Is Achieved Economic Development Environment Friendly? A New Insight From Central and Eastern Europe

Over the last few years, the linkage between economic development and environmental 12 degradation has become a provocative question. Although this nexus has been studied vastly, 13 some of the critical variables of economic development and their impacts on the environment 14 need more focus. The present study explores the association between economic development, 15 outward foreign direct investment, financial development, renewable energy consumption, 16 natural resource rents, trade openness, and ecological footprint in Central and Eastern European 17 economies. The panel data estimators such as augmented mean group and common correlated 18 effect mean group are employed from 1990 to 2017. Empirical findings document that outward 19 foreign direct investment, financial development, trade openness, natural resource rents, and 20 renewable energy consumption increase economic development, implying that they positively 21 affect economic development. Findings validate the inverted U-shaped EKC for concerned 22 economies in case of the ecological footprint. The results show that the interaction term of 23 GDPC with NR, outward foreign direct investment, and RE are eco-friendly indicators. The 24 study results develop imperative policy implications for the selected region to attain sustainable 25 development goals. 26


29
Considering global and regional economic assimilation, multinational enterprises as participants 30 of international direct investment have gradually become organizers and watchdogs of universal 31 resource provision. Foreign direct investment has quickly developed since the 1980s. Outward 32 foreign direct investment (OFDI) raised from 243.8 billion USD in 1990 to 1014 billion in 2018, 33 and Raddatz 2008), and the bureaucratic system (Dutt 2009). Generally, TO plays a vital role in 64 altering NR into a blessing rather than a curse. 65 Domestic finance creation can be seen as one innovation category that is capable to accelerate shocks. However, it can be challenging to move from non-RE to RE production. The high initial 75 expense is one of the most significant issues for RE. In contrast with non-RE-based energy 76 expenditure, many financial barriers need to be met, including higher start-up infrastructure and 77 operating costs. Therefore, economies want to attain economic development across the world 78 while ignoring its harmful impact on environment. is not yet addressed by literature, but researchers and practitioners are eager to know more about 94 it. Countries worry about per capita income to contend with other countries but ignore the 95 adverse environmental effects. This paper addresses the per capita income in the CEE economies 96 with environmental aspects and suggests how harmful consequences of economic development 97 can be reduced. In addition, this work is prevailing literature in three ways; firstly, this is the first   Africa and supported the evidence of the positive association between the growth and RE. A 128 study by Li et al. (2017) for emerging economies and found OFDI as the main driver of growth.

142
In the light of previous case studies, this section is divided into two different faces 1) deals with 143 those studies, which try to estimate the impact of growth on carbon emissions, 2) deals with past 144 studies, which attempted to evaluate the effect of growth on EFP. As a global issue, carbon      185 This study used the yearly panel data for the period 1990-2017 for 16 selected CEE countries 186 (Appendix Table 10). These selected economies decided to ratify both global and regional goals   In Table 2, correlation coefficients of the study variables are given. In the light of both models, 208 the income per capita is positively correlated to EFP at the one percent statistical significance.

209
The correlation matrix also discloses that LEFP is negatively correlated with OFDI, and 210 likewise, this behavior with RE consumption at one percent level of significance. We also 211 described a positive influence of TO on EFP at a one percent level of significance, while a 212 negative correlation among NR, FD, and per person EFP variables is also statistically significant.

213
Furthermore, OFDI, RE consumption, LFD, and TO have a positive correlation with income per 214 capita, while NR income has a negative association at a 1% level of significance. According to 215 the given correlation coefficient outcomes, there is no correlation among the variables as they 216 have moderate values, and to confirm this, we have used the variance inflation factor (VIF) (see 217 Appendix Table 11 for details).

228
LGDPCi,t=α0+α1LOFDIi,t+α2LNRi,t+α3LTOi,t+α4LFDi,t+α5LREi,t+µi,t (1) 229 Here, i and t denote the economy and time in the panel estimation, as i = 1, 2, ……….., n and t = 230 1, 2, ……., T and µ is random error. The effect of OFDI and NR are determined by α 1 and α 2 , 231 respectively. Accordingly, statistically significant and positive α 3 directs that TO enhance the The CD test has a drawback, such as lacking power under a situation where pair-wise statistics. The bias-adjusted LM statistics are given as Equation (6): as Equation (8): Where ̅ −1 represents the mean across each cross-section. Further, the CIPS test can be 314 presented as given in Equation (9): cointegration can be expressed as Equation (10): Where δˊid t and η i are the deterministic and coefficient of error correction terms. This test is 327 based on two statistics, i.e., group (G τ , G a ) and panel (P τ , P a ). The null and alternate hypothesis 328 can be expressed as: The rejection of H o means all the panel is cointegrated.

350
In this section, the empirical results of the relationship between economic development and 351 environmental quality are given for CEE economies for two different empirical models as 352 described above. Thus, for the empirical analysis, descriptive statistics plays an important role.

353
Descriptive statistics 354 The following Table 3 gives descriptive statistics of the variable for the selected panel under   355 inquiry. Table 3 depicts no significant difference between the mean and median of all concern 356 factors, and all variables show a considerable extent of attention. Here, the findings of the CD tests are provided. Findings of CD tests have been given in Table 4.

362
Results validate early perceptions since the different techniques assumed can strongly reject H 0 363 of cross-sectional independence for both panels. As a result, our specific income groups showed 364 CD, and thus, 2 nd generation unit root and cointegration tests that accommodate the dependence 365 problem are applied to attain consistent findings. In addition, findings of the test for homogeneity 366 are given in Table 5.   Note: *** and ** represent the one and five percent significance level. 378 379 380 Cointegration association between the variables in models 1 and 2 by following the (Westerlund 381 and statistics 2007) error correction model cointegration method are provided in Table 7. This 382 cointegration test is in line with our selected data since it permits for the CD. Table 7 represents

392
In this sub-section, findings of the AMG and CCE-MG approaches are given for both models for 393 the CEE economies. Discussion of analysis is given in two parts for both models separately.    469 quality 470 The second panel of Table 8 represents the estimations through AMG and CCE-MG to achieve LGDNR. For understanding the effect of economic development on environmental degradation, 475 we have used income per capita as a proxy for achieved economic development. To answer this 476 question raised at the start of the study, we have taken help from Environment Kuznets curve 477 hypothesis theory, which is well known. For the understanding of EKC-hypothesis, the GDPC 2 is 478 introduced for the EFP.

479
Likewise, the co-efficient of GDPC (β 1 > 0) and the GDPC 2 (β 2 < 0) recommended that one 480 percent increase in this factor can cause an escalation of 1.169% in the dependent variable while 481 the square of GDPC can cause a reduction of 0.7278% in the EFP for the AMG-specification. A 482 one percent increase in concern factors [GDPC and GDPC 2 ] would lead to an increase of 3.275%

The causal association between economic development and its factors 542
The D-H panel causality test outcomes are provided in the following Table 9. This table is    LGDFD, and from LGDPC to LGDFD. In the last LGDPC, Granger causes interaction term of 617 LGDNR.

618
The investigated results give some practical implications that may be imperative for policy There is no financial support for this work that can influence its outcome.

662
The authors declare that they have no competing interests.

664
The data will provide the corresponding author on demand.