Was Mahalanobis stupid?

 Was Mahalanobis stupid? No. He was a good mathematician, a creative thinker and he had a strong religious as well as an entrepreneurial streak.  He started the Indian Statistical Institute by himself. Within a decade, it had won international acclaim. It is the only academic institute from the Third World which was reduplicated in the the US.

Sadly, Mahalanobis didn't understand that incentives matter. Sampling is all very well and good, but how do you know the people paid to do it won't fudge the numbers? Only if those paying for the survey have a strong mercenary motive to get accurate results will the exercise be trustworthy. In other words, a Government funded Statistical Survey will be shit more particularly if it the result of the Survey is known to be- everybody is very very fucking poor!

Consider Mahalanobis's idea of a Labour Reserve Service-

 run by government... When a particular worker is thrown out of a job, the enterprise should make a payment to L.R.S. Under the circumstances, a person will be thrown out only when the anticipated benefit is more than the payment to be made.

Which means the marginal cost of hiring a worker is the wage plus the present value of the 'payment to the LRS'. Mahalanobis has just reduced the demand for labour. The other side of the coin is that the Government has an incentive to prop up 'sick industries'. If they go bankrupt, the contingent liability will not be realizable. But if tax revenue is eaten up by accumulated losses from sick industries, then there will be less money left over for infrastructure investment. 

All such workers will be maintained by the L.R.S., financed by the government and payments received from enterprises.

So, it isn't just employers but the Government too which now has shouldered a heavy liability. A labour surplus country can't afford to 'maintain workers' who aren't working. 

The workers in L.R.S. will receive training and be engaged on productive work, the average wage in L.R.S. being slightly lower than the industrial wage.

In which case, that wage can't be an 'efficiency wage'. In simple terms, the worker has no incentive to have a good attitude. If the Boss isn't nice to him he scarcely loses any money by quitting while the Boss class has to shell out big bucks to maintain workers who don't want to take orders. Mahalanobis has just destroyed productivity.  

With industrial expansion, the enterprise will provide employment to members of L.R.S. at nominal wage rates,

they would avoid such workers like the plague. They have gotten used to being paid for pretending to receive training or pretending to do 'productive work'.  

the L.R.S. thus serving as a buffer against unemployment.

Only for those who get on it- i.e. those who already had a job but quit so as to live a leisured life on the LRS. But this also means nobody would get jobs save those whom bosses are sure won't apply to the LRS. This means, either you have Mafia dons as employers or employees have strong kinship ties with the employer.  

Since enterprises will be able to fire a worker on grounds of economy and efficiency,

But will have to shell out big bucks for the pleasure of doing so 

it will be possible for them to attain high productivity.

Boss says 'please darling, do some work.' You reply, 'fuck you, Boss! If you fire me, you will have to shell out big bucks. You are too much of a miser to do so.'  

On the other hand, the worker will find a regime free from any apprehension of unemployment.

No. They will find a regime where there is no danger that they will be employed in the organized sector- i.e. one where productivity can rise rapidly permitting higher real wages in the future.  

Most workers will find they have no chance of regular employment.  They will subsist as casual labour or eke out a miserable existence in the informal sector.