For all latest news, follow the Daily Sun's Google News channel.
Social protection systems are designed to reduce poverty, protect vulnerable groups and stabilise livelihoods during shocks. Yet, in many developing countries, including Bangladesh, these systems often fail to deliver meaningful results. The problem is rarely the absence of programmes, but rather deep structural weaknesses in design, governance and implementation.
Bangladesh’s social protection system is a life-cycle-based approach that comprises multiple types of programmes, including social assistance (cash transfers such as old-age, widow and disability allowances), social insurance (limited contributory schemes mainly for formal workers), and labour market interventions (skills training, employment generation and public works). The nature of the system has historically been relief-orientated and fragmented, but recent reforms have shifted it towards more consolidated, digitalised and cash-based delivery mechanisms, particularly through government-to-person (G2P) transfers. In terms of funding trends, public expenditure on social protection has expanded significantly over time, now accounting for around 14–15% of the national budget (around 2% of GDP). The proposed budget for social protection in Bangladesh for FY-26-27 includes a significant allocation of Tk 144,338 crore, which is a 14.4% increase from the previous year.
Despite its positive impact, the system is not beyond criticism. A core reason for failure is incorrect identification of beneficiaries. Many poor households are excluded due to lack of documentation, awareness, or bureaucratic barriers. At the same time, non-poor households often receive benefits through local influence or weak verification systems. In Bangladesh, studies show that a large share of the poorest households remains uncovered, while significant leakage goes to non-poor groups.
Even when targeting works, the support is often too small to matter. Cash transfers in many developing countries are symbolic rather than transformative, insufficient to cover food inflation, health costs, or education expenses. As a result, social protection reduces short-term distress but does not lift people out of poverty.
Another major factor is political economy distortions. Social protection is often treated as a tool for political visibility, a mechanism for constituency building and a channel for local-level patronage. This leads to bias in beneficiary selection, short-term populist schemes instead of structural reforms and resistance to rationalisation (because duplication creates political rents). In developing states, these incentives often outweigh technical efficiency considerations.
Besides, governments often fall into a “programme proliferation trap” where, instead of reforming weak programmes, governments tend to create new schemes on top of old ones and avoid eliminating politically sensitive programmes. This results in administrative overload and diluted impact per programme. This is not just inefficiency; it is institutional accumulation without consolidation.
Additionally, most failures originate from fragmented and outdated data systems, manifested by no real-time poverty tracking, weak interoperability between ministries and the absence of dynamic household profiling. Without data integration, targeting remains static while poverty is dynamic.
Moreover, in many developing countries, social protection is not a unified system but a collection of overlapping schemes run by multiple ministries. Bangladesh is not an exception in this regard. Instead of a lifecycle-based system (childhood, working age, and elderly), programmes often emerge in an ad hoc manner, driven by political priorities or donor projects.
Furthermore, traditional systems often focus on rural poverty while neglecting rapidly growing urban informal populations. Urban migrants, slum dwellers, and informal workers frequently fall outside eligibility criteria or registration systems.
Based on the scenario, countries need to transition from scattered programmes to a single national social protection architecture focusing on a unified beneficiary registry, integrated databases across ministries and lifecycle-based design (child, working age, elderly, and disability).
Besides, Bangladesh can introduce “dynamic poverty scoring", which can move beyond static poverty lists. Satellite data (housing quality, land use), mobile phone usage patterns (mobility, consumption proxy), school attendance and health data and seasonal income modelling can be used. This produces a living poverty index, updated quarterly or monthly.
Alongside this, modern tools such as national ID-linked databases, digital payments, and data cross-checking with tax and poverty records can significantly reduce errors. However, digitalisation must be inclusive to avoid excluding the poor due to technological barriers. Additionally, support must be large enough to meaningfully affect welfare. Instead of spreading resources thinly across many schemes, governments should consolidate programmes and raise per-beneficiary payments. Moreover, independent audit and monitoring systems, effective grievance redress mechanisms and community oversight can be instrumental. It is necessary to build platforms where citizens can verify beneficiary lists, report exclusion/inclusion errors, and track local fund distribution. This initiative can be combined with anonymous reporting systems and public dashboards at the union/municipality level.
At the same time AI-assisted fraud detection and leakage control can be beneficial. Machine learning can be used to identify duplicate beneficiaries, abnormal payment patterns, ghost recipients and politically biased allocations. But also need to ensure human oversight, transparent algorithms and grievance mechanisms. In addition to that, a mandatory rule should be introduced under which every social protection programme must be reviewed every 3 to 5 years. It should be integrated with employment and human capital policies. Meanwhile, social protection alone cannot eliminate poverty completely. It must be linked with skills development, job creation programmes and health and education services.
A shift towards unified, data-driven, adequately funded, and rights-based social protection integrated with broader development policy is essential if these systems are to move from symbolic relief to real poverty reduction.
____________________________
Writer is a public policy analyst