The LOO of public spending

JADU ECONOMICS

There is a new welcome buzz in town. It is that the government will no longer automatically approve, and spend, money on welfare programmes. As a beginning, the government will begin to monitor the outcomes, and hopefully, after all the information is in, it will begin to act.

As an articulation of its policy, the UPA government said this via its finance minister in Budget 2005-06: "Outlays do not necessarily mean outcomes. The people of the country are concerned with outcomes."

It is ironic that this statement comes in the middle of a tax-and-spend Budget, with both some new taxes and new spending, being hugely questionable. A large amount of noise has justifiably been made about the loony fringe taxes that have been suggested: a tax on cash transactions and the loony squared "fringe benefits tax".

Little commentary has appeared on what probably necessitated the over-the-cliff attempt by the UPA government to raise tax revenues. One mad policy begets another""so if you have loony spending, then you probably need loony taxes, no?

The reason relatively little noise has been made about such spending is that is not done in polite company. How can one question something as noble as an Employment Guarantee Programme, or education for the poor, or mid-day meals for the malnourished. Oh, and we forgot""how can one object to expenditures that improve the well-being of the downtrodden, especially when they belong to the oppressed opposite sex? Can you object? We can't.

But then what is the import behind the caution that "outlays do not necessarily mean outcomes"? After all, the outlays are noble. Why wasn't thinking devoted to outcomes when outlays were being decided? Or is it the case that mother's milk betterment outlays are the holy cows of social expenditure?

We hope not, because it can be argued that precisely the noble goal expenditures are where leakages are the most. (Hence, our title""leakages, outlays and outcomes, LOO.) A leakage is also a polite term for corruption.

The Budget document agrees: "We shall also ensure that programmes and schemes are not allowed to continue indefinitely from one Plan period to the next without an independent and in-depth evaluation. Civil society should also engage Government in a healthy debate on the efficiency of the delivery mechanism."

As members of civil society, we wish to take up the government's kind request to engage it in debate. We have a paper evaluating a wide spectrum of past expenditure policies (Surjit S Bhalla and Tirthatanmoy Das, "Policy Leakages: The Gap between Outlays and Outcomes," forthcoming, www.oxusresearch.com ).

Here, we wish to question what possibly is the least thought-filled noble outlay project ever conceived either by this or any other government""the Employment Guarantee Act (EGA). This Act, which some have termed the flagship of "human face" endeavours, purports to guarantee 100 days of employment to every able-bodied poor person in need of employment, especially in rural areas, where most of the poor reside.

This social security measure has nobility written all over it. The poor are meant to do back-breaking work, creating assets for India, especially trendy infrastructure assets. In exchange, they benefit by obtaining meagre wages. The total expenditure for this innovative programme was estimated to be Rs 40,000 crore to Rs 60,000 crore; the originators of this idea are equally worthy""the National Advisory Council (NAC) of the government of India.

The finance ministry appears to have shortchanged the NAC. It is "only" spending about Rs 11,400 crore on rural development and employment this year, compared to Rs 8,525 crore last year""a hefty 33 per cent increase but far short of what the NAC says is "needed". Asking the NAC to be honest about its expenditure policies is like asking Ganguly whether he deserves to be captain of India.

The NAC did not discover an Employment Guarantee Act""it has been implemented in India, in different states, and under different names, for at least 30 years. We (civil society) also have data on how much unemployment is present, via the National Sample Survey on employment and unemployment for 1999-00.

The economy has grown at a healthy fashion since then, and presumably unemployment today is lower than in 1999-00. Nevertheless, the 1999-00 survey can provide upper-bound estimates of unemployment in India.

We want to err in the direction of higher expenditures; so we have chosen the largest of the unemployment estimates contained in the survey. (The NSS estimates are according to daily, weekly, and usual status; we have used the daily status for analysis.)

This number is 142 crore person-days of unemployment among poor households in rural India in 1999-00. This number is for total unemployment among poor rural households. For example, if I am poor and want to work 365 days a year and did not find work for 50 days, then these 50 days are part of the total of 142 crore man days.

The average wage in 1999-00 for poor rural workers was Rs 37 per day; given the inflation since then, the equivalent wage today is Rs 42 per person per day. So, the total wage cost of eradicating unemployment among the rural poor is (142X42) Rs 5,964 crore, or Rs 6,000 crore. Given the fact that the wage bill constitutes about 60 per cent of the total cost of worthy employment generation projects, the total cost to the government should be no more than Rs 10,000 crore annually.

But the government does not intend""yet ""to eradicate unemployment among the poor. It believes it can provide only 100 days of employment to the unemployed. So instead of 142 crore mandays, its target is considerably less""64 crore mandays. So the total wage expenditure involved is Rs 2,704 crore (64X6,000/142); including capital costs and overheads (40 per cent of the total), the EGA bill should be a maximum of Rs 4,507 crore. But wait""the NAC wants to pay above market minimum wages, not market wages. Even if you assume close to the upper bound of minimum wages or Rs 60 per day, the total cost comes out to be Rs 6,440 crore.

Let this be the first humble offering of civil society to the gods that formulate pro-poor policies: at best, the government will reach the Rajiv Gandhi index of government effectiveness (the ratio of benefits reaching the poor to the expenditure involved) of only 24 per cent (2,704 crore divided by 11,360 crore).

Some additional dimensions of the unemployment problem in rural India: despite its much-touted land reform, and its popularity among glitterati radical chic economists, West Bengal has the highest unemployment days among the poor""29.5 crore, or more than a fifth of the all-India figure! Bihar is close behind""these two UPA states with the longest running single-party governments together account for 40 per cent of rural poor unemployment but only 20 per cent of the rural population.

Indeed, they account for substantially more of the rural unemployed than the following seven large poor states: Andhra Pradesh, Madhya Pradesh, Chhattisgarh, Orissa, Rajasthan, Uttar Pradesh, and Uttaranchal.

We believe that a major purpose of taxation is to redistribute""whether through provisions of public goods, or redistribution of income. We also believe, however, that the structure of governments makes good governance extremely difficult. It is time we did things directly rather than circuitously. The best way of helping the poor is to provide cash transfers""and the poor, like all of us, will ensure good outcomes to their outlays.

ssbhalla@gmail.com , tirtha.das@gmail.com