Dynamic Effects of Islamic Finance and Corruption on Environmental Quality in Malaysia
Hengchao Zhanga, Riasat Amin Imonb, Rafia Afrozb
aSchool of Graduate and Professional Studies, INCEIF University, Malaysia. (Email: zhang@inceif.edu.my)
bDepartment of Economics, Faculty of Economics and Management Sciences, International Islamic University Malaysia
15th International Conference on Islamic Economics and Finance (15th ICIEF)
20th -22nd February 2024, Sasana Kijang, Kuala Lumpur
Executive Summary
INTRODUCTION
-2nd highest per capital CO2 emitter in ASEAN
-temperature increased nearly 0.2 °C every decade since 1970.
-29% of trees coverage (8.67 million hectares) loss since 2000.
Malaysia: CO2 Emissions Over the Time
LITERATURE REVIEW
LITERATURE REVIEW
Theme | Explanation | Evidence |
Corruption & Environmental Degradation | Corruption and environment: a) Direct: ↑Corruption 🡺 ineffective implementation of of environmental regulations 🡺↑ CO2 emissions b) Indirect: ↑Corruption 🡺 reduced productivity🡺 lower income 🡺 less energy consumptions🡺 ↓ CO2 emissions. ↓ Corruption🡺 ↑compliance costs 🡺 ↑ CO2 emissions | |
Population & Environmental Degradation | Population growth and increased consumption levels can exacerbate greenhouse gas emissions and climate changes (Dietz & Rosa, 1997) | |
Islamic Finance & Environmental Degradation | Preservation of the ecological equilibrium on Earth is a significant responsibility of humanity, and as such, it is an integral aspect of the divine objectives outlined in the Shariah (Obaidullah, 2017). Mixed Belief:
| |
METHODOLOGY
METHODOLOGY
Variables | Measurement | Data Source | Symbol | Reference |
Environment Impact (I) | Per capita CO2 emission in metric tons | WDI | CO2 | Khan et al. (2020); Nosheen et al. (2020); Shahbaz et al. (2016) |
Population (P) | Urban population (% of total population) | WDI | U | (Bekhet & Othman (2017); Nosheen et al. (2020) |
Affluence (A) | Real GDP per capita (constant 2015 US$) | WDI | Y | Jamel & Maktouf (2017); Nosheen et al. (2020); Shahbaz et al. (2016) |
Islamic finance development (IFD) | Index generated by applying Principal Component Analysis from total savings and total financing in Islamic banking system, and total valuation of new Sukuk issued | Owner’s own calculation based on data extracted from Bank Negara Malaysia | IFD | Grassa & Gazdar (2014); Iskandar et al. (2020); Kassim (2016) |
Corruption | Corruption Perception Index: 0 (most corrupted)-100 (least corrupted) | Transparency International | CPI | Balsalobre-Lorente et al. (2023); Butt et al. (2023); Yahaya et al. (2020) |
TABLE 1
Description of Variables
EMPIRICAL RESULTS
UNIT ROOT TEST
TABLE 3 Unit Root Test | ||||||||
| Level | First Difference | ||||||
| ADF | PP | ADF | PP | ||||
| Constant | Constant & Trend | Constant | Constant & Trend | Constant | Constant & Trend | Constant | Constant & Trend |
lnCO2 | -1.446* | -2.059 | -1.448 | -1.986 | -5.769*** | -6.018*** | -5.769*** | -6.018*** |
lnU | -2.498 ** | -2.125 | -16.363*** | -5.205*** | -2.444** | -2.423 | -2.580* | -2.393 |
lnY | 0.270 | -5.045 | 0.328 | -4.916*** | -6.610*** | -6.406*** | -6.610*** | -6.406*** |
IFD | 0.407 | -2.015 | 0.487 | -1.967 | -4.808*** | -5.433*** | -4.808*** | -5.433*** |
lnCPI | -2.774*** | -2.846 | -2.869** | -2.955 | -5.116*** | -4.999*** | -5.116*** | -4.999*** |
Note. *,**,*** indicates significance at 10%, 5%, and 1%. | | | | |||||
COINTEGRATION
TABLE 4 Optimal Lag Selection | |||||||
Sample: | 2001-2022 |
|
|
| Obs = 22 |
| |
Lag | LL | LR | df | p | AIC | HQIC | SBIC |
0 | 143.8 | | | | -12.618 | -12.5596 | -12.3701 |
1 | 288.97 | 290.35 | 25 | 0 | -23.5429 | -23.1925 | -22.0552 |
2 | 321.84 | 65.744 | 25 | 0 | -24.2586 | -23.616 | -21.5309 |
3 | 410.03 | 176.38* | 25 | 0 | -30.0029* | -29.0683* | -26.0355* |
4 | . | . | 25 | . | . | . | . |
TABLE 5 ARDL-bounds Cointegration Test | ||||||||||
| Test-Statistics | 10% | 5% | 1% | P-Value | |||||
| I(0) | I(1) | I(0) | I(1) | I(0) | I(1) | I(0) | I(1) | ||
ARDL (2,2,1,1,1) | F | 6.860** | 2.894 | 4.426 | 3.69 | 5.522 | 5.861 | 8.466 | 0.005 | 0.023 |
t | -4.901** | -2.491 | -3.636 | -2.922 | -4.16 | -3.854 | -5.3 | 0.002 | 0.018 | |
Note: ∗∗denotes the rejection of the null hypothesis of no-cointegration relationship at 5% significance level. | ||||||||||
DIAGNOSTIC TESTS
ABLE 6A Breusch–Godfrey LM Test for Autocorrelation | |||
lags(p) | df | F | Prob > F |
1 | (1,11) | 4.774 | 0.0514* |
2 | (2,10) | 2.399 | 0.1409 |
3 | (3,9) | 1.987 | 0.1865 |
4 | (4,8) | 1.909 | 0.2023 |
Note: ∗denotes the rejection of the null hypothesis of no-autocorrelation at 10% significance level. | |||
TABLE 6B: White's Homoskedasticity Test | |||
Source | Chi2 | df | P-value |
Heteroskedasticity | 24.00 | 23 | 0.4038 |
Skewness | 9.40 | 11 | 0.5846 |
Kurtosis | 0.91 | 1 | 0.3402 |
Total | 34.31 | 35 | 0.5010 |
Note: null hypothesis of homoscedasticity has failed to be rejected at 10% significant level | |||
TABLE 6C Skewness and Kurtosis Tests for Normality | ||||
|
|
| Joint Test | |
Obs | Pr(skewness) | Pr(kurtosis) | Adj chi2 | Prob>chi2 |
32 | 0.2755 | 0.1613 | 3.53 | 0.1713 |
Note: null hypothesis of normality in residual cannot be rejected at any level of significance | ||||
DYNAMIC ARDL ESTIMATION
TABLE 7 Dynamic ARDL Model Estimation | ||||
Obs=24 | | | Prob > F=0.0000 | |
R2= 0.9792 | | | Adj R2=0.9601 | |
Variable | Coefficient | Std. Err. | t | P>t |
lnCO2t-1 | -0.30 | 0.26 | -1.12 | 0.284 |
lnUt-1 | 0.19 | 0.17 | 1.15 | 0.273 |
lnYt-1 | 8.87 | 9.81 | 0.90 | 0.384 |
IFDt-1 | 5.88 | 1.31 | 4.49*** | 0.001 |
lnCPIt-1 | 24.15 | 9.39 | 2.57** | 0.024 |
ΔlnCO2t-1 | 0.63 | 0.25 | 2.55** | 0.026 |
ΔlnUt | -0.04 | 0.03 | -1.61 | 0.134 |
ΔlnUt-1 | 0.33 | 0.17 | 1.97* | 0.072 |
ΔlnYt | -0.35 | 0.29 | -1.23 | 0.241 |
ΔIFDt | -0.08 | 0.02 | -3.27*** | 0.007 |
ΔlnCPIt | 0.75 | 0.19 | 3.99*** | 0.002 |
Cons | -22.64 | 5.01 | -4.52*** | 0.001 |
Notes: *,**,*** indicates significance at 10%, 5%, and 1%. | ||||
IMPULSE RESPONSE ANALYSIS
Dynamic Effects of ±2% Shock in lnU on lnCO2
Dynamic Effect of ±2% Shock in lnCPI on lnCO2
Dynamic Effect of ±2% Shock in IFD on lnCO2
CONCLUSION & IMPLICATIONS
Objective: To examine how IFD, corruption (CPI), economic growth (Y), and urbanization (U) affect CO2 emissions in Malaysia using the STIRPAT framework and DYARDL analysis on data from 1997 to 2022.
Findings:
Implications:
Thank you �for your time
Dr. Hengchao, Zhang (Ali)
INCEIF University, Malaysia
zhang@inceif.org