Excerpt

## Table of Contents

Introduction

Literature Review

Objectives

Methodology

Results and Findings

Limitations

Conclusion

References

Appendices

## Abstract

Over a quarter of a century has elapsed since the disintegration of former Union of Soviet Socialist Republics (USSR) and former socialist nations from the Eastern Bloc. The erstwhile countries from East Europe, Central and East Asia have gradually adopted economic globalization for transition towards market economies. There has been a plethora of research works in the academic world regarding the *apparent* success or failure of globalization and transition economies as a whole containing strong arguments in favor and against of globalization. This research paper is aimed at identifying the major prospective markets among the economies in transition across the world. For this purpose, an attempt has been made to formulate an index of transition from socialist to market economy by means of a combination of conventional econometrical and statistical methods based on time - series study for a period of 21 years (1995 – 2015) on 19 major countries in the East European, Central and East Asia (including Russian Federation) which adopted the path of economic globalization by shedding the rigid structures of socialistic or command economy since 1990.

The *Transitional Impact Index* (TII) would help to classify the countries in transition broadly into four categories, viz., countries with *Low*, *Medium*, *High* and *Very High* levels of ‘transitional effect’ en - route to transformation from socialist to market economy - which will enable us to detect the potentialities of the economies in transition to emerge as the future market leaders in an era of continuing global recession.

## Keywords

Trade to GDP Ratio (TGDPR); Gross Capital Formation Ratio (GCFR); Net Foreign Direct Investment to GDP Ratio (FDIGDPR); Product Concentration Index (PCI); Product Diversification Index (PDI); Transitional Impact Index (TII).

### JEL Classifications

C10; C32; C43; C53; P20; P27

## Introduction

Since globalization is an inseparable part of transition process from planned economy to market economy, hence an introductory discussion about the globalization process would be relevant in this context. The issue of globalization is a much debated one. There is an abundance of critics and supporters of globalization among the authors, economists, thinkers and intellectuals. Thus in this context, we would like to put forth some of the views and findings of a group of eminent researchers across the world as a mixed bag of reactions.

Nobel laureate economist Joseph Stiglitz (2008) has openly criticized globalization for the disastrous economic performances of regions like South America and sub – Saharan Africa. Brown (2008) has affirmed that globalization in its current form has clearly failed to serve the poor as the absolute poverty has increased across the world in the post – globalization scenario. Collier (2008) has admitted that globalization has helped to increase the income levels of developing countries which are homes for five billion people, but he also added that the world’s poorest one billion people who live in sub – Saharan Africa and a few land – locked countries elsewhere, have suffered the worst from globalization.

Bhagwati (2008) has shown his strong support for globalization to preserve its positive effects for tackling the three major thrust areas, viz., gender discrimination, poverty reduction and child labor – which are often cited as the major drawback areas for globalization.

De Soto (2008) has suggested that the countries fail to benefit from the globalization process because they lack the proper infrastructural framework for smooth transitioning into market economies.

Keeping the merits and demerits of globalization in mind, an attempt has been made to formulate an index (with the conventional statistical merits and demerits of an index in general) which will be able to provide a simplified and holistic measurement of the impact / effect of transition for the different countries in transition regardless of their demographical positioning in the world. In order to approach that goal, the most important issue is related to choosing of the appropriate (independent) variables which will encompass and highlight the major features related to transition to market economies from planned economies. These features are described below as:

- **Increased share of trade in GDP** – An increased share of trade in GDP (as a percentage) would denote that the country in question is proceeding strongly in the path of transition to market economy,

- **Increased share of Net Foreign Direct Investment (FDI) in GDP** – An increased share of *positive* Net Foreign Direct Investment (FDI) inflow in GDP (BOP surplus at current US Dollar, as a percentage) would also denote that the country in question is proceeding strongly in the path of transition to market economy,

- **Increased share of Gross Capital Formation in GDP** – One of the major issues related to transition to market economy from a planned economy is the *formation* of private physical capital (K) in the said economy which helps to appreciate the level of net investment demand (I) in the long run. Following Samuelson’s (1939) *accelerator model* of investment (by ignoring the rate of depreciation of capital), we may express:

illustration not visible in this excerpt

- **Product Concentration Index:** Product Concentration Index, ‘also named Herfindahl - Hirschmann Index (Product HHI), is a measure of the degree of product concentration. The following normalized HHI is used in order to obtain values between 0 and 1. It can be calculated as:

illustration not visible in this excerpt

where H_{j} = Country or Country Group Index j,

x_{ij} = Value of export for country j and product i, and

n = Number of products (SITC Revision 3 at 3-digit group level).

An index value closer to 1 indicates a country's exports or imports are highly concentrated on a few products. On the contrary, values closer to 0 reflect exports or imports are more homogeneously distributed among a series of products (UNCTAD)”, and

- **Product Diversification Index:** “The diversification index is computed by measuring the absolute deviation of the trade structure of a country from world structure:

illustration not visible in this excerpt

where hij = Share of product i in total exports or imports of country or country group j,

h_{i} = Share of product i in total world exports or imports.

The diversification index takes values between 0 and 1. A value closer to 1 indicates greater divergence from the world pattern.

This index is a modified Finger-Kreinin measure of similarity in trade. For more information, please consult the article of Finger, J. M. and M. E. Kreinin (1979), “A measure of ‘export similarity’ and its possible uses” in The Economic Journal, 89: 905-12 (UNCTAD)’.

The Product Concentration Index and Product Diversification Index have been included in formulation of Transitional Impact Index (TII) primarily to gauge the effects of technological innovation on the traded exportable goods of sample countries in transition during the post – globalization era. For example, a country having a very low value (close to ‘0’) for Product Concentration Index combined with a very high value (close to ‘1’) for Product Diversification Index would be an ideal example of *successful* transition from planned to market economy.

Thus a sample-size of 19 countries has been chosen consisting of Albania, Armenia, Azerbaijan, Bulgaria, etc. which are spread across Eastern Europe, Central and Eastern Asia in order to assess the impacts of transition by formulation of Transitional Impact Index (TII) by using a time – series analysis for a period of 21 years (1995 – 2015) based on the data from World Bank for the variables Gross Domestic Product (GDP) at current US Dollar, Trade to GDP Ratio (TGDPR), Net Foreign Direct Investment Inflow to GDP Ratio (FDIGDPR) [BOP *surplus* at current US Dollar], and Gross Capital Formation Ratio (GCFR) while the data for Product Concentration Index (PCI) and Product Diversification Index were culled from UNCTAD (United Nations Conference for Trade and Development) respectively.

When the Transitional Impact Index (TII) is estimated for each individual country under transition, those countries would be classified into **four** categories on a scale from 0 to 1 where ‘0’ represents *no* transitional impact and ‘1’ represents *complete* transitional impact.

The classification of countries based on TII would be as follows:

illustration not visible in this excerpt

## Literature Review

The transition phase from the planned economy to market oriented economy had been always marked with various short and long term economic problems, especially moderate to severe inflationary pressure coupled with contraction of GDP.

Dollar *et al*. (2008) have presented an empirical study that showed a positive correlation between a country’s investment climate and trade liberalization. A favorable investment climate is supported by the number of its customs clearances, infrastructural and logistics facilities.

Schultz (2008) used a cross – country analysis to show that trade liberalization is associated with strong gains in women’s economic status – which eventually helps to reduce the gender gaps in terms of life expectancy, years of schooling, and earnings.

Fan Gang’s (2008) study attempts to decode the success story of China – which is often treated as the “brand ambassador” of successful globalization, where he emphasizes on the fact that without internal institutional reforms, China’s transition from a planned to a market economy would have failed miserably. He also points out that the governmental policies which are conducive to transformation must be local and country – specific.

Fischer and Sahay (2000) claimed that a mix of price – stabilization policy along with structural reforms, e.g., privatization, helps to channelize the transition economies back to the tracks of growth. Thus they concluded that higher the rate of reform, quicker is the rate of recovery and growth.

De Melo, Denzier, and Gelb (1996) formulated the Liberalization Index (LI) primarily in order to assess the impact of transition on the liberalized economies with respect to **three** crucial factors:

a) Liberalization of *private* sector of the home country,

b) Liberalization of *external* sector of the home country, and

c) Liberalization of *internal* sector of the home country.

They formulated LI consisting of Private Sector Conditions Index (LIP), External Liberalization Index (LIE), and Internal Liberalization Index (LII) by using the formula of weighted average with weights of 40 percent, 30 percent, and 30 percent respectively.

In another study by De Melo, Denzier, Gelb, and Tenev (1997), they argued that the effect of transition depends upon initial conditions, clusters and other factors.

Hoskisson *et al.* (2000) have argued that the emerging economies in post – transition phase are progressing in divergent manner; hence the real challenge lies ahead for the policymakers to design and implement successful market – based strategies.

Bao *et al.* (2014) conducted a study on nearly 4000 employees in 180 Chinese manufacturing firms regarding market orientation. The findings of the survey showed that the *consensus* between the top – level management and non – supervisory level employees positively affects the firms’ performance levels.

Lin *et al.* (2014) showed in their study that for 1 percent increase in trade openness (Trade to GDP ratio) for Small Developing Countries (SDC) under transition leads to an increase in Government expenditure (G) to GDP (Y) ratio within the range of 0.1 – 0.2 percentage points on average.

Misch *et al.* (2014) in their study have assessed the merits of using business perceptions of growth constraints as a guide to growth – enhancing fiscal reforms. The study has revealed the “over – estimation” of growth – enhancing effects of lower income tax regime relative to public services and public capital.

Klimek (2012) in his study has shown that the net inflow of FDI has increased for the emerging and transitional economies in the post – 2008 depression era.

Caporale *et al.* (2009) by means of a dynamic panel study over the period 1994 – 2007 have showed that for the 10 new European Union (EU) member countries, the stock and credit markets are underdeveloped and their contribution to economic growth is limited.

Poghosyan and Poghosyan (2010) in their study have shown that domestic and Foreign Greenfield banks enjoy more financial power compared to foreign – acquired banks and Foreign Greenfield banks are superior in terms of efficiency compared to domestic and foreign – acquired banks.

Perugini and Selezneva (2015) used quintile regression method for studying gender pay - inequality in 10 Central and Eastern European EU countries before 2007 and during the economic recession of 2009. Their study revealed the fact that labor market deregulation helps to increase inequality in the distribution of payments at the middle and top levels of management and reduces inequality at the bottom level.

Babecky and Havranek (2014) performed their research on the effectiveness of structural reforms on 26 countries in transition across the world. The findings showed that although structural reforms have a high short – term cost, however, the reforms help to attain strong positive growths in the long run.

Thus by summing up the findings and analyses of all the major works mentioned above, it may be interpreted that globalization, indeed, has very strong points in its favor – although it requires a very strong country–specific, internal framework for it successful implementation.

## Objectives

The major objectives of formulation of Transitional Impact Index (TII) are discussed below:

a) To construct an indicator which would be able to provide a simplified and holistic picture about the effects of transition from a socialist to market economy,

b) To readily evaluate the impact of transition for a sample country under *four* broad categories, that is, low, medium, high and very high respectively,

c) To analyze the economic performance of the countries in transition by means of ranking those countries in terms of the TII values, and

d) To identify the countries which have gained most from economic transition and those which have lagged in doing so, or simply, the ‘poor performers’.

In order to fulfill the objectives mentioned above, the *time - series* datasets for *five* major variables for 19 countries^{[1]} are considered and the necessary model for analysis is supplemented by rigorous mathematical approach.

## Methodology

**First step:**

In an attempt to formulate Transitional Impact Index (TII), the traditional statistical method is used to express it as the *Geometric Mean (GM)* of *five* individual independent variables, that is, Trade to GDP Ratio (TGDPR), Net Foreign Direct Investment to GDP Ratio (FDIGDPR), Gross Capital Formation Ratio (GCFR), Product Concentration Index (PCI), and Product Diversification Index (PDI).

Formally, we may express:

TII = (TGDPR * FDIGDPR * GCFR * PCI * PDI)1 ∕ 5 (1)

**Second step:**

In the second step, by using equation: (1), the *individual country – wise* values of TII are calculated;

**Third step:**

Since by assumption, TII is *directly* correlated to the individual independent variables, hence we may express it in a Cobb – Douglas *functional* form as below:

TII = TGDPRβ1 FDIGDPR β2 GCFR β3 PCI β4 PDI β5 (2)

where β1 = Coefficient of TGDPR, β2 = Coefficient of FDIGDPR,

β3 = Coefficient of GCFR, β4 = Coefficient of PCI, and

β5 = Coefficient of PDI respectively;

**Fourth step:**

The next step requires formulation of a *log – linear* regression equation for equation: (2) as:

ln TII = α + β1 ln TGDPR + β2 ln FDIGDPR + β3 ln GCFR + β4 ln PCI + β5 ln PDI + ε

(3)

where α = Slope / Intercept of the regression line,

ε = Residual term (βi ≠ 0, i = 1, 2, ., 5).

If the assumptions of Classical Linear Regression Model (CLRM) are fulfilled, the parameters of equation: (3) can be estimated by Ordinary Least Squares (OLS) method as below (Gujarati *et al*. 2009):

TII* = α + β1 TGDPR* + β2 FDIGDPR* + β3 GCFR* + β4 PCI* + β5 PDI* + ε . (4)

where TII* = ln TII, TGDPR* = ln TGDPR, FDIGDPR* = ln FDIGDPR,

GCFR* = ln GCFR, PCI* = ln PCI, and PDI* = ln PDI.

The individual country – wise econometrical and statistical summarized results derived by using the **software GRETL v.1.9.4** for equation: (4) are shown on Table: 1 – Table: 19, and Table: 1(a) – Table: 19(a) respectively.

From the data of Table: 1 to Table: 19, it may be clearly interpreted that *all* the values of the coefficients of the regressors along with the intercept (wherever applicable) are statistically significant. From the derived *p – values*, it may be inferred that the coefficients of the slope / intercept of the regression line, TGDPR, FDIGDPR, GCFR, PCI and PDI are *statistically significant* at 5 percent level or better. No regression equation is formulated with coefficients of the regressors (with derived *p – values*) at less than 5 percent level of significance.

An additional independent variable SAFDIGDPR (Statistically Adjusted net FDI to GDP Ratio) is included in the analysis for the countries Kyrgyzstan and Slovenia in order to tackle the problem of *negative* net FDI inflow or BOP *deficit* at current US Dollar.

**Note:**

In order to comply with the feature of Geometric Mean (GM), *only* positive values of the independent variables are considered for calculation – especially, for the variable Net Foreign Direct Investment to GDP Ratio (FDIGDPR). In case of net FDI inflow, only the BOP *surplus* values at current US Dollar (in millions) for the sample nations are taken into account. Some countries’ datasets are also duly *modified* to take care of the missing observations. For example, datasets for Bosnia & Herzegovina, Georgia, Kazakhstan, Lithuania, Turkmenistan and Uzbekistan are *truncated* for the time – periods (1998 – 2014), (1997 – 2015), (1995 – 2014), (1995 – 2014), and (1996 – 2012) respectively since the omitted years in the time – series of 21 years (1995 – 2015) did not contain data for one or more independent variables like GDP, Trade to GDP Ratio (TGDPR), and Gross Capital Formation Ratio (GCFR) [Source: Data bank of World Bank].

In this manner, TIIs’ are computed and log – linear regression equations are formed for 17 countries. However, for the remaining 2 countries, that is, Kyrgyzstan and Slovenia, the two *negative* entries under FDI columns (for BOP *deficits* at current US$) are converted into *exactly equal positive* entries by means of mathematical *adjustments* without changing the sum of net FDI values for the given time – series by an experimental method.

The adjustment technique is fairly simple to execute. First, exactly the double *positive* value of the negative observation is added to it in order to yield an *absolute* positive value of the negative observation.

In the second step, the double positive value of the negative observation is subtracted from the value of FDI *preceding* or *succeeding* the negative observation so that the sum of the observations remains unchanged.

For example, let there be 3 entries under the net FDI column as $300, (-) $100, and $400 respectively and hence the sum of observations is $600.

Now when (+) $200 is added to the negative observation, it becomes (+) $100. So in terms of adjustment, $200 may be deducted from either the preceding or succeeding observation, that is, $300 or $400 so that the *adjusted* new set of observations becomes $300, $100, and $200 respectively (in case of adjusted *third* observation) and the sum of entries remain unchanged at $600. This technique works fine for a *single* negative entry, however, for *multiple* negative entries; the necessary adjustments could be done by choosing multiple positive observations randomly under the net FDI column with values greater than the *adjusting factor* ($200 in the cited example).

In order to do the necessary adjustments and related calculations, two additional columns are created for the variables **SAFDI** (Statistically Adjusted Net Foreign Direct Investment) and **SAFDIGDPR** (Statistically Adjusted Net Foreign Direct Investment to GDP Ratio) for Kyrgyzstan and Slovenia (Refer to TII).

## Results and Findings

1. For **Albania**, all the coefficients of five independent variables, that is, TGDPR, FDIGDPR, GCFR, PCI and PDI are statistically significant at 1 percent level or better. Even the coefficient term for the slope / intercept is also statistically significant at 1 percent level or better (Refer to Table: 1).

Hence the log – linear form of regression equation for **Albania** may be expressed as:

TII* = -0.155667 + 0.0960945 TGDPR* + 1.06346 FDIGDPR* + 0.198219 GCFR* + 0.215789 PCI* + 0.276472 PDI* (1)

Equation: (1) may be *approximated* as:

TII* = -0.155 + 0.1 TGDPR* + 1.06 FDIGDPR* + 0.2 GCFR* + 0.2 PCI* + 0.28 PDI* 1(a)

The Cobb – Douglas *functional* form of equation: 1(a) may be expressed as:

TII = -0.155 TGDPR0.1 FDIGDPR1.06 GCFR0.2 PCI0.2 PDI0.28 (2)

The statistical summary for the TII dataset of Albania may be expressed as below [Refer to Table: 1(a)]:

- The TII dataset for Armenia reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Albania are 0.278706 and 0.296161 respectively with a moderate Coefficient of Variation (C.V.) at approximately 18 percent;

- From equation: (2) it may be inferred that FDIGDPR, GCFR, PCI, and PDI combined play more significant role in influencing the value of TII for the said country; and

- The straight line trend equation for Albania (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 1]:

TII = 0.207 + 1.31 FDIGDPR . (3)

2. For **Armenia**, all the coefficients of five independent variables, that is, TGDPR, FDIGDPR, GCFR, PCI and PDI are statistically significant at 5 percent level or better. The coefficient term for the slope / intercept is also statistically significant at 10 percent level or better and hence it is omitted (Refer to Table: 2).

Hence the log – linear form of regression equation for **Armenia** may be expressed as:

TII* = 0.114259 TGDPR* + 1.0378 FDIGDPR* + 0.256282 GCFR* + 0.236956 PCI* + 0.145964 PDI* (4)

Equation: (4) may be *approximated* as:

TII* = 0.11TGDPR* + 1.04 FDIGDPR* + 0.26 GCFR* + 0.24 PCI* + 0.15 PDI* 4(a)

The Cobb – Douglas *functional* form of equation: 1(a) may be expressed as:

TII = TGDPR0.11 FDIGDPR1.04 GCFR0.26 PCI0.24 PDI0.15 (5)

The statistical summary for the TII dataset of Armenia may be expressed as below [Refer to Table: 2(a)]:

- The TII dataset for Armenia reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Armenia are 0.274112 and 0.291562 respectively with a moderate Coefficient of Variation (C.V.) at approximately 17 percent;

- From equation: (5) it may be inferred that FDIGDPR, GCFR, PCI, and PDI combined play more significant role in influencing the value of TII for the said country; and

- The straight line trend equation for Armenia (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 2]:

TII = 0.196 + 1.47 FDIGDPR . (6)

3. For **Azerbaijan**, the coefficients of only two independent variables, that is, TGDPR, and FDIGDPR are statistically significant at 5 percent level or better (Refer to Table: 3).

Hence the log – linear form of regression equation for **Azerbaijan** may be expressed as:

TII* = 0.126913 TGDPR* + 0.645186 FDIGDPR* (7)

Equation: (7) may be *approximated* as:

TII* = 0.127 TGDPR* + 0.645 FDIGDPR* . 7(a)

The Cobb – Douglas *functional* form of equation: 7(a) may be expressed as:

TII = TGDPR0.127 FDIGDPR0.645. (8)

The statistical summary for the TII dataset of Azerbaijan may be expressed as below [Refer to Table: 3(a)]:

- The TII dataset for Azerbaijan reflects a *positively skewed leptokurtic distribution*;

- The mean and median of TII for Azerbaijan are 0.444632 and 0.402563 respectively with a high Coefficient of Variation (C.V.) at approximately 26 percent;

- From equation: (8) it may be inferred that TGDPR and FDIGDPR combined play significant role in influencing the value of TII for the said country; and

- The straight line trend equation for Azerbaijan (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 3]:

TII = 0.315 + 0.727 FDIGDPR . (9)

4. For **Belarus**, the coefficients of only three independent variables, that is, TGDPR, FDIGDPR and GCFR are statistically significant at 1 percent level or better (Refer to Table: 4).

Hence the log – linear form of regression equation for **Belarus** may be expressed as:

TII* = 0.0550711 TGDPR* + 2.03273 FDIGDPR* + 0.439749 GCFR* (10)

Equation: (10) may be *approximated* as:

TII* = 0.055 TGDPR* + 2.032 FDIGDPR* + 0.44 GCFR*. 10(a)

The Cobb – Douglas *functional* form of equation: 11(a) may be expressed as:

TII = TGDPR0.055 FDIGDPR2.032 GCFR0.44 . (11)

The statistical summary for the TII dataset of Belarus may be expressed as below [Refer to Table: 4(a)]:

- The TII dataset for Belarus reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Belarus are 0.241344 and 0.226357 respectively with a high Coefficient of Variation (C.V.) at approximately 24 percent;

- From equation: (11) it may be inferred that FDIGDPR and GCFR combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Belarus (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 4]:

TII = 0.168 + 3.28 FDIGDPR . (12)

5. For **Bosnia & Herzegovina**, the coefficients of only three independent variables, that is, TGDPR, FDIGDPR and PCI are statistically significant at 5 percent level or better. The coefficient term for GCFR is also statistically significant at 10 percent level or better and hence it is omitted (Refer to Table: 5).

Hence the log – linear form of regression equation for **Bosnia & Herzegovina** may be expressed as:

TII* = 0.0984349 TGDPR* + 1.06661 FDIGDPR* + 0.214219 GCFR* (13)

Equation: (13) may be *approximated* as:

TII* = 0.1TGDPR* + 1.07 FDIGDPR* + 0.21 PCI* 13(a)

The Cobb – Douglas *functional* form of equation: 13(a) may be expressed as:

TII = TGDPR0.1 FDIGDPR1.07 PCI 0.21 . (14)

The statistical summary for the TII dataset of Bosnia & Herzegovina may be expressed as below [Refer to Table: 5(a)]:

- The TII dataset for Bosnia & Herzegovina reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Bosnia & Herzegovina are 0.230309 and 0.241946 respectively with a moderate Coefficient of Variation (C.V.) at approximately 18 percent;

- From equation: (14) it may be inferred that FDIGDPR and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Bosnia & Herzegovina (1998 – 2014) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 5]:

TII = 0.184 + 1.14 FDIGDPR . (15)

6. For **Bulgaria**, the coefficients of only three independent variables, that is, FDIGDPR, GCFR, and PCI are statistically significant at 1 percent level or better (Refer to Table: 6).

Hence the log – linear form of regression equation for **Bulgaria** may be expressed as:

TII* = 0.367199 FDIGDPR* + 0.506029 GCFR* + 0.569658 PCI* (16)

Equation: (16) may be *approximated* as:

TII* = 0.37 FDIGDPR* + 0.51 GCFR* + 0.57 PCI*. 16(a)

The Cobb – Douglas *functional* form of equation: 16(a) may be expressed as:

TII = FDIGDPR0.37 GCFR0.51 PCI 0.57. (17)

The statistical summary for the TII dataset of Bulgaria may be expressed as below [Refer to Table: 6(a)]:

- The TII dataset for Bulgaria reflects a *positively skewed leptokurtic distribution*;

- The mean and median of TII for Bulgaria are 0.234969 and 0.226160 respectively with a high Coefficient of Variation (C.V.) at approximately 31 percent;

- From equation: (17) it may be inferred that GCFR and PCI combined play major role in influencing the value of TII whereas FDIGDPR plays a minor role for the said country; and

- The straight line trend equation for Bulgaria (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 6]:

TII = 0.167 + 0.833 FDIGDPR . (18)

7. For **Croatia**, the coefficients of only three independent variables, that is, FDIGDPR, GCFR, and PCI are statistically significant at 1 percent level or better. The coefficient term of PDI is also statistically significant at 10 percent level or better and hence it is omitted (Refer to Table: 7).

Hence the log – linear form of regression equation for **Croatia** may be expressed as:

TII* = 1.09736 FDIGDPR* + 0.237765 GCFR* + 0.568565 PCI* (19)

Equation: (19) may be *approximated* as:

TII* = 1.09 FDIGDPR* + 0.24 GCFR* + 0.57 PCI*. 19(a)

The Cobb – Douglas *functional* form of equation: 20(a) may be expressed as:

TII = FDIGDPR1.09 GCFR0.24 PCI 0.57 . (20)

The statistical summary for the TII dataset of Croatia may be expressed as below [Refer to Table: 7(a)]:

- The TII dataset for Croatia reflects a *negatively skewed leptokurtic distribution*;

- The mean and median of TII for Croatia are 0.202712 and 0.213258 respectively with a moderate Coefficient of Variation (C.V.) at approximately 19 percent;

- From equation: (20) it may be inferred that FDIGDPR and PCI combined play major role in influencing the value of TII whereas GCFR plays a minor role for the said country; and

- The straight line trend equation for Croatia (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 7]:

TII = 0.141 + 1.57 FDIGDPR . (21)

8. For **Czech Republic**, the coefficients of four independent variables, that is, TGDPR, FDIGDPR, GCFR, and PCI are statistically significant at 5 percent level or better. (Refer to Table: 8).

Hence the log – linear form of regression equation for **Czech Republic** may be expressed as:

TII* = 0.0364828 TGDPR* + 0.94631 FDIGDPR* + 0.192829 GCFR* + 0.342488 PCI* (22)

Equation: (22) may be *approximated* as:

TII* = 0.04 TGDPR* + 0.95 FDIGDPR* + 0.19 GCFR* + 0.34 PCI* 22(a)

The Cobb – Douglas *functional* form of equation: 23(a) may be expressed as:

TII = TGDPR0.04 FDIGDPR0.95 GCFR0.19 PCI 0.34 . (23)

The statistical summary for the TII dataset of Czech Republic may be expressed as below [Refer to Table: 8(a)]:

- The TII dataset for Czech Republic reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Czech Republic are 0.218929 and 0.223075 respectively with a relatively low Coefficient of Variation (C.V.) at approximately 13.5 percent;

- From equation: (23) it may be inferred that FDIGDPR, GCFR and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Czech Republic (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 8]:

TII = 0.179 + 0.833 FDIGDPR . (24)

9. For **Georgia**, the coefficients of four independent variables, that is, TGDPR, FDIGDPR, GCFR, and PCI are statistically significant at 1 percent level or better. The coefficient term of PDI is also statistically significant at 10 percent level or better and hence it is omitted (Refer to Table: 9).

Hence the log – linear form of regression equation for **Georgia** may be expressed as:

TII* = 0.0752038 TGDPR* + 0.703173 FDIGDPR* + 0.213648 GCFR* + 0.306701 PCI* (25)

Equation: (25) may be *approximated* as:

TII* = 0.075 TGDPR* + 0.7 FDIGDPR* + 0.21 GCFR* + 0.31 PCI* 25(a)

The Cobb – Douglas *functional* form of equation: 25(a) may be expressed as:

TII = TGDPR0.075 FDIGDPR0.7 GCFR0.21 PCI0.31. (26)

The statistical summary for the TII dataset of Georgia may be expressed as below [Refer to Table: 9(a)]:

- The TII dataset for Georgia reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Georgia are 0.295458 and 0.303104 respectively with a relatively low Coefficient of Variation (C.V.) at approximately 15 percent;

- From equation: (26) it may be inferred that FDIGDPR, GCFR and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Georgia (1997 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 9]:

TII = 0.221 + 0.93 FDIGDPR . (27)

10. For **Kazakhstan**, the coefficients of all the five independent variables, that is, TGDPR, FDIGDPR, GCFR, PCI, and PDI are statistically significant at 1 percent level or better. (Refer to Table: 10).

Hence the log – linear form of regression equation for **Kazakhstan** may be expressed as:

TII* = 0.0873197 TGDPR* + 0.978229 FDIGDPR* + 0.308857 GCFR* + 0.138684 PCI* + 0.114267 PDI* (28)

Equation: (28) may be *approximated* as:

TII* = 0.09 TGDPR* + 0.98 FDIGDPR* + 0.31 GCFR* + 0.14 PCI* + 0.11 PDI* 28(a)

The Cobb – Douglas *functional* form of equation: 28(a) may be expressed as:

TII = TGDPR0.09 FDIGDPR0.98 GCFR0.31 PCI0.14 PDI0.11. (29)

The statistical summary for the TII dataset of Kazakhstan may be expressed as below [Refer to Table: 10(a)]:

- The TII dataset for Kazakhstan reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Kazakhstan are 0.354030 and 0.348098 respectively with a moderate Coefficient of Variation (C.V.) at approximately 17 percent;

- From equation: (29) it may be inferred that FDIGDPR, GCFR and PCI combined play major role in influencing the value of TII whereas TGDPR and PDI play a minor role for the said country; and

- The straight line trend equation for Kazakhstan (1995 – 2014) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 10]:

TII = 0.233 + 1.53 FDIGDPR . (30)

11. For **Kyrgyzstan**, the coefficients of only two independent variables, that is, TGDPR, and SAFDIGDPR are statistically significant at 5 percent level or better. (Refer to Table: 11).

Hence the log – linear form of regression equation for **Kazakhstan** may be expressed as:

TII* = 0.0834905 TGDPR* + 1.57897 SAFDIGDPR* (31)

Equation: (31) may be *approximated* as:

TII* = 0.08 TGDPR* + 1.58 SAFDIGDPR* (31a)

The Cobb – Douglas *functional* form of equation: 31(a) may be expressed as:

TII = TGDPR0.08 SAFDIGDPR1.58 (32)

The statistical summary for the TII dataset of Kyrgyzstan may be expressed as below [Refer to Table: 11(a)]:

- The TII dataset for Kyrgyzstan reflects a *negatively skewed leptokurtic distribution*;

- The mean and median of TII for Kyrgyzstan are 0.270686 and 0.298058 respectively with a high Coefficient of Variation (C.V.) at approximately 26 percent;

- From equation: (32) it may be inferred that SAFDIGDPR (Statistically Adjusted net FDI to GDP Ratio) plays the major role in influencing the value of TII ; and

- The straight line trend equation for Kazakhstan (1995 – 2014) between TII (dependent variable) and SAFDIGDPR (independent variable) may be expressed as [Refer to Figure: 11]:

TII = 0.178 + 1.92 SAFDIGDPR (33)

12. For **Lithuania**, the coefficients of four out of five independent variables, that is, TGDPR, FDIGDPR, GCFR, and PCI are statistically significant at 5 percent level or better. The coefficients of the intercept and PDI are statistically significant at 10 percent level or better and hence they are omitted. (Refer to Table: 12).

Hence the log – linear form of regression equation for **Lithuania** may be expressed as:

TII* = 0.0802996 TGDPR* + 1.36827 FDIGDPR* + 0.419284 GCFR* + 0.271051 PCI* . (34)

Equation: (34) may be *approximated* as:

TII* = 0.08 TGDPR* + 1.37 FDIGDPR* + 0.42 GCFR* + 0.27 PCI* . 34(a)

The Cobb – Douglas *functional* form of equation: 34(a) may be expressed as:

TII = TGDPR0.08 FDIGDPR1.37 GCFR0.42 PCI0.27. (35)

The statistical summary for the TII dataset of Lithuania may be expressed as below [Refer to Table: 12(a)]:

- The TII dataset for Lithuania reflects a *negatively skewed leptokurtic distribution*;

- The mean and median of TII for Lithuania are 0.218599 and 0.221026 respectively with a high Coefficient of Variation (C.V.) at approximately 21.5 percent;

- From equation: (35) it may be inferred that FDIGDPR, GCFR and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Lithuania (1995 – 2014) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 12]:

TII = 0.161 + 1.8 FDIGDPR . (36)

13. For **Republic of Moldova**, the coefficients of three independent variables, that is, TGDPR, FDIGDPR, and PDI are statistically significant at 5 percent level or better. The coefficient of PCI is statistically significant at 10 percent level or better and hence it is omitted. (Refer to Table: 13).

Hence the log – linear form of regression equation for **Republic of Moldova** may be expressed as:

TII* = 0.092481 TGDPR* + 1.11945 FDIGDPR* + 0.220981 PDI* . (37)

Equation: (37) may be *approximated* as:

TII* = 0.09 TGDPR* + 1.12 FDIGDPR* + 0.22 PDI* . 37(a)

The Cobb – Douglas *functional* form of equation: 37(a) may be expressed as:

TII = TGDPR0.09 FDIGDPR1.12 PDI0.22 (38)

The statistical summary for the TII dataset of Republic of Moldova may be expressed as below [Refer to Table: 13(a)]:

- The TII dataset for Republic of Moldova reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Republic of Moldova are 0.288995 and 0.277341 respectively with a low Coefficient of Variation (C.V.) at approximately 15 percent;

- From equation: (38) it may be inferred that FDIGDPR, and PDI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Republic of Moldova (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 13]:

TII = 0.221 + 1.34 FDIGDPR . (39)

14. For **Russian Federation**, the coefficients of all the five independent variables, that is, TGDPR, FDIGDPR, GCFR, PCI and PDI along with the coefficient of the intercept are statistically significant at 5 percent level or better. (Refer to Table: 14).

Hence the log – linear form of regression equation for **Russian Federation** may be expressed as:

TII* = -0.132689 + 0.0679618 TGDPR* + 2.14355 FDIGDPR* + 0.0716983 GCFR*+ 0.263258 PCI* + 0.25053 PDI* . (40)

Equation: (40) may be *approximated* as:

TII* = -0.132 + 0.07 TGDPR* + 2.14 FDIGDPR* + 0.07 GCFR* + 0.26 PCI* + 0.25 PDI* . 40(a)

The Cobb – Douglas *functional* form of equation: 40(a) may be expressed as:

TII = -0.132 TGDPR0.07 FDIGDPR2.14 GCFR0.07 PCI0.26 PDI0.25 . (41)

The statistical summary for the TII dataset of Russian Federation may be expressed as below [Refer to Table: 14(a)]:

- The TII dataset for Russian Federation reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Russian Federation are 0.210104 and 0.216194 respectively with a low Coefficient of Variation (C.V.) at approximately 16 percent;

- From equation: (41) it may be inferred that FDIGDPR, PCI and PDI combined play major role in influencing the value of TII whereas TGDPR and GCFR play a minor role for the said country; and

- The straight line trend equation for Russian Federation (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 14]:

TII = 0.158 + 2.57 FDIGDPR . (42)

15. For **Slovakia**, the coefficients of only three independent variables, that is, FDIGDPR, GCFR, and PCI are statistically significant at 5 percent level or better (Refer to Table: 15).

Hence the log – linear form of regression equation for **Slovakia** may be expressed as:

TII* = 1.34849 FDIGDPR* + 0.267603 GCFR* + 0.599934 PCI* (43)

Equation: (43) may be *approximated* as:

TII* = 1.35 FDIGDPR* + 0.27 GCFR* + 0.6 PCI*. 43(a)

The Cobb – Douglas *functional* form of equation: 43(a) may be expressed as:

TII = FDIGDPR1.35 GCFR0.27 PCI0.6. (44)

The statistical summary for the TII dataset of Slovakia may be expressed as below [Refer to Table: 15(a)]:

- The TII dataset for Slovakia reflects a *negatively skewed platykurtic distribution*;

- The mean and median of TII for Slovakia are 0.232004 and 0.230091 respectively with a high Coefficient of Variation (C.V.) at approximately 23 percent;

- From equation: (44) it may be inferred that FDIGDPR and PCI combined play major role in influencing the value of TII whereas GCFR plays a minor role for the said country; and

- The straight line trend equation for Slovakia (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 15]:

TII = 0.176 + 1.49 FDIGDPR . (45)

16. For **Slovenia**, the coefficients of only two independent variables, that is, AFDIGDPR and GCFR are statistically significant at 5 percent level or better. (Refer to Table: 16).

Hence the log – linear form of regression equation for **Slovenia** may be expressed as:

TII* = 1.77744 SAFDIGDPR* + 0.308238 GCFR* (46)

Equation: (46) may be *approximated* as:

TII* = 1.77 SAFDIGDPR* + 0.31 GCFR* (46a)

The Cobb – Douglas *functional* form of equation: 46(a) may be expressed as:

TII = SAFDIGDPR1.77 GCFR0.31 (47)

The statistical summary for the TII dataset of Slovenia may be expressed as below [Refer to Table: 16(a)]:

- The TII dataset for Slovenia reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Slovenia are 0.186317 and 0.185216 respectively with a high Coefficient of Variation (C.V.) at approximately 21 percent;

- From equation: (47) it may be inferred that SAFDIGDPR (Statistically Adjusted net FDI to GDP Ratio) plays the major role in influencing the value of TII ; and

- The straight line trend equation for Slovenia (1995 – 2015) between TII (dependent variable) and SAFDIGDPR (independent variable) may be expressed as [Refer to Figure: 16]:

TII = 0.153 + 1.87 SAFDIGDPR (48)

17. For **Turkmenistan**, the coefficients of four out of five independent variables, that is, TGDPR, FDIGDPR, GCFR, and PCI are statistically significant at 1 percent level or better (Refer to Table: 17).

Hence the log – linear form of regression equation for **Turkmenistan** may be expressed as:

TII* = 0.0742306 TGDPR* + 0.921317 FDIGDPR* + 0.271738 GCFR* + 0.219328 PCI* . (49)

Equation: (49) may be *approximated* as:

TII* = 0.07 TGDPR* + 0.92 FDIGDPR* + 0.27 GCFR* + 0.22 PCI* . 49(a)

The Cobb – Douglas *functional* form of equation: 50(a) may be expressed as:

TII = TGDPR0.07 FDIGDPR0.92 GCFR0.27 PCI 0.22 . (50)

The statistical summary for the TII dataset of Turkmenistan may be expressed as below [Refer to Table: 17(a)]:

- The TII dataset for Turkmenistan reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Turkmenistan are 0.418803 and 0.400431 respectively with a low Coefficient of Variation (C.V.) at approximately 15 percent;

- From equation: (50) it may be inferred that FDIGDPR, GCFR and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Turkmenistan (1996 – 2012) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 17]:

TII = 0.338 + 1.10 FDIGDPR . (51)

18. For **Ukraine**, the coefficients of three independent variables, that is, TGDPR, FDIGDPR, and PCI are statistically significant at 5 percent level or better. The slope - coefficient is statistically significant at 10 percent level or better and hence it is omitted. (Refer to Table: 18).

Hence the log – linear regression equation for **Ukraine** may be expressed as:

TII* = 0.0537992 TGDPR* + 1.29367 FDIGDPR* + 0.689195 PCI* . (52)

Equation: (52) may be *approximated* as:

TII* = 0.05 TGDPR* + 1.29 FDIGDPR* + 0.69 PCI* . 52(a)

The Cobb – Douglas *functional* form of equation: 52(a) may be expressed as:

TII = TGDPR0.05 FDIGDPR1.29 PCI 0.69 . (53)

The statistical summary for the TII dataset of Ukraine may be expressed as below [Refer to Table: 18(a)]:

- The TII dataset for Ukraine reflects a *positively skewed platykurtic distribution*;

- The mean and median of TII for Ukraine are 0.213791 and 0.214543 respectively with a moderate Coefficient of Variation (C.V.) at approximately 17 percent;

- From equation: (54) it may be inferred that FDIGDPR, and PCI combined play major role in influencing the value of TII whereas TGDPR plays a minor role for the said country; and

- The straight line trend equation for Ukraine (1995 – 2015) between TII (dependent variable) and FDIGDPR (independent variable) may be expressed as [Refer to Figure: 18]:

TII = 0.162 + 1.58 FDIGDPR . (54)

19. For **Uzbekistan**, the coefficients of four out of five independent variables, that is, TGDPR, FDIGDPR, GCFR, and PCI are statistically significant at 1 percent level or better (Refer to Table: 19).

Hence the log – linear form of regression equation for **Uzbekistan** may be expressed as:

TII* = 0.0906194 TGDPR* + 2.08964 FDIGDPR* + 0.205548 GCFR* + 0.0870925 PCI* . (55)

Equation: (55) may be *approximated* as:

TII* = 0.09 TGDPR* + 2.08 FDIGDPR* + 0.21 GCFR* + 0.09 PCI* . 55(a)

The Cobb – Douglas *functional* form of equation: 55(a) may be expressed as:

TII = TGDPR0.09 FDIGDPR2.08 GCFR0.21 PCI 0.09 . (56)

The statistical summary for the TII dataset of Uzbekistan may be expressed as below [Refer to Table: 19(a)]:

**[.]**

^{[1]} 1. Data available at: http://databank.worldbank.org/data/home.aspx; http://unctadstat.unctad.org/wds/ReportFolders/reportFolders.aspx?sCS_ChosenLang=en

- Quote paper
- Debasish Roy (Author), 2017, Formulation of Transitional Impact Index (TII). A Cross–Country Analysis, Munich, GRIN Verlag, https://www.grin.com/document/379825

Publish now - it's free

Comments