Poverty & Inequality

DUET re-examined

  • Blog Post Date 09 March, 2021
  • Perspectives
  • Print Page
Author Image

Jean Drèze

Ranchi University; Delhi School of Economics


In September 2020, Jean Drèze’s proposal for an urban work programme called DUET (Decentralised Urban Employment and Training) was presented on I4I. This was followed by an extensive symposium in which eminent economists and practitioners provided their perspectives on the proposal. In this post, Drèze updates his original proposal in one important respect, by suggesting that the programme should be run by and for urban women. Responding to some of the key comments and questions of the symposium contributors, he argues that the best way to assess the efficacy of DUET in practice is to give it a chance.   

I am grateful to all the contributors of the DUET (Decentralised Urban Employment and Training) symposium on I4I for their valuable comments. Many of them seem to think that this “this DUET is worth playing”, as Debraj Ray nicely puts it. That is encouraging, even if it is just a nod for trying it out on a pilot basis. As mentioned in the proposal, this could be done quite easily in specific districts or municipalities. There is little to lose – if DUET does not work, we shall learn from it at least.

Most of the contributors’ comments are well-taken. Before I respond to some of them (especially the more sceptical ones), I would like to update the proposal in one important respect.

A women’s DUET

Here is a variant of DUET1 that merits special consideration – how about giving priority to women workers? I am not thinking of a minimum quota for women, like the one-third quota under the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA)2, but of an absolute priority: as long as women workers are available, they get all the work. In fact, women could also run the placement agencies, or the entire programme for that matter.

To facilitate the involvement of women, most of the work could be organised on a part-time basis – say four hours a day. A part-time employment option would be attractive for many poor women in urban areas. Full-time employment tends to be very difficult for them, especially if they have young children. Wage employment for a few hours a day would be much easier to manage. It would give them some economic independence and bargaining power within the family, and help them to acquire new skills. Remember, the economic dependence of women on men is a major factor in the perpetuation of extreme gender inequality and female oppression in India.

Giving priority to women would have two further merits. First, it would reinforce the self-targetting feature of DUET, because women in relatively well-off households are unlikely to go (or be allowed to go) for casual labour at the minimum wage. Second, it would promote women’s general participation in the labour force. India has one of the lowest rates of female labour force participation in the world. According to National Sample Survey (NSS) data for 2019, only 20% of urban women in the age group of 15-59 years spend time in “employment and related activities” on an average day. On average, they spend less than 5% of their time in “paid activities”. This is a loss not only for women – who live at the mercy of men – but also for society as a whole, in so far as it stifles women’s contributions to economic activity and public life.

I would add, on a more tentative note, that giving priority to women may help to prevent corruption. If wages are paid directly to the workers’ accounts, siphoning off DUET funds would require collusion with workers, real or dummy. Women may be more reluctant than men to participate in a scam, if only out of fear. For the same reason, putting women (or women’s collectives) in charge of the entire scheme may be a useful safeguard against embezzlement, aside from having some value in its own right.

Many creative ideas and activities could be built around a ‘women’s DUET’. Women-managed placement agencies could help to promote better facilities for women workers such as safe transport, protective gear, crèches, and toilets – setting an example for all workplaces. They could also provide various services to women workers (for instance, help centres, health check-ups, canteens), and organise special training programmes that challenge the traditional division of labour. In short, DUET could become a springboard of collective action and mutual support among urban women. This may sound idealistic, but there are precedents of this sort of initiative, such as the Kudumbashree programme in Kerala.

Loose ends

As it happens, this modified DUET (I would be happy to let it override the original) makes it easier to answer some of the queries and criticisms that came up in the symposium. I shall focus on four issues that have come up more than once.

Why before how

Some contributors, particularly Ashok Kotwal, asked for greater clarity about the case for launching a DUET-like scheme. In fact, Ashok asked so many questions under this heading that it may seem impossible to answer them. Fortunately, he gave lucid answers to some of his own questions, making my job easier.

The Covid-19 crisis has drawn attention to the insecurities that haunt the lives of the urban poor. Generally, they are less insecure than the rural poor, partly because fallback work is easier to find in urban areas – if only pulling a rickshaw or selling snacks. Still, the urban poor are exposed to serious contingencies, both individual (like illness and unemployment) and collective (lockdowns, floods, cyclones, financial crises, and so on). Social networks and mutual support can help, but they have obvious limits, especially in times of collective crisis.

There is, thus, a need for better social protection in urban areas. There are not many options. Universalising the public distribution system (PDS) in urban slums would be a step forward (and it can be done under the National Food Security Act), but foodgrain rations do not take people very far. Employment-based support is one way of doing more. It has two major advantages: self-targetting, and the possibility of generating valuable assets or services.

Kotwal questioned whether self-targetting would really work if DUET pays minimum wages. Informal-sector earnings, he argues, are often much lower, so the demand for DUET work could be very large. I am not so sure about this. As of now, minimum wages in India do not distinguish between rural and urban areas. In rural areas, they tend to be higher than market wages, but this is not necessarily the case in urban areas. If DUET gives priority to women, it is likely that most workers will come from poor households, because casual labour is unlikely to be of much interest to better-off women.

Even then, of course, there may be some excess demand for DUET work. That is not a major issue, as long as those who get the jobs come from poor households. After all, DUET is not envisaged as an employment guarantee scheme, let alone a law. Employment guarantee is a more ambitious idea.

The fact that DUET is not an employment guarantee detracts from its value as a form of social protection. But it still has some value. It would give an additional source of income, or at least potential income, to workers who register under the scheme. DUET could also be a useful step towards an employment guarantee scheme in urban areas – or perhaps for urban women at least.

DUET versus maintenance funds

Ashok Kotwal and others also questioned the case for DUET as opposed to providing more generous maintenance funds to municipalities or other public institutions. On this, let me first clarify that DUET need not be restricted to maintenance work. The list of permissible works could also include other types of productive, labour-intensive, short-term jobs that may be required from time to time by the approved public institutions – for instance extra help for a public function, temporary security services, assistance with periodic cleanliness drives, environmental improvements, public health campaigns, or debris removal after a storm. Still, maintenance work is likely to be an important part of DUET since there is so much scope for it (I am writing this in a university office with falling plaster and dust-caked windows).

No doubt, adequate funds and arrangements for routine maintenance of public spaces and institutions would be a very good thing. They are already in place – in substantial measure – in some of India’s better-off and better-governed states. But in the poorer states (where DUET would be particularly useful), shoddy maintenance remains endemic, and putting in place better maintenance arrangements will take a long time – I would say 10 years at the very least, as a ballpark timeline. One reason why this problem is not easy to fix is that some maintenance needs are difficult to anticipate. In the absence of contingency funds, they require a series of administrative steps – evaluating the need, sending a proposal, sanctioning the funds, hiring a contractor, and so on. I have seen how the simplest construction, maintenance and repair works in states like Jharkhand can take years. Pending a better system, why not supplement existing maintenance arrangements with a more flexible option like DUET?

Migration effects

As expected, Martin Ravallion (among others) raised the question whether DUET might induce some rural-urban migration. If DUET attracts more workers to the cities, would that not defeat the purpose of alleviating urban unemployment? On this, a few points.

First, this would be a serious question for an urban employment guarantee act, but perhaps not for DUET, unless and until the scheme employs large numbers. Remember, DUET is not open-ended – the scale of the scheme is a matter of public policy.

Second, a DUET for women is unlikely to have major rural-urban migration effects. Migration decisions are typically made by men, and the vague possibility that their wives (or other female members of the household) might get some casual DUET work from time to time is unlikely to be a major factor.

Third, if this is an important concern, it is best dealt with by restricting DUET to local residents rather than by not having DUET at all. Permanent migration, involving a change of residence, is unlikely to be influenced by DUET, because this sort of migration typically occurs after the household has been able to find sustained work opportunities in urban areas. Temporary migrants could be excluded from DUET if need be, but this would significantly detract from the purpose of the scheme.

Finally, if DUET does facilitate rural-urban migration, would that be a bad thing? According to Dilip Mookherjee (in his contribution to the symposium), “there is steady accumulation of research showing that growth in India is hampered by an exceptionally slow rate of structural transformation from rural to urban sectors”. This is consistent with the view that there tends to be too little rather than too much labour migration, because of the “stickiness” of labour markets (Banerjee and Duflo 2019). If rural-urban migration is not a bad thing, the possible influence of DUET on migration decisions may not be a major concern. One qualification is that India’s large metropolitan cities are probably big enough as it is. As suggested by Pranab Bardhan and Ashwini Kulkarni, it would probably be a good idea for DUET to focus mainly on small towns, to start with at least.

It seems to me that any effort to improve the lives of the urban poor potentially induces some rural-urban migration. That is not generally accepted as a valid argument against assisting the urban poor, for example, by giving them ration cards or providing basic amenities in slums. The reason is that we consider these facilities as a matter of right. It is for the same reason that the government provides water and sanitation facilities to protesting farmers on the outskirts of Delhi, even if their presence there is considered a nuisance. Whether the same principle can be extended to guaranteed employment is the basic dilemma of an urban employment guarantee act. For DUET, and especially a DUET for women, induced migration is unlikely to be more than a minor downside of the scheme. Hopefully, the benefits will make up for it.

Finally, several commentators have expressed concerns about possible corruption. That, of course, is a challenge for almost any welfare scheme in India – even the Aadhaar3-based cash transfers that were supposed to be largely corruption-proof.4 Some contributors have proposed various safeguards, such as independent inspections, social audits, and cost-sharing. Further possibilities are mentioned in the urban employment guarantee proposal prepared by the Centre for Sustainable Employment (2019). These and other safeguards are indeed essential.

In the DUET proposal, the placement agency is presented as another safeguard, aimed at preventing collusion between employers and workers. I do think that this could really help. When wages are paid directly to the workers’ bank accounts, a corrupt entrepreneur is constrained to ‘manage’ them (the workers) in some fashion if they want to siphon off wage money. The simplest approach is collusion:5 the entrepreneur picks the workers, handles the paperwork, and asks them for a share of the wages in return for helping them to earn without doing any work. The workers, typically, are the entrepreneur’s own friends, relatives, clients, or cronies. This corruption model, so to speak, is still very common in MNREGA, and is proving hard to eradicate. Social audits, for one thing, are of limited use, because the workers (or presumed workers), who would normally help to expose corruption, are part of the scam.

A placement agency can help to break this nexus by preventing the entrepreneur from picking the workers. In the DUET system, a corrupt entrepreneur will have to ‘manage’ not only the workers but also the placement agency, to ensure that his/her cronies are hired. That is bound to be harder than just managing the workers – not impossible perhaps (depending on the reliability of the placement agency) – but harder at least. Thus, I do not quite agree with Ashwini Kulkarni that “If two people can collude so can three”.

In the DUET framework, the key thing is to ensure that the placement agency is reliable and accountable. That is likely to be easier than keeping a tab on thousands of dispersed entrepreneurs, as MNREGA requires. There are other reasons why corruption may be easier to contain under DUET than MNREGA. For instance, worksite inspections (ex-post or on-the-spot) would be easier to conduct, because of shorter distances and better transport facilities. Similarly, higher education levels in urban areas would make it easier to ensure that workers are aware of their rights and understand how the scheme works. As mentioned earlier, it is possible that putting women in charge of DUET would also help.

Concluding remarks

Let me conclude by sharing a doubt of my own about DUET – will the concerned public institutions make active use of the job stamps? Think of the head of a university department who sees that the walls need whitewashing. It would not take much initiative for him/her to reach for the job stamps and get the job done. Still, it is easier to do nothing.

This is where there is a big difference between DUET and the ‘service voucher’ schemes that have proved so popular in some European countries.6 The service vouchers are much like job stamps, except that they are used by households instead of public institutions, for the purpose of securing domestic services such as cooking and cleaning. The service vouchers are not free, but they are highly subsidised, and households have an incentive to use them since that is a way of buying domestic services very cheap. In the DUET scheme, the use of job stamps relies on a sense of responsibility among the heads of public institutions, not on their self-interest.

It is, thus, not easy to guess how intensively job stamps will be used. Once again, the best way to find out is to give DUET a chance.

I4I is now on Telegram. Please click here (@Ideas4India) to subscribe to our channel for quick updates on our content


  1. This variant was initially proposed in Drèze (2020).
  2. MNREGA provides guaranteed employment on local public works to any adult in rural areas who is willing to do unskilled manual work at the prescribed wage, up to 100 days per household per year.
  3. Aadhaar or Unique Identification (UID) number is a 12-digit identification number linked to an individual’s biometrics (fingerprints, iris, and photographs), issued to Indian residents by the Unique Identification Authority of India (UIDAI) on behalf of the Government of India.
  4. The recent ‘scholarship scam’ in Jharkhand is one useful reminder of these vulnerabilities (Angad 2020).
  5. Two other approaches are coercion (forcing the workers to part with some of their wages) and deception (extracting money from their bank accounts without their knowledge) – see Adhikari and Bhatia (2010). In these approaches, workers are victims of the scam rather than accomplices. When workers are reasonably aware of their rights and how the system works, coercion and deception become difficult, and collusion becomes the chief method of embezzlement. That, it seems, is what happened with MNREGA in recent years.
  6. For an introduction to these schemes, see for example, Raz-Yurovich and Marx (2018), Desiere and Goesaert (2019), and the literature cited there.

Further Reading

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