50

8,000

3,000

6,000

2,500

2,000

30

feed-in Tariff (eurocents/KWh) Annual Installed PV Capacity (MW)

Annual Installed PV Capacity (MW)

feed-in Tariff (eurocents/KWh)

1,500

4,000

20

2,000

10

1,000

500

0

0

for installed capacity. Subsequently, they have to select the projects based on certain criteria such as first-come-first-serve basis, random selection, or through price-based competitive bidding. The last option also helps “discover” the price. We look at recent experiences of some countries and that of India.

Germany: Germany fixes the price for solar power in the form of feed-in tariffs (FiTs) over 20 years. Higher FiTs are offered for smaller size systems, usually rooftop, while the least are offered to ground-mounted systems greater than one megawatt. The quantity is somewhat controlled by imposing a strict annual degression rate, which is a percentage reduction in feed-in tariffs based on the quantity or solar capacity installed during the previous year. Figure 1(a) shows the reducing feed-in tariffs against the annual installed solar PV capacity. Due to recent drop in solar PV prices, Germany reduced its feedin tariffs twice during 2010. Even then, the annual installed solar PV capacity exceeded 7,400 MWs, equal to about a third of solar capacity addition expected under JNNSM in the coming decade. This translates to a large financial commitment for Germany’s electricity consumers over the next 20 years.

Spain: Spain set the price for solar procurement by offering FiTs for 25 years. How ever, unlike Germany, it also fixed the quantity in the form of a cap, to limit the financial impac t on its utilities. To circumvent the issue of project selection, as noted earlier, the Spanish government decided to accept all projects till one year after 85% of the annual cap was met. When the Spanish government increased its FiTs for PV by 75% in 2007 to provide a boost to its solar sector, 2,661 MW of PV were installed, exceeding the annual cap of 1,200 MW two times over (Figure 1b). The additional capa city of 1,461 MW meant a large unexpected financial commitment of a net present value of several billion euros over the next 25 years.

Further, the Spanish government had and continues to keep electricity consumer tariffs low and reimburses utilities for the deficit by paying through the national budget, i e, taxpayer monies. Spain was one of the worst hit countries during the financial crisis with a high budget deficit. Although the deficit was not all due to support for renewable energy, the government could not keep offering high FiTs for solar energy generation (Craig 2009). In September 2008, it slashed the FiTs by 23%. The Spanish PV market collapsed with only 70 MW of installed capacity being added in 2009. Further, the Spanish government is even considering retroactive cuts to FiTs for existing projects, a move that breaches contracts and provides considerable uncertainty to the Spanish solar sector.

California, USA: In December 2010, the California Public Utilities Commission in the United States introduced the Renewable Auction Mechanism to procure renewable energy projects of less than 20 MW, which mainly include solar. Under this mechanism, the required installed capa city will be fixed and projects selected based on least cost rather than first-come-firstserved basis at a set feed-in tariff (CPUC 2010). The programme aims to use standard terms and conditions to lower transactional costs and provide con tractua l transparency needed for effective financing.

India’s JNNSM

5F

4GIKUVTCT

EPW

COMMENTARY

ted CST projects was Rs 11.41 per unit. Figures 3 and 4 show the capacity-wise bids of the selected projects and the range of the tariffs in comparison to the CERC FiT.4 These tariffs were computed

400 500

by assumin g that the

project promoters are not availing the benefit of accelerated depreciation.5 The PV and CST tariffs were on an average 32% and 25% lower than the 2010-11 CERC set feed-in tariffs (Rs 17.91/ kWh for PV and Rs 15.31/kWh for CSP; (CERC 2010)) res pectivel y. This red uc tio n in tariffs for the first 620 MW of solar projects will result in a total savings of Rs 4,700 crore (NPV at 10% discount rate over 25 years) for the consumers, indicating success of the auction.

Low Tariff Bidding Concerns

The large discounts in tariffs have led to several concerns being raised within and outside the solar industry. These include underbidding by firms, which may hinder financial closure and timely completion of projects. It may lead to subsequent underperformance due to the usage of substandard equipment. Critics point to “inexperienced” players such as knitwear and animation firms figuring in the list of successful bidders (Pearson 2010).

(Crore)

2,000

0 -2,000 -4,000 -6,000 -8,000

-10,000

Source: PFC (2010).

(PV Tariffs in/kWh) 20

(CST tariffs in kWh)

(without accounting for state government subsidies) touching a phenomenal Rs 53,000 crore in 2008-09 (Figure 2 for losses of utilities in major states) (PFC 2010). The reason is similar to the Spanish case where all the costs incurred by utilities are not passed on to the consumers. Hence, any further impact on utility expenses can put a strain on not just electricity consumers but also taxpayers (in the form of government subsidies).2

Initially, the GoI intended to offer a fixed feed-in tariff over 25 years, set by the Central Electricity Regulatory Commission (CERC). However, seeing the large response from the industry, the GoI chose to select projects using a reverse auction mechanism. This mechanism is similar to the one adopted later by California as discussed earlier.

Auction Results

Under the reverse auction mechanism, the first 150 MW of PV (Figure 3) and 470 MW (Figure 4) of CST projects were selected based on the maximum “discount” that they offered on the CERC declared feed-in tariff. The auction received a very high response with applications received for 5,126 MW capacity, which was eight times more than the 620 MW target (MNRE 2011).3 The weighted average of quoted tariffs for the selected PV projects was Rs 12.16 per unit, while that for the seven selec

random lots would not have ensured fair play, nor would it have addressed the concerns of timely completion or adequate performance of projects.

A drawback of the auction mechanism is that the process may keep out small project developers. However, the auctions are being held only for large MW-scale solar plants (5-100 MW), with require d investment over Rs 10-12 cr/MW. Smaller gridconnected PV projects (2MW) including rooftop PV are being procured at CERC set feed-in tariffs and have a target of 100 MW for phase I of the JNNSM.

Other Solar Programmes

Apart from the JNNSM, the state government of Gujarat has signed power purchase agreements for 968 MW of solar projects6 under its own Solar Power Policy 2009 (GoG 2009). Gujarat is procuring the solar power at a fixed “levelised” feed-in tariff of Rs 12.54 per kWh and Rs 9.29 per kWh for PV and CSP respectively, which includes the benefit of accelerated depreciation (GERC 2010). These tariffs are higher than the average quoted tariffs under the JNNSM (especially considering the benefit of accelerated depreciation), but significantly lower than the CERC 2010 tariffs. Rajasthan too is planning to develop an additional 300 MW of MW-scale solar projects by 2013 and another 400 MW by 2017 under its own solar energy policy (GoR 2010). The state plans to use competitive bidding for this solar procurement. Maharashtra is also forming its own solar policy to develop 500 MW of MWscale solar over the next three years (Pearson 2011).

Given this significant push for solar power presently underway in India, the debate on solar procurement mechanisms is important. However, while solar power costs are high, it is even more important to ask what strategic objectives will India achieve by providing this deployment support for the proposed quantum of solar capacity as well as the type of solar applications. Pushing for grid-connected solar power through solar-specific Renewable Purchase Obligation (RPO) targets, as was recently done through the tariff policy amendment7 without considering the rate of solar power cost decline will unnecessarily cost the Indian consumer while providing little strategic advantage to the country.

Solar Deployment Support

India’s deployment support for solar power should be strategic, not only in terms of procurement but also applications. There is a disproportionate focus on MW-scale solar projects in all of India’s solar programmes. Focus on MW-scale CST plants is justified due to the limitations of project size for that technology. However the emphasis on MW-scale PV plants in India is questionable mainly for the following reasons.

Creating large domestic demand for PV when its costs are still much higher than other RE options is not necessary for the development of India’s domestic PV manufacturing industry. China and Taiwan developed their PV industries without providing any significant deployment support and in 2010, accounted for roughly 60% of the world’s PV cells manufacturing (Hering 2011). Their industries rely almost entirely on export markets, PV equipment being easily shipped across continents. In fact, not mandating domestic content in India’s early phase of solar deployment could lead to Indian subsidies going towards imports.

Further, the reduction in PV costs depend s on the size of the global PV market and most importantly on research breakthroughs. Domestic demand for PV will not have any significant effect on global PV prices since the Indian PV market is very small compared to the global market.8

Finally, PV technology’s biggest advantage is its use in small-scale and decentralised applications. Most countries encourage decentralised PV applications by providing higher incentives compared to MWscale applications. More than 99% of Germany’s PV installations between January 2009 and August 2010 (accounting for 85% of the 8.7 GW installed capacity during that time period) were less than 1 MW in size.9 India, with its 70 million unelectrified rural households10 and 0.8 million un-electrified schools (DISE 2010), should focus on subsidising decentralised PV applications that provide much needed access to electricity and clean lighting, in a way that the rural poor consumer’s tariffs are not more than their grid connected counterparts. PV demand from such decentralised applications, which would be in the range of several gigawatts, should be India’s contribution to the global PV market. does have a 200 MW target for offgrid solar PV under the phase I of JNNSM, with 40 MW allocated in 2010-11 (Ministry of New and Renewable Energy (MNRE) 2011a). However, without appropriate performance-based incentives, robust monitoring and verification procedures, consumer grievance redressal mechanism and ensuring future grid-interaction as and when the grid is extended, the longterm sustainability of these projects and their effectiveness in providing electricity services remains to be seen.

Conclusions

Notes