Tuesday, July 24, 2007

Further details about the implementation of the NREGA

Following up on his recent piece in the Hindu on the implementation of the National Rural Employment Guarantee Act (see this earlier post for a link and some basic facts about the NREGA), Jean Dreze has a more elaborate piece on the issue in the latest issue of Frontline. Written with a different co-author, Drez's Frontline piece has an interesting opening statement:

"THE National Rural Employment Guarantee Act (NREGA) has been a subject of lively debate in the past two years or so. Unfortunately, the factual basis of this debate has been, so far, rather thin. This has made it possible for extremist positions to flourish without being put to the test of careful evidence. While the Act is regularly pilloried in the corporate-sponsored media as an "expensive gravy train" (as a former Chief Economic Adviser described it), the government gets away with extravagant claims of success.

Fortunately, the scope for informed analysis is rapidly growing as reports are beginning to pour in from various parts of the country. "

The rest of the piece consists of fairly close analysis of the statistics available on the NREGA website maintained by the Ministry of Rural Development. The authors note that the website and the data it provides have problematic aspects, but parse it nevertheless to come to interesting conclusions about the relative progress achieved in different states. Some of these would appear to go against conventional wisdom, such as the following bit:

"Is NREGA doing better in States ruled by particular political parties? No obvious pattern emerges in this respect. Nevertheless there is a hint that, if any national political party is taking the NREGA seriously, it is the Bharatiya Janata Party (BJP). Indeed, among the major States, the four best performers in terms of employment generation under NREGA are Rajasthan, Assam, Madhya Pradesh and Chhattisgarh, all of which, except Assam, had BJP governments in 2006-07. As it happens, Assembly elections are due relatively soon in Chhattisgarh, Madhya Pradesh and Rajasthan. One wonders whether active implementation of the NREGA in these States is part of a deliberate electoral strategy of the BJP. It would be a cruel irony if the BJP were to reap the political benefits of a programme initially championed by the Congress and the Left parties."

This is how the authors conclude:

"It would be naive to think that the long history of fraud in public works programmes has already come to an end. But recent experience shows that it is possible to remove mass corruption from NREGA. This calls for strict implementation of the transparency safeguards, as well as firm action whenever corruption is exposed. In these simple steps lies the future of the Act, and of all those for whom it is a new ray of hope."


Dilip said...

The article is informative but not insightful . Table 2 as he notes, shows major interstate differences some of which he highlights, but does not seek to explain what they mean. (1) For example, some states have much higher average cost/person-day (Kerala for one as he points out) and appear to have a very high share of wages out of total expenditure whereas the reverse is true for other states – is that a good or a bad thing? He doesn’t say and the indicator of success that he uses, the number of person-days of employment per rural household generated, does not factor in the non-salary project expenditure at all which makes one wonder: if work done is of no consequence at all to this scheme, why not just hand out the money as welfare rather than have an employment guarantee in the first place? Either I am missing something here or this is not the complete picture or at least, not the way to go about analyzing its outcome. (2) Rajasthan has generated the most number of person days but as he notes, average wage cost per person-day is lower than minimum wage that ought to have been paid. Other states have very low employment generated but much higher wages paid. Again what are the implications of this trade off? No answer is provided. (3) Some states seem to have much lower employment generation despite low wage cost and wage share of total expenditure. Does this indicate pilferage? More silence.

The most important question is whether these districts that have benefited from this scheme have shown improvement in their economic indices that is attributable to this scheme and its value relative to other measures or schemes. The article does not mention anything about this and I could not find anything pertaining to this question at the NREGA website either. Sure if you spend several thousand crores on a scheme, you will find some good come out of it but is it bang for the buck? The title of the piece seems to suggest so but I see nothing in the content to justify that belief.

Anonymous said...

I find Dreze- Oldiges paper unique in its approach as it tries to put together different dimensions of the NREGA quagmire together. As Dilip notes, there are vast inter-state differences and there are many counts on which the success of the Act can be rated. To name a few benchmarks:
- Asset creation by way of NREGA works
- Culture of Transparency finding its roots at government works (rural gamut of RTI Act)
- Employment generation
- helping in reducing migration and making survival in village a feasible option
- Activating Panchayti Raj and making people participate in government processes generally
- mobilization of rural population

Some of them are qualitative parameters which are difficult to gather. Nevertheless, if one concentrates on the numerous Social Audits (people conduct audit and present the findings before the govt.) of NREGA works throughout the country, one gets a feeling of how the Act has been more than a mere palliative. For instance in Rajasthan and Chattisgarh it is so refreshing to notice that the same doomed delivery framework is on its toes to provide work and meet the high expectations of the Act. Corruption has been found to be minimal. It is sea change from the times when other schemes such as NFFWP and SGRY were in existence. I don’t think that it will be possible to apply the infamous Rajiv Gandhi’s 15 paise in a rupee claim to NREGA now.

As for queries put forwarded by Dilip my understanding is the following:

1. Ideally under the Act 60:40 ratio should be maintained between the labour and material component. If the states can manage more than 60% for wages, that is even better (ie. greater percentage is going towards labour)

2. Asset creation is an integral part of NREGA not just for the sake of long term potential of livelihood creation from the same but also Panchayats can now prepare shelf of projects which are to be later pursued under the Act. I do not get Dilip’s query relating to “number of person-days of employment per rural household generated, does not factor in the non-salary project expenditure”. The entire movement behind NREGA was for Right to Employment and never for doles. Demand Driven (employment on demand i.e. factoring in the entitlement philosophy in the Act is most intelligent drafting) aspect is crucial to the Act.

3. Wage payment is I think the most contentious issue relating to the implementation of the Act. Actually payment of Minimum Wages is subject to completion of Minimum Task decided according to Schedule of Rates. Every state has its own schedule and Minimum Wage thresholds. In Rajasthan, for a variety of reasons, average wage rate is coming out to be lower than the minimum wage. One simplistic conclusion is that the minimum task decided is not being completed by people at worksites and therefore proportionate cut is imposed on the minimum wage threshold. This aspect warrants further inquiry.

4. States like Kerala where higher wages are paid and still there is low demand implies that people do not want to get employed for manual unskilled labour for earning just the minimum wage. NREGA in that scenario would be availed by the most economically distressed section.

5. It is to be noted that the demand driven character of NREGA has not yet percolated down amongst people. Therefore by and large for employment generation, State is the prime mover. Creating an ambience in villages that there is employment on demand under the Act would only ensure employment generation. Low employment generation signifies low administrative preparedness and low awareness levels (and not that there is no actual demand for work) in a typical NREGA district (as it is the chosen 200 poor districts where the Act is in operation).

6. Qualitative aspects are difficult to gauge and as is being pointed out in the piece that it is only “year zero” for the Act. Still from some places reports are coming indicating reduced migration.
It is implied that there will be some economic benefit if we can show that there is little pilferage. Therefore in a typical NREGA district, if there has been massive NREGA related income generation amongst people, and there is no pilferage, it is implied that money will help improve people’s quality of life.

I find the piece to be one of the best figure based stories on NREGA which attempts to showcase the country perspective on the Act. It is true that conclusions are not easy to dissect and causations are not direct nevertheless we get some points to ponder upon from the quoted parts:

“Behind these facts and figures is a simple yet powerful message about the NREGA. Within a year of the Act coming into force, the programme has been actively taken up in a small but significant number of districts (20 of them spent more than Rs.100 crore on NREGA in 2006-07). Further, field reports suggest that many of the anticipated benefits of employment guarantee are beginning to show in these pioneer districts: there is greater economic security, agricultural wages are rising, migration is slowing down, productive assets are being created, women have more economic independence, power equations are changing, and so on. The need of the hour is to extend these positive experiences to other districts. If Sarguja or Mandla or Banswara are able to spend more than Rs.100 crore on this programme in a single year, there is no reason why (say) Palamau or Kalahandi should not be able to do it.
Of course, it is also important to ensure that the reported expenditure levels actually correspond to "real" work and wages. Earlier employment programmes have left a long trail of fake muster rolls and embezzled money. However, there is growing evidence that firm enforcement of NREGA's extensive transparency safeguards can go a long way in preventing corruption.”
“These large differences raise the question whether it is better to have State-specific wages or a national norm. This complex matter is yet to be adequately debated. Indeed, wage payments raise a host of interesting and complex questions that have been lost in the din of arguments for and against the Act: how NREGA wages should be determined; whether there should be a national norm; whether piece-rate payments are better than daily-wage payments; how work should be measured; whether the "schedule of rates" should be gender-specific; how to avoid long delays in the payment of wages; and so on. It is not too late to initiate an informed debate on these issues.”

Dilip said...

Dear Anonymous,
Thank you for the insightful clarification. I agree that the figures themselves are suggestive of certain trends. My only point was that the interstate disparities themselves might be more revealing about the strengths and weaknesses of the program and suggest ways of improvement.

I understand that asset creation is an integral part of NREGA. My question is if ~40% (share of wages in total expenditure from table 1 suggests that the actual figure here is 100 – 66 = 34%) is being spent on the material component, what is the index of measurement used to evaluate it? Person-days of employment per household does not account for this. He points out that Maharashtra and Tamil Nadu have managed to implement the program with virtually no expenditure other than wages – I wonder whether this is akin to a dole. That being so, how does one compare performance in these states with that of other states where a much higher percentage is spent on material expenditure?

He does note without elaboration that field reports suggest a variety of benefits. I suppose we would have to wait for the details to evaluate it better.