Negotiating power purchase agreements for solar projects

The article highlights the main points to be considered while negotiating and signing Power Purchase Agreements for solar projects.

In order to achieve its target of 1,000 MW solar based generation capacity in phase I of the National Solar Mission (NSM), the mission has nominated NTPC Vyapar Vidyut Nigam Limited (NVVN) to bundle conventional power procured from its principal NTPC Limited with solar power and distribute it to state power distribution companies (Discoms). The proposal envisages procurement of solar power at rates determined by Central Electricity Regulatory Commission (CERC) guidelines for solar tariff, combining it with NTPC’s unallocated generation and selling it to the states at rates determined by CERC.

NVVN has proposed a tripartite arrangement for purchase of power from Solar Project Developers (SPD) and sale to Discoms. The proposal includes a Power Sale Agreement (PSA) with Discoms and Power Purchase Agreement with Solar Power Developers. In its draft PPA, NVVN has implied that the final responsibility of payment rests on Discoms and in the eventuality Discoms fail to pay NVVN, then it ‘may’ make payment to SPDs from a separate fund, currently under consideration by the Govt. of India.

This paper discusses the broad principles of the PPA and key aspects that Solar Project Developerss need to be cognizant of as they plan forward.

Payment guarantees

For any power project to achieve financial closure, bankability of PPA is considered paramount. Solar power projects are amongst the first of their kind in India and significant emphasis is likely to be placed on the bankability of the project parameters. Good technology partner with proven track record, financial capability of the company add to the bankability factor. However, high project costs (about 3 – 4 times of conventional power projects) outweigh all factors and an attractive Feed-in-Tariff mechanism backed by bankable PPA with a financially sound company and qualified and experienced technology partner seems to be the clinching factor.

The NVVN proposed PPA document in its present form can neither be considered bankable nor does it provide any guarantees of payments.

Power Purchase Agreement Purchase Price

Central Electricity Regulatory Commission (CERC) has proposed a Levelized Cost of Electricity (LCOE) valid for 25 years and NVVN has proposed this as the PPA purchase price. However SPDs are concerned about the clauses that could cause PPA prices to be reduced over the tenure of the agreement. Such fears must be allayed through specific clauses that prevent any future law or act to change this price during the entire tenure of the agreement.

The cost of electricity from solar power projects constructed at today’s costs and efficiencies has been levelized for 25 years based on prevailing conditions and assumptions. Unlike conventional power projects where the major component of generation cost is the changing fuel costs, solar electricity costs depend largely on the cost of capital, operation and maintenance costs; cost of capital tends to more or less fixed unless the rate of interest changes (floating interest rates) or inflationary trends reduce the internal rate of return of the project. If the capital invested is in foreign currency, then foreign exchange rate fluctuations affect the cost of electricity.

It is therefore important to view tariff for solar power projects differently and take into consideration capital costs, inflation rate, foreign exchange fluctuations, operation and maintenance (O & M) costs where the major component of O&M costs is replacement parts followed by fixed costs of maintaining miscellaneous equipment, labor and water treatment costs similar to those of conventional power plants.

Table 1 highlights the proportional (levelized) generation costs, as determined by CERC based on their guidelines for determining generic tariff for solar PV and solar thermal power projects. This example pertains to solar thermal projects.

Table 1
Breakdown of cost of generation (levelized)

Note: Capital costs include cost of debt, interest on working capital and return on equity, levelized, as per CERC norms

Linkages of PPA Purchase Price with Inflation

While CERC in its guidelines has taken into consideration an annual increase in O & M costs at the rate of 5.72% per annum, it does not include the impact of inflation on capital. Return on equity may be significantly eroded over the 25 year period if the rates of inflation are significant. It is suggested that PPA may include a provision that allows for increase in PPA price if inflation exceeds a certain level. 

Table 2 shows the erosion of equity return from Govt. recommended 18% for 1st 10 years and 24% until year 25. The example pertains to solar thermal projects.

Table 2
Inflation adjusted return on equity (weighted average)
Rate of Inflation Return of Equity

Part payment of PPA Purchase Price with Foreign Exchange Rates

If the solar project is financed through foreign commercial borrowings or equity, any fluctuations in foreign exchange rates would directly impact the cost of electricity, as the cost of capital is the single largest cost component of electricity generated from solar power projects.

It is impossible to predict the foreign exchange rates over a 25 year period and to insulate SPDs in such cases; it may be worthwhile to consider partially paying electricity costs in foreign exchange.

A mechanism that protects foreign investors from currency fluctuation risks serves to broaden the investor base and may be instrumental in attracting a larger base of fund inflow for solar projects.

Non Firm Power

Renewable energy projects are dependent on the vagaries of nature and therefore its generation output cannot be predicted accurately. NVVN suggestion to penalize lower or excess generation of electricity based on CERC suggested Capacity Utilization Factor (CUF) is considered contrary to the principles of power generation through renewable sources.

Solar resource, despite sophistication in techniques employed to forecast hourly solar irradiation using Typical Meteorological Year (TMY), may vary by 15% on a year-to-year basis.

Moreover surprisingly, technology or operations that are more efficient and generate more electricity than the specified CUF penalize the SPD under this proposed PPA.

There is a strong case for rewarding excess generation through better CUF, especially if it caters to peak load demand. It is possible, in solar thermal power projects, to increase the capacity utilization by adding thermal storage at additional costs. Thermal storage provides consistent output and even provides power during non-sun hours, catering to peak load needs. Therefore additional generation to cater to peak lead demand must not be penalized.

There is unanimity amongst SPDs on the need to remove the penalty clauses for exceeding CUF norms and limiting penalties for under-generation in case generation is lower by more than 15%.

Deemed Generation

Power from renewable sources is (currently) costly and based on a variable resource. During periods of generation if Discoms are unable to withdraw power for any reason, SPDs must not be penalized.

Generation from renewable sources must be measured at SPD generation point and not at Discom off-take point. It is a norm for conventional power projects to supply and bill power on the basis of Discom off-take. If Discom do not purchase on generation basis IPP loses. Power generated is non-firm power due to natural source; hence these relaxations are essential initially until solar power becomes dispatchable power source through extensive use of thermal storage technologies.

This is an important clause for renewable projects and must be included in PPAs for the benefits of SPDs.

Penalties for Delays

Solar thermal power projects take more than 3 years from financial closure to complete. This is due to the complexities involved in the project implementation schedule. Accurately determining site specific solar resource data is the most important parameter in determining the project feasibility and engineering of the project. Land required for solar thermal projects are in excess of 3 – 5 hectares per MW (depending on the solar resource, whether thermal storage is included or not etc.). Site selection itself involves estimating solar resource from several potential sites. Solar resource estimation involves meteorological ground measurements of irradiation, wind, temperature, humidity etc over a minimum period of one year and then calibrating these readings against estimations from 12-year satellite data using sophisticated modeling techniques. Site selection and solar estimation studies are unlikely to be commenced prior to project approvals and tariff finalization.

Meteorological ground measurement stations are established immediately after project approvals and studies carried out for minimum one year. Therefore project work cannot commence before the completion of these studies, analysis of readings and determining solar resource and expected electricity output. This exercise generally takes up to 15 months from project approvals. Assuming that while the ground measurements are being recorded, all other project approvals up to pre-financial closure could be achieved; 15 months have already elapsed from the project approval date. A time period of minimum 6 months must be allocated for due diligence towards financial closure, especially due to the complexity of solar thermal power projects. Moreover, solar thermal projects are being commissioned in India for the very first time and therefore unforeseen delays in due diligence and financial closure may be expected.

Due to the nature of solar irradiation and its varying pattern, detailed engineering of the project and its optimization takes up to 3 months and includes preparation of detailed bill of materials for procurement. Vendor qualification, contract negotiation and order placements may take up an additional 3 months from completion of detailed engineering. This can commence only after financial closure and based on solar resource data validated for at least one year period.

Current deliveries for solar thermal components are one year (on an average) with turbine deliveries exceeding 18 months. Installation of all the solar components may take up to one year. Synchronization of the plant from installation completion generally takes up to 3 months. For initial projects another 3 months may be allocated for rectification of plant problems that may unexpectedly crop up.

Initial projects in Spain have not been completed within 3 years from financial closure date despite robust industry presence, availability of experience and expertise. Initial projects in India would definitely take longer.

Stringent penalty clauses for project delays could discourage SPDs and their inexperience would result in project cancellation and/or heavy penalties, making projects unviable.

Initial PPAs must not include stringent clauses for delays and allow sufficient time for project commissioning based on international experience.

Lavleen Singal is a serial entrepreneur and Founder of Acira Solar, a solar thermal power producer. Acira Solar has technology partnerships with international technology companies that have been involved in the development and operation of solar thermal power projects. Lavleen has a Master’s in Physics and MBA from USA. Lavleen may be contacted at Lavleen@acirasolar.com.

The author is a member of CII expert committee that deliberated on NVVN PPA draft. The views offered are personal and do not reflect the views of this CII committee.

Image Courtsey
JK5854
Lif