Meeting-Indias-Renewable-Targets-The-Financing-Challenge.pdf

8/10/2019 Meeting-Indias-Renewable-Targets-The-Financing-Challenge.pdf

1/43

Meeting Indias Renewable Energy argets:Te Financing ChallengeClimate Policy Initiative

CPI-ISB Report

David NelsonGireesh ShrimaliShobhit GoelCharith KondaRaj Kumar

December 2012


2/43

About CPIClimate Policy Initiative (CPI) is a policy effectiveness analysis and advisory organization whose mission isto assess, diagnose, and support the efforts of key governments around the world to achieve low-carbongrowth.CPI is headquartered in San Francisco and has offices around the world, which are affiliated with distin-guished research institutions. Offices include: CPI Beijing affiliated with the School of Public Policy andManagement at singhua University; CPI Berlin; CPI Hyderabad, affiliated with the Indian School of Busi-ness; CPI Rio, affiliated with Pontical Catholic University of Rio (PUC-Rio); and CPI Venice, affiliated withFondazione Eni Enrico Mattei (FEEM). CPI is an independent, not-for-prot organization that receiveslong-term funding from George Soros.

DescriptorsSector Power

Region India

Keywords Renewable Energy, Finance, Solar, Wind

Related CPIReports

Te Impact of Policy on the Financing of Renewable Projects; Supporting RenewablesWhile Saving axpayers Money; Improving Effectiveness of Climate Finance

Contact Gireesh Shrimali, Hyderabad [email protected]

David Nelson, San Francisco [email protected]

AcknowledgementsTe authors thank the following for their collaboration and input: Alipt Sharma of GEF, Amit Oke of ICICIBank, Charlie Donovan of EADA Business School, Don Purka and Sujata Gupta of ADB, Gaurav Malikof Olympus Capital, Harsh Shah of L& Infra, Robert Collier, Saravanan Nattanmai of Nereus Capital,

Srinivas Kara of IDFC, and developers of renewable projects analyzed in this report. Te perspectivesexpressed here are CPIs own.

Copyright 2012 Climate Policy Initiativewww.climatepolicyinitiative.orgAll rights reserved. CPI welcomes the use of its material for noncommercial purposes, such as policy discus-sions or educational activities, under aCreative Commons Attribution-NonCommercial-ShareAlike 3.0 UnportedLicense.For commercial use, please [email protected].

About the Indian School of BusinessTe Indian School of Business (ISB) was established in 2001 with an aspiration to put India on the globalmap of management education. In less than a decade since its inception, the ISB has successfully pio-neered several new trends in management education in India, and rmly established itself as a world-class management institution. In 2008, the ISB became the youngest institution to be ranked among the

op 20, in the Global B-school Rankings by the Financial imes, London, and since then has been rankedconsistently among the top B-schools globally. Te ISB today has a strong pool of research-orientedresident faculty and invites high caliber international faculty from reputed B-schools to teach in its PostGraduate Programme in Management (PGP), Post Graduate Programme in Management for SeniorExecutives (PGPMAX), and Executive Education Programmes. In addition to teaching, the visiting facultyalso participates in collaborative research with the resident faculty. Te school has over 4000 PGP Alumniand 13,000 Executive Education Alumni making an impact on business and society across the world.
mailto:uday%40cpisf.org?subject=http://www.climatepolicyinitiative.org/http://creativecommons.org/licenses/by-nc-sa/3.0/http://creativecommons.org/licenses/by-nc-sa/3.0/mailto:admin%40cpisf.org?subject=mailto:admin%40cpisf.org?subject=http://creativecommons.org/licenses/by-nc-sa/3.0/http://creativecommons.org/licenses/by-nc-sa/3.0/http://www.climatepolicyinitiative.org/mailto:uday%40cpisf.org?subject=


3/43

I A CPI R

M I R E T : T F CN 2012

Executive summaryIndias power sector has two overlapping, historic chal-lenges one that has grabbed international headlinesand another that has largely own below the radar.

The widespread blackouts that brought much of Indiato a sputtering halt in 2012 were a dramatic signal of apower sector that requires attention. But a challenge noless central to Indias future, and arguably much moreso, is that of the countrys goals for renewable energy.

The national governments ambitious goals for solarenergy, coupled with the countrys rapid progress indeveloping wind energy, raise many questions regard-ing the sources and costs of the investment that will beneeded to install and operate this infrastructure.

Under the Jawaharlal Nehru National Solar Mission(JNNSM), grid-connected solar PV capacity increasedby 165% in 2011 to reach 427 MW. However, theambitious targets for 4,000-10,000MW by 2017 and20,000 MW by 2022 may be hard to achieve undercurrent policies and programs, and nancing may be thebiggest obstacle. Likewise, with 16 GW installed, Indiahad already become the worlds fth largest market forwind by 2011, but ambitious plans for a further expan-sion to 31 GW by 2017 will face similarly daunting policyand nancing problems.

In this report, Climate Policy Initiative (CPI) analyzes thechallenges for designing national policy that will attractthe investment needed to spur rapid growth in wind andsolar energy at a reasonable cost. CPI has conducteddetailed nancial modeling of actual Indian renewableprojects; numerous interviews with developers, nan-ciers, and policy makers; and examined, in depth, theidiosyncrasies of Indian nancial markets. This reportdescribes and analyzes the impact of national and statepolicies on various classes of renewable energy inves-tors, as well as the overall relative costs or benets of

policies on the nal cost of renewable energy projects.We focus particularly on the cost and availability ofequity and debt, respectively, and the consequent impli-cations for Indian renewable and nancial policy.

Table ES-1 summarizes the main nancing issues facingrenewable energy policy currently and evaluates howthese issues might grow or diminish in importance inthe medium to long term. To summarize its ndings:

In the short term: The high cost of debt that is, high interest

rates is the most pressing problem currentlyfacing the nancing of renewable energy. Ournancial modeling of actual renewable energyprojects in India and elsewhere indicates thatthe higher cost and inferior terms of debt inIndia may raise the cost of renewable energy by24-32% compared to similar projects nancedin the U.S. or Europe.

Interviews with investors and developerssuggest that neither the cost nor availabilityof equity is currently a major problem. In fact,our analysis suggests that when adjusted fordifferences due to the less attractive nature ofdebt in India, expected returns on equity (ROE)India may actually be lower than in the U.S. orEurope, despite potentially higher country risks.

General Indian nancial market conditionsare the main cause of high interest rates forrenewable energy. Growth, high ination,competing investment needs, and countryrisks all contribute. A shallow bond marketand regulatory restrictions on foreign capitalows also adds to the problem, while the costof currency swaps and country risk negate the

advantages that could come from access tolower cost foreign debt.

In the medium to long term: A declining availability of debt for renewable

energy projects, whether high cost or not, maybecome an impediment. Interviews with lendersand analysis of debt markets indicate thatmany lenders may be reaching the limit for theamount of money they will lend to the sector.Their withdrawal from the market may restrict

project development. Continued high borrowing by the Government

of India and related regulatory restrictions arelikely to keep interest rates high.

Even if the cost of debt goes down, our analysissuggests that loan terms including shorttenors and variable interest rates will becomemore signicant impediments, especially inlower interest rate environments.

Finally, attractive, low cost equity may be less

available in the future. First, as debt becomesless available, current equity investors may not


4/43

II A CPI R

M I R E T : T F CN 2012

be able to recycle their investment capital intonew projects by borrowing against the operatingprojects. Second, as the market matures, manyinvestors may no longer be willing to invest atrelatively low returns in order to establish a

strategic foothold, as they currently appear tobe doing in solar photovoltaic (PV) projects.

Beyond Debt and Equity: Regulation and the structure of the Indian power

sector also raise signicant issues. State-levelpolicies including the nancial weakness ofthe state electricity boards that buy much of theoutput from renewable generators increaseproject risk. National policies designed to weavestate policies together, particularly the recently

established Renewable Energy Certicate (REC)market, do not adequately reect the realities ofnancial markets or state-level risks.

Our analysis shows that renewable energypolicy lessons from the U.S. and Europe may notapply to Indias nancial realities. The economyand nancial markets in which renewableenergy policies operate partly determine theeffectiveness of these policies. The signicantdifferences between India and developed worldnancial markets, including the high cost of debt

and what that means for the impact of policy,means that policy levers used to decreasenancing costs in the U.S. and Europe have aless impact in India.

Other developing countries have bridged the

nancing gap in unorthodox but successfulways. The Brazilian Development Bank (BNDES)is an especially promising example thatdeserves further study and consideration byIndian policymakers.

As a next step, CPI proposes to investigate potentialpolicy solutions to bring lower cost, long-term debt intothe Indian renewable energy market. Given that renew-able energy will require some nancial support in themedium term, until renewable energy costs reach adegree of parity with conventional sources of electricity,we will ask the question: Within a portfolio of policies,is it more effective to close the gap between renew-able energy costs and alternatives through subsidies,tax benets, or other support mechanisms, or throughconcessional debt, or a mix of two or more?

As part of this investigation, we will explore the mecha-nisms used to nance renewable energy in other rapidlydeveloping economies including China and Brazil. TheBNDES example should be particularly instructive.


5/43

III A CPI R

M I R E T : T F CN 2012

NEGATIVE IMPACT OF ISSUE:

ISSUE NOWMEDIUM TO LONG TERM

E Q U I T Y

C O S T A N D

T E R M S

Equity returns required by investors are reasonable for good projects NONE LOW

Required returns depend on technologymaturity and developer strategy NONE MEDIUM

A V A I L A B I L I T Y

Equity is generally availablefor good renewable energy projects NONE MEDIUM

Equity is available from foreign investorsbut facing a shortage of attractiveprojects

LOW MEDIUM

Equity availability is heavily skewed towards a few states with good policy

regimes and attractive business environments

MEDIUM MEDIUM

Lack of debt may reduce available equity in the medium term as equity investorscannot recycle investments from current projects to future investments

LOW HIGH

Table ES-1: Current and future impact of major issues in Indian renewable energy nance

D E B T

C O S T A N D T E R M S

High general Indian interest rateenvironment raises renewable project debt cost VERY HIGH VERY HIGH

Longer tenor debt is generally unavailable MEDIUM HIGH

Fixed interest rate debt is rare; market relies on variable rate debt LOW HIGH

Debt is not strictly non-recourseas it is usually offered on a relationship basis MEDIUM MEDIUM

Shortage of debt may raise its costin the longer term LOW MEDIUM


Banks limit lending to any one sectorfor risk purposes LOW HIGH

Renewable energy debt is often classied within power or energy sectors, which

are already nearing sector limitsMEDIUM HIGH

Some banks are not lending to renewable projectsdue to unfamiliarity withrenewable energy policy and markets

MEDIUM HIGH

Limits on foreign debtmay restrict foreign lending LOW HIGH

Technology riskis limiting debt to some technologies MEDIUM MEDIUM

State-level policy, the poor nancial condition of the State Electricity Boards, andunfavorable local business environments restrict lending to some states

MEDIUM HIGH


6/43

IV A CPI R

M I R E T : T F CN 2012

C1. I 1

2. R 22.1 Renewable energy development in India 22.2 Sources of nance 22.3 The cost of renewable energy in India 3

3. F T 5

4. I A H 94.1 The high general interest rate environment in India 94.2 Longer tenor debt is generally unavailable 104.3 Fixed interest rate debt is rare 114.4 Debt is not strictly non-recourse 114.5 Availability of debt 124.6 Many banks do not lend, or limit lending, to renewable energy projects 124.7 Limits on foreign lending - Capital controls on foreign debt 134.8 Limits on foreign lending - Interest rate ceilings 134.9 Technology risks 14

4.10 State-level issues 155. E A , ? 16

5.1 Cost of equity or required return on equity (ROE) and differentiation between technologies165.2 The availability of equity 17

6. S 18

7. G 227.1 Renewable Energy Certicate markets 227.2 State-level policy issues 24

8. P - C 278.1 Case study analysis India versus the U.S. and Europe 278.2 Brazil and China 29

9. C N S 319.1 Next Steps for Research and Analysis 31


7/43

1 A CPI R

M I R E T : T F CN 2012

1. IntroductionAs India wrestles with the historic challenge of pro-viding enough electricity for its huge population andbooming economy, the national government is pursuing

another equally difficult and impressive goal mark-edly expanding the share of renewable sources in itsenergy supply mix.

The government has embarked upon an ambitiousplan, Jawaharlal Nehru National Solar Mission, to build20,000 MW of solar power nationwide by 2020. At thesame time, India is already the fth-largest market forwind turbines, with cumulative installations of 16GW atthe end of 2011. 1

This urry of activity demonstrates the seriousness

of Indias intentions. But the scale of these ambitions,and the nancial resources required, raise many ques-tions regarding the sources and costs of the neededinvestment, about the adequacy of the investment toreach these goals, and the efficacy of that investment inreaching the most suitable projects.

The question of whether Indias nancial system isadequately nancing the renewable energy sector and if not, why is especially poignant in light of thedifferences between Indias nancing system and thoseof Europe and North America as well as developingnations such as China and Brazil.

Of course, the challenges of the renewable energysector cannot be completely divorced from Indiasoverall infrastructure nancing challenges. Accordingto the Planning Commission of India, infrastructureinvestment will need to increase from about 8% of GDPin 2011-12 to approximately 10% of GDP by 2016-17. Thetotal investment in infrastructure is required to be overINR 45 lakh crore (USD 1 trillion) during the 12 th FiveYear Plan period (2012-17). 2 However, the Indian gov-ernment estimates a 30%, or USD 300 billion, gap inthe targeted infrastructure investments by 2017, largelydue to the lack of long-term nance. 3

Given the emerging nature of renewable energy tech-nologies and business models, nancial and regulatorysectors must adapt quickly a task they have carriedout somewhat unevenly. For this reason, the renewableenergy sectors current challenges are especially acute.

1 BP Statistical Review of World Energy, 20122 An Approach to the Twelfth Five Year Plan, Government of India, Planning

Commission, October 20113 India estimates a USD 300 billion infrastructure funding gap, Infrastructure

Investor, October 2010

In India, the role of development nancial institutions(DFIs), a major source of long-term funding for energyinfrastructure prior to the 1991 nancial reforms, hasmarkedly diminished in recent years. Because of thederegulation of nancial markets, DFIs have been

obliged to compete with commercial banks to raisefunds at market rates. This undermined the businessmodel of the DFIs; many were forced to convert them-selves into commercial banks. This change, combinedwith the shallowness of the corporate bond market, hasleft a gap in funding for renewable energy.

In Brazil, by contrast, the Brazilian Development Bank(BNDES) plays a major role in almost every renew-able project in the country, often by offering long-termloans at below market rates. BNDESs role in nancingrenewable energy has led to lower prices and, arguably,

enhanced the performance of other policies such aswind energy auctions designed to create competitionamongst potential wind park developers.

In this paper we analyze the impact of Indian policies onthe attractiveness of renewable energy investments tovarious classes of investors, and we assess the overallrelative costs or benets of polices on the fully nancedcost of renewable projects. We have used detailednancial modeling of actual Indian renewable projectsas a key element of this analysis.

This study does not address the subject of off-gridrenewable energy projects, which present differentnancing challenges and require different policy solu-tions than on-grid renewable energy projects. Nor doesit discuss in depth the overlap between the nancing ofrenewable energy and conventional power generation.CPI will address these topics in future research.


8/43

2 A CPI R

M I R E T : T F CN 2012

2. Renewable energy industrytrends2.1 Renewable energy development in India

Historically, the growth of wind power in India wasprimarily driven by state-level incentives in conjunc-tion with an accelerated depreciation benet extendedby the national Government of India beginning in 1995.Wind power installed capacity increased from 230 MWin 1994-95 to 16,078 MW by 2011, reaching approxi-mately 94% of the 11 th plan (2007-12) target.

In 2007-11, wind capacity installations increasedat a compound annual growth rate of 19.7% as theGovernment of India introduced additional incentivessuch as the generation-based incentive (GBI) as well asaccelerated depreciation. However, both the acceler-ated depreciation and the GBI, which comprised 10%of wind energy incentives, 4 expired in March 2012.Reaching Indias ambitious target of total wind installedcapacity of 31,078MW by 2017, will now depend uponnew mechanisms, including the Renewable EnergyCerticate (REC) market discussed in Section 7.

Indias solar power industry began signicant growthonly in 2010, when the Jawaharlal Nehru National SolarMission (JNNSM) was announced. After the announce-

ment of JNNSM, grid-connected solar PV capacityincreased by 165% in 2011 alone to reach 427MW.However, a failure to address the remaining nanc-ing challenges will make the targets set under JNNSM Phase 2 (4,000-10,000MW by 2017) and Phase 3(20,000MW by 2022) difficult to achieve.

4 See section 7.

Figure 2-1: India wind power installed capacity and future targets

Source: BP Statistical Review, 2012; Central Electricity Authority; Ministry ofNew and Renewable Energy; India Infoline. Note: Yearly data is at the end ofDecember every year.

17,164

31,078

11th plan target

12th plan target

2006 2011 2012 2017

3 0 , 0

0 0

2 0 , 0

0 0

1 0 , 0

0 0

0

I n s t a l

l e d c a p a c i

t y ( M W )

2 , 0 0 0

2009 2011 2013

1 0 0 0

0

I n s t a l l e d c a p a c i t y ( M W

)

JNNSM

Phase I

target

Figure 2-2: India grid-connected solar PV installed capacityvs. Phase 1 target

2007

Source: BP Statistical Review, 2012 Note: Yearly data is at the end of Decemberevery year.


9/43

3 A CPI R

M I R E T : T F CN 2012

TYPE OF INVESTOR CATEGORYTOTAL

REGISTEREDIN INDIA

ACTIVE INRENEWABLE

SECTOR

Commercial banks

Public sector banks 26 9

Private sector banks 30 6

Foreign banks 37 -

Equity investorsPrivate equity 51 16

Venture capital 180 21

Institutional investors Insurance funds 24 11

Development Banks Development nancial institutions*3 3*DFIs include national level institutions IREDA, IFCI, SIDBI

Table 2-1: Renewable energy investors (number of institutions)

2.2 Sources of nanceA variety of investors nance renewable energy projectsin India, including institutions, banks, and registeredcompanies (Table 2-1). Institutional investors are eitherstate-owned or bilateral and multilateral institutions.Among banks, both private sector and public sectorbanks are involved. In addition to registered companies,venture capital and private equity investors contributeequity investment. Return expectations of the investorsvary according to the sources of their funds and the riskattached to specic projects.

During 2006-09, Indias annual total renewable energyinvestment remained between USD 4 billion and USD5 billion. Investment has risen rapidly since then, fromUSD 4.2 billion in 2009 to USD 12.3 billion in 2011

(Figure 2-3).While wind continues to receive the majority of invest-ment, solar has seen the highest growth, and the gapbetween the two is falling rapidly, as shown in Figure2-4.

2007 2008 2009 2010 20112006

4

1 2

8

Figure 2-3: Renewable energy investment trends in India

Source: Global trends in renewable energy investment 2012

I n v e s t m e n

t ( b i l l i o n s o f

U S D )

0

2009 2010 2011

2

6

4

I n v e s t m e n t ( b i l l i o n s o f U S D )

Figure 2-4: Renewable energy investment trends in India bytechnology

Source: Bloomberg NEF- UNEP Reports, (i) Global trends in sustainable energyinvestment 2010. (ii) Global trends in RE investment 2011 and (iii) Global trendsin RE investment 2012

Solar

Wind

0


10/43


11/43

5 A CPI R

M I R E T : T F CN 2012

3. Finding equity and raisingdebt Two very different

issuesConditions for renewable nance can be very differentdepending on the technology employed, the devel-oper, geography, or the requirements of the investorsthemselves. The most important distinction is betweeninvestors in the debt markets (lenders) and those in theequity markets (owners).

Generally speaking, debt investors are more conserva-tive, accepting lower returns in exchange for lower risk.As such, their primary concern is that downsides arelimited; that is, that the project does not fail. Equityinvestors are willing to take more risk in exchange forhigher returns, and therefore focus equally on risk andthe prospects of a project performing even better thanexpected. Under most circumstances, a project will beleast expensive when it is funded by a mix of debt andequity, either at the project level, or through debt andequity secured at the corporate level.

Renewable energy nancing can become costly wheneither debt or equity investors demand too high a returnor when either is simply unavailable. Thus, for both debt

and equity we pose two sets of questions: Cost and terms: Are the returns investors are

demanding and the conditions they are placingon their investment so onerous as to make theproject economically unattractive? or;

Availability: Is debt or equity just not available?That is, are there enough investors willing toinvest or lend to renewable projects in India?

Signicantly, while policy can inuence the returnsrequired by equity and debt investors and the availabil-ity of either, different policies are likely to be importantto different classes of investors.

In India, the differences between debt and equity areparticularly striking. In general, we nd that equityappears to be readily available at a reasonable cost,while renewable energy debt is both limited and expen-sive. Figure 3-1 highlights the differences betweenrenewable energy debt and equity markets in India anddeveloped markets. Note how equity returns in India aresimilar to those in the U.S. and Europe, but interest rateson debt are signicantly higher.

To compound the problem, access to potentially lowercost international debt is limited due to regulatorybarriers, the cost and risks associated with long-termcurrency swaps, and perceived country risks (Section5.6). As a result, the cost of debt to a renewable energyproject in India will typically be in the 10-14% range,

as compared to the 5-7% range typical in the UnitedStates. Despite the higher cost, debt in India also suffersfrom inferior terms, including shorter tenors and vari-able rather than xed interest rates.

Figure 2-5 in Section 2 demonstrates that the nancingcosts added 28% and 22% to the cost of solar PV andwind projects, respectively, in India. In Figure 3-2, wetake a closer look at the nancing component. We beginby noting that there are many factors that inuence thetotal nance cost. Here we highlight ve:

The cost of debt The tenor of debt that is, the length of time

over which the debt is repaid

Whether the debt is variable or xed

Extra risk that will be taken on by equity in theevent that debt rates are variable

The cost of equity, or the required return onequity (ROE)

Using the same typical projects described in Figure 2-5,

Figure 3-2 compares just the nancing differences of

Low cost nancing of renewable energy projects usually requires both debt and equity investors.However, in India debt and equity investors face very different conditions. When compared tothe nancing of renewable energy projects in the developed world, in India the relatively high costand low availability of debt may add 24-32% to the cost of renewable energy projects compared tosimilar projects in the US, while equity returns have a comparatively minor impact on relative costs.

In this section, we analyze the relative impact of the cost and terms of debt and equity onrenewable energy project costs and identify a set of issues beyond debt and equity marketsthat may concern policy makers. This section paves the way for more detailed discussions ofdebt (section 4), equity (section 5) and other potential policymaker concerns (section 6).


12/43

6 A CPI R

M I R E T : T F CN 2012

SOLAR PV ONSHORE WIND SOLAR THERMAL*

E Q U I T Y

20%

15%

10%

5%

D E B T

SOLAR PV ONSHORE WIND SOLAR THERMAL*

20%

15%

10%

5% E q u i t y R e q u i r e

d R e t u r n o r

D e b t r a t e

( % p e r y e a r ) E q u i t y

R e q u i r e d R e t u r n o r D e b t r a t e ( %

p e r y e a r )

IndiaUS

Europe

Figure 3-1: Range of returns on equity and debt costs for renewable energy projects India versus U.S. and Europe

* No data for solar thermal in Europe. Note: Equity is levered equity. Source: Projects in India and US CPI nance paper.

the U.S. and Indian projects. To summarize, we nd thefollowing:

In the case of the solar PV projects, the higher

interest rate on the debt alone added 19% tothe project cost, while it added 10% to the windproject.

Indian debt tenors are typically much shorterthan in the U.S. or Europe. Overall, we havefound the shorter debt tenors add between3% and 10% to the cost of the project, byforcing more rapid amortization of the loan,and therefore reducing the effective leverageover the life of the project. 5 Stated another way,with shorter-term debt, the average amountof debt over the project life goes down andproject debt has a relatively smaller inuenceon bringing nance costs down over the life ofthe project. It is important to point out that debttenors are relatively less important in India thanthey would be elsewhere given that the spreadbetween the cost of debt and equity is smaller.Therefore, the value of maintaining a higherlevel of debt throughout the life of the project

5 The solar project used for our analysis had an uncharacteristically long tenor.

Therefore, we have adjusted the debt tenor down to 13 years, which wouldbe much more typical of Indian PV projects. The result is to increase theimpact of shorter debt tenors from 3% to 6%.

decreases. If the cost of debt was to decreaseand the spread between debt and equityexpand, we would expect the impact of shorter

debt tenors to increase signicantly. Despite being more costly, the debt terms

in India are less attractive, particularly thevariable interest rate of most debt in India. Inmore developed markets, project developersnearly always seek xed interest debt. Whencombined with feed-in tariffs, or long-termpower purchase agreements (PPAs) with axed price, a xed interest rate leads to a highdegree of certainty around future cash ows. Aninvestment with well-dened cash ows is less

risky, and therefore attracts lower cost nance.Developers will typically use interest rate swapsto convert variable rate debt into xed ratedebt. In India there is no liquid swap market.Therefore, to calculate the cost of the greateruncertainty, we have used the current interestswap rates in the U.S. and applied them to thedebt. At 2%, 10 year swap rates are currentlylow by historical standards. Nevertheless, thiscost adds approximately 7% to the nance costof solar PV and 4% to the cost of wind on acomparable basis. Why do Indian developersaccept variable rate debt? See section 4.3 forfurther discussion.

0 0


13/43

7 A CPI R

M I R E T : T F CN 2012

The corollary to the lower value of the variablerate debt is that equity is actually taking onmore risk. With a xed price PPA, it is easyto see how an unexpected rise in the interestrates could consume all of the cash ow from

a project and wipe out the equity investor.While we have no accurate way of measuringthe cost of the additional risk assumed by theequity investor, we assume that the risk theytake on is equal to the value of the swap used tocompare India debt to U.S. debt. The result is anexact offset between debt and equity, movingexpected return from the equity column to thedebt column.

Finally, as shown in gure 3-1, the cost of equityin India is only slightly higher than in the U.S. orEurope, depending on the technology, despitethe higher underlying country risks. The slightlyhigher Returns on Equity add only 3% and 2%

respectively to the total cost of nancing thesolar and wind project.

When all of the adjustments are made to account forthe differences in terms and tenors, equity ends upbeing less expensive than in the U.S. or Europe, despitethe higher country risks. Meanwhile, the total impact ofdebt, including terms and costs, is 24-32% added to thecost of the project.

Figure 3-2: The impact of debt and equity costs and terms in India on overall nancing costs compared to a U.S. baselin

P e r c e n t a g e i n c r e a s e i n

I n d i a p r o j e c t c o s t s

a b o v e

U S p r o j e c t c o s t s

( % L C O E )

US LCOE

+30%

+20%

+10%

Debt cost

+19%

Shorterdebttenor

+6%Variable toxed debt

+7%

US LCOE

+20%

+10%

Debt cost and terms in Indiaadd 32% to cost of energy

-7%

Equityabsorbsvariabledebt risk

+3%Equitycost

Overall, equity cost reducesLCOE by 4% in India

Financing in Indiaadds 28% to aprojects LCOE

+28%

S O L A R P V

Debt cost and terms in India

add 24% to cost of energy

Financing in Indiaadds 22% to aprojects LCOE

+22%

O N S H O R E W I N D

P e r c e n t a g e i n c r e a s e i n I n d i a p r o j e c t c o s t s

a b o v e

U S p r o j e c t c o s t s ( % L C O E )

+10%

Debt cost

+10%

Shorterdebttenor

+4%

Variable toxed debt

-4%

Equityabsorbsvariabledebt risk

+2%

Equitycost

Overall, equity cost reduces

LCOE by 2% in India


14/43

8 A CPI R

M I R E T : T F CN 2012

It is signicant that, on an adjusted basis, projectReturns on Equity (ROEs) in India are below those inthe United States. This implies that equity investors arebuying into the market and accepting below rationalreturns for strategic reasons thus suggesting that

ROEs may rise once the market matures.Beyond the cost and availability of debt lie other distinc-tions with signicant implications for policymakers:

Are the cost and availability drivers for Indianrenewable energy a function of Indian nancialmarkets in general, or are they specic torenewable energy?

Do the drivers affect all investors, or onlyforeign investors?

Does the impact depend on the renewabletechnology, for instance affecting only wind, butnot biomass or vice versa?

Does the impact vary by state? That is, dostate-level policies and business environmentsaccount for differences in cost or availability ofinvestment capital?

Sections 4 and 5 focus on important issues that impactthe nancing of renewable energy in India now and inthe future. Eleven of these issues relate to the cost andavailability of debt (discussed further in section 4) andsix relate to the cost and availability of equity (Section5). In Section 6 we return to the other issues outlinedhere, including whether these issues are specic torenewable energy or characteristic of Indian Financialmarkets in general, whether foreign investors are spe-cically affected, and whether the impact is differentdepending upon the technology employed or the statein which a project is built.


15/43

9 A CPI R

M I R E T : T F CN 2012

4. Indias Achilles heel Highcost of debt4.1 The high general interest rate

environment in IndiaIndia is a rapidly growing country. As is typical forrapidly developing countries, growth in India comeswith a need for investment in new infrastructure, creat-ing competition to raise debt; and general inationarypressures that need to be controlled through higherinterest rates. As a result, benchmark interest rates inIndia are signicantly higher than in developed coun-tries. Furthermore, uncertainty around the Governmentof Indias future borrowing needs and the value of theRupee create a longer-term uncertainty that constrainsthe development of longer-term debt markets.

NEGATIVE IMPACT OF ISSUE:SEE

SECTIONNOW MEDIUM TO LONG TERM

C O S T A N D T E R

M S

High general Indian interest rateenvironment raises renewable project debt cost VERY HIGH VERY HIGH4.1

Longer tenor debt is generally unavailable MEDIUM HIGH 4.2

Fixed interest rate debt is rare; market relies on variable rate debt LOW HIGH 4.3

Debt is not strictly non-recourseas it is usually offered on a relationship basis MEDIUM MEDIUM4.4

Shortage of debt may raise its costin the longer term LOW MEDIUM4.5-4.10


Banks limit lending to any one sectorfor risk purposes LOW HIGH 4.6

Renewable energy debt is often classied within power or energy sectors, whichare already nearing sector limits

MEDIUM HIGH 4.6

Some banks are not lending to renewable projectsdue to unfamiliarity withrenewable energy policy and markets

MEDIUM HIGH 4.6

Limits on foreign debtmay restrict foreign lending LOW HIGH 4.7, 4.8

Technology riskis limiting debt to some technologies MEDIUM MEDIUM4.9

State-level policy, the poor nancial condition of the State Electricity Boards, andunfavorable local business environments restrict lending to some states

MEDIUM HIGH4.10, 7.2

Table 4-1: Summary of issues affecting debt cost and availability

The starting point for our analysis is the underlyingbenchmark interest rates for India versus other coun-tries. The impact of the nancial crisis is clear. Since2007, benchmark interest rates have fallen signicantlyin the developed world to stimulate the economy, but

have stayed relatively at in the rapidly developingeconomies. India is the only country whose benchmarkinterest rates are higher than they were in 2007. Indianinterest rates are now higher than each of the coun-tries in Figure 4-1 save Brazil and the gap with Brazilhas fallen from 12% in 2005 to 0.5% today. The higherbenchmark rates are, themselves, indicative of the fastgrowth in India and the multiple competing demandsfor borrowing to meet infrastructure needs and othergovernment borrowing.

The roughly 7-8% difference between benchmark inter-

est rates in India and the U.S. and Europe account for


16/43


17/43

11 A CPI R

M I R E T : T F CN 2012

government stopped supporting their fund-raising activities. After the liberalization, DFIshave had to compete with commercial banksto raise funds at market-determined rates. Thenancial market deregulation made the DFI

business model unviable and many were forcedto convert themselves into commercial banks.Given that the corporate bond market hasremained shallow, the reduced role of DFIs hasleft a gap in infrastructure funding.

One exception to this is the Indian RenewableDevelopment Agency (IREDA). IREDA, theinvestment arm of the Ministry of New andRenewable Energy (MNRE) is a governmentagency that offers loans to renewable energyprojects at favorable rates compared to

commercial lending. IREDA was established in1987 with objective of developing and nancingrenewable energy projects. IREDA has deployedover a billion U.S. dollars and has further com-mitments from the Asian Development Bank(ADB), Kreditanstalt fr Wiederaufbau (KfW),and development banks from Japan and Francefor another billion dollars. However, the scaleof IREDA is small compared to the challenge ofnancing Indian renewable energy ambitions.

4.3 Fixed interest rate debt is rareIn India, loans commonly have variable, rather thanxed, interest rates, primarily due to the issues outlinedin section 3.2: short-term lending by the banks due toasset-liability mismatch and the near-absence of bondmarkets. Long-term hedging instruments term-swapsand bonds are typically unavailable due to imma-ture nancial markets and risks related to a growingeconomy. Variable rate debt makes cash ows to equityholders, which include project cash ows minus theinterest payments to debt holders, less certain as they

are subject to changing interest expenses.While outside India, equity investors would not investin a project with variable rate debt or, at least, wouldexpect a premium equity return, it is not clear that thisis the case in India. Interviews with developers andpotential providers of xed rate loans suggest that thedemand in India for xed rate loans is not high. Part ofthe reason may be a general expectation that interestrates in India will fall presumably because they areso high already however, unlike in Brazil, forwardmarkets suggest that interest rates are far from certain

to fall in the future. Indeed, long-term borrowing bythe Indian government might be enough to ensure that

interest rates do not fall for some time. Thus, theremay be false expectations baked into the investmentassumptions regarding renewable energy projects.The fear would be that if these expectations change,required ROEs would need to increase or equity might

become otherwise scarcer or more expensive.4.4 Debt is not strictly non-recourseOne area of controversy is the extent to which projectnance exists for renewable energy in India. Projectnance, or non-recourse nance, is where money is lentor bonds are sold solely on the basis of the projectscash ows. Typically, the project is set up as a sepa-rate company and the loan is made to that company.Since the lender only has recourse to that project, if theproject company fails, the lender loses their investment.

The project owner, meanwhile, loses only their invest-ment in that specic project company. That is, thereis no recourse to the parent company of the projectdeveloper. Developers often prefer this type of nanc-ing because it is less risky to their company and theycan leverage the balance sheet of the parent companymore aggressively. Developers typically pay a signi-cant premium for project debt over recourse debt atthe parent company level, because the lower risk to theparent enables them to take on additional projects.

We had extensive discussions with stakeholders onwhether project nance exists in India. One extremeview is that pure project nance (i.e., non-recoursebased nancing) has not become popular in India. 9 Onthe other hand, many equity as well as debt investorsclaim that, though not common in the past, it is becom-ing more popular. 10 IREDAs lending practices supportthe view that non-recourse lending is becoming morecommon. In 2007 only 6% of IREDA lending was non-recourse, by 2011 this had grown to 55%.

Discussions with private-sector banks indicated that in

most cases some guarantees are required.11

However,these discussions also indicated that these guaranteesare mostly given by promoters in particular, for windprojects that may not have large balance-sheets.It looks like projects get approved only when there isa good promoter. Once loans are approved, they areevaluated on their own strength with some furtherdiscount for a better promoter. This is no different fromhow project nance happens in the West and it is safe

9 Discussion with Global Environment Fund (GEF)

10 Discussion with Infrastructure Development Finance Company (IDFC)Project Finance as well as L&T Infra

11 Discussions with ADB, ICICI Bank, Deutsche Bank (DB), IDFC, and L&T


18/43

12 A CPI R

M I R E T : T F CN 2012

to assume that project nance exists to some degree at least in practice in India.

Since relationships play a stronger role in Indian bankingthan in Western banking, exercising a non-recourseoption could have a very strong impact on a companysability to borrow money in the future and, maybe, evensurvive. However, again, this situation is not dissimilarto that in the U.S. or Europe. Therefore, although webelieve that the Indian debt probably has more recourseand, therefore, is less valuable than in the U.S., we havenot attempted to value this particular difference inIndian debt markets.

4.5 Availability of debtWe heard conicting accounts regarding the availabilityof debt in India. In general, larger companies with goodrelationships with banks have access to debt. Superiorprojects also seem to have access. However, marginalprojects with smaller, unconnected developers mightnd it more difficult. Likewise, some foreign developerswithout existing relationships might nd it difficult toraise debt in local markets.

However, our conversations indicate that access todebt is becoming more difficult now and will verylikely become even more difficult in the longer term forseveral reasons:

Banks have sector limits to limit their exposureto any one market, sector, or technology.As renewable deployment increases, morebanks are nearing their sector exposure limits.(Section 4.6)

Further, most banks in India include renewableenergy in their power, utility, or energy sectorlimits. These sectors have heavy borrowingdemand, to the point that many banks may havebeen near their limits even before lending torenewable energy could begin. (Section 4.6)

While many banks have sector limits, otherswill not lend to the renewable energy sector atall due to the novelty of the sector, immaturetechnology, and uncertain regulation. (Section4.6)

Sourcing debt from foreign lenders is restrictedby regulations that set caps on the amount ofmoney that can ow in as well as the pricing ofthat debt, which in some cases may be too lowfor international lenders. (Sections 4.7 and 4.8)

Finally, state-level issues, including the poornancial condition of many of the State

Electricity Boards (SEBs) who are the counter-parties to many of the contracts that pay therenewable energy projects, will restrict lendingin those states. (Section 4.10)

4.6 Many banks do not lend, or limitlending, to renewable energy projectsDomestic banks restrict lending ows to renewablepower projects which limits the availability of debt forrenewable energy projects. Our analysis reveals thatless than one third of public sector banks lend to renew-able energy projects. The situation is worse for privatesector banks where less than one fth lend to renew-able projects (Table 2-1). Banks cite non-familiarity withthe renewable energy sector as well as the perceivedriskiness as the major reason for not lending to renew-

able energy projects. Even among banks that lend tothese projects, the amount is restricted due the reasonsdiscussed below.

Commercial banks in India cap investments in infra-structure at 10-15% of their total domestic advancesbased on the Reserve Bank of India (RBI) prudentiallending norms. At present, the renewable energy sectoris coupled with the power sector, which is governedby (implicit and self-enforced) sub-sector limits inthe range of 4.5-5.0%. During the last few years, dueto large capacity additions primarily of coal basedpower projects commercial banks in India almostreached their lending limits for the power sector (Figure

0%

4%

8%

12%

16%

2005-2006

2006-2007

2007-2008

2008-2009

2009-2010*

2010-2011

Totalinfrastructuresector lending

Power sectorlending L

e n d i n g a s a

% o f t o t a l a d v a n c e s

Figure 4-2: Gross bank credit of scheduled commercial banks

*2009-10 data is not available. Source: Reserve Bank of India. Data are provisional and relate to select scheduled commercial banks which account for mor

than 95 per cent of bank credit of all scheduled commercial banks. Power seclending includes lending to generation, transmission, and distribution projects


19/43

13 A CPI R

M I R E T : T F CN 2012

4-2), potentially leaving limited funds for renewablepower projects.

Conversations with banks and developers conrmedthat they are worried not only about banks not lendingto renewable power due to power sector lending limitsbut also about the (resulting) increased cost of debt. InFebruary 2011, Andhra Bank noted that many banks areclose to exhausting their internal limits set for infra-structure rms. In April 2012, SBI Capital added thatthere is a need to categorize the solar power sector asa priority sector to increase funds available for solarpower development. Further, Solarsis noted that short-age of debt capital in the country is adding 50-100 basispoints to the variable component of the interest rate.

4.7 Limits on foreign lending - Capitalcontrols on foreign debtCompanies operating in the infrastructure sector and hence the renewable energy sector can obtainlong-term ECB loans of up to USD 500 million per year.Infrastructure companies can raise an additional USD250 million per year with government approval. Ouranalysis shows that the limit on ECB may not be impact-ing renewable energy projects at the moment, but theymay bind in the future, and the ECB ceiling may becomea barrier for renewable investments.

At the project level, assuming a typical debt-equityratio of 70:30, Indian wind and solar renewable projectshave reached only 55% and 63%, respectively, of theUSD 500 million ECB limit. 12 The largest wind projectin India so far (until March 2012) has been 300MW. Atan average cost of USD 1.3 million per MW, a 550MWwind project would surpass the ECB limit. Similarly, thelargest solar PV power project in India so far has been125MW. At an average cost of USD 3.6 million per MW,a project size of approximately 200MW would exceedthe ECB limit.

However, the situation is different at the holdingcompany level. In at least one case Beta Wind Farmsin 2010 the nancing requirement could have hit theECB limit if the company had tried to obtain cheaperforeign debt (Figure 4-3). Similarly, in 2011, MytrahEnergy got very close to the ECB limit. To the best of ourknowledge, neither of these companies has attemptedto obtain foreign loans yet.

12 Bloomberg New Energy Finance

4.8 Limits on foreign lending - Interest rateceilingsThe ow of foreign funds may be constrained due tointerest rate ceilings imposed by the government ofIndia on foreign loans. The ECB interest rate (all-in-cost

ceiling) is capped at six-month LIBOR + 300 bps13

forthree to ve year loans and six-month LIBOR + 500bpsfor loans longer than ve years. Our analysis indicatesthe ceiling for typical renewable project loans to beapproximately 11.8% (Figure 4-4). For many investors,these conditions may be so stringent as to make invest-ing in Indian renewable energy unattractive.

Against this effective 11.8% limit for the interest ratethat could be charged by foreign lenders, we comparethe cost of the foreign investor making that loan (inthe second column of gure 4-4). To the current Libor

13 LIBOR the London Inter Bank Offer Rate is used as a benchmark rate towhich a cap of 300 bps or basis points (or 3%) is added

800

0

400

200

600

0

400

200

600

0

400

200

600

2007 2008 2009 2010 2011 2012

I n s t a l l e d c a p a c i t y ( M W )

B W F P L

C L P

M Y T R A H

Estimated project sizeat which ECB amountceilings could bind

Figure 4-3: Sample of large wind power investors inIndia versus ECB ceiling, 2007-12

Note: Installed capacity is all projects combined (that attainednancial closure) undertaken by a company in a specic year.BWFPL: Beta Wind Farms Pvt. Ltd. (a subsidiary of Orient GreenPower Company Ltd.). CLP: CLP Power India Ltd. (a subsidiary ofCLP Holdings). Mytrah: Mytrah Energy Ltd. Source: BloombergNew Energy Finance.


20/43

14 A CPI R

M I R E T : T F CN 2012

interest rate, the lender would need to add 5.5% fora currency swap from Dollars to Rupees over the lifeof the project, and 2% for a term swap, to convert theshort-term Libor loan to a longer term (in this case, a10-year) loan. A typical project premium for an Indialender in India is 3.3%. Adding the impact of withhold-ing tax, the foreign lender would need 12.1% to make theloan attractive. The Indian lender could make the sameloan at 11.6%, but offering a variable rather than xedrate, or approximately 13.6% if it could offer a xed rateloan.

These numbers are central estimates, and there issome range, particularly around project premium. Thus,some foreign debt has been able to enter below the cap.However, we will note that given the Indian marketscurrent low value placed on xed debt, the foreignlending cannot compete with the variable rate debt oflocal borrowing.

The limits could increase the cost of renewable energyprojects if the ceilings restrict foreign loans even whenthese loans are less expensive than available domesticxed rate loans.

According to project developers and banks, the impactis real: a typical foreign loan may cost approximately 2%less than an equivalent domestic loan, and loans pro-vided by banks, such as U.S. EXIM bank and OverseasPrivate Investment Corporation (OPIC) could be even

cheaper. However, not many renewable projects inIndia are being nanced by foreign loans due to high

hedging costs, 14 as well as the interest rate ceiling. Theavailability of foreign loans is further limited by interna-tional lenders requiring a company (or project) ratingequivalent to Indias sovereign rating (BBB), in additionto requiring credit guarantees from the holding compa-nies. Higher international funding transaction costs alsomake it difficult for small project developers to opt forforeign loans. 15, 16

4.9 Technology risksThe availability of nance depends on the technol-ogy. Wind is typically easily funded, given its trackrecord and increasing competitiveness multiple largeprojects are getting nanced, and project nancingis becoming more prevalent after the introduction ofgeneration-based incentives.

On the other hand, solar is still new, so many banks are

cautious, and many banks are adopting an invest-a-littleand then wait-and-see strategy. As a result, projectsnanced solely by equity are more common in the solarPV space. The question that arises is whether solar PVwill mature fast enough such that banks will lend tothe projects before developers exhaust their patience,strategic intent, and nancial reserves.

14 Doors wide open for external commercial borrowings, Business Standard,March 17, 2012

15 Discussion with Acciona, a wind power developer.16 Further, at present Indian companies are reluctant to go for xed rate foreign

loan due to expectations of a fall in domestic interest rates as the bench-mark interest rate is currently at a high level last seen in 2000-01 (Source:tradingeconomics.com).

15%

10%

5%

0%

6 monthLIBOR

Interest rateceiling on

foreign debt Projectpremium

Term swap

15%

10%

5%

0%

F O R E I G N D E B T I N D I A N D E B T

Allowed rate Project debt cost

Term swap

Comparablexed rate

Variable rate

Figure 4-4: Estimated cost of debt for Indian renewable energy projects, 2012

CurrencySwap

Withholdingtax

Projectpremium

India RiskFree Rate


21/43


22/43

16 A CPI R

M I R E T : T F CN 2012

5. Equity Ample availability,but can it last?5.1 Cost of equity or required return on

equity (ROE) and differentiation betweentechnologiesIn a rational world, expected return on equity should behigher than the cost of debt outside exceptional cir-cumstances. Equity has decidedly more risk, and thusshould attract higher returns. The role of debt, in fact,is normally to concentrate the risks to the equity holderand therefore enhance equity returns. Another role is toreplace the equity capital of the developer so that theycan recycle the money into the next project.

The spread between equity costs and debt costsrepresents the allocation of risks and returns betweendebt and equity. As more debt is taken on, that is, asa project has more leverage, we should expect equityreturns to rise as risks are more concentrated. Weshould also expect to see some difference in debt equityspreads between technologies and markets.

For Indian wind projects, we see healthy spreads of4-7% between the cost of debt and the expectedreturn on equity that are similar to those found in othercountries. Wind seems to be the easiest technology toget funded, given that wind is competitive and relativelyrisk-free.

For solar, the observed spreads appear to be low(0-3%). Thus equity investors are taking on more riskby assuming debt, but, due to the high cost of debt, dolittle to enhance their returns. On the face of it, at thesereturn levels, equity investors might not be interested in

the solar PV sector at all. However, interviews suggestthat a few considerations might be driving investment:

Strategic positioning. 18 Equity investorsin solar appear to be trying to gain marketshare to emerge as a dominant player oncethe technology matures in the long term.However, the strategy of accepting lowerreturns in the short term to capture the marketis not sustainable over time, because the highrisk investment environment in India will requirecorrespondingly higher expected returns. In the

long run, the ow of information in the market about the risk prole of projects will helpin obtaining more rational risk pricing.

Equity returns could be increased as theproject progresses. 19 Developers sometimesintentionally delay buying modules to enhanceequity returns when they expect a futurereduction in cost of solar modules. This hasresulted in increasing the internal rate ofreturn from the initial expected value of 15%to 20-25%. Developers may also invest in a

project with lower initial returns expecting

18 Discussion with Astoneld Renewables.19 Discussion with Astoneld Renewables.

NEGATIVE IMPACT OF ISSUE:SEE

SECTIONISSUE NOW

MEDIUM TO LONG TERM

C O

S T A N D

T E R M S Equity returns required by investors are reasonable for good projects NONE LOW 5.1

Required returns depend on technologymaturity and developer strategy NONE MEDIUM 5.1


Equity is generally availablefor good renewable energy projects NONE MEDIUM5.2

Equity is available from foreign investorsbut facing a shortage of attractiveprojects

LOW MEDIUM 5.2

Equity availability is heavily skewed towards a few states with good policyregimes and attractive business environments

MEDIUM MEDIUM4.10, 7.2

Lack of debt may reduce available equity in the medium term as equity investorscannot recycle investments from current projects to future investments LOW HIGH 4, 5.2

Table 5-1: Summary of issues affecting equity cost and availability


23/43


24/43

18 A CPI R

M I R E T : T F CN 2012

6. Summary of the state ofrenewable energy nanceand rst indications of wherepolicy mattersSection 5 suggested that, at least for now, there are

few problems with the cost or availability of equity.However, equity for solar projects may become moreexpensive as markets mature and strategic investingdiminishes; meanwhile, the availability of equity for alltypes of renewable energy may decrease if a lack ofdebt causes equity providers to exhaust their capitaldue to the lack of recycling options.

The high cost of debt, however, as in Section 4, is animmediate problem that is exacerbated by the shortdebt tenors and variable interest rates. If interest ratesfall, tenors and variable rates are likely to becomeincreasingly important. The availability of debt is not yeta major problem, but signs are that debt will becomeincreasingly hard to secure as sector limits and riskcontrols begin to bite.

In section 3, we suggested that policy makers shouldidentify the most signicant policy areas affectingrenewable energy nance and focus their attentionthere. We identied ve sets of policy areas that affectthe cost and availability of debt and equity for renew-able energy:

Policies related to general Indian nancialmarket characteristics

Specic renewable energy policies or having asingular impact on renewable energy

Policies relevant to foreign investment inrenewable energy

Policies with different impacts on differentrenewable energy technologies

Indian state-level policies

In tables 6-1 and 6-2, we repeat summaries of equityand debt issues laid out in chapters 4 and 5. However,here we compare the importance of these issuesagainst the policy areas that they most inuence. Table6-1, which looks at the near-term issues, shows that thesingle biggest issue lies with general Indian nancialmarkets. In other words, improving renewable energynance cannot be done through renewable energypolicy alone. Concerted effort dedicated to reduc-ing renewable energy debt costs is a critical step that,today, overwhelms other policy concerns.

Table 6-2 moves us to the longer term. In the longerterm, even if, or especially if, the problem of high inter-est rates is solved, we see a number of policy areasincreasing in importance.

Policy makers will need to focus on specicrenewable energy policy, for instance, by xinguncertainty and risk surrounding renewablespecic policies such as the Renewable EnergyCerticate market. (see section 7.1)

As the current debt providers reach their limits,foreign debt providers and equity investorswill become more important, suggesting alook at the impact of debt cost and quantityrestrictions.

As investors become overexposed or otherwiseexhaust opportunities in the states withattractive policies and business environments,growth in renewable energy will depend on

states with less attractive regimes, suggestingthat these states need to be made moreattractive either through state-level actionor through Government of India policy. (Seesection 7.2)

Where should policy makers turn to rst to solve policy dilemma related to debt and equity costand availability for nancing renewable? Building on the analysis of Sections 3, 4 and 5, we ndthat the most pressing problems lie with characteristics general to all Indian nancial markets.However, in the longer term, if solutions to bridge general Indian market conditions are found,then specic renewable energy policy and state-level policy issues will also need to be addressed.


25/43

19 A CPI R

M I R E T : T F CN 2012

This page intentionally left blank.


26/43

20 A CPI R

M I R E T : T F CN 2012

ISSUE PERTAINS TO:

G E N E R A L I N D I A N

F I N A N C I A L M A R K E T S

S P E C I F I C

T O

R E N E W A B L E E N E R G Y

F O R E I G N I N V E S T M E N T

I N R E N E W A B L E S

S P E C I F I C R E N E W A B L E

T E C H N O L O

G I E S

S P E C I F I C I N

D I A N

S T A T E S

NEGATIVE IMPACT NOWEQUITY

C O S T A N D

T E R M S Equity returns appear reasonable for good projects None

Technology maturity and strategic positioning affectsrequired project returns

None


Equity is generally available None

Equity from foreign investors is also available Low

Equity availability is heavily skewed towards a fewstates

Medium

Lack of debt may reduce available equity in themedium term

Low

Table 6-1: Relevance of major issues to specic policy arenas Current Impact

DEBT


High general Indian interest rate environment Very High

Longer tenor debt is generally unavailable Medium

Fixed interest rate debt is difficult to nd Low

Debt is usually offered on a relationship basis Medium

Shortage of debt Low


Banks place limits Low

Renewable energy debt is often included within poweror energy sector limits

Medium

Some banks are not lending to renewable energy Medium

Limits on foreign debt Low

Technology risk Medium

State-level policy Medium


27/43

21 A CPI R

M I R E T : T F CN 2012

ISSUE PERTAINS TO:

G E N E R A L I N D I A N

F I N A N C I A L M A R K E T S

S P E C I F I C

T O

R E N E W A B L E E N E R G Y

F O R E I G N I N V E S T M E N T

I N R E N E W A B L E S

S P E C I F I C R E N E W A B L E

T E C H N O L O

G I E S

S P E C I F I C I N

D I A N

S T A T E S

NEGATIVE IMPACT IN

MEDIUM TO LONG TERMEQUITY

C O S T A N D

T E R M S Equity returns appear reasonable for good projects Low

Technology maturity and strategic positioning affectsrequired project returns

Medium


Equity is generally available Medium

Equity from foreign investors is also available Medium

Equity availability is heavily skewed towards a fewstates

Medium

Lack of debt may reduce available equity in themedium term

High

Table 6-2: Relevance of major issues to specic policy arenas Medium Term to Long Term Impact

DEBT


High general Indian interest rate environment Very High

Longer tenor debt is generally unavailable High

Fixed interest rate debt is difficult to nd High

Debt is usually offered on a relationship basis Medium

Shortage of debt Medium


Banks place limits High

Renewable energy debt is often included within poweror energy sector limits

High

Some banks are not lending to renewable energy High

Limits on foreign debt High

Technology risk Medium

State-level policyHigh


28/43

22 A CPI R

M I R E T : T F CN 2012

So what can policy do about the renewable energy nancing challenge? To begin to answer that question,in this section we summarize what has been done so far and whether those policies have been successful.

7. Government policyframeworkThe Electricity Act of 2003 transformed the powersector in India by driving changes such as deregulat-ing power generation, opening access in transmission,and allowing the state electricity regulatory commis-sions to x the level of renewable energy procurement.The National Electricity Policy (2005) and Tariff Policy(2006) followed, with the goal of increasing the share ofrenewable energy in the total energy supply mix.

Currently, the Indian government supports the develop-ment of renewable energy through a variety of incen-tives and mandates. These are described in Figure 7-1,which also maps each policy to the applicable technol-ogy wind or solar according to their relevance.

As introduced in Section 2.1, we have assessed theimpact of policies discussed above on the delivered costof electricity for actual wind and solar projects (Figure7.1). We rst calculate the counterfactual LCOE in

absence of any incentives using underlying project-levelparameters (e.g., debt-rate, debt-tenor, and hurdle rate)while maximizing debt. This counterfactual LCOE repre-sents the lowest cost of electricity possible from theseprojects. We then add various incentives in a sequentialmanner and recalculate LCOE, using the same method i.e., debt maximization as before.

Given that we have focused on actual projects, we startwith accelerated depreciation for solar PV and the gen-eration based incentive (GBI) for wind. We observe thatthese policies reduce the LCOE of representative solar

PV and wind projects by 18% and 10%, respectively.The approximate impact of both income tax benetsand CDM turns out to be 5%. Given that the support ofaverage power procurement cost (APPC) towards LCOEis 16% and 50% for solar PV and wind, respectively;the rest of the LCOE 67% and 40% for solar PV andwind, respectively is supported by the preferentialtariffs.

7.1 Renewable Energy Certicate marketsWith the expiration of the generation based incentive(for wind and solar) and the lapse of accelerated depre-ciation (for wind), the Indian government is expectingthe Renewable Energy Certicate (REC) market to takeup the slack. The rationale for the new market mecha-nism is that, although India has huge renewable energypotential, 21 its sources are geographically dispersed; forexample, in Delhi, the potential for renewable energy isinsignicant, whereas some states have excess renew-able energy sources (e.g., wind in Gujarat). However, instates with abundant renewable sources, the incremen-tal cost of renewable energy above other, conventional,sources discourages local distribution licensees frompurchasing renewable power beyond the level man-dated by state policy. Thus, there should be an oppor-tunity to reduce the costs of renewable energy for Indiaas a whole by tapping the additional, relatively low-costopportunities in states rich with renewable resources.

To address this mismatch and to reach the ambi-tious national level targets, The Government of India

launched a market based mechanism in 2011 in the formof Renewable Energy Certicates (REC). Under thismechanism, certicates are issued to renewable energypower generators, which can be sold later in recognizedpower exchanges. 22 RECs have been widely touted bymany analysts as the solution to drive investment intorenewable energy generation.

However, our look into the actual performance ofREC market trading shows that the current number ofcerticates issued is less than 4% of the technical RECdemand potential, indicating that the full potential ofREC markets is far from being realized. Our analysis,detailed in a companion report entitled, Falling Short:An Evaluation of the Indian Renewable CerticateMarket, clearly indicates that RECs are not consideredviable nancial instruments by investors yet, and severalchanges are needed if the REC mechanism is to deliverits intended results (Table 7-1). In our analysis, we con-clude that the REC mechanism, as currently structured

21 Renewable Energy Potential of India is estimated at 84,776 MW, includingWind (45,195 MW), Small Hydro (15,000 MW), Biomass Power (16,881MW), Cogeneration (5,000 MW), and Waste to Energy (2,700 MW) (IREDA2009).

22 Feed-in tariff also co-exist in India and developers have to choose either FITsor RECs


29/43

23 A CPI R

M I R E T : T F CN 2012

Figure 7-1: Contribution of policies and market prices to overall renewable energy remuneration

Renewable energy certicates (RECs):RECs are market-based instruments to address the mismatch betweenavailability and requirement of the obligated entities to meet their state-level renewable purchase obligation (RPO).Developers have a choice between using preferential tariffs or RECs.

Other benets (excise, wheeling charges):The Government of India provides concessional rates for excise (reducedfrom 8% to 0%) and customs duty (reduced by 2.5%-5%) for specic renewable sources of energy including wind,solar, and biomass. Several states in India levy relatively lower wheeling or transmission charges for renewable energy.

Feed-in (or preferential) tariffs (FIT): FITs aredetermined in a cost plus manner; and involve longcontracts (20-25 years), priority purchase, and priorityaccess to the grid. With the exception of JNNSM, FITs aredeclared by State Electricity Regulatory Commissions(SERCs).

Income tax exemption:TheGovernment of India allows a 100%tax waiver on prots for any single10-year period during the rst 15years of the operational life of apower generation project.

ACCEL.DEP. GBI

INCOME TAXEXEMPTION

CLEANDEVELOPMENT

MECHANISM

0

-25%

-75%

-50%

PREFERENTIALTARIFFS (E.G. FIT)

Solar

Wind

Accelerated depreciation:The Government ofIndia allows renewable (including wind and solar)projects to depreciate 80% in the rst year.

Generation based incentive:The Government ofIndia offered a generation based incentive (GBI)as an alternative to accelerated depreciation.

Clean DevelopmentMechanism: Projectdevelopers are free to gettheir project registered withMinistry of Environmentand Forests to participate incertied emission reduction(CER) credits markets.

E ff e c t s o f p o

l i c y m e c

h a n i s m s o n

t h e c o s t o f e n e r g y

( % L C O E )

-67%

-40%

-18%-10% -5% -6%

-4% -5%

ACCEL. DEP.

GBI

1995 2000 2010

FIT

RECS

INC. TAX EXEM.

OTHER

CDM

CentralState

Policy TimelineSolarWind

State

and executed, is unlikely to achieve most of the govern-ments objectives for the mechanism.

In particular, to create a stable market for RECs andreduce demand uncertainty, stricter compliance lawsand enforcement of state-level RPOs will be required.

Further, to make RECs viable nancial instruments, thegovernment will need to not only declare state-level,long-term targets along with their annual targets but

also enable well-functioning secondary markets.


30/43

24 A CPI R

M I R E T : T F CN 2012

7.2 State-level policy issuesThe Electricity Act of 2003 empowers State ElectricityRegulatory Commissions (SERCs) to establish poli-cies and rules for renewable energy development. TheGovernment of India has mandated renewable purchaseobligations (RPO) targets and has established guide-lines for setting RPOs. Accordingly, SERCs determinethe RPO targets (a solar RPO and a non-solar RPO) andalso establish state agencies responsible for enforce-ment of these targets. The obligated entities usuallyinclude distribution companies, 23 captive consum-ers, and open access users. 24 They can meet the RPOthrough generating renewable power themselves, bybuying from other renewable power generators, orthrough REC certicates.

Despite this obligation and a host of supporting policiesat the state and national level, there has been a widevariation between states. For example, wind capacityis more than 1,000 MW in states such as Tamil Nadu,Maharashtra, Gujarat and Karnataka, but less than50 MW in states such as West Bengal and Kerala. Formore detailed analysis, see Appendix 3.

Variation between states is explained by at least threefactors as indicated by our analysis in Appendix 3:

Resource potential Some states have morerenewable energy potential than others. Forexample, Rajasthan and Gujarat are the toptwo states in terms of solar irradiation in thecountry; they also hold the top two spots interms of solar deployment, accounting forthe majority portion of the solar PV installedcapacity in the country. Conversely, resourcesonly explain part of the story for example,Gujarat has the highest wind energy potentialbut less than half the installed wind capacity ofTamil Nadu.

Policy environment States have taken a rangeof stances towards renewable policy and RPOtargets. Table 7-2 shows the wide variationof RPO targets between states, and indicateshow some states have yet to approve targets,while others have reduced their targets. Further,enforcement of RPO targets has been scattered,as have state-level policy support mechanisms.

23 State owned distribution companies operate 95% of the power distribution.24 Many electricity intensive industries like Cement, Steel, Ferro Alloys, Paper &

Pulpare operating their own power plants run by either thermal generation

or generation from other resources including renewable energy. In 2007-08nearly 15% of the installed capacity in the country was in captive powerplants.

However, there are also many examples ofstrong and proactive policies towards devel-opment of renewable energy. For example,Gujarats preferential tariffs for solar powerhelped the state install approximately two thirds

of the total solar PV in the country by 2012,before the announcement of JNNSM.

Perceived counterparty risk State electricityboards (SEBs) or similar institutions typicallybuy power from renewable energy projectsthrough PPAs. However, many SEBs are inbad shape nancially. According to CRISIL, ananalytical rm and rating agency, the aggregateaccumulated losses of state-owned powerutilities are estimated to reach INR 2 trillion bythe end of March 2012, 25 with three fourths of

these losses in the last ve years (2007-12).The result is that developers and lenders alikeperceive risk associated with PPAs becauseof the potential for non-payment by the SEBor renegotiation of the terms of the contract.Indian banks continue to be wary of lendingto renewable energy projects that signedPPAs with state-owned electricity distribu-tion companies in states where SEBs are in badshape nancially.

Other factors, undoubtedly, also contribute to varia-tion between states. However, our analysis, set outin Appendix 3, indicates a signicant correlation witheach of these three factors. The result is that Indiarenewable energy deployment is concentrated in a fewstates. While the performance of these states has beensufficient to drive growth at the national level, achievingtargets will require signicant deployment across moststates in India. We have had several conversations withinvestors that indicated that their exposure to well-performing states such as Gujurat was too high and thatthey were actively looking to expand the geographic

range of their Indian portfolio, but were nding it hard todo so.

25 Power distribution companies losses cross Rs 2 lakh crore, says Crisil, TheEconomic Times, May 7, 2012


31/43

25 A CPI R

M I R E T : T F CN 2012

Government Objectives /Evaluation Criteria

Findings

Encourage cost reduction in renewable energy projectsby promoting market forces and competition

Unlikely to achieveParticipation in the REC markets has been too low to drive any costreduction

Provide incentives to drive capital investment inrenewable energy projects

Unlikely to achieveTime frame of RECs is much shorter than the investment horizon;investors discount RECs due to perceived uncertainty and risk overproject life

Provide a mechanism to limit boom and bust cyclesNot clear yetCannot test as the renewable energy market has not yet overheated;current participation and incentive levels suggest REC mechanism insuf-cient to dampen cycles

Weave together various state-level incentive and policy

regimes within a national structure

Unlikely to achieve

Does not incentivize states to work towards reaching national goalsProvide incentives incremental to other relevant policiesNot clear yet

REC cash ows are supplementary to other policies and can work inconjunction with state policies, but have not to date

Allow for technologically differentiated incentives tosupport new and diversesources of energy

Not clear yetDesign allows for differentiation, however support for new renewablesources has been weak to date

Accomplish the above at a reasonable additional transac-tion cost

Not clear yetDirect transaction cost is low to date; it is unclear how costs will evolveas market matures

Accomplish the above with a reasonable additional costdue to higher perceived or real risks to developer

Unlikely to achieveRisk perception of using RECs has been high, failing to drive anyinvestment

Table 7-1: REC policy analysis

Source: Falling Short: An Evaluation of the Indian Renewable Certicate Market, CPI (2012)


32/43

26 A CPI R

M I R E T : T F CN 2012

STATE YEAR OF

FIRSTREGULATION

PRE-REC TARGET POST-RPO TARGETS

2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16

Andhra Pradesh (Draft)* 2005 5.0% 5.0% 5.0% 5.0% 5.0%

Assam 2010 1.4% 2.80% 4.20% 5.6% 7.0%Bihar 4.0% 1.5% 2.50% 4.00% 4.5% 5.0%

Chhattisgarh 2008 10.0% 10.0% 10.0% 5.0% 5.25% 5.75%

Delhi (Draft)* 1.0% 1.00% 1.00% 2.0% 3.40% 4.80% 6.2% 7.6% 9.0%

Gujarat 2005 1.0% 2.0% 5.0% 6.0% 7.0%

Haryana 2007 2.0% 5.0% 10.0% 1.5% 2.0% 3.00%

Himachal Pradesh 20.0% 20.0% 20.0% 10.01% 10.01% 10.25% 10.25% 10.25% 10.25%

Jammu & Kashmir 1.0% 3.0% 5.00%

Goa & UT 1.0% 2.0% 3.00%

Jharkhand 2.0% 2.50% 3.10%

Karnataka 2008 1.0% 1.00% 1.0% 0.25% 0.25% 7.25% 7.25% 7.25% 7.25%

Kerala 2006 5.0% 5.0% 5.0% 5.25% 5.25% 5.25% 5.25% 5.25% 5.25%

Madhya Pradesh 2008 10.0% 11.0% 0.80% 2.50% 4.0% 5.50% 7.0%

Maharashtra 2006 4.0% 5.0% 6.0% 6.0% 7.0% 8.0% 9.0% 9.0% 9.0%

Manipur 2010 2.0% 3.0% 5.0%

Mizoram 2010 5.0% 6.0% 7.0%

Meghalaya 2010 0.50% 0.75% 1.0%

Nagaland 6.0% 7.0% 8.0%

Orissa 3.0% 3.0% 4.0% 5.0% 5.0% 5.50% 6.0% 6.50% 7.0%

Punjab 2007 1.0% 1.0% 2.0% 2.4% 2.86% 3.44% 3.94% 4.0%

Rajasthan 2006# 4.28% 6.25% 7.5% 8.5% 9.50%

Tamil Nadu 2006 10.0% 10.0% 13.0% 10.15% 9.05%

Tripura 2010 1.0% 1.0% 2.0%

Uttarakhand 5.0% 5.0% 8.0% 10.0% 11.0%

Uttar Pradesh 7.5% 7.5% 7.5% 4.0% 5.0% 6.0%

West Bengal 2005 4-6.8% 2.0% 3.0% 4.0%

Table 7-2: RPO targets by state


33/43

27 A CPI R

M I R E T : T F CN 2012

8. Policy analysis Casestudies and internationalcomparisons

8.1 Case study analysis India versus theU.S. and EuropeIn a 2011 study entitled The Impact of Policy on theFinancing of Renewables A Case Study Analysis CPIstudied a series of renewable projects in Europe andthe U.S. to evaluate the way policy can impact the costof nancing renewable energy through different path-ways. In this context, we dened pathways as generalcharacteristics of policy that could affect investor cashows or perceptions. Specically we look at pathwaysthat included (see Appendix 1 and Appendix 2):

Duration of revenue support That is, how longa preferential tariff, power purchase agreement,or other nancial support mechanism wouldlast. Generally, a longer support mechanismallows debt to be amortized more slowly withthe result of higher effective leverage and lowernancing costs across the project life;

Revenue certainty Here we looked specicallyat the impact of having a xed level of supportor one that varied as a function of markets or

commodity prices. A xed level of supportoffers more certainty of cash ows and allowsgreater leverage;

Risk perceptions We looked at the rangeof risk premia applied to different renewableprojects to ascertain the nancial costsassociated with riskier perceptions;

Completion certainty Here we looked at thecost of delays to a project caused by policy orregulatory hurdles delaying project completion.

A project delay increases nancing costs due tothe delay in receiving the nancial return.

Cost certainty Policy could also lead touncertainty in costs; for example, by imposingadditional requirements during the constructionphase.

We analyzed each of these pathways through detailed

nancial modeling of representative onshore wind, solarPV, and a more innovative technology each in Europeand the U.S. and modeled alternative policy scenariosfor each of the projects to determine the impact thatchanges to key policy pathways would have had toproject nancing costs. Figure 8-1 summarizes theresults of the study.

Our analysis indicated that in the U.S. and Europe,extending the length of a support policy could lowercosts. We also found that mechanisms such as feed-in(or preferential) tariffs offering constant, stable prices

could also lower costs, as could mechanisms withvariable support, but appropriate and well-designedprice oors, such as feed-in premia with oor prices.Meanwhile, cost and completion certainty could besolved mainly through already commercially availablecontracting arrangements.

In 2012, CPI decided to repeat this analysis for a setof India projects to see if the lessons were universallyapplicable, or whether there were additional insightsto emerge from the analysis. Much of what is includedearlier in this paper stems from this analysis. For India,we studied three projects in detail (see Table 8-1).Figure 8-2 presents the results of the same analysis forthe Indian projects listed in Table 8-1.

Two points emerge from the analysis:

Differences between the policy impactpathways are smaller in India. This is becauseone important mechanism for reducingnancing costs is reducing risks to allowincreased debt and project leverage. With debtcosts so high in India, the value of leverage (andtherefore reducing risks) is lower. Secondly,some of t

Documents

Meeting-Indias-Renewable-Targets-The-Financing-Challenge.pdf