Does financial development reinforce ecological footprint in singapore? Evidence from ardl and bayesian analysis

doi:10.21203/rs.3.rs-988321/v1

Download PDF

Research Article

Does financial development reinforce ecological footprint in singapore? Evidence from ardl and bayesian analysis

https://doi.org/10.21203/rs.3.rs-988321/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 25 Nov, 2021

Read the published version in Environmental Science and Pollution Research →

You are reading this latest preprint version

Singapore has been ranked in the most dynamic financial market and the highest ecological deficit country, indicating that the trade-off hypothesis may exist. The main goal of the present study is to probe the impact of financial development, economic growth, and human capital on ecological footprint in Singapore from 1980 to 2016. The outcomes obtained from the Auto Regressive Distributed Lag (ARDL) method have failed to provide a clear impact of financial sector development on ecological footprint. However, the Bayesian analysis reveals that both financial development and economic growth have a harmful influence on EF, while the impact of human capital is beneficial. A theoretical conclusion derived is that monetary expansion policies should be associated with improving human capital to achieve the United Nations SDGs in the context of Singapore. The findings of the study are of particular interest to policymakers for developing sound policy decisions for sustainable economic progress which is not at the cost of environment.

Environmental Engineering

Agricultural Economics & Policy

Financial development

Ecological footprint

ARDL

Bayesian analysis

Singapore

Table 1. Summary of previous studies

No.	Authors	Duration	Countries/Areas	Econometric approach	Findings
Financial development (FD) – EF nexus
1	(Uddin et al., 2017)	1991-2012	27 countries	FMOLS, DOLS	FD → EF (-);
2	(Ahmed et al., 2019)	1971-2014	Malaysia	ARDL
3	(Destek & Sarkodie, 2019)	1977-2013	11 countries	Augmented Mean Group (AMG)
4	(Omoke et al., 2020)	1971-2014	Nigeria	NARDL
5	(Pata & Yilanci, 2020)	1980-2015	G7 countries	Panel Threshold regression
6	(Saud et al., 2020)	1990-2014	45 countries	Pooled Mean Group (PMG)
7	(Charfeddine, 2017)	1970-2015	Quatar	Markov Switching models	FD → EF (+)
8	(Mrabet & Alsamara, 2017)	1980-2011	Quatar	ARDL
9	(Usman et al., 2020a)	1995-2017	20 countries	DCCEMG, AMG
10	(Naqvi et al., 2020)	1990-2017	152 countries	AMG
11	(Godil et al., 2020)	1986-2018	Turkey	Quantile ARDL
12	(Shujah Ur et al., 2019)	1991-2014	CCE countries	SUR regression
13	(Baloch et al., 2019)	1990-2016	59 countries	Driscoll-Kraay regression
14	(Khan et al., 2019a)	1990-2016	5 countries	AMG, CCEMG
Economic growth (EG) – EF nexus
1	Alola et al. (2019)	1997-2014	16 EU countries	Panel ARDL	EG " EF (+)
2	Danish et al. (2019)	1971-2014	Pakistan	ARDL
3	Zafar et al. (2019)	1970-2015	United States	ARDL
4	Nathaniel et al. (2020a)	1990–2016	MENA countries	AMG
5	Destek and Sinha (2020)	1980-2014	24 OECD countries	FMOLS, DOLS
6	Ahmed et al. (2020a)	1970-2016	China	Bootstrap ARDL
7	Nathaniel and Khan (2020)	1990-2016	ASEAN countries	AMG
8	Ahmed et al. (2020b)	1971-2014	G7	CUP-FM and CUP-BC
9	Usman et al. (2020b)	1994-2017	33 countries	FMOLS, DOLS, AMG	EG " EF (-) (Africa and Europe)
10	Baz et al. (2020)	1971-2014	Pakistan	NARDL	EG " EF (^ns)
11	Hassan et al. (2019)	1970–2014	Pakistan	ARDL	EG " EF (+) Then EG " EF (-)

Notes: (+): positive impact; (-) negative impact; ^ns: no-significant

Table 2. Descriptive statistic

Variables	Obs	Mean	Std.Error	Minimum	Maximum
EF	37	5.843	1.465	3.441	8.310
lnFD	37	15.262	0.996	11.167	15.894
lnGDP	37	10.287	0.422	9.497	10.881
HC	37	2.528	0.572	1.651	3.809

Table 3. Results of the unit root test

Variables	Test statistic	1% Critical value	5% Critical value	10% Critical value
ADF test
EF	-0.605	-4.297	-3.564	-3.218
lnFD	-2.019
lnGDP	-1.328
HC	-1.628
Δ.EF	-4.335***	-4.306	-3.568	-3.221
Δ.lnFD	-4.427***
Δ.lnGDP	-3.187*
Δ.HC	-1.542
PP test
EF	-1.502	-4.279	-3.556	-3.214
lnFD	-7.197***
lnGDP	-1.722
HC	0.267
Δ.EF	-6.719***	-4.288	-3.560	-3.216
Δ.lnFD	-14.431***
Δ.lnGDP	-5.338***
Δ.HC	-1.562
GLS-ADF test
EF	-1.652	-3.770	-3.283	-2.968
lnFD	-3.946***
lnGDP	-1.645
HC	-2.460
Δ.EF	-5.776***	-3.770	-3.293	-2.976
Δ.lnFD	-5.262***
Δ.lnGDP	-4.887***
Δ.HC	-2.993*

Note: All tests are chosen based on the Akaike Information Criterion with intercept and trend, without structural breaks. ***, ** and * respectively denote significance levels of 1%; 5% and 10%.

Table 4. Result of cointegration test

Test Statistic	Value	Significantly level	I(0)	I(1)
F-statistic	4.602	10%	2.72	3.77
k	3	5%	3.23	4.35
		1%	4.29	5.61
t-statistic	-3.671	10%	-2.57	-3.46
k	3	5%	-2.86	-3.78
		1%	-3.43	-4.37

Table 5. The optimal lag of each variable

Lag	AIC	HQIC	SBIC
EF variable
0	3.491	3.507	3.537
1	2.083*	2.113*	2.174*
2	2.139	2.185	2.275
3	2.142	2.203	2.323
4	2.198	2.274	2.425
lnFD variable
0	2.975*	2.991*	3.021*
1	1.173	2.997	3.027
2	1.245	3.057	3.103
3	1.252	3.062	3.123
4	1.307	3.103	3.179
lnGDP variable
0	0.851	0.867	0.897
1	0.002*	-3.618*	-3.587*
2	0.002	-3.562	-3.516
3	0.002	-3.587	-3.526
4	0.002	-3.540	-3.464
HC variable
0	1.592	1.607	1.637
1	0.001	-3.668	-3.638
2	0.001*	-4.672*	-4.627*
3	0.001	-4.625	-4.564
4	0.001	-4.582	-4.506

Note: * denotes the optimal lag

Table 6. The short- and long-run estimated coefficients

Variables	Coefficient	Standard error	T-ratio [Prob]
CoinEq(-1)	-0.5647	0.1538	-3.67 [0.001]
lnFD(-1)	0.2631	0.2901	0.91 [0.372]
lnGDP(-1)	7.7833	1.6444	4.73 [0.000]
HC(-1)	-3.7203	1.2682	-2.93 [0.006]
ΔlnFD	0.0044	0.1096	0.04 [0.968]
ΔlnGDP	4.3956	1.1761	3.74 [0.001]
ΔHC	-2.1011	0.6705	-3.13 [0.004]
Intercept	-38.851	10.791	-3.60 [0.001]
R²	0.42	R²-adj	0.41
	3.501 [0.174]		0.746 [0.395]
	14.56 [0.439]		3.695 [0.594]
CUSUM test	Stability	CUSUMSQ test	Stability

Note: , , , denote the LM tests for serial correlation, functional form, normality, and heteroskedasticity. The value in brackets is the corresponding p-value, respectively.

Table 7. The empirical results by Bayesian analysis

Variables	Mean	Std. Dev.	MCSE	Prob of mean > 0	[95% Cred. Interval]
Dependent variable: EF
lnFD	0.0886	0.1145	0.0053	0.793	[-0.1329; 0.3213]
lnGDP	7.0411	0.9407	0.0687	1	[5.0955; 8.9038]
HC	-3.1321	0.7001	0.0512	1*	[-4.5258; -1.6912]
Intercept	-60.026	7.8752	0.5864	1*	[-75.158; -43.902]
e.EF sigma2	0.4305	0.1242	0.0044		[0.2601; 0.7158]
MCMC sample size = 10,000				Acceptance rate = 0.373

Note: * denote probability of mean < 0

Table 8. Results of the causality test

Null hypothesis: No causality	Chi_square	p-value
lnFD does not Granger cause EF	1.9601	0.5807
EF does not Granger cause lnFD	13.6307	0.0035
lnGDP does not Granger cause EF	11.6575	0.0087
EF does not Granger cause lnGDP	9.0829	0.0282
HC does not Granger cause EF	10.0785	0.0179
EF does not Granger cause HC	7.8212	0.0499
lnGDP does not Granger cause lnFD	17.3014	0.0006
lnFD does not Granger cause lnGDP	5.0863	0.1656
HC does not Granger cause lnFD	14.6027	0.0022
lnFD does not Granger cause HC	1.3994	0.7057
lnGDP does not Granger cause HC	5.0463	0.1684
HC does not Granger cause lnGDP	7.8931	0.0483

Download PDF

Journal Publication

published 25 Nov, 2021

Read the published version in Environmental Science and Pollution Research →

Reviews received at journal
29 Oct, 2021
Reviewers invited by journal
29 Oct, 2021
Editor invited by journal
29 Oct, 2021
Editor assigned by journal
18 Oct, 2021
First submitted to journal
14 Oct, 2021

You are reading this latest preprint version

Does financial development reinforce ecological footprint in singapore? Evidence from ardl and bayesian analysis

Status:

Journal Publication

Version 1

Abstract

Figures

Full Text

Tables

Status:

Journal Publication

Version 1