Poverty & Inequality

DUET: Flexible implementation is key

  • Blog Post Date 07 October, 2020
  • Perspectives
  • Print Page
Author Image

Yamini Aiyar

Centre for Policy Research


Commenting on Drèze’s DUET proposal for an urban work programme, Yamini Aiyar contends that it is a welcome opportunity to reclaim the importance of robust social protection for urban India, while debating the contours of the proposal itself. In her view, this DUET will be best performed when states find their own tunes synchronised with outcomes.

Jean Drèze’s DUET (Decentralised Urban Employment and Training) proposal and the Ideas for India symposium fills a long ignored but critical gap in debates on India’s welfare system – the near exclusion of the urban poor in our current social protection architecture. For all its flaws, India’s welfare system is far more robust when it comes to responding to the needs of chronically poor, rural populations. In urban India, on the other hand, with the exception of the Public Distribution System (available to city residents only) and a smattering of insurance and pensions programmes (which together account for approximately 6% of total central government spending on social protection), government-sponsored social-protection schemes are conspicuously absent. This fact has rarely elicited much policy debate but the urban (largely casual, daily wage) worker – as was evident through the COVID 19 induced lockdown in India – has paid a heavy price for this absence. Yet, the horrific images of millions of workers walking home, and surveys repeatedly highlighting the sharp income drop amongst poor and vulnerable urban workers, have failed to move the policy debate on urban social protection beyond the issue of portability for ration cards, and in an ironic twist, the policy focus has now shifted in the direction of dismantling rather than reforming the formal labour-protection regime. Against this background, Jean Drèze’s DUET is a welcome opportunity to reclaim the importance of robust social protection for urban India whilst debating the contours of the DUET proposal itself. 

Before delving into the details of DUET, I would like to articulate two key principles of what should shape our debates on the design of social protection – universality (not just portability), decentralisation and agility (so that the social protection system reflects the diversity of risk profiles in India). As the World Bank’s 2019 Social Protection Report highlights, 50% of India’s population is vulnerable (one income shock away from poverty), and their social protection needs a framework that empowers them to anticipate and manage risks. But the degree of risk varies in time and location, thus agility and decentralisation rather than a one-size-fits-all approach is a better model for India today. This requires designing a national social protection framework that states can apply to their specific contexts. Now to DUET.

The first question that the proposal raises, and one that several interlocuters in this symposium have engaged with is – why should employment provision be the framework for an urban social-protection programme? Like Debraj Ray, I too am drawn to the idea of a universal ‘right to employment’ as a normative aspiration. But its feasibility is a question that needs debate. In my reading, DUET is far more modest in its ambition. It is not a right to employment or an antidote to chronic unemployment. Rather it performs the function, as Ashok Kotwal suggests, of an unemployment insurance, designed to offer temporary succor to those out of the workforce. It is too small to impact existent patterns of rural-urban migration or significantly influence the urban labour market. The question that arises then is whether this goal is better achieved through ‘employment’-based insurance or a direct cash transfer/income support scheme.

The first obvious advantage of DUET is like the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA) it avoids the targeting conundrum. Under DUET, Workers have to present themselves to a placement agency, get registered and show up to work when called. The State does not play a role in determining eligibility criterion. As we know from the experience with MNREGA, self-targeting has proved far more efficient than targeted transfers, when it comes to inclusion. However, as I discuss below, the process of enrolling workers and capturing demand for work in urban India is non-trivial. 

Second, in the short term at least, DUET does not require any expansive new spending by government – something that the central government has stubbornly refused to do. In my reading, work is only to be provided for physical infrastructure repair and maintenance undertaken by public institutions. Some of this is budgeted, and costs of new employment are offset by the benefits of asset maintenance (this is an advantage over MNREGA where the focus was on new assets and the quality varied).  The ‘T’ training component can be dovetailed with the existent skilling programme, thus enabling budgetary ‘convergence’. What DUET would do is to ensure that government honours its budget commitments (rather than delay or substitute these expenses, as it is prone to doing even in ordinary times) and enable convergence (which bureaucrats would welcome) rather than significant new spending. Hence, DUET may have surprisingly more political appeal, compared with a brand-new cash transfer programme and therefore be a more pragmatic approach to building an urban social-protection system.

Third, a possible advantage of DUET is whether, like MNREGA, it could provide a wage floor, thus shifting the general equilibrium of market wages in favour of casual workers. But as Sandip Sukhtankar has pointed out, the urban labour market is far more dynamic than rural India where employers have monopsony powers. As a result, the wage effect of DUET is unlikely to be significant. However, DUET has the potential to impact wages through a different route, at least for public works. At present, public works are undertaken by the public works department or the urban local body (ULB) through petty, often exploitative, contractors. Under DUET, this role is to be performed by a placement agency. Importantly, payments will be made directly to workers accounts which ensures minimum wages. 

But for these advantages to be realised, the success of DUET will hinge on knotty implementation challenges. First, demand registration or, put differently, linking workers to potential employers – the role petty contractors perform. This is where the DUET design is weakest. It anticipates that workers will ‘register’ through an independent placement agency either run by the local government, a worker cooperative or a non-profit. But unless handled carefully, this registration can become the first point of exclusion. Cumbersome paperwork, a bureaucratic passion, as Debraj Ray pointed out, can be the first hurdle. But equally important there is no guarantee that the local government agency or the non-profit may not be captured by the very contractor the DUET seeks to weed out. There are some interesting experiments at the state level that DUET can learn from. My colleagues at the Centre for Policy Research have been closely observing the Odisha governments’ Urban Wage Employment Initiative, which has mobilised the self-help groups (SHGs) and slum dweller associations to serve as the bridge between workers and employers (the ULBs, in this instance). Odisha government has also limited the shelf of works to civic works performed by ULBs, thus avoiding a direct clash with the powerful Public Works Department and their nexus of government contractors. However, this may not be replicable in other parts of the country. Replicating this model requires an important precondition – investments in animators (local community workers) and local organisations. In smaller towns, this may be viable but even this does not necessarily displace the contractor, as these animators and local organisations may also be part of the nexus. However, it does bring new workers in the fold, as the size of the pie increases. Another option is to create a ward-level employment register, with multiple points of registration to avoid institutional monopoly.1 This could be maintained at the ULB which in turn performs the function of linking employees to employers. But can local governments perform this role? 

This brings us to the second challenge the DUET will have to confront – the variable capacity of urban local governments across the country. Many interlocutors in this symposium have viewed with excitement the use of the term ‘decentralised’ as central to the design of DUET. Could this be the pathway to reviving our defunct local bodies? Drawing from the MNREGA experience, I am sceptical. MNREGA too was implemented in an environment of variable local government capacity. Where capacity was low, it was bureaucrats who took over implementation relegating panchayats to the role of post offices, relaying orders from the bureaucracy. The resultant bureaucratic capture broke the virtuous accountability cycle embedded in the ideal of local governance – governments closest to people are most responsive to bottom-up pressures for accountability. This is not an argument against decentralisation through DUET, rather a caution and reminder that this will not happen without need serious investment.

So where does this leave us on the debate of DUET versus direct cash? The benefits of DUET are significant, if, the implementation knots can be ironed out. I remain cautiously optimistic but as a way forward, I recommend a state-led approach. Let states assess their readiness and design pathways for implementation. The Centre can finance this but not as a one-size-fits-all centrally sponsored scheme. DUET will need to be implemented in ways that are state-specific and responsive to state-specific urban institutions. In some cases, cash transfers may still be easier and more effective to serve the goal of unemployment insurance. For others, especially those with stronger local bodies and community organisations, the benefits of DUET will outweigh those of cash transfers. Large metropolitan cities and towns may require a completely different approach. In conclusion, this DUET will be best performed when states find their own tunes synchronised with outcomes. And for those who prefer, cash can be an option. This may unleash new innovation that we can learn from. In conclusion, this DUET will be best performed when states find their own tunes synchronised with outcomes. 

I am thankful to Rakshita Swamy, Partha Mukhopadhyay and Shrayana Bhattacharya for patiently discussing this proposal with me and commenting on an earlier draft. Any errors are mine alone.  


  1. I am grateful to Rakshita Swamy for this important operational insight.
No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter