Water Resources and Industry 18 (2017) 1–8

Contents lists available at ScienceDirect

Water Resources and Industry

journal homepage: www.elsevier.com/locate/wri

Exploratory analysis of fines for water pollution in Bangladesh MARK

Nabil Haque
PhD Student in Technology Policy & Innovation, Department of Technology & Society, Old Computer Science Building, Stony Brook University, Stony
Brook, NY 11790, USA

AR TI CLE I NF O AB S T R A CT

Keywords: As a consequence of rapid industrialization, the waterbodies of Bangladesh have transformed into
Water pollution seasonal dead zones from the ensuing pollution. Despite having environmental regulations to
Textile control industrial pollution, lack of effective enforcement has jeopardized environmental quality.
Effluent Treatment Plant (ETP) Evaluation of enforcement mechanisms have not gained attention from researchers until recently.
Clearance
This qualitative and descriptive analysis illustrates the current enforcement regime for environ-
Violation
Bangladesh
mental compliance in Bangladesh focusing on fines levied on polluters. Although there are no
official guidelines for fines based on type of violation, this paper identified that there are
differences of fines among violation based on historical data. It was also found that textile
factories are not penalized heavily compared to non-textile factories. Repeat offenders were
found to be penalized at the same rate. This study can be used to design appropriate penalty
structure based on violation types, and reform the enforcement system so that polluters pay
principle is actually implemented.

1. Introduction

Scenes and images of industrial pollution in Bangladesh always follow a call for action to the government to go tough on polluters.
Even without considering administrative and political discretion, the task is monumental due to the growth of industries over the last
decade. Fines or fiduciary penalties for non-compliance with environmental regulations are levied in Bangladesh by the Department
of Environment (DoE). While there are generally no policies governing these penalties rates [3], the status quo is usually fining
arbitrarily. Enforcement is also weak and intermittent due to shortage of inspectors of the DoE [11]. The Department of Environment
publishes these penalties in the press so that it serves as a deterrent. A cursory look at these press reports show the variety of
violations that polluters are charged with. Since the fines are levied arbitrarily, the learning outcome of enforcement through levying
fines have been missing in the policy process.
There is a dearth of academic literature on the policy aspects of water pollution and industrial environmental economics in
Bangladesh. Some empirical work has been done focusing mostly on the abatement costs for polluters [7,13,17]. Empirical studies on
monitoring and enforcement is limited not only for developed countries, but more for developing countries due to difficulties in
accessing data [5]. Appropriate decision support tools to assess and compare the efficiency and cost effectiveness of command and
control based instruments in the context of Bangladesh are currently non-existent [17].
Among the existing literature on environmental governance, most of them have qualitatively stated the capacity constraints of the
enforcement conducted by DoE due to shortage of manpower as well as unambiguity of existing rules [11,14,19]. This has led to a
situation where the benefits from non-compliance far outweighs the potential fines polluters will face if they get caught. Ideally,
polluters must realize that non-compliance has material consequences that outweigh the benefits of breaking the rules, and that the
system for pursuing non-compliance functions efficiently and effectively. Currently, polluting factory owners are not motivated to

E-mail addresses: mdnabil.haque@stonybrook.edu, nabilhaq@gwmail.gwu.edu.

http://dx.doi.org/10.1016/j.wri.2017.05.001
Received 24 February 2017; Received in revised form 4 May 2017; Accepted 22 May 2017
2212-3717/ © 2017 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
N. Haque Water Resources and Industry 18 (2017) 1–8

comply with water regulations, as the penalty imposed by DoE for noncompliance is negligible [20]. Although this has been a well-
known phenomenon, no empirical study has been done till date to study the penalties imposed or the effects it has on polluters.
There is a growing recognition that only traditional sanctions based approach will not help develop incentives to reduce untreated
wastewater discharge. Although the Department of Environment has made substantial efforts in recent years to increase staff and
strengthen its enforcement practices, the focus on ad hoc fines and closures alone remains a rather blunt instrument [19]. Even the
finding that the fines are made on an ad hoc basis comes from surveys and interviews of stakeholders [14]. Although many
establishments have installed the mandated chemical and biological effluent treatment plants, they are often kept idle due to their
significant operating costs—combined with a relatively low probability of getting caught [19]. The later assertion on the probability
of getting caught is also based on stakeholder feedback, as currently there are no credible research focusing exclusively on the
pollution fine or the probability of getting fined.
Recognizing this important gap in the knowledge of enforcement effectiveness, this paper seeks to shed light on the current
structure of pollution fines, for which there are no official guidelines in Bangladesh. Since there were no previous studies supporting
directions or hypothesis development in the subject of environmental enforcement in Bangladesh, the study took the form of an
exploratory analysis. This paper aims to understand the prevailing status of environmental fines in Bangladesh using historical data as
it relates to surface water pollution. This qualitative study is based on collection of the press records of fines from May 2011 to April
2016, and the basis of analysis developed during the period of data collection as key variables fell into its place. The novelty of the
study is in the methodology, and findings in the context of Bangladesh, which policymakers can use to review the effectiveness of its
monitoring and violation approaches, and explore options for strengthening its impact.

1.1. Research objectives

The study is first of its kind in the context of Bangladesh and will add to existing literature of environmental economics. Some of
the questions have direct policy implications that can direct future path by assessing whether the fines are effective or not. The key
research questions it tries to address are–

a) Summarize the descriptive statistics of the fines and their distribution.

b) Are the environmental fines geographically representative of industrial clusters?
c) Since textile sector is a dominating industry, do fines for the textile sector differ from other sectors?
d) What are the most common violations that results in environmental fines?
e) Are there repeat offenders? Do the fines increase for second time violations?

2. Methodology

This qualitative study is descriptive, and aims to portray existing scenario of water pollution fines based on historical records. The
research is an example of grounded theory where data collection was done in isolation from any predetermined theory or conceptual
framework. The aim of this research is assessing phenomenon rather than testing theories.

2.1. Sampling

Although the data collected is not exhaustive, it represents a sample of environmental fines that have been publicized by the DoE.
Every time reports of industrial water pollution came in Bengali and English newspaper, the news clips were stored digitally, and a
list of records is updated using key information that were common in all reports. Only one person was responsible for the sampling, so
the possibilities for missing data exist. However, bi-annually internet searches of all Bangladeshi newspaper in both languages were
conducted to verify existing records minimizing sampling error. Other sampling error would be the use of news media to collect
information, rather than going direct to the source. Complete information of environmental fines is not shared by Department of
Environment due to internal bureaucracy and pressure from business community for not harming reputation. However, the
government also wants to portray that they are tough on polluters and therefore these press releases are made from time-to-time.
Newspapers may choose to not report some of these, and these differences were observed over a number of newspaper. This sampling
error was minimized by following publication of 5–6 separate news agencies. Fig. 1 shows the number of factories fined over each
quarters of the five-year period.
The methodology is an innovative approach for countries where such data repository is missing. The initial dataset had 309
observations, but some of the observations were derived. In some cases, newspaper articles reported that 3 factories were fined a total
of 5.8 million Taka, thereby splitting that amount by 3. For this analysis, such cases were dropped because it didn’t reflect the fines
based on types of violation which is an element for this analysis. The final dataset contains 290 observations with 7 variables
(Table 1).

2.2. Analysis

Apart from the presentation of descriptive statistics in a policy relevant manner, statistical tools were employed to test for
differences of means between groups. Two samples t-test was used to assess differences between textile and non-textile factors, and
also for fines among repeat offenders. A one-way analysis of means was conducted to determine if there are any differences among

2
N. Haque Water Resources and Industry 18 (2017) 1–8

Fig. 1. Quarterly pollution fines (2011–2016).

Table 1
Dataset characteristics (n=290).

Variable Type Comments

Date Ordinal Required for time-series analysis

Name of Factory Nominal Identifying unit
Location of Factory Nominal Useful for locating geographical clusters
District of Factory Nominal Useful for district level planning
Type of factory Nominal – to be converted to Given the dominance of textile factories in the list, it would be easier to analyze “textile” vs. “non-
Binary textile” observations
Violation Type Categories Key variable for grouping observation according to types of violations
Fined Amount (BDT) Interval/Ratio The penalty charged expressed in Bangladeshi taka (US $1 = 80 taka)

Table 2
Properties of variable “fined amount”.

Minimum fine 10000 taka

Maximum fine 30000000 taka
Median 800000 taka
Mean 1965000 taka
Mode 200000 taka
Standard deviation 3531116

violation groups, following data transformation.

3. Overall picture of water pollution fines in Bangladesh

Before segregating the factories into groups of violations, the fined amount as a whole is analyzed to assess some basic
characteristics. Table 2 presents descriptive statistics of fines for all 290 records.
As seen in Table 2, the maximum fine is 3000 times the minimum fine and the standard deviation is very high when expressed in
local currency of Taka ($1 = 80 taka). This finding substantiates the arbitrariness of the fines. However, presenting the fines for all
290 records in a histogram shows that most of the fines are under 2500000 which is closer to the mean and mode values. The
histogram shows that the pollution fine records may not be normally distributed. Further analysis was conducted to establish the
normality of the records.
The correlation coefficient of the QQ plot in Fig. 2 was also found to be 0.7134, which falls below all critical values for QQ plot
correlation coefficient test for normality (Fig. 3).
Fig. 4 shows the district-wise distribution of factories that were present in the sample. This distribution should not be compared
with other spatial distribution of manufacturing industries in Bangladesh, as the sampling for this distribution is solely composed of
those factories who have been fined for water pollution. It will also differ from DoE's list of permits issued, since lack of clearance/
permit is also considered a violation in this study. Nonetheless, this distribution can help regulators to prioritize resources among hot-
spot districts.

3.1. Industrial sectors covered in the dataset

Out of the 290 observations or factories, 35 factories belong to other sectors and remaining belongs to the apparel or textile sector.

3
N. Haque Water Resources and Industry 18 (2017) 1–8

Fig. 2. Histogram of Fined Amount.

Fig. 3. Q–Q plot of Fined amount showing that data is not normal.

Fig. 4. District-wise distribution of factories from sample (n=290).

Table 3
Sectors covered within sample of polluting factories.

Sector Number in sample

Metal (aluminum & steel) 3

Battery 2
Utility 1
Board mill 2
Tannery 1
Food Processing 7
Pharmaceutical 9
Paper 10
Textile 255

Table 3 outlines the sectors covered within the sample. The dominance of the textile sector is not surprising as it is the main pillar of
the Bangladesh economy. In the news items, further disaggregation was made in terms of identifying the units as dyeing, washing or
spinning mill, but for this analysis all of them were lumped as the textile sector for ease of analysis. Since other sectors made up only
35 records in the sample, they were collectively grouped as non-textile.

3.2. Difference between textile and non-textile factories

Since textile sector is more dominant and export oriented, one would expect that pollution fines will be heavier for the textile
sector compared to the non-textile sector. A Welch's two-sample t-test was conducted for the two groups with the null hypothesis
being that there was no difference of means between textile and non-textile factories. Considering the high values of all fines, they
were transformed by taking log of environmental fine. The boxplot in Fig. 5 shows the differences between the two groups.
The Welch's two-sample t-test is an appropriate test since it takes into account heterogeneity of variances which is likely since the
sample sizes of non-textile and textiles are unequal. Results show that textile factories (m = 13.6, n = 255) and non-textile factories

4
N. Haque Water Resources and Industry 18 (2017) 1–8

Fig. 5. Boxplot of environmental fines between groups.

Table 4
The observations are grouped according to the following violations.

Group Violation Frequency

1 No clearance, No ETP 89
2 Faulty ETP 73
3 No ETP 42
4 ETP turned off 19
5 Bypass Drain 13
6 Bypass Drain and Ineffective ETP 11
7 Fines Based on Discharge Quality Report 11
8 Clearance not renewed; faulty ETP 8
9 Clearance not renewed; No ETP 5
10 ETP not operating 4
11 ETP turned off and bypass drain used 4
12 No Clearance 4

*ETP refers to Effluent Treatment Plant or Industrial Wastewater Treatment Plant

Table 5
Descriptive statistics of each violation group.

Group Frequency Mean Fine (BDT) Median Fine (BDT) Standard Deviation

1 89 1605029 500000 2830976

2 73 1364603 700000 3230243
3 42 2329107 656000 5070339
4 19 1826316 800000 2739798
5 13 3545385 2900000 3868347
6 11 5190909 2000000 6294832
7 11 2286818 1075000 2867911
8 8 2000000 1150000 2541091
9 5 1000000 600000 1133578
10 4 3274500 2800000 2381638
11 4 3100000 2750000 2469818
12 4 1450000 1300000 574456.3

Fines are expressed in nominal values.

(m = 13.2, n = 35) did not differ in terms of environmental fines according to Welch's t-test, t(40.96) = 1.47, p = 0.14.
The 95% confidence interval for difference of the mean log of environmental fine is between −0.15 and 0.98. Since the difference
between means crosses zero, it can be deduced that there is no difference at all, and we fail to reject the null hypothesis that the two
groups are equal. Textile factories are not fined heavily compared to non-textile factories.

3.3. Types of violations

The analysis evolved during the phase of data collection as groups were identified for the variable “Reason”. Newspaper reports
differed in how these violations were reported, but in general some common groups were identified. While many of factories were
charged for two violations, such cases were categorized as a different group making all the groups independent and mutually
exclusive. Table 4 shows the groups identified according to the violations, and Table 5 shows the descriptive statistics of the groups
(identified and labelled from Table 4).

3.4. Data transformation

Even though it appears that mean fines of the groups differ, statistical tools were used to ensure that these groups were actually

5
N. Haque Water Resources and Industry 18 (2017) 1–8

Fig. 6. Histogram of transformed fines (transformed using natural log).

Fig. 7. Q–Q plot of transformed environmental fines.

Fig. 8. Box-plot of fines for repeat offenders.

different. However, use of such tools will be difficult considering these fines are not normally distributed. Therefore, the fines were
transformed for further analysis by taking their natural log. Use of natural log for analysis was emulated from the empirical study
conducted by Oljaca et al. [15]. Fig. 6 shows the histogram of fines when they have been transformed for both frequency and density.
Fig. 7 shows the QQ plot showing normality of the transformed fines.
The correlation coefficient of the transformed QQ plot was found to be 0.9959, which is greater than critical value of 0.9953 for
QQ plot correlation coefficient test for normality at 0.05 significance level.

3.5. Differences among violation groups

Although the transformed fines are now normally distributed, the unequal variance among groups violated the assumptions for
analysis of variance (ANOVA) that could have established difference of means. Instead a one-way analysis of means was conducted
which assumes unequal variances. Results show that there is a significant difference of fines at the p < 0.05 level among twelve
groups [F(11, 29.9) = 3.91, p = 0.001447]. Therefore, we reject the null hypothesis that the fines for groups of violations are equal.
Fines vary among the violations even in the absence of a policy or guidelines for regulators. However, given the variances among
groups, it will be unrealistic to put the violations in any type of order without maintaining equal sizes.

3.6. Repeat offenders

Out of the 290 list of offenders, one factory has been fined thrice over the five years. Additional 25 factories were fined at least
twice over the years. The fines for the 26 factories were compared between the two instances they were fined.
Although Fig. 8 uses log of fine to compare two instances, it is clear that 2nd time fines were lower than first time even though the
means look the same. Therefore, for the t-test, actual fine value, instead of log of fine, was used to compare the means of the two
groups. The Welch t-test results show that fines for the 1st instance (m = 3417769 Tk., n = 26) and 2nd instance (m = 2426854, n =
26) did not differ in terms of magnitude, t(37.75) = 0.59, p = 0.55. The 95% confidence interval for difference of the mean of
environmental fine is between −23,66,571 Tk. to +43,48,401 Tk. Since the difference between means crosses zero, it can be

6
N. Haque Water Resources and Industry 18 (2017) 1–8

deduced that there is no difference at all, and we fail to reject the null hypothesis that the two groups are equal. Fines don’t differ
between first instance and second instance, and this finding has huge policy implication because it implies that enforcement, in terms
of levying environmental fines, will not be enough to make these factories comply and not pollute. This result confirms earlier
research findings on the limitations of uniform discharge reduction system along with the high total abatement cost the system entails
[17]. Increased fines for repeat offenders is a common-sense enforcement strategy backed up by theory and is empirically found to be
a significant factor in determining size of penalty [15]. Targeting enforcement on repeat offenders will not only help reduce total
enforcement costs, but it also gives violators stronger incentives to comply, especially for the violation of not having Effluent
Treatment Plant (ETP), which is a technology adoption decision by firms [4].

4. Discussion

This empirical study is the first of its kind focusing on evaluating enforcement in Bangladesh. However, news media have
highlighted the shortcomings of the existing enforcement regime. Given the lack of data availability, this methodological innovation
of using media reports of pollution fines was used as an alternative pathway to answer some vital policy evaluation questions. It needs
to be mentioned that the fines mentioned in each of the news item does not indicate that those fines were paid. There have been
instances where the Ministry of Environment and Forests (MoEF) intervened and reduced or cancelled the fine, sometimes for
political reasons [12]. There is an appeal process for fines levied by the DoE, and it is headed by the additional secretary of the MoEF.
Many of the fines have been forgiven, reduced and even returned following the appeals process regardless of the violation, further
encouraging the culture of impunity [9]. According to Mahmud [12], during the years 2010–2013, 1371 industrial units were fined
for environmental violations (water, air pollution combined) a total of 1,278,365,000 Taka. Out of this amount, MoEF collected only
880,993,000 Taka which is 69% of levied fine. Another newspaper article reports that during 2010–2014, 1788 industrial units were
fined a total of 1,516,100,000 Taka, out of which MoEF collected 1,027,900,000 Taka which is 68% of the levied fine [9]. Khan [10]
also reported a collection rate of 66.7%. This analysis is based on the initial levied fines and evaluates them as a compliance tool, even
though political and bureaucratic forces sometimes hinder the effectiveness of this tool [6]. Nevertheless, such an analysis is overdue
for both polluting factories and the regulatory agency. Regulatory design should be about consequences focusing on what works, by
how much, and how much costs and side-effects there will be [18].
For polluting firms, this analysis is a step towards assessing cost of non-compliance. It gives a good indication of the fine amount
based on historical figures, in case they are ever fined for non-compliance or other glaring violations. For policymakers, this analysis
could be a baseline study for reforming existing practices and establish responsive regulations, where DoE selects enforcement
strategies reacting to the actions chosen by the polluting firms [1]. Such evaluation will enable the regulator to design better
enforcement strategies to affect both general and specific deterrence [6]. The arbitrary nature of the fines must be replaced with
progressive fines, the magnitude of which must vary with the ranked level of violation. These fines also need to be applied uniformly
across industrial sectors, and size of factories. The violations listed in Table 4 should be taken seriously in determining enforcement
strategy. It substantiates earlier claims made by environmentalists, that DoE runs enforcement on the basis of checklists only [2].
Taking samples of wastewater, verifying the effluent quality and levying fines based on those report makes up only 0.04% of all
fines. This is a severe problem since many of the existing effluent treatment plants could be under-performing by not meeting effluent
standards, and there seems to be no monitoring and enforcement for such hidden violations. DoE is fully aware of the non-operation
of installed ETPs and the resulting illegal discharge of pollutants into water bodies [14]. Together with the lack of manpower, the
existing enforcement structure is just not equipped to detect midnight dumping and other anomalies. Moreover, the DoE does not
maintain a proper database for recording inspections, and as such it cannot be verified whether any records are updated following
penalties [14].
As mentioned earlier, credible research in the field of industrial environmental economics in Bangladesh is very limited. However,
investigative journalism has exposed some of the policy relevant analysis. It has been estimated that the average cost of wastewater
treatment for textile dyeing factory is around 32 Taka per cubic meter, which can translate into an annual cost of Tk. 1,33,50,000 for
a small factory. Even if the factory is fined twice a year an average of Tk. 10,00,000, it would be more economical to pay the fine
rather than running a wastewater treatment plant [16]. Other estimates of unit cost of effluent treatment ranges from 14.43 to 25.66
Taka per cubic meter [7]. The annual operational cost of chemical ETP is estimated to be 87.86 million Taka [13], which is
significantly higher than the average fine of 19.6 million Taka reported in Table 2. The research community needs to build up more
on the empirical estimates of operational costs in order to develop trade-offs and firm-level decision choice models. This process will
facilitate restructuring of the penalty rates based on the abatement costs so that policy changes can ensure widespread compliance
through effective enforcement.

4.1. Further scope of research

Although the dataset was limited in having only one continuous variable, it can be expanded using identifier variable of factory
name and finding the annual turnover rate of the factory in order to test whether firm size have any influence on the size of the fine
[15]. If the violations can be ranked objectively, it will also be possible to run a regression analysis to see how much the fine will
increase with an increase in ordinal violation group. In addition, more statistical tools need to be employed to find out how much
fines are charged for repeat offenders for repeating the same offence. Current sample size of repeat offenders (n=26) is too small for
making these predictions. If this dataset can be used to find expected value of pollution fine, which is the product of probability of
inspection rates and mean/median nominal penalty, it can be compared with abatement costs to assess options for polluters which

7
N. Haque Water Resources and Industry 18 (2017) 1–8

will be to comply or violate. Finding an optimal expected fine will also be useful since theoretically full compliance can be achieved
by setting this expected fine at an arbitrarily high level by manipulating either inspection rates or nominal penalties or both [8].

5. Conclusion

Policy evaluation in general is an unheard practice for environmental governance in Bangladesh. However, it is critical that
enforcement mechanisms are evaluated so that elements that are successful are scaled up, and corrective actions are taken for
processes that don’t work. This paper provides the first estimate of the fines levied for water pollution and how these fines are
distributed across violations and sectors using information published in news media. The purpose of this paper in offering these
empirical estimates is to enable evaluation of whether enforcement is affecting any changes on polluters behavior to comply with
existing regulations and standards. Using news reports of environmental fines of last five years, some patterns are identified
qualitatively to illustrate the situation of environmental enforcement in Bangladesh. The sampling methodology is the key limitation
of the study since it was based mostly on press records of environmental fines for the last five years. The arbitrary nature of
environmental fine is evident from the analysis, and it should serve as a basis for developing progressive penalty structure. It was also
observed that repeat offenders are not penalized heavily, thereby giving wrong incentives to polluters. DoE should review
mechanisms and penalties for non-compliance to pollution regulations, and create more stringent sanctions and more effective
mechanisms for penalizing noncompliant parties. This analysis is the first step towards reforming enforcement rules and policies, and
it is essential that such changes in future are data-driven.

Acknowledgement

The author is grateful to Dr. Thomas Woodson from the Department of Technology & Society of Stony Brook University for his
encouragement and thoughtful review. The author wishes to thank all the colleagues working in Promotion of Social & Environmental
Standards (PSES) project, implemented by GIZ in Bangladesh, for their continuous encouragement.

References

[1] C. Arguedas, Pollution standards, technology investment and fines for non-compliance, J. Regul. Econ. 44 (2) (2013) 156–176, http://dx.doi.org/10.1007/
s11149-013-9217-8.
[2] E. Azad, Environmental pollution in Bangladesh: need for a legal overhaul, Bangladesh Law Assoc. J. 1 (1) (2014) 81–100.
[3] Bangladesh Center for Advanced Studies (BCAS), Guide to the Environmental Conservation Act 1995 and Rules 1997. BCAS: Dhaka, 1999.
[4] J. Coria, X.B. Zhang, State-dependent enforcement to foster the adoption of new technologies, Environ. Resour. Econ. 62 (2) (2015) 359–381, http://dx.doi.org/
10.1007/s10640-015-9943-8.
[5] S. Dasgupta, B. Laplante, N. Mamingi, H. Wang, Inspections, pollution prices, and environmental performance: evidence from China, Ecol. Econ. 36 (3) (2001)
487–498, http://dx.doi.org/10.1016/S0921-8009(00)00249-4.
[6] W.B. Gray, J.P. Shimshack, The effectiveness of environmental monitoring and enforcement: a review of the empirical evidence, Rev. Environ. Econ. Policy, Req.
017 (2011) 017, http://dx.doi.org/10.1093/reep/req.
[7] E. Haque, M. Sumsuddha, T. Ashrafee, Project report on study on Chemicals Consumption and Operational Cost of Textile ETP Plants. Retrieved from: 〈http://
dspace.library.daffodilvarsity.edu.bd:8080/handle/20.500.11948/1371〉 (accessed 23 February 2017), 2014.
[8] A. Heyes, Implementing environmental regulation: enforcement and compliance, J. Regul. Econ. 17 (2) (2000) 107–129, http://dx.doi.org/10.1023/
A:1008157410380.
[9] A. Hossain, 80 lakh taka fine forgiven with one appeal. Daily Samakal. Retrieved from: 〈http://archive.samakal.net/2014/05/21/60832〉, 21 May, 2014.
[10] M.R. Khan, Environmental pollution: fine collection reduced to half. Daily Janakantha, 1, September 23, 2014.
[11] M.S. Islam, Implementation of policy to control industrial pollution in Bangladesh: major drawbacks (In Public Administration and Policy in the Middle East),
Springer, New York, 2015, pp. 297–317.
[12] I. Mahmud, Environmental fines are publicity stunts Prothom Alo, 1, 14 February 2014.
[13] M.S. Miah, Cost-effective treatment technology on textile industrial wastewater in Bangladesh, J. Chem. Eng. 27 (1) (2013) 32–36.
[14] K.G. Moazzem, M.I. Chowdhury, Green growth in Bangladesh–Improving regulations, monitoring and enforcement in the textile sector. (Policy Brief Number 96-
16). South Asian Network for Development and Environmental Economics (SANDEE). Retrieved from: 〈http://www.sandeeonline.org/uploads/documents/
publication/1080_PUB_Policy_Brief_96_Mehruna.pdf〉 (accessed 22 May 2016), 2016.
[15] N. Oljaca, A.G. Keeler, J. Dorfman, Penalty functions for environmental violations: evidence from water quality enforcement, J. Regul. Econ. 14 (3) (1998)
255–264, http://dx.doi.org/10.1023/A:1008031307103.
[16] Prothom Alo, No punishment for river polluters, Daily Prothom Alo, 14, 28 October, 2013.
[17] M.A. Rahman, T. Ancev, Economic analysis of alternative pollution abatement policies: the case of Buriganga River, Bangladesh, Interdiscip. Environ. Rev. 15 (1)
(2014) 66–87, http://dx.doi.org/10.1504/IER.2014.062138.
[18] J.B. Wiener, The regulation of technology, and the technology of regulation, Technol. Soc. 26 (2) (2004) 483–500, http://dx.doi.org/10.1016/j.techsoc.2004.01.
033.
[19] World Bank, The Bangladesh Responsible Sourcing Initiative: A new model for green growth. Washington, DC: World Bank Group. Retrieved from: 〈http://
documents.worldbank.org/curated/en/614901468768707543/The-Bangladesh-Responsible-Sourcing-Initiative-A-new-model-for-green-growth〉 (accessed 1
January 2017), 2014.
[20] WWF and H & M, Water Governance in Bangladesh: Challenges and Opportunities around Policy, Institutional Function and Implementation for a Sustainable
Water Future. Retrieved from: 〈http://d2ouvy59p0dg6k.cloudfront.net/downloads/wwf_hm___water_governance_in_bangladesh_published.pdf〉 (accessed
3 February 2016), 2016.