Reimagining Borrower Protection in Bangladesh: Why UK’s FCA Consumer Duty Framework Should Guide Regulatory Reform of our Micro Credit Sector?

By Samiur Rahman

Once heralded as a beacon of poverty alleviation by targeting the achievement of Sustainable Development Goal (SDG) 1, Bangladesh has made significant strides in the consumer lending sector since 1990 due to rapid growth in Micro Finance Institutions (MFIs), which are estimated more than 1,000 in numbers and among these 724 licensed MFIs are serving 32.18 million borrowers. An early 2005 empirical study on the consumer lending practices of mainstream MFIs, i.e., Grameen Bank, Brac, etc., indicates a reduction in subjective poverty through microcredit lending of 6.5% per programme a year. Likewise, a recent 2025 study revealed that the micro-credit sector has expanded dramatically showing that over the span of 5 years between 2018 and 2022 , Bangladesh obtained 29.74 million customers against 25.40 million during 2018 with an annual loan disbursement of BDT 1.9 trillion. 

These figures demonstrate the growing significance of micro-finance in our economy particularly in agriculture and rural economy. However, the overall regulatory landscape still reveals structural problems. For instance, the lack of effective regulation of interest rates, as most of the MFIs use a flat rate method typically between 10 to 30% , may conceal the true cost of credit and risk, inflating the interest rate for certain lending products, thereby harming the borrowers in the long run. Recently, a Supreme Court ruling also indicated this issue , questioning the exploitative interest rates of the micro credit services offered by Grameen Bank.

Furthermore, the rapid expansion of micro-credit institutions and recently introduced MFS (Mobile Financial Service) micro lending services may weaken the macro prudential regulatory oversight and create systemic risk within the consumer lending sector. However, the 2005 study also revealed that the top three NGOs in Bangladesh held 85% of total cumulative credit, leading to a high concentration of lending capital. The 2005 study revealed a strong picture of vulnerability for the new borrowers who are objectively poor, around 65.3%. Finally, drawing from these, there is an indication of  an anti-client-based lending regime which lacks transparency thereby raising a critical question, is the micro-credit regulation properly optimized to ensure fairness to our consumers? In this context, the regulatory framework of the micro-credit regulation in Bangladesh needs to be unpacked.

The regulatory framework for micro-credit regime in Bangladesh comprises of two organs. First, the Micro Credit Regulatory Authority (MRA) under the remit of the Micro Credit Regulatory Authority Act 2006, which is the primary legislation regulating entities like NGO and NBFIs (Non-Bank Financial Institutions) in Bangladesh. The act itself holds prudential regulation of the institution’s ability to lend rather than embracing any consumer duties and fairness obligations for the regulated entities working under its remit. Especially, section 9 of the act specifies supervisory duties of MRA to ensure proper regulation of the MFIs, but no consumer-friendly guidelines exist for ensuring fair treatment, fair value, transparency, and vulnerable borrower protection against unscrupulous lending practices of the MFIs. Moreover, other provisions of this legislation  focus on maintaining the prudential stability of the regulated institutions, with no provision  for borrower protection, which is alarming. The enforcement provisions of the Act similarly fall short, resembling sections 35 to 37 of Chapter 6 of the act. These only apply to licensing and certification-related issues not irresponsible lending or consumer protections. Therefore, the primary legislation governing the micro-credit sector only concerns about the governance of the MFIs (lenders) rather than promoting borrower protection.

 To remedy this, the MRA rules 2010 make matters to some extent favorable for the marginalized borrower. Circular 5 of the rules states interest rate caps of up-to 27%, which is still enormous for the most vulnerable borrowers. Additionally, it also ensures 15 days grace period before the first installment which is an attempt to ease the repayment-related burden of the borrowers. However, the compliance of such regulation is questionable because MRA lacks human resource capacity to supervise an enormous number of MFIs that exist in our country, thereby giving enough chance to the unscrupulous lenders to exploit the loopholes of the current regulatory regime and get away with unfair lending practices, as this regime fails to regulate borrowers’ understanding, affordability test, and MFIs attempt to justify the value of their debt financing. Finally, the rules of deposit mobilization enumerated in the MRA rules only justify liquidity, capital reserve, etc., which are prudential safeguards but overlook conduct regulation relating to responsible consumer lending. However, Bangladesh Bank (BB), also published some guidelines in the form of soft laws, which are non-binding instruments with zero accountability for non-compliance of the MFIs actively engaged in consumer lending. In this regard, Bangladesh Bank (BB) Guidelines for Customer Services and Complaint Management (2014), could be scrutinized as Bangladesh Bank is responsible  for macro prudential oversight of our financial sector.

The guideline as a soft law, lacks enforcement as it is a service and quality guideline rather than a conduct-of-business-centric rule book. Therefore, even if the MFIs breach this, there is no substantial disciplinary outcome as guidelines lack legislative enforcement power. Moreover, although the guidelines focus on administrative and operational resilience of the MFIs, it is does not cover borrower protection. The consumer protection regime of Bangladesh, i.e. the Consumer Rights Protection Act 2009, also paints a grim picture as the act itself regulates general consumer-related products rather than regulated lending activities like micro-credit services. Although the act recognizes financial services, the definition in sec.2(22) is too narrow to include regulated financial activities and does not contain any provisions concerning borrower protections. Therefore, drawing from the above concludes that the borrower protection regime is underdeveloped therefore facilitating a regime where the MFIs take advantage of uninformed borrowers .  Although the borrower can file a writ petition under Article 102 of our constitution as a judicial recourse for unfair lending practices, this pathway itself is unreliable because of cost, litigation backlog in the SC, and delay in the pronouncement of the judgments. Moreover, Artha Rin Adalat sec. 6(2)  (a special court in Bangladesh specializing in fast loan recovery from financial institutions pursuant to section 2),  also restricts individual access to financial redress as the forum only allows regulated corporations to file litigation. 

So, regulatory reform of the borrower protection regime is  inevitable in Bangladesh and should follow reform initiatives in light of the United Kingdom’s recent overhaul of the consumer lending framework. In the face of growing criticism of Pay-Day and Wonga lending scandal, which took the high cost short-term credit (micro-credit) interest rate to APR 4214% and 31.37% a month, led to the creation of FCA’s (Financial Conduct Authority consumer lending guideline  with stringent enforcement outcome,  safeguarding the sector-wide borrowers. Although the financial regulation in the UK is principle-based, the country has moved away from box-ticking and light-touch regulation to a more law- and society-based regulation . The new FCA consumer duty framework is revolutionary as it consists of three pillars. First ,  lending entities must follow strict retail and consumer-focused principles, i.e. the main agenda behind lending would be borrowers. Secondly, cross-cutting rules enabling the consumers to pursue their financial objectives, act in good faith, and avoid causing foreseeable harm to the retail customers. Finally, the financial services should promote four outcomes aligned with the previous objectives.

The four outcomes are: first, products and services that meet customers’ needs. Second, price and value that is fair for what customers get. Third, consumer understanding, meaning clear information that helps customers decide and fourth, consumer support when customers need assistance. Moreover, if the borrower is entering a regulated credit agreement, must coincide with the FCA Consumer Credit Sourcebook (CONC) 5A, with default charges not exceeding 15 pounds, an interest rate cap of 0.8% per day, and the borrower must never pay more in interest than the borrowed amount. Clifford Chance (2023) reported that this consumer duty regime in the UK is ensuring responsible lending and allowing board-level accountability. Moreover, in case of consumer complaints,  the legal aid through the Financial Ombudsman Service (FOS) allows the complainant quasi-judicial recourse to their grievances, provided FCA can also impose sanctions on non-compliant firms for breach of the consumer duty framework.  

So, why would Bangladesh care to model their borrower protection regime in light of FCA?  The straight answer is that enabling borrower protection would make corporate lending more accountable and would allow the vulnerable borrower, i.e. rural farmers, SME entrepreneurs make more informed choices and exert control over their lending or debt outcomes. To achieveborrower-level fairness, the MRA needs exclusive clauses of borrower protection and fairness outcomes of the MFIs as the MRA is responsible for the conduct and prudential regulation of the NGOs and MFIs. Later, Bangladesh Bank should enable strict consumer lending guidelines like the FCA and impose stricter sanctions for non-compliance. Most notably, the flat-rate interest should be abolished or should be made transparent by the Bangladesh Bank, allowing the impoverished borrowers a sigh of relief when borrowing money for their personal financial endeavors. 

Therefore, I suggest that the legal redress regime should mirror the UK and creation of a separate tribunal for financial consumer protection is inevitable or the government could empower the Ombudsman office quasi judicial power to tackle complaints relating to consumer lending products. Additionally, the legal aid must be made individually accessible to ensure access to legal justice which is a fundamental right of our citizens. In conclusion, the financial regulation of  Bangladesh is certainly rule-based regulation (RBR) and lacks conduct oversight where box-ticking is pretty much evident and a law and society-based approach focusing on consumer well-being seems like a misnomer. Therefore, a regulatory divergence in light of the twin-peak structure of the UK, separating conduct and prudential regulation, enabling outcome-based lending practices, would allow the micro-credit sector of BD to become more compliant, responsible and consumer-focused which will allow  borrowers to remain at the center of our financial regulation and allow them to pursue their financial objective as a result giving them a proper chance to alleviate from the vicious cycle of poverty.

Environmental Constitutionalism and EU AI Regulation (Part II) 

Nífémi Adéșínà – PhD Student (Digital Constitutionalism, Newcastle University) – o.adesina2@ncl.ac.uk

Introduction

Part I of this series explored what environmental constitutionalism is and the missed opportunity by the EU AI Act’s draftsmen to adequately enshrine environmental protections into its provisions. The final part of this series will seek a path forward through environmental constitutionalist principles and the forthcoming technical standards of the AI Act.

Technical Standards: A Path to Strengthen Environmental Constitutionalism? 

Article 2 of the Standardisation Regulation defines ‘standard’ as a technical specification, adopted by a recognised standardisation body, for repeated or continuous application, with which compliance is not compulsory. This includes ‘harmonised standards’ which are European standards adopted on the basis of a request made by the EU Commission for the application of Union harmonisation legislation.

The EU Commission has tasked two European Standardization Organizations (ESOs), i.e. European Committee for Standardization (CEN) and the European Committee for Electrotechnical Standardization (CENELEC), forming the Joint Technical Committee (JTC21), to develop these standards. These standards will allow for the Act to simply specify that certain safety outcomes must be achieved by AI models or system providers; however, the Act itself does not go into technical detail about the methods of measuring compliance with these safety outcomes. This is where the technical experts come in. These experts are expected to fill in the blanks and update these methods of measurement to keep up with the evolving state of AI capabilities.

Although the standards being developed to support the EU AI Act are not legally binding in themselves, they provide a means to demonstrate compliance with the Act’s requirements, and adhering to them offers a presumption of conformity, as stated in Article 32 of the Act. This can be used as a tool of environmental constitutionalism by establishing a normative compliance mechanism which can shape the behaviour of corporations in a way that incentivises the incorporation of environmental considerations into technical practices. One of the primary ways these standards bolster environmental constitutionalism is by fostering regulatory cohesion, clarity and accountability. When harmonised standards set clear benchmarks for AI-related environmental impact assessments, risk mitigation measures, and sustainability criteria, they create predictable guidelines for developers and users of AI systems. This clarity enhances compliance and accountability, ensuring that AI operations are in line with existing constitutional environmental protections. Such standards essentially embed environmental considerations into the core of AI governance, making it easier to monitor and evaluate.

These standards can also serve as a pioneering means for cross-border cooperation beyond the EU. The “Brussels effect” has previously been demonstrated through EU constitutionalism and its related climate legislation and has been seen to incentivise governments and non-governmental organisations outside of the Union to increase their environmental protection standards. For example, the EU introduced the Carbon Border Adjustment Mechanism (CBAM), which was officially adopted in May 2023. Companies that imported products in sectors with high emissions were mandated to report the amount of carbon used in making those products and the carbon prices they’ve paid in their home countries. From 2027, these companies will need to buy certificates that cover the same amount of carbon emissions as the EU’s carbon prices, effectively putting a cost on imported carbon-intensive goods. Importers can get some relief by showing that they have paid higher carbon prices in their home countries, which can reduce the certificates they need to buy. This serves as an incentive for other countries to adopt greener policies and help companies reduce their emissions without the EU losing competitiveness. As foreign AI systems and products are launched in the EU, the standards can become invaluable in terms of the reduction of ambiguities, but also in maintaining consistent environmental safeguards and promoting consistency across jurisdictions. Although constitutionalism is often defined within a state model and therefore tied to a specific jurisdiction, environmental challenges transcend borders. Thus, for environmental constitutionalism to supersede mere legislative greening, it must be applied not only within defined national borders, but beyond.

Despite these promising prospects, there are a few challenges that influence the extent to which technical standards can be effectively employed to reinforce environmental constitutionalism. The voluntary nature of the standards may ostensibly be seen as a shortfall in the quest for environmental constitutionalism within AI regulation. Nonetheless, this is not necessarily the case because compliance with these standards holds an irrefutable incentive for AI companies that would be “impractical to ignore” in that a presumption of legal compliance is conferred along with voluntary compliance. 

Additionally, ensuring that standards are flexible enough to accommodate evolving AI technologies while robust enough to enforce environmental protections would constantly remain a balancing act for the standardisation committee(s). Lastly, the standard-setting process has been critiqued for lacking adequate scrutiny and accountability. The closed-off nature of the standard-setting process has been noted to potentially serve the commercial interests of industry players, therefore weakening the protection for affected stakeholders. There is, however, a means to overcome this obstacle in what has been termed “the twin-track approach”, which calls for the European Commission’s active participation and oversight in the process, therefore leading to socio-technical rather than just technical standards that address both the perspectives of AI experts and other stakeholders.

Substantive Elements of the Technical Standards

The task of defining technical standards for AI demands a critical approach to environmental considerations. This approach must transcend mere superficial greening and requires a deliberate shift towards the weaving of environmental protections into the very fabric of AI governance structures.

Firstly, JTC21 must embrace the precautionary principle in drafting these standards. Within the environmental constitutional discourse, the precautionary principle mandates that new technologies ought to be rejected until deemed safe; it is a presumption of avoidance when presented with uncertain risks. The precautionary principle, or the foresight principle, acts to prevent environmental harm even when all the scientific evidence is not yet fully established. Article 191of the Treaty on the Function of the EU also stipulates that the Union policy on the environment shall aim at a high level of protection taking into account the diversity of situations in the various regions of the Union. It shall be based on the precautionary principle and on the principles that preventive action should be taken, that environmental damage should as a priority be rectified at source and that the polluter should pay. 

JTC21 must look beyond the present impacts of AI and prioritise potential ecological harms. This would require developers and deployers to preemptively consider the full life cycle of AI systems, from the extraction of raw materials and manufacturing processes to their eventual use and disposal. The integration of circular economics will also be beneficial here, i.e. minimising waste and the carbon footprint of AI computing, maximising resource efficiency, and promoting intrinsic recyclability and reusability principles within the AI lifecycle.

These technical standards must also provide for environmental impact assessments of AI systems, incorporating comprehensive metrics that encompass energy consumption, carbon emissions, water usage, resource depletion, and electronic/electric waste. These assessments could ensure consistent evaluation across sectors and would provide a clear means to adequately scrutinise AI developers’ assessment results to promote accountability. 

The Future of Environmental Constitutionalism

As global discourse converges on establishing robust AI regulatory frameworks, the EU’s approach, exemplified by the AI Act and its supporting technical standards, offers a crucial opportunity to move beyond conventional technology governance. While the EU’s current framework reflects some environmental values within its constitutional tradition, there isn’t much evidence to suggest that the EU as a whole operates as an environmental constitutional model. However, the EU could serve as a compelling case for adopting such a paradigm, which would deepen the integration of ecological considerations into AI regulation.

It is important to note that the Act, although designed to facilitate compliance and accountability, does not embody a genuine commitment to ecological well-being, and it appears to have missed an important opportunity to embed environmental concerns at its core. For true environmental constitutionalism to be realised within EU AI regulation, there must be a normative shift that no longer places the environmental consequences of AI at the periphery but rightfully entrenches such considerations into the very foundations of the legislation. 

Environmental constitutionalism is arguably the most suitable framework for addressing the environmental risks and harms associated with AI development and deployment, particularly in the context of the EU, because it has the ability to build on and extend the existing discourse on transnational constitutionalism, and if adopted it could offer an adaptive and resilient pathway to guide AI regulation by way of the forthcoming technical standards.

It is too soon to say how the standards will fare, as JTC21 is currently in the process of drafting them and the AI Act itself is still navigating the turbulent waters of balancing innovation and regulation. But this opportunity should not be overlooked. The focus should be on navigating the ways in which these standards can strengthen environmental safeguards in AI development through the core tenets of environmental constitutionalism.

Environmental Constitutionalism and EU AI Regulation (Part I)

 

Nífémi Adéșínà – PhD Student (Digital Constitutionalism, Newcastle University) – o.adesina2@ncl.ac.uk

Introduction

There has been a wave of discourse over what a regulatory framework for Artificial Intelligence (AI) would look like across various jurisdictions around the globe. Stakeholders have continued to centre legislative AI debates on the balance of innovation with regulatory protection. Conversely, within the realm of constitutionalism, there has been momentum and a continuous push for the development of the environmental/climate component of the discourse. On the surface, these two domains may appear to be unrelated, the latter being grounded in ecological rights-based advocacy within constitutional law, and the former in technological governance and legislation. However, a space for potential overlap has emerged through the introduction of the European Union (EU) AI Act (‘the Act’) and the forthcoming technical standards. The EU’s regulatory framework on AI has the ability not only to serve as a frontier for similar governance structures but as a platform for advancing environmental constitutionalism. 

Environmental Constitutionalism

Environmental constitutionalism is a multi-perspective legal and philosophical movement aimed at integrating and embedding environmental protections within constitutional frameworks. It envisions environmental protection as both substantive and procedural, but its understanding is often wrongly limited to fundamental rights and judicial enforcement through the right to a healthy environment. Environmental constitutionalism, which ought to extend beyond mere codification and environmental rights adjudication, encompasses a broader commitment to ecological sustainability rooted within foundational legal and political structures and involves a shift in governance paradigms. It seeks to foster a societal ethos that prioritises ecological well-being, driving behavioural changes and collective action. Several countries have explicitly enshrined climate clauses into their constitutional texts. According to the United Nations (UN) Environment Programme, over 150 countries have incorporated environmental provisions within their constitutions, and the UN Human Rights Council has also recognised the ‘Right to a healthy environment’ in Resolution 48/13. But despite its transformative potential and emerging acknowledgement, environmental constitutionalism is sometimes viewed as symbolic or aspirational, and its strengths are often overshadowed by national or international political and economic interests, specifically the prioritisation of industries that guarantee shorter-term gains over environmental protections, such as the tech industry.

Furthermore, a restrictive evaluative position on the constitutionality of environmental preservation and justice to an enforcement/rights-based model, and an exclusively or primarily ‘rights and courts’-centric approach may “suffer from a legitimacy deficit” due to lacklustre enforceability mechanisms. Courts have been hesitant to engage for fear of justiciability and technical complexity, and this, in turn, limits the impact of environmental constitutionalism on an adjudicatory level. It is, however, important to note that this understanding is flawed and is ill-equipped to grasp the true significance and mechanisms of environmental constitutionalism. This narrow interpretation of what constitutionalism is, places a high degree of reliance on the judiciary while missing out on the potential impact of regulations and lawmaking on constitutional theory. Afterall, constitutions are not the only scenario in which environmental constitutionalism can thrive.

The Relevance of Non-constitutional Legislative Instruments 

Constitutionalism need not be confined to constitutional documents and can, in fact, encompass other instruments within the broader legal and political landscape. As much as environmental constitutionalism involves utilising the law to protect the environment by recognising environmental rights, duties, and principles, it also prioritises the influence of constitutionalist norms on the interpretation of other laws and policies. Instruments such as the Paris Agreements and the UN Framework Convention on Climate Change are not constitutional in name, nor do they outrightly create constitutional provisions. They, however, epitomise environmental constitutional values that seek to foster a national or global culture of responsibility and instill a duty to protect and preserve the environment. 

In the age of technology, constitutionalism has become an integral means of safeguarding fundamental rights and limiting political and private powers in a rapidly evolving digital landscape. It has been applied to non-constitutional techno-centric regulations such as the General Data Protection Act (GDPR), Digital Services Act, the Digital Markets Act, and even the EU AI Act. However, this has centred on digital constitutionalism, and the environment has yet to make an appearance in the discourse.

AI Production, EU Constitutionalism and Environmentalism 

The production and deployment of AI tools and systems have a significant impact on the environment. From mining rare earth minerals used in manufacturing hardware to increased demands on energy consumption from the data centres that are used to train AI models, the environmental costs are quite steep. Thus, it would follow that such considerations ought to be reflected in AI regulation. The EU’s AI Act aims to ensure that AI systems are safe and ethical. With the objective of promoting transparency, accountability and human-oversight within AI innovation in Europe, the Act places great emphasis on the classification of systems in accordance with their risk thresholds, i.e. unacceptable, high, limited, minimal or no risk. However, the Act and its ancillary recitals make negligible efforts to adequately bolster its provisions with environmental risk obligations.

Within EU constitutionalism, legal principles and practices have been established to ensure that democratic values, governmental legitimacy, fundamental rights and the rule of law are preserved, through constitutional documents like the Treaty on European Union (TEU), Treaty on the Functioning of the European Union (TFEU), and the Charter of Fundamental Rights of the European Union. These treaties outline the EU’s core principles, institutional structure, and decision-making processes, similar to a national constitution, and codify fundamental rights and freedoms while reinforcing their constitutional status through enforceability. Amongst these core principles and fundamental rights is the protection of the environment and the promotion of sustainable practices. Within the EU, environmental considerations are not just a choice, but rather a legal, political and even moral obligation that holds up the EU’s constitutional framework and reflects a commitment to environmental sustainability.

Article 3(3) of the consolidated TEU states that the Union shall work for the sustainable development of Europe based on balanced economic growth…and a high level of protection and improvement of the quality of the environment. Article 11of the TFEU also states that Environmental protection requirements must be integrated into the definition and implementation of the Union’s policies and activities, in particular with a view to promoting sustainable development. Furthermore, Article 37 of the EU Charter of Fundamental Rights also requires this consideration to be integrated into EU policies in accordance with the principle of sustainable development. However, when it comes to ensuring environmental justice as a fundamental aspect of the AI Act, the provisions of the Act appear to be lacking.

Environmental Constitutionalism or Legislative Greening in the AI Act

Environmental protection/preservation is duly mentioned in Article 1(1) and Recitals 1,2,4,8,27,46,48,121,130,142,165, and 176. The Act also attempts to address environmental risks and opportunities through the use of tools like fundamental rights impact assessments (FRIA) and codes of conduct. These mechanisms, however, are less prominent aspects of the legislation and the nature of the codes of conduct and the potential for limited application of the FRIA due to scope constraints are clear shortcomings of their objectives.

The lack of sufficient environmental provisions in the final draft of the AI Act has been considered a missed opportunity. The EU Commission’s Paper on Artificial Intelligence, which predates the Act, lays emphasis on the importance of considering the environmental impact of AI systems throughout their life cycle, including resource usage for algorithmic training and data storage. However, these expectations have significantly been watered down, and much of the environmental considerations in the Act are through inference. 

In a bid to correct this, it is important to avoid taking the mere legislative greening approach that already exists within the Act. For legislation to possess the elements of environmental constitutionalism, it needs to be transformed through environmentally conscious norms; in this case, it would mandate a shift in the way AI regulation is interpreted and enforced. It is not just about environmental rights but as Louis Kotzé posits, provoking consciousness and responsibilityby defining and implementing duties for not just state actors but non-state actors, such as AI developers and deployers. 

The second and final part of this series will explore how the EU AI Act’s technical standards and environmental constitutionalist principles can mutually strengthen one another.