The ICOR is really a very blunt measure of the productivity of investment and has been quite volatile between years, as pointed out in a recent op-ed in this paper by Subir Gokarn ("Bang for the buck", March 23, 2014). All the same, we have to ask whether there is anything in the fabric of our taxation structure which actually encourages infructuous investment, and gives a policy-induced upward push to the ICOR. Any such push would of course be a structural feature, and would not explain the year-to-year variation in ICOR.
One such in our income tax law is accelerated depreciation (AD), by which investment in certain chosen sectors is allowed depreciation at higher than normal rates upfront (section 32 of the Income Tax Act, rule 5). There is a long list of types of investment eligible for AD. Some of it is equipment with a naturally short life, like rollers in sugar or flour mills. Others carry positive externalities, like air or water pollution control equipment, or solid waste treatment plants, or water supply and purification, or renewable energy.
Wind energy was withdrawn from the list at the end of March 2012 (it still includes solar and other renewable sources). Until that date, against taxable profit in any other sector, you could write off depreciation at 80 per cent of the value of your wind energy plant as expenditure, and immediately gain the one-third share of that expenditure which would have gone as taxes to the government. It amounts to cash upfront right then and there which would otherwise have been surrendered as taxes to the government.
So what is wrong with tax incentives for investments that work to control industrial effluents, or generate clean energy? Surely we need more, not less, of those. The problem is that there is nothing in the incentive structure to ensure that the objective sought is actually achieved. The tax authorities cannot monitor equipment (effluent treatment, say) that is being incentivised with a tax break, to ensure that it actually is in working condition. Effluent monitoring is the responsibility of local authorities and frequently degenerates into a rent-collection opportunity. The investor has no incentive to keep it in working order because his profits are unaffected (that was the whole point of why he was tax incentivised to install it in the first place).
A renewable energy plant, wind or solar, by contrast is an income-generating proposition - whether sold to the grid, or held for captive use. So these are bank-loan worthy, up to 70 per cent of the cost of the plant. This then means that, since the bank loan has to be serviced, the promoter cannot afford to let the plant lie idle. So argues the powerful lobby that is trying to restore wind energy to the AD list, buttressing their case with anecdotal evidence that there are no defaults on bank loans to wind energy plants.
There remain two problems with AD as an incentive mechanism even for income-generating equipment. The plant may be kept functional enough to service the bank loan, but there is no incentive to maximise generation. And, because the incentive calls for speed in installation, the investor typically gets a package deal which is likely to carry a higher cost than necessary. This padded investment is incentivised into the numerator of the ICOR, and although some output does go into the denominator, there is no push to maximise that.
A generation-based incentive (GBI) was introduced in 2009-10, to do precisely what its name intends, by offering 50 paise for every kilowatt-hour of wind energy generation. This was a Plan scheme, with a short life up to the end of March 2012. It excluded AD beneficiaries, so that the incremental investment effect under the scheme became neatly identifiable, and amounted to 1,000 MW per year. Because AD was dropped at the same time that the GBI came to an end, at the close of March 2012, there was a sharp fall after that date in investment in wind energy.
The reasons for the sudden cessation of AD for wind energy are not known. Perhaps it was the revenue loss, which cannot be discerned from the aggregate revenue loss from AD as reported in the Budget documents, but can be worked out at roughly 1.6 crore per megawatt of installed capacity. The GBI has subsequently been restored in November 2013, with retrospective effect, although the funds have not yet begun to flow. The GBI payout cap has been raised to Rs 1 crore per megawatt over 10 years. Even at this higher cap it remains much cheaper than AD, in terms of capacity per rupee of (tax or subsidy) expenditure. Most of all, it is better structured to get us what we really want, which is renewable energy output, not just capacity installed.
The aggregate revenue loss from AD, across all sectors, is around one-third of one per cent of GDP. It is impossible to get a sectoral break-up of that revenue loss, because the sector box to be ticked in corporate tax returns does not indicate clearly whether it should be the major activity of the taxpayer, or the sector for which AD is being claimed. We need to get a better idea of what we are paying for so that, as with the GBI, we might devise a superior route to what we want to achieve in the end -more output from investment.
The writer is a retired professor of economics