization manifests itself in the activities of organizations that, in accordance with the principles of international law, make it possible to unite the forces of the countries of the world in the fight against global problems. The positive effects of globalization include the optimization of production due to the cross-country division of labor and access to innovation [3], the diversification of financial risks with the attraction of foreign investment [4], the increase in the level of human capital development due to the development of information technology and the international education system [5]. Researchers usually identify the main problem of globalization as an increase in income inequality [6]. Among the negative effects, there are also risks to economic security [7], violation of human rights [8] and loss of ethnic identity [9].
To measure globalization, index systems reflecting its structure are used [10]. The most popular of them is the system of the KOF-index of globalization from the Swiss Economic Institute1. The appearance in 2018 of the latest edition of the methodology of its calculation with a radical expansion of the structure and informa-
tion base significantly expanded the possibilities of studying the problems of globalization [11]. In addition to the integral indicator, the system of this index contains sub-indices of economic, social and political globalization, which in turn are divided into separate components. The economic sub-index includes sub-indices of trade and financial globalization, and the social subindex includes personal, informational, and cultural globalization. Each of these indicators is further divided into de facto and de jure categories. The indicators de facto measure the actual flows between countries (for example, the amount of imports), while de jure — their institutional capabilities (for example, import taxes). Each indicator is formed according to world official statistics since 1970, published with a delay of two years, measured on a 100-point scale. A total of 42 variables are used. The principal component method is used for calculation, as well as the panel normalization method2.
Most of the works on the study of globalization are devoted to analysis of its impact on the well-being of the population. The most interesting of them are the works [12—18], which use econometric tools for data analysis. However, it should be noted that these studies are limited by lack of results from studying the structure of the globalization process itself, the relationship between its individual components. This reorganization of the KOF-index of globalization in 2018 marks noticeable progress in the development of the statistical methodology for measuring globalization and expands the possibilities of its systematic analysis. Taking into account this reorganization, the authors set a goal to analyze the links between institutional and actual levels of globalization. Two main tasks were put forward. The first is to investigate the trends of dynamic relationships between de
1 KOF (KonjunkturforschungssteUe) Globalisation Index: https://www.kof.ethz.ch/en/ forecasts—and—indicators/indicators/kof—globalisation—index.html
2 The structure of the KOF index indicating the weights of individual indicators is given in the Appendix ( Table A1)
jure and de facto sub-indices. And the second is to study the degree of influence of the institutional component and the effectiveness of its application to form the actual component of globalization for the countries of the world.
1. Methods 1.1. Data
Globalization was measured by the KOF-sub-indices — the de facto and de jure globalization index. All indicators were considered in the period from 1970 to 20183. Figures 1—4 show graphs of the analyzed time series averaged by countries of the world for the Globalization KOF-index (gl) and its components: economic (ec), social (soc) and political (pol) globalization; trade (tr) and financial (fin) globalization; personal (per), information (inf) and cultural (cul) globalization. The corresponding subindices are de facto and de jure denoted with the addition of the symbols _df, _dj.
Figure 1 shows that the acceleration of globalization in the world began in 1994. At the same time, if previously the sub-index was de facto superior to de jure, now the de jure subindex prevails. There is a noticeable divergence of trends. De jure, in general, is growing faster. Figure 2 shows that this is due to the social and political index de jure.
At the same time, Figure 2 shows that the de facto political sub-index remains noticeably smaller. Among economic sub-indices, on the contrary, the de facto sub-index is superior. The connection of the components of economic globalization is noticeable. It can be seen how the de facto sub-index follows the de jure subindex. The connection of the sub-indices of social globalization is also visible, but not so noticeable. The connection between the components of political globalization is even less noticeable.
Fig. 1. Sub-indices of the de facto and de jure KOF-index of globalization
Fig. 2. De facto and de jure sub-Indices of the components of the globalization KOF-index
The de facto predominance in the economic index is due to the relatively strong growth of the corresponding component of financial globalization. As part of the trade globalization index, in recent years, the values of the de facto and de jure sub-indices have been converging.
Figure 4 shows that the prevalence of the de jure sub-index in the indicator of social globalization is achieved at the expense of all components. However, a strong acceleration is noticeable for the dynamics of the de facto information globalization index in recent years. At the same time, the relationship of this index with the de jure index is visible. The cultural component of globalization is characterized by a relatively low value of the de facto sub-index.
3 https://kof.ethz.ch/en/forecasts-and-indicators/indicators/kof-globalisation-index.html
70
65 -\
60
55
50
45
40
35
30
-tr df
fin df
fin_dj
• ■■"'*' Vi
1—n—I—r~n—n—I—n—I—I—n—I—I—I—I—I—I—n—r
Fig. 3. De facto and de jure sub-indices of the components of the economic sub-index of the KOF-index of globalization
80 ■
75
70
65
60
55
50
45
40
35
30
............per_df per dj - inf df
---cul_df - cul_dj
...............................'.....
c^ c^ c^ c^
Fig. 4. Sub-Indices of de facto and de jure components of the social sub-index of the KOF-index of globalization
Figure 5 shows the values of the de jure (abscissa axis) and de facto (ordinate axis) subindices of the KOF-index of globalization at the time of 2018 for 196 countries of the world. It can be seen from the figure that the sub-indices are strongly correlated. The sample value of the paired coefficient of correlation was 0.87.
Additionally, when analyzing the spatial sample at the time of 2018, the income index was used, obtained on the basis of the logarithm of
gross national income (GNI) per capita at purchasing power parity (PPP) in 2017 prices in US dollars4 and measured on a 100-point scale. The indicator was calculated as the ratio of the logarithm of the GNI growth index to the logarithm of its maximum value5. The minimum value of GNI was assumed to be equal to $100, as the minimum fixed in official statistics. The maximum value was set at $75,000 in accordance with the phenomenon of the immutability of the level of well-being for countries with a higher level of GNI [19]. The Appendix contains a list of countries used in the analysis for which the index values are available. Figure 5shows its values for these countries at the time of 2018. The figure shows that the spread of values is quite large and amounts to almost 70%, which indicates a high differentiation of well-being in the countries of the world.
1.2. Procedures
The methodology for the study of dynamic relationships was based on the idea of cointegra-tion analysis of random processes using the Vector Error Correction Model (VECM) [20]. Nine models of the relationship between the de facto and de jure sub-indices were constructed for the KOF-index of globalization and its components.
100 90 80 70 -60 -50 -40 30 -I 20
.....i.................I"
.....
........................!......
;
• • V *
......
...............................
.....
f ■i.....
......I........•......h-.........i............f-
0 40 50
60
70
80
90 100
Fig. 5. De jure and de facto sub-indices of globalization 2018 for the countries of the world
4 https://data.worldbank.org/indicator/NY.GNP. PCAP. PP. KD?view=chart
5 Income index=ln(GNI/GNI . )/ln(GNI /GNI . )x100
v ' min'' v max' mm'
100 90 -\ 80 70 60 50 40 30 20
• .......'......*......> • • '.. •
. • • • .. .
...*.. .«...•. ;........................... ?............,..;...
. ' ' . ' # . ' .........••...............*..........**..............?.......•..................•.......:
-i-i-i-i—
0 50 100 150 200
Fig. 6. Income index in the countries of the world, 2018
The general form of the model:
AX^fi.+afl'X^+f^TjX^+U,,
M
t -1>2, ...,T.
AXt=X-X1_l,a=\\aiJ\\kxr,fl = №J\Lr, U~N(O,It®I:u),U = (u1,...,Ut),
z ,. = k
where the components of the vector Xf are the processes analyzed in the work; the vector fit contains deterministic components for each of these processes: a trend and a constant; r — cointegration rank; Ut — errors vector.
For each time series, tests were applied for the presence of a single root of the characteristic equation of the corresponding process in accordance with the algorithm of the Dolado procedure [21]. ADF (Augmented Dickey—Fuller) and KPSS (Kwiatkowski—Schmidt—Shin) tests [20] were used. The Johansen approach [22] was used to evaluate the parameters of the coin-tegration ratio and the error correction model. The optimal specification of the models was selected based on the Bayesian Information Criterion (BIC) and the model's compliance with its assumptions. The remnants of the mod-
els were tested for the absence of autocorrelation and compliance with the normal distribution law (multidimensional analogues of the Breush—Godfrey and Jarque—Bera LM test). Durnik—Hansen orthogonalization [23] was used for the remnants of the VEC model.
The characteristics of long-term dynamic relationships for cointegrated processes were determined using testing of variables for weak exogeneity relative to the parameters of the error correction model. For this purpose, the statistical significance of the estimates of the components of the correction matrix a was analyzed, since the insignificance of the estimate a„ means that when the processes deviate from long-term equilibrium, the corresponding i-th variable is not corrected. To draw conclusions about strict exogeneity, the Toda—Yamamoto approach [24] was used with the choice of the number of lags in the test model according to the BIC criterion. The degree of impact of the analyzed processes on each individual process was measured using the decomposition of the variance of its prediction error, in which the selected process occupied the last position in the recursive order of causality by Wold (Wold-causality) [25]. As an impact measure for each individual process, the corresponding proportion of the estimate of the variance of the forecast error, the maximum for 10 years, was considered.
To analyze the degree of influence of the institutional factors of globalization on the actual globalization, regression models were used, estimated from the prolog data of the de facto and de jure sub-indices for 2018, averaged across the countries of the world. Taking into account the heteroscedasticity of the errors of the usual regression
= *,./?(/= 1,2,...,«)
quantile regression [26] was used for quantiles corresponding to probabilities 0.25, 0.5 and 0.75:
qr\Yi\Xi-] = XiPv,P{Yi<qr} = r.
To analyze the heterogeneity of the influence of the factors of the formation of actual financial globalization in the world, the Stochastic Frontier Model (SFM) [27] was evaluated:
Yt=Xtp + Vt-Ut ,Ut ~ iidN+(0,a2u), Ut > 0,cov[i7(.,^.] = 0,cov[CA,.,F:] = 0.
Additionally, the logarithm of the income index was taken into account as a regressor. To justify the use of the model, the remnants of the usual regression were tested for the statistical significance of the asymmetry coefficient; its sign was checked. The hypothesis about the ineffective influence of factors H0 =0 was also tested. The evaluation of the distribution of the efficiency indicator (Ef) of the factors under consideration for the countries of the world was interpreted:
w =
-S£i<7« a2
a
f
exp
<
№
Y2"
Ji dx.
~~2
1
■—<
2
—GO
The efficiency distribution was compared with the ranking of the de facto index.
2. Results
2.1. Analysis of dynamic relationships between de jure and de facto sub-indices
Based on the results of testing processes for stationarity (Table 1), with a probability of 0.95, it can be argued that all the series under consideration are realizations of random processes that are stationary in the first differences.
Table 2 shows the results of testing the de facto and de jure sub-indices for cointegration. The table shows that the hypothesis of the absence of cointegration is rejected at the significance level
of 0.05 for most sub-indices with the definition of one cointegration ratio for them. The presence of a long-term relationship is not evident for the global index and sub-indices of cultural and political globalization.
The results of quality control of the evaluated VEC models showed their sufficiently high validity and allowed them to be used for analysis and interpretation. In particular, the hypothesis of the absence of autocorrelation of residues up to and including the 3-rd order was not rejected at the significance level of 0.05 for all models. Table 3 shows the results of testing cointegrated sub-indices for weak exogeneity. Statistically insignificant at the level of 0.05 estimates for the short-term ratio for the de jure sub-indices of financial and information globalization indicate a weak exogeneity of these values relative to the parameters of VEC models.
Tables 4 and 5 present the results of an analysis of the short-term relationship between the de facto and de jure sub-indices. Table 4shows that such a relationship, as well as a long-term one, is practically not manifested for the political and cultural sub-indeices. The economic subindex is characterized by the influence of the de jure sub-index on the de facto. This is especially noticeable for the financial sub-index. For trade globalization, the de facto influence on de jure prevails a little. For the sub-index of social globalization, there is a relatively weak mutual influence of the de facto and de jure sub-indices. However, for the indices of personal and information globalization, the priority influence of de jure on de facto is noticeable.
The data in Table 5 confirm the fact that there is no short-term connection between the de facto and de jure sub-indices for political and cultural globalization. For both economic and social globalization, the defining role of the de jure sub-index is manifested. For financial and information globalization, this indicator is a highly exogenous variable relative to the parameters of the VEC model.
Table 1.
Results of the analysis of processes for stationarity
Y dY
ADF KPSS ADF KPSS
1 gl_df 0.99 0.21 0.00 0.34
2 gl_dj 0.42 0.16 0.03 0.17
3 ec_df 0.58 0.08 0.00 0.17
4 ec_dj 0.99 0.13 0.00 0.20
5 soc_df 0.52 0.22 0.06 0.40
6 soc_dj 0.36 0.21 0.25 0.40
7 pol_df 0.62 0.21 0.00 0.25
8 pol_dj 0.46 0.12 0.01 0.09
9 tr_df 0.32 0.09 0.00 0.16
10 tr_dj 0.19 0.19 0.15 0.29
11 fi_df 0.99 0.06 0.00 0.09
12 fi_dj 0.999 0.149 0.000 0.214
13 in_df 0.53 0.22 0.06 0.44
14 in_dj 0.10 0.20 0.22 0.34
15 per_df 0.42 0.19 0.01 0.17
16 per_dj 0.03 0.21 0.32 0.30
17 cul_df 0.62 0.19 0.00 0.27
18 cul_dj 0.09 0.08 0.00 0.11
Notes:
1. For all series It Is Indicated: Y Is the Initial series, dY Is the first difference.
2. For the ADF test, one-sided MacKinnon P-values are given.
3. For the KPSS test, test statistics are compared with critical values at the significance level of 0.05: 0.146 - for the Initial series (taking Into account the trend) and 0.463 - for the first difference (taking Into account the constant).
2.2. Analysis of the degree and heterogeneity of the influence of institutional factors on the actual globalization
Further, using the example of the financial globalization sub-index, the results of the analysis of the degree of influence of the de jure index on the de facto for the 2018 globalization KOF-
Table 2.
Results of the analysis of processes for cointegration*
Test Trace Max-eigenvalue
Grade 0 1 0 1
gl 0.54 0.76 0.76 0.43
ec 0.04 0.07 0.11 0.07
soc 0.02 0.22 0.02 0.22
pol 0.44 0.98 0.36 0.98
tr 0.06 0.71 0.04 0.71
fin 0.01 0.92 0.01 0.92
in 0.03 0.62 0.02 0.62
per 0.06 0.69 0.03 0.69
cul 0.13 0.44 0.14 0.44
* MacKInnon-Haug-MIchelis P-values with minimum BIC value
Table 3.
Results of the analysis of sub-indices for weak exogeneity*
■ ec soc tr fin inf per
df -2.92 -2.04 -2.98 -4.22 -3.94 -2.37
dj -2.05 -3.90 2.80 -1.57 1.01 -4.83
* The value of T-statistics for estimating the coefficient at the remainder of the cointegration ratio In the ECM (error correction model) for the de facto (df) and de jure (dj) sub-Indices Is given.
index with the involvement of the income index are demonstrated. Taking into account the omissions in the data, 145 observations were used in the models. The data was used in logarithms. There was no multicollinearity problem, since the correlation coefficient between the regres-sors turned out to be statistically insignificant and equal to 0.03. Table 6shows some results of estimating the usual regression model, quan-tile regression with quantiles corresponding to probabilities 0.25, 0.5, 0.75, and the stochastic frontier model.
Table 4.
Estimates of the decomposition of the variance of the forecast error in VAR/VEC models, %*
Endogenous d(gi) ec soc d(pol) tr fin inf per d(cu)
df 14.98 55.81 27.76 2.15 24.18 54.50 44.00 62.01 10.23
dj 4.24 15.12 29.76 2.58 29.87 6.83 15.75 27.85 3.85
* The maximum value for a 10-year period is given. For each of the de facto (df) and de jure (dj) sub-indices, in accordance with the Cholesky Ordering, the proportion of variation due to a change in the alternative variable va and he is shown.
Table 5.
Results of the analysis of processes for causality by Granger*
Endogenous gl ec soc pol tr fin inf per cul
df 0.00 0.01 0.00 0.45 0.00 0.03 0.00 0.01 0.99
dj 0.00 0.45 0.44 0.50 0.00 0.87 0.95 0.06 0.61
* The P-value of X2 -statistics is given to test the hypothesis that each of the de facto and de jure sub-indices is not a Granger reason for an alternative de jure and de facto sub-index.
The use of quantile regression was due to the high heteroscedasticity of the remnants of the usual regression — the P-value of the X2-statis-tics of the Breusch—Pagan test was equal to 0.000. This is also noticeable in Figure 5. It can be seen that there are countries with a low de jure and high de facto sub-index, but with a high de jure, high de facto is determined almost unambiguously.
Table 6 shows that all models describe the direct statistically significant impact of institutional factors, together with the indicator of the state of health, on the actual globalization in the financial sector. According to estimates of conventional regression, the de facto index value changes by almost a third of a percent on average when the de jure index changes by 1 percent. A comparison of the estimates for the dj indicator and their graphic illustration in Figure 7 shows that for countries with a higher level of globalization, this influence is weaker than for countries with a low level.
The application of the stochastic frontier model was facilitated by a statistically significant (at the level of 0.001), rather high modulo negative value of the residual asymmetry coefficient (—0.788). The relatively low BIC value compared to its value for conventional regression also supports the use of this model. The hypothesis of the absence of inefficiency of factors was rejected at the significance level of 0.05. Figures 8and 9 show the results of calculating the effectiveness of the factors of the formation of actual globalization in the financial sphere for 145 countries.
3. Discussion
The results of the calculations showed that there is both a long-term and a short-term dynamic relationship between the processes of globalization in the institutional sphere and its actual manifestation. And although it does not manifest itself at the global level, for the general index of globalization — perhaps due to its com-
Table 6.
Results of evaluation of regression models*
Normal regression Quantile regression Stochastic frontier model
0.25 0.5 0.75
dj 0.29*** 0.51*** 0.27*** 0.16** 0.15**
gni 0.44*** 0.61*** 0.39*** 0.41*** 0.37***
const 1.10*** -0.62 1.03** 1.94*** 2 27***
Prob F/y2 0.00 0.00 0.00 0.00 0.00
R2/Pseudo R2 0.38 0.28 0.24 0.23 -
BIC 35.58 - - - 27.48
* The statistical significance of coefficient estimates at the level of 0.01 Is marked ***, 0.05 - **, 0.1 - '
4.64.44.24.03.83.63.43.23.02.8
¿if ..,.•••,........••......
2.7 3.2 3.7 4.2 . Ln(fin_df) -0.5 ......0.25---0.75
Fig. 7. Initial and model values of the logarithm of the de facto financial globalization index for quantile regression, excluding the income index
posite structure — for most of its components it is clearly traceable. The special role of institutional factors for the process of globalization has manifested itself for the financial sphere and the sphere of information technology. It turned out that their formation in these areas occurs relatively independently, without significant reliance on the results of de facto globalization in them, and they are the determining basis for actual internationalization.
The analysis of dynamic relations between de jure and de facto factors revealed the ambiguity of mutual influence for individual components of the globalization process. If economic
90 -80 -70 -60 -50 -40 -30 -20 -10 -0
[,285, ,511]
, 511, , 738]
, 738, , 964]
Fig. 8. Distribution of efficiency of factors of formation of actual globalization in the financial sphere
1.00 0.80 0.80 0.70 0.60 0.50 0.40 0.30 0.20
. 1 ,
-i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—
3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.6
Fig. 9. The relationship of the indicator
of the effectiveness of the factors of the formation of actual globalization with the sub-index of de facto financial globalization
globalization is characterized by a noticeable influence of institutional factors on de facto globalization, then for social globalization this apparent orientation is replaced by their mutual influence. For the process of political globalization, the connection between de jure and de facto factors has not been found, both in the long term and in the short term.
The analysis of the spatial sample according to the data of the financial globalization KOF-index at the time of 2018 indicated that in countries with a lower manifestation of actual globalization, the role of institutional factors in shaping the level of globalization is higher than in more globalized countries. Perhaps this indicates a tendency to equalize the level of financial globalization in the countries of the world.
According to the results of calculating the effectiveness of the factors of the formation of actual globalization in the financial sphere, Figure 8 shows a high heterogeneity of countries in this indicator: the spread of values occurs from 0.285 to 0.964. Almost 10% (8.28%) of countries have low efficiency up to 50%. These include countries such as Iran, Bangladesh, Ethiopia. Figure 7 shows that these are countries with a low level of actual financial globalization — noticeably lower than the first quan-tile of the logarithm of this indicator (3.937). A third of the countries (30.34%) have average efficiency (50—75%), for example, Turkey, Russia, Brazil. These are mainly countries with an average level of actual financial globalization that does not exceed its median (4.199). More than 60% (61.38%) of countries have high efficiency. Note that among the countries with high efficiency there are representatives from all groups according to the de facto index level. A complete list of countries with the valu-
es of efficiency and the logarithm of the corresponding globalization index is given in the Appendix (Table A2).
Conclusion
The application of the methodology of coin-tegration analysis to the 2020 data of the globalization KOF-index system made it possible to determine the dynamic relationships between the actual and institutional components of globalization. It is shown that the institutional factors of globalization are the determining basis for the actual internationalization of the financial sphere and the sphere of information technology.
Modeling of the relationship between the de facto and de jure sub-indices of globalization revealed a significant heterogeneity of influence for the countries of the world. Using the example of financial globalization, the difference in the degree of influence of the institutional component on actual globalization is shown, in particular, its predominant importance for less globalized countries with a level of financial globalization less than the first quartile.
The application of the stochastic frontier model to the data of financial globalization made it possible to analyze the degree of effectiveness of the impact of institutional measures together with the overall level of well-being on the actual globalization in the financial sphere. It is shown that the level of efficiency varies in the range from 28 to 96 percent. Almost 10% of countries have low efficiency of up to 50%. One third of the countries has average efficiency (50—75%). Thus, the share of countries with high efficiency over 75% is about 60%. ■
References
1. Akhter S.H. (2004) Is globalization what it's cracked up to be? Economic freedom, corruption, and human development. Journal of World Business, vol. 39, no 3, pp. 283-295. DOI: 10.1016/j.jwb.2004.04.007.
2. Weiß J., Sachs A., Weinelt H. (2018) Globalization report 2018: Who benefits mostfromglobalization. Bertelsmann Stiftung. Available at: https://ged-project.de/globalization/globalization-report-2018-who-benefits-most-from-globalization/ (accessed 01 July 2021).
3. Tsai C. (2007) Does globalization affect human well-being? Social Indicators Research, no 81, pp. 103—126. DOI: 10.1007/s11205-006-0017-8.
4. Ghosh A. (2017) How does banking sector globalization affect economic growth? International Review of Economics & Finance, no 48, pp. 83-97. DOI: 10.1016/j.iref.2016.11.011.
5. Manyika J., Lund S., Bughin J., Woetzel J., Stamenov K., Dhingra D. (2016) Digital globalization: The new era of global flows. Report. McKinsey Global Institute. Available at: https://www.mckinsey.com/ business-functions/digital-mckinsey/our-insights/digital-globalization-the-new-era-of-global-flows (accessed 01 July 2021).
6. Bergh A., Nilsson T. (2010) Do liberalization and globalization increase income inequality? European Journal of Political Economy, vol. 26, no 4, pp. 488-505. DOI: 10.1016/j.ejpoleco.2010.03.002.
7. Krylova I.A. (2016) Russia in the context of globalization: New threats. Russian Journal of Philosophical Sciences, no 4, pp. 30-44 (in Russian).
8. Dreher A., Gassebner M., Siemers L. (2012) Globalization, economic freedom, and human rights. Journal of Conflict Resolution, no 56, pp. 516-546. DOI: 10.2139/ssrn.1695446.
9. Yanitzky O.N. (2019) Challenges and risks of globalization. Seven Theses. Sociological Studies, no 1, pp. 29-39 (in Russian). DOI 10.31857/S013216250003745-2.
10. Cherkashina T.Yu. (2011) Indices of globalization: Indicators and the calculation scheme. Sociology: methodology, methods, mathematical modeling, no 33, pp. 136-165 (in Russian).
11. Gygli S., Haelg F., Potrafke N. (2019) The KOF globalisation index - revisited. Review of International Organizations, no 14, pp. 543-574. DOI: 10.1007/s11558-019-09344-2.
12. Rajan R. G., Zingales L. (2003) The great reversals: The politics of financial development in the twentieth century. Journal of Financial Economics, vol. 69, no 1, pp. 5-50.
DOI: 10.1016/S0304-405X(03)00125-9.
13. Baltagi B.H., Demetriades P.O., Law S.H. (2009) Financial development and openness: Evidence from panel data. Journal of Development Economics, vol. 89, no 2, pp. 285-296. DOI: 10.2139/ssrn.1808903.
14. Tovar-García E. (2012) Financial globalization and financial development in transition countries. Economía: Teoría y Práctica, no 36, pp. 155-178. DOI: 10.24275/ETYPUAM/NE/362012/Tovar.
15. Atif S.M., Srivastav M., Sauytbekova M., Arachchige U.K. (2012) Globalization and income inequality: A panel data analysis of 68 countries. EconStor Preprints, no 65664. ZBW - Leibniz Information Centre for Economics. Available at: http://hdl.handle.net/10419/65664 (accessed 01 July 2021).
16. Law S.H., Azman-Saini W.N.W., Tan H.B. (2014) Economic globalization and financial development in East Asia: A panel cointegration and causality analysis. Emerging Markets Finance and Trade, vol. 50, no 1, pp. 210-225. DOI: 10.2753/REE1540-496X500112.
17. Le T.H., Kim J., Lee M. (2016) Institutional quality, trade openness, and financial sector development in Asia: An empirical investigation. Emerging Markets Finance and Trade, vol. 52, no 5, pp. 1047-1059. DOI: 10.1080/1540496X.2015.1103138.
18. Muye I.M., Muye I.Y. (2017) Testing for causality among globalization, institution and financial development: Further evidence from three economic blocs. Borsa Istanbul Review, vol. 17, no 2, pp. 117-132. DOI: 10.1016/j.bir.2016.10.001.
19. Kahneman D., Deaton A. (2014) High income improves evaluation of life but not emotional well-wbeing. Proceedings of National Academy of Sciences, vol. 107, no 38, pp. 16489-16493. DOI: 10.1073/pnas.1011492107.
20. Mills T.C. (2008) The econometric modeling of financial time series. Cambridge. New York. DOI: 10.2307/2329254.
21. Dolado H., Jenkinson T., Sosvilla-Rivero S. (1990) Cointegration and unit roots. Journal of Economic Surveys, no 4, pp. 243-273.
22. Johansen S. (1988) Statistical analysis of cointegration vectors. Journal of Economic Dynamics and Control, vol. 12, no 2—3, pp. 231—254.
23. Doornik J.A., Hansen H. (2008) An omnibus test for univariate and multivariate normality. Oxford Bulletin of Economics and Statistics, no 70, pp. 927-939. DOI: 10.1111/j.1468-0084.2008.00537.x.
24. Toda H.Y., Yamamoto T. (1995) Statistical inferences in vector autoregression with possible integrated processes. Journal of Econometrics, no 66, pp. 225-250.
25. Lutkepohl H. (2007) New introduction to multiple time series analysis. New York: Springer-Verlag. DOI: 10.1017/S0266466606000442.
26. Koenker R., Hallock K.F. (2021) Quantile regression. Journal of Economic Perspectives, vol. 15, no 4, pp. 143-156. DOI: 10.1257/jep.15.4.143.
27. Kumbhakar S., Parmeter C., Zelenyuk V. (2017) Stochastic frontier analysis: Foundations and advances. Working Papers 2017-10, University of Miami, Department of Economics. DOI: 10.1017/CB09781139174411.
Appendix
Table A1.
Structure of the KOF index of globalization
Index Weight Index Weight
1. Economic globalization 33.3
1.1. Trade globalization 50.0
De facto 50.0 De jure 50.0
Trade in goods 38.8 Import barriers 26.8
Trade in services 44.7 Average tariff level 25.6
Variety of trading partners 16.5 Taxes on trade 24.4
Trade agreements 23.2
1.2. Financial globalization 50.0
De facto 50.0 De jure 50.0
Foreign direct investment 26.7 Barriers to investment 33.3
Portfolio investments 16.5 Capital account openness 38.5
International debt 27.6 Investment agreements 28.2
International reserves 2.1
International income payments 27.1
2. Social globalization 33.3
2.1. Personal globalization 33.3
De facto 50.0 De jure 50.0
Phone traffic 20.8 Access to telephone communication 39.9
Money transfers 21.9 Freedom of visits 32.7
International tourism 21.0 Number of airports 27.4
Migration International education 17.2 19.1
2.2. Information globalization 333
De facto 50.0 De jure 50.0
Internet Usage 37.2 Access to television 36.8
Export of high technologies 34.5 Internet access 42.6
International patents 28.3 Freedom of the press 20.6
2.3. Cultural globalization 33.3
De facto 50.0 De jure 50.0
Trade in cultural goods 28.1 Gender equality 24.7
Exchange of services 24.6 Human capital 41.4
International trademarks 9.7 Civil liberties 33.9
Number of McDonalds 21.6
Number of IKEA stores 16.0
3. Political globalization 33.3
De facto
Embassies in the country Participation in UN missions Public organizations
50.0 36.5
25.7
37.8
De jure
Membership in organizations International agreements Variety of partners
50.0 36.2 33.4
30.4
Table A2.
The effectiveness of the formation of financial globalization in the countries of the world and the logarithm of the corresponding de facto globalization sub-index*
No country ef ln_d
1 Iran. Islamic Republic 0.28 3.030
2 Comoros 0.33 3.075
3 Bangladesh 0.33 3.147
4 Ethiopia 0.42 3.271
5 Kenya 0.43 3.479
6 Pakistan 0.44 3.437
7 Haiti 0.44 3.461
8 Chad 0.45 3.284
9 Nepal 0.45 3.363
10 Algeria 0.46 3.532
11 Iraq 0.47 3.573
12 India 0.5 3.624
13 Guatemala 0.51 3.744
14 Myanmar 0.54 3.600
15 Sri Lanka 0.56 3.783
16 Paraguay 0.56 3.866
No country ef ln_d
17 China 0.56 3.835
18 Turkey 0.58 3.896
19 Dominican Republic 0.58 3.930
20 Bolivia 0.59 3.839
21 Cameroon 0.59 3.722
22 Gabon 0.59 3.866
23 Rwanda 0.59 3.754
24 Korea. Rep 0.59 4.020
25 Romania 0.61 4.033
26 Ecuador 0.61 3.896
27 Indonesia 0.62 3.943
28 Tanzania 0.63 3.738
29 Botswana 0.64 4.027
30 Egypt. Arab Republic 0.65 3.971
31 Bhutan 0.65 3.806
32 Morocco 0.65 3.933
No country ef ln_d
33 Nigeria 0.65 3.924
34 Brazil 0.65 3.938
35 Benin 0.65 3.806
36 Burundi 0.66 3.559
37 Philippines 0.66 4.003
38 Mexico 0.67 4.104
39 Argentina 0.68 4.108
40 Peru 0.68 4.089
41 Sudan 0.69 3.999
42 El Salvador 0.69 4.031
43 Mali 0.69 3.807
44 Uganda 0.69 3.926
45 Russian Federation 0.7 4.139
46 Belarus 0.71 4.090
47 Maldives 0.71 4.124
co Poland 0.72 4.199
49 Costa Rica 0.72 4.179
50 Djibouti 0.72 4.061
51 Gambia 0.72 3.976
52 Israel 0.72 4.237
53 Saudi Arabia 0.73 4.206
54 Zimbabwe 0.73 3.948
55 Albania 0.74 4.152
56 Uruguay 0.74 4.216
57 Oman 0.74 4.211
58 Bulgaria 0.74 4.200
59 Guinea 0.75 3.934
60 Cote d'Ivoire 0.75 4.035
61 Bosnia and Herzegovina 0.76 4.102
62 North Macedonia 0.76 4.198
63 Ghana 0.76 4.027
64 Lithuania 0.77 4.282
65 Sierra Leone 0.77 3.871
66 Kyrgyz Republic 0.77 4.095
67 Croatia 0.77 4.255
68 Colombia 0.78 4.189
69 Tunisia 0.78 4.130
70 Congo. Dem. Republic 0.78 3.859
71 Mauritania 0.79 4.027
72 Japan 0.79 4.343
No country ef ln_d
73 United States 0.79 4.371
74 New Zealand 0.79 4.330
75 Chile 0.79 4.310
76 Greece 0.8 4.303
77 Madagascar 0.8 3.937
78 Serbia 0.8 4.261
79 Armenia 0.81 4.256
80 Jordan 0.81 4.250
81 Czech Republic 0.81 4.381
82 Slovenia 0.81 4.333
83 Nicaragua 0.81 4.205
CO Italy 0.82 4.360
85 Niger 0.82 3.971
86 Vietnam 0.82 4.205
87 Malaysia 0.82 4.336
88 United Arab Emirates 0.83 4.432
89 Solomon Islands 0.83 3.984
90 South Africa 0.83 4.220
91 Moldova 0.84 4.224
92 Estonia 0.84 4.416
93 Panama 0.84 4.400
94 Latvia 0.84 4.397
95 Honduras 0.84 4.164
96 Germany 0.85 4.446
97 Spain 0.85 4.429
98 Hungary 0.85 4.411
99 Canada 0.86 4.450
100 Equatorial Guinea 0.86 4.253
101 Namibia 0.86 4.248
102 Liberia 0.86 4.089
103 Congo 0.86 4.15
104 Eswatini 0.87 4.232
105 Austria 0.87 4.479
106 Portugal 0.87 4.456
107 Vanuatu 0.87 4.199
108 France 0.87 4.483
109 Finland 0.87 4.491
110 Denmark 0.87 4.499
111 Cabo Verde 0.88 4.315
112 Sweden 0.88 4.508
No country ef ln_d
113 Norway 0.88 4.506
114 Bahrain 0.89 4.507
115 Belgium 0.89 4.533
116 Jamaica 0.89 4.369
117 United Kingdom 0.89 4.533
118 Kazakhstan 0.89 4.369
119 Senegal 0.89 4.493
120 Georgia 0.89 3.274
121 Switzerland 0.9 4.493
122 Burkina Faso 0.9 3.274
123 Brunei Darussalam 0.9 4.493
124 Belize 0.9 3.274
125 Luxembourg 0.9 4.493
126 Netherlands 0.9 3.274
127 Kiribati 0.9 4.493
128 Kuwait 0.91 3.274
129 Ireland 0.91 4.493
No country ef ln_d
130 Montenegro 0.91 3.274
131 Mongolia 0.91 4.493
132 Cambodia 0.91 3.274
133 Lebanon 0.91 4.493
134 Guinea-Bissau 0.91 3.274
135 Angola 0.92 4.493
136 Cyprus 0.92 3.274
137 Malta 0.92 4.493
138 Bahamas 0.92 3.274
139 Ukraine 0.93 4.493
140 Lesotho 0.94 3.274
141 Mauritius 0.94 4.493
142 Marshall Islands 0.95 3.274
143 Togo 0.96 4.493
144 Mozambique 0.96 3.274
145 Timor-Leste 0.96 4.493
* The quartiles of the de facto index are marked in bold
About the authors
Elena D. Kopnova
Cand. Sci. (Tech.);
Associate Professor, Department of Statistics and Data Analysis, National Research University Higher School of Economics, 11 Pokrovsky Bulvar, Moscow 109028, Russia;
E-mail: ekopnova@hse.ru
ORCID: 0000-0002-8429-141X
Lilia A. Rodionova
Cand. Sci. (Econ.);
Associate Professor, Department of Statistics and Data Analysis, National Research University Higher School of Economics, 11 Pokrovsky Bulvar, Moscow 109028, Russia;
E-mail: lrodionova@hse.ru
ORCID: 0000-0002-0310-6359
