Agricultural Research Management Information System

Home
Research Summary
Search
About Us
- ARMIS
- Brochure
Contact Us
Report
User Request
Data Input
Help
- Operation Manual
  - PDF
  - Video
- Program Area & Commodity
We have reached 37600 number of research entries at this moment.

- Logout

Research Detail

Home
Research
Detail

Carbonomics of the Bangladesh Agricultural Output: Causality and Long-run Equilibrium

Md. Wahid Murad
Business School, Faculty of the Professions, The University of Adelaide, Adelaide, Australia,

Janek Ratnatunga
School of Commerce, Division of Business, University of South Australia, Adelaide, Australia

Abstract/ Executive Summary:

Purpose – The key purpose of this paper is to examine the causality and long-run relationship between CO2 emission and agricultural output for an agriculture-dependent developing country, namely Bangladesh. Design/methodology/approach – In order to attain the objective, this study has used long-time series data and employed advanced econometric techniques of unit root test, nonlinear least square estimation, Vector Error Correction estimation and Granger causality test. Findings – The empirical results of the study reveal that Bangladesh's agricultural output is not a Granger causal for Bangladesh's CO2 emission, but the country’s CO2 emission is a Granger causal for its agricultural output. The results also reveal for Bangladesh that any disequilibrium between CO2 emissions and agricultural output could take approximately 17 years to converge to the long-run equilibrium. The results further reveal that the adjustment rate for Bangladesh's agricultural output is positive and quite fast at the rate of 69 percent a year. So any disequilibrium will be corrected mostly by the adjustment in Bangladesh agricultural output. Practical implications – The current CO2 emission in Bangladesh is still below the equilibrium level, which is considered to be an advantage for the country, particularly its agriculture sector which will reasonably not face any stricter CO2 emissions controlling policies and regulations in the near future. Originality/value – The originality of this study lies on the extent to which an agriculture-dependent developing country such as Bangladesh does not have greater concern about the CO2 emission for now and the near future. The originality does also lie on the fact that no other study has yet examined this issue.

Keywords: Bangladesh, Agriculture, Developing countries, Econometrics, CO2 emission, Agricultural output, Causality, Long-run equilibrium, Carbonomics

Location: In Bangladesh

Start Date:

End Date:

Program Area: Socio-economic and Policy

Commodity: Socio-economic status

Objectives:

Bangladesh is essentially an agriculture-dependent developing country, and it is one of the world’s most vulnerable countries to climate change. Climatic variations in terms of heavy rainfall and excessive temperature have been observed to be significantly affecting the agriculture sector in the country. Since about 19 percent of the country’s GDP is contributed by the agricultural sector, which employs 45 percent of its total labor force, the implication of climate change to Bangladesh’s agricultural growth is unquestionably significant. This study is thus an effort to empirically investigate the causality and long-run relationship between CO2 emission and agricultural output for Bangladesh.

Methodology:

Keeping the research hypothesis and objective in mind, this study has gathered relevant time series data and information from secondary sources. The data on Bangladesh agriculture output and Bangladesh carbon dioxide emissions covering the period from 1990 to 2007 were collected from the United Nations (UN) online database (http://data.un.org/). It is worth noting to mention that the data on the above two-time series variables were found available on the UN database only from 1990 until 2007, and thus we had to employ the time span of 17 years in the present study. The variables were coded and entered accordingly into EViews 4.1 data analytical base. For example, the variable of Bangladesh carbon dioxide emissions (thousands metric tons) was coded as BDCO2 while Bangladesh agriculture output (thousands US$) was coded as BDAO. Both series of variables were tested for stationarity or non-stationarity using the widely used augmented Dickey-Fuller (ADF) and Phillips-Perron (PP) tests for a unit root. In fact, the ultimate objective to conduct a unit root test in time series variables is to defend the econometric models against the characteristics of spurious time-series regressions.

At the first stage of the model's development, a simple ordinary least square (OLS) method was employed for both time-series variables. This was done with the view of observing how and at what extent the results of OLS models will ultimately vary when the series are transformed into a numerical log. The log-linear regression model was then employed to observe if the paired variables provide any meaningful relationship. Hence, the logarithmic dynamic equation was developed and an actual fitted residual graph of the model was observed. Other features of the model such as serial correlation LM test, histogram normality test, and white heteroskedasticity (no cross-terms) test were also observed here. A non-linear estimation for all the dependent variables was then conducted in their de-trended (first difference) form. This was actually conducted to estimate the partial adjustment of the model and to find the estimates of the long-run parameters and the speed of adjustment directly. An ARIMA (1,1,1) estimation of all the dependent and independent variables was also conducted to forecast them and view their divergence or convergence graphically.

At the next and final stage, a vector error correction (VEC) estimation of the paired time series was conducted in order to observe the long-run adjustment coefficients on the cointegrated equation of the model. Since the VEC model is a special type of restricted vector autoregression (VAR), which is an econometric model used to capture the evolution and the interdependencies between multiple time series, the unrestricted cointegration test and VEC pairwise Granger causality test were also reasonably conducted for the model. Other features of the combined model such as inverse roots of AR characteristics polynomial were checked and VEC lag exclusion Wald test, VEC residual serial correlation LM test, VEC residual normality test, VEC residual heteroskedasticity test and impulse responses of the variables to Cholesky test were also conducted.

Source of Information: Management of Environmental Quality: An International Journal Vol. 24 No. 2, 2013 pp. 256-271

Article Link (Online Journal): DOI 10.1108/14777831311303128

Funding Source:

Budget:

Total Budget (Tk.) :

Achievement/Findings:

Based on the findings obtained from the above models, this study concludes that Bangladesh's agricultural output and the related policy do not have any important effects on the country’s CO2 emissions. This study also concludes that any disequilibrium between CO2 emissions and agricultural output could take approximately 17 years to converge to the long-run equilibrium. The adjustment rate for Bangladesh's agricultural output is positive and quite fast at the rate of 69 percent a year. So any disequilibrium will be corrected mostly by the adjustment in Bangladesh agricultural output. This implies the huge potential for Bangladesh agriculture even under adverse climatic conditions. This study further concludes that the country’s present CO2 emissions are still below the equilibrium level. This situation is considered as an advantage for the country as the agriculture sector will not face stricter CO2 regulations in the near future. This study also concludes that Bangladesh's agricultural output is not Granger Causal for Bangladesh's CO2 emission. But the Bangladesh CO2 emission is a Granger causal for Bangladesh agricultural output. While the nexus between CO2 emission and agricultural output is being debated widely at both country and global levels, a universal conclusion on their causal relationship requires further investigation using multi-disciplinary approaches. This is due to the fact that agricultural practices vary from country to country and they are actually influenced largely by the respective country’s agricultural policies and regulations. Thus the relationship between the agricultural output and CO2 emissions is felt differently by different stakeholders in different countries.

Knowledge Management: Journal