Winning the Yieldco Confidence Game: Long and Short Opportunities Among the Yieldco Stocks

admin
Print Friendly, PDF & Email

______________________________________________________________________________

Eric Selmon Hugh Wynne

Office: +1-646-843-7200 Office: +1-917-999-8556

Email: eselmon@ssrllc.com Email: hwynne@ssrllc.com

SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES

_________________________________________________________________________________________

June 20, 2017

Winning the Yieldco Confidence Game:

Long and Short Opportunities Among the Yieldco Stocks

In this note, we assess the value of the various yieldcos in two ways: first, based on the cash available for distribution, after taxes and debt service, from their existing portfolio of assets assuming no growth and, second, based on the assumption of continued asset acquisitions and cash flow growth, for those companies that have the ability to fund asset purchases through the issuance of equity on terms that are accretive to cash flow per share. For the sake of brevity, we refer to the first valuation as DCF valuation and the second as DDM. In both sets of valuations, we have taken care to model explicitly the limited duration of the tax shield available to the yieldcos from the accelerated depreciation of their acquired assets; when asset portfolios cease to grow, tax rates can be expected to rise, and cash available for distribution to fall materially, once the last asset acquired has been fully depreciated. Despite the conservatism of our assumptions, we believe the present value of future cash distributions materially exceeds the price of NextEra Energy Partners (NEP) at discount rates of 8% or less, and are keeping the stock on our list of favored names among the yieldcos. Conversely, we find 8Point3 Energy (CAFD) to have little potential for accretive asset acquisitions, and believe the stock to be materially overvalued based on distributions from existing assets. We have added CAFD to our list of least favored yieldcos. For a summary of our valuation analyses of the yieldcos, please see pages 9-11and Exhibit 5.

Portfolio Manager’s Summary

  • We have assessed the value of each of the various yieldcos in two ways:
  • First, in a no growth scenario, where our valuation is based upon the cash available for distribution, or operating cash flow after taxes, maintenance capex and debt service, from each yieldco’s existing portfolio of assets and given its current outstanding debt.
    • We found no yieldco to be attractive on this basis at any discount rate of 7% or higher.
  • Second, based on the assumption of continued asset acquisitions and cash flow growth for those companies that have the ability fund asset purchases through the issuance of equity on terms that are accretive to cash flow per share.
    • NEP’s DDM value is attractive at discount rates of 8% or lower.
    • CAFD is extremely overvalued on both DDM and DCF valuations. We forecast its dividends will total less than its current share price, even without discounting.
  • Among yieldcos, currently only NEP and Pattern Energy Group (PEGI) are able to issue equity to fund the acquisition of projects on terms that are accretive to cash flow per share. NEP can do so in respect of projects whose CAFD (cash available for distribution) yields are as low as of 8%, while PEGI can do so in respect of projects whose CAFD yield is 9% or higher.
  • In both our DCF and DDM valuations, we have taken care to model explicitly the limited duration of the tax shield available to the yieldcos from the accelerated depreciation of their acquired assets.
  • Most yieldcos are subchapter C corporations that must pay tax at the corporate level. However, as long as yieldcos are in a position to fund their growth with new share issuance, and to invest the proceeds in acquisitions of assets from their sponsors, their cash taxes are likely to be near zero.
  • Yieldcos’ tax basis in the assets they acquire is the purchase price. By applying MACRS depreciation schedules, yieldcos can depreciate this tax basis on an accelerated basis.
  • Yieldcos that are continuing to acquire assets can thus shelter their income from taxation for many years. However, when yieldcos cease to grow, their existing assets will eventually be fully depreciated. Generally, yieldcos can shield their income from taxes for ~10 years from their last significant acquisition.
  • Once a yieldco has fully used up the tax benefits from accelerated depreciation, and the carry forward of prior years’ tax losses, its taxes will rise to reflect the federal corporate tax rate of 35% applied to their pre-tax income, with state taxes on top of that. The impact of these rising taxes on yieldcos’ cash distributions will be even greater than their impact on pre-tax income: we have found that taxes can reduced cash available for distribution by as much as 80% and for the companies in our yieldco universe it generally causes a 50-60% reduction at the time when cash taxes become significant.
  • The increase in taxation faced by yieldcos when their current tax shield expires has a material impact on their value. Discounted over 20 years at 8%, the present value of the yieldco’s cash distributions, after adjustment for the increase in taxes, is 18% lower than their present value absent any taxation. Adding the impact of taxation in the calculation of the yieldco’s terminal value widens the gap to 22%.
  • Despite the conservatism of our assumptions, we believe the present value of future cash distributions materially exceeds the price of NextEra Energy Partners (NEP) at discount rates of 8% or less (please see Exhibit 5 and our analysis of NEP on page 10). We are keeping the stock on our list of favored names among the yieldcos.
    • Atlantica Yield (ABY) is moderately attractive on DDM at discount rates of 8% of lower (Exhibit 5), but the additional risk from lower transparency of project economics and its holdings in projects in several less developed economies prevent us from adding it to our list of preferred yieldcos.
    • Pattern Energy (PEGI) could also be attractive at discount rates of less than 8% were it willing to issue equity more aggressively to fund more rapid growth. Such a change in strategy is not anticipated by the market; were PEGI to adopt it, therefore, the stock could offer potentially greater medium term returns than either NEP or ABY as PEGI is revalued to reflect its improved growth potential. The company’s current strategy is more conservative, however, with very limited equity issuance, resulting in a significantly lower DDM valuation (please see our discussion on page 11, and compare the rows labeled “PEGI” and “PEGI – shares” in Exhibit 5).
  • Conversely, we find 8Point3 Energy (CAFD) to have little potential for accretive asset acquisitions, and believe the stock to be materially overvalued based on distributions from existing assets (see Exhibit 5). We have added CAFD to our list of least favored yieldcos.

Exhibit 1: Heat Map: Preferences Among Utilities, IPP and Clean Technology


Source: SSR analysis

Introduction to Yieldcos: Virtuous or Vicious Cycles

Yieldcos are a recent investment vehicle, introduced with the initial public offering of NRG Yield (NYLD) in June, 2013. They were created by power project developers active in the renewable energy sector, including NRG Energy (creator and sponsor of NRG Yield), NextEra Energy (NextEra Energy Partners), FirstSolar and SunPower (8Point3 Energy) and Abengoa (Atlantica Yield, formerly Abengoa Yield). In each of these cases, the developer sponsoring the yieldco did so by creating a subsidiary and selling a minority interest in its stock through an initial public offering. The yieldco in turn used the proceeds of the offering to acquire from its parent a portfolio of operating assets, generally renewable power projects whose output had been sold under long term contracts. A key objective of this structure was to permit the sponsors to sell portfolios of

very low risk assets to a special purpose vehicle that, due to its low risk profile, could raise capital at a much lower cost than would be available to the yieldco’s sponsor. The sponsors could then redeploy the capital received from the sale of their lowest risk assets to higher risk but also higher return assets and businesses, including the development of additional renewable projects.

The difference in the risk profile of yieldcos and their sponsors is stark, and helps to explain the attractiveness of the yieldco structure. As a general matter, yieldcos are engaged in the business of operating renewable power plants and selling their output under long term contracts; their sponsors, by contrast, may hold large portfolios of higher risk generating assets, including nuclear and coal fired power plants, and engage in materially higher risk business activities, including merchant generation, energy marketing and power project development. Because most yieldcos focus on renewable generation assets, they are free of key operating risks faced by fossil fuel generators, including volatility in fuel prices, consequent changes in dispatch, the risk of disruption of fuel supply, and the cost of complying with future environmental regulations governing air emissions, cooling water intake or the disposal of combustion residuals. Yieldcos’ portfolios of renewable projects generally also enjoy much lower maintenance expense and materially lower operating risk than the nuclear, coal and gas fired power plants operated by some of their sponsors. Construction risk is a further distinguishing factor: yieldcos tend to own power plants that have already entered operation, and so are no longer exposed to the risk of construction delays or cost overruns. Finally, most renewable generation projects sell their output under long term contracts, often with terms of 15 years or more, at predetermined prices, and thus take no merchant power risk. Given their low construction, operating and market risks, the cash flows of yieldcos’ renewable energy portfolios can be predicted with a high degree of confidence, minimizing the risk to yieldco investors and, consequently, yieldcos’ cost of capital.

However, yieldco investors bear one critical risk that investors in their sponsors do not. Yieldcos are designed to access the capital market repeatedly at a lower cost than their parents, and to use the proceeds to make repeated purchases of assets from their parents’ project portfolios. Yieldcos do not have the capability to develop, design and build power projects on their own, and thus rely for their growth on their ability to fund the acquisition of existing renewable assets at an attractive cost. To the extent these assets can be acquired at cash yields that are expected to be higher than the expected cash yields on the yieldcos’ shares, the cycle of share issuance and asset purchase can be repeated on a basis that is accretive to the yieldco’s shareholders, enhancing their expected returns. But if the expected cash returns on these asset acquisitions falls below the cash yields expected on yieldco shares, the cycle of share issuance and asset purchases stops. No longer able to fund asset purchases on an accretive basis, yieldcos’ portfolio cease to grow, diminishing the expected returns to their shareholders. As yieldco share prices fall to compensate, the potential for accretive acquisitions becomes more remote. Under these circumstances, yieldco share prices have tended to stabilize at levels that reflect the present value of the cash flows to the yieldco from its current portfolio of assets. With the exception of NextEra Energy Partners, all yieldcos find themselves in this situation today.

The fact that yieldcos’ growth is contingent upon their ability to fund asset acquisitions on an accretive basis thus results in a counter-intuitive pattern of financial risk for their shareholders. A yieldco stock whose price is rising, and whose cash yield is falling as a result, becomes more rather than less attractive, as its declining yield increases the opportunities for the yieldco to raise capital and acquire new assets on an accretive basis. Conversely, a yieldco stock whose price is falling, and whose cash yield is rising, becomes less attractive, as the yieldco may find itself unable to raise and invest capital in new assets.

As a result, yieldco stocks can be highly volatile. A yieldco whose share price is rising is in a position to accelerate the growth of its assets through accretive acquisitions, and thus accelerate the expected growth in its cash distributions to shareholders – contributing to further share price increases. A yieldco whose share price is falling will see its opportunities for investment constrained and its growth rate fall, contributing to further price declines. (To date, such downward spirals in yieldco share prices have tended to reflect increases in their cost of capital, triggered, for example, by rising long term interest rates, or reduced expectations as to the growth of their renewable generation portfolios, reflecting, for example, lower forward prices for power, a material weakening of their sponsor or reduced tax incentives for wind and solar projects.) This polarization of potential outcomes, and the volatility it implies, may in time undermine the raison d’etre of the sector: to the extent that investors perceive the benefit from the stability of their cash flows is insufficient to offset the financial risk implicit in the yieldco business model, yieldcos could lose their ability to attract low cost capital to the renewable energy sector.

Valuation of Yieldcos

Yieldcos that can no longer issue shares and acquire assets on an accretive basis, such as 8Point3 Energy, are unable grow their portfolios; investors in these “busted yieldcos” must be content, therefore, with the cash available for distribution from their existing portfolio of assets. Because the renewable power projects in yieldco portfolios tend to be highly leveraged, these distributions must be made from after-tax operating cash flows net of principal repayment. The value of a busted yieldco’s stock is thus equivalent to the present value of the distributions from all the renewable projects owned by the yieldco over the remaining useful life of these assets.

Valuing yieldcos on this basis differs in certain important respects from a traditional discounted cash flow (DCF) valuation of a company. In the case of yieldcos, the key assumption underlying a DCF valuation, that the firm can distribute to shareholders its after-tax operating cash flow net of capex, is not met; because the assets of a busted yieldco are not being replaced, its existing debt cannot remain outstanding but must be repaid over time, implying that the yieldco must limit its distributions to after-tax operating cash flow net of maintenance capex and principal repayments.

Even in the case of a growing yieldco, the use of certain assumptions common to DCF valuations can lead to distortions. In most DCF analyses, for example, the terminal value of a company is a function of the organic growth rate it can achieve in perpetuity by reinvesting its free cash flow. As noted above, yieldcos lack the capacity to develop and build new renewable projects. It is industry practice, therefore, to pay out to the shareholders the bulk if not all of cash available for distribution. The growth of yieldcos’ asset portfolios is not a function, therefore, of their ability to reinvest cash to develop new projects; rather, it is a function of their ability to issue stock at a low cash yield and to acquire assets at a higher cash return. Because this cycle of financing and acquisition drives the growth in cash distributions, it needs to be modeled explicitly.

Recognizing the constraints described above, the valuation analyses in this note begin with 20 year forecasts of the cash available for distribution from each yieldco’s existing portfolio of assets and assuming the amortization of its outstanding debt. To estimate the value of the yieldco, we then calculate the present value of these forecast distributions, plus a no-growth terminal value based on estimated distributions beyond year 20 over the remaining useful life of the yieldco’s assets, at a range of discount rates reflective of the yieldco’s cost of equity. We estimate the cost of equity for yieldcos to be ~8% based on current 20 year US Treasury yields (2.52%), the current market equity risk premium of ~5.5% and a sector beta of ~1.0.

A further complication in the valuation of yieldcos is the fact that the cash taxes paid by yieldcos are likely to rise materially over time, substantially reducing the cash that can be distributed to shareholders. We believe these expected tax increases to be a key but often overlooked element in the valuation of yieldco stocks.

Taxes Can Reduce Yieldco CAFD by up to 80%

Most yieldcos are regular subchapter C corporations that must pay tax at the corporate level. However, as long as yieldcos are in a position to fund their growth with new share issuance, and to invest the proceeds in the acquisition of assets, their cash taxes are likely to be low. Yieldcos’ tax basis in the assets they acquire is the purchase price; by depreciating this tax basis on an accelerated basis, applying the Modified Accelerated Cost Recovery System (MACRS) permitted by the IRS, yieldcos can shelter their income from taxation for many years. Generally, allowing for the carry forward of net operating losses, yieldcos can shield their income from taxes for about 10 years from their last substantial acquisition.

Thereafter, yieldcos’ taxes can be expected to rise significantly, with a material negative impact on cash distributions to shareholders. Once a yieldco has used up the tax benefits from accelerated depreciation, and the carry forward of prior years’ tax losses, its taxes will rise to reflect the federal corporate tax rate of 35% applied to their pre-tax income, with state taxes on top of that.

The impact of these rising taxes on yieldcos’ cash distributions will be even greater than their impact on pre-tax income. Yieldco’s cash available for distribution is equivalent to their after-tax earnings, plus depreciation, less principal repayments. Among the yieldcos, debt amortization schedules show principal repayments to be equivalent to ~60-100% of current CAFD, suggesting that after-tax earnings plus depreciation must equal ~160-200% of CAFD. As long as the accelerated depreciation allows a yieldco to shield its income from taxation, the yieldco’s taxable income and after-tax earnings are the same. This implies that taxable income plus depreciation equals 160-200% of CAFD. When accelerated depreciation ceases, tax depreciation will fall to zero, and taxable income will rise commensurately; at that point, taxable income is equal to 160-200% of current CAFD. If taxes must now be paid at a 35-40% rate on taxable income, CAFD could be reduced by as much as 80% to 85%. Generally, as amortization of debt, and therefore the gap between CAFD and taxable income, declines over time, taxes reduce CAFD by 50-60% for the companies in our yieldco universe.

The increase in taxation faced by busted yieldcos when their current tax shield expires has a material impact on their value. As an example, consider a busted yieldco with flat pre-tax CAFD, debt amortization equal to 60% of CAFD, about the level of most yieldcos in year 11 absent growth, and a 35% tax rate. Discounted over 20 years at 8%, the present value of the yieldco’s cash distributions, after adjustment for the increase in taxes, is 18% lower than their present value absent any taxation even though the impact is not until year 11 and later. Incorporating the impact of taxation in the calculation of the yieldco’s terminal value widens the gap to 22%. The gap is larger at lower discount rates, which increase the present value of later year cash flows, and smaller at higher discount rates.

Takeover Targets: The Discounted Cash Flow Valuation

In this note we estimated the value of the yieldcos in two different ways:

  • First, in a no growth scenario, where our valuation is based upon the cash available for distribution, or operating cash flow after taxes, maintenance capex and debt service, from each yieldco’s existing portfolio of assets and given its current outstanding debt.
  • Second, based on the assumption of continued asset acquisitions and cash flow growth for those companies that have the ability fund asset purchases through the issuance of equity on terms that are accretive to cash flow per share.

For the sake of brevity, we refer to the first valuation as DCF valuation and the second as DDM.

The first step in calculating the DCF was to prepare a 20-year forecast of pre-tax cash available for distribution (CAFD) without growth. We started with each company’s disclosures of its current CAFD run rate adjusted for any acquisitions they were in the process of completing. We assumed that CAFD would change going forward to reflect (i) repricing of power sales after the termination of existing power purchase agreements (PPAs) and (ii) changes in debt service (debt amortization and interest) at both the projects and the holding companies. Term debt and convertible debt at the holdco were assumed to be refinanced and amortized over the remainder of the 20 year forecast period.

Next we calculated after-tax CAFD. We began by estimating the year in which the tax shield for each yieldco would be fully consumed, in the absence of additional acquisitions, exposing future earnings to taxes. When companies stated the remaining life of the deferred tax assets we used their guidance, otherwise we used 10 years. After that date, we estimated taxable income as pre-tax CAFD plus debt amortization. We then applied a 35% tax rate to the estimated taxable income and subtracted the result from pre-tax CAFD to obtain an after-tax CAFD estimate in future years. For Atlantica Yield (ABY), which is a UK corporation and has assets around the world, we used its average statutory tax rate of 30%.

We also estimated a terminal value, assuming no growth, to capture the value of distributions over the remaining life of the yieldco’s assets. For the terminal value, we used 50% of the 20th year’s CAFD divided by the discount rate, because by this time many of the PPAs will have been repriced to market and, based on our analysis, this reflects a reasonable approximation of the value of the assets over their remaining useful life (10-20 years).

We then calculated for each yieldco the present value of the 20-year forecast of CAFD, and the terminal value, at discount rates of 7%, 8,% and 9%. We estimate the cost of equity for yieldcos to be ~8% based on current 20 year US Treasury yields (2.52%), the current market equity risk premium of ~5.5% and a sector beta of ~1.0.

In Exhibit 2 we show our estimated DCF valuation for the yieldcos at the different discount rates. As noted earlier, the need to pay taxes starting around 2026 reduces the value of the companies by ~20%. At current prices, no yieldcos offer positive net present values on our DCF analysis. Within the past six months, however, ABY, NYLD and Pattern Energy (PEGI) have all traded at levels that would be attractive at discount rates below 7%.

In light of these valuations, we think it would be difficult for any yieldco to find a private buyer for the entire company. However, if a private buyer were willing to place additional value on the yieldco platform, itself, beyond the assets alone, and could obtain attractive financing, ABY, NYLD and PEGI would be the most attractive acquisition candidates. NEP is significantly overvalued on a no-growth DCF, implying that the current valuation is pricing in substantial continued growth.

Exhibit 2: Discounted Cash Flow Valuations

___________________________________________________________________________________________________________________________________________________________________________________________

Source: Company reports, SSR analysis

Which Yieldcos can make accretive acquisitions?

Because our DDM analysis is predicated on the potential for continued growth, we had first to determine which yieldcos could currently issue stock and use the proceeds to make accretive acquisitions. (See Exhibit 3) We calculated for each yieldco the accretion in CAFD per share from acquiring a project funded by equity issuance that would increase CAFD by 10%, using valuations based on 8%, 9% and 10% CAFD yields on the project. We assumed the yieldco’s shares would be issued at a 5% discount to the current share price.

Exhibit 3: CAFD/Share Accretion from Acquisition of 10% of Current CAFD

___________________________________________________________________________________________________________________________________________________________________________________________

Source: Company reports, SSR analysis

At CAFD yields of 9%, the level we assume for acquisitions in our forecast, we believe NEP is the best positioned to make accretive acquisitions. Allowing for the use of some additional holdco debt, however, NYLD and PEGI could be more aggressive in issuing equity to fund acquisitions than they currently are. We expect that, if they decided to pursue such a policy, they markets would likely react negatively initially, but with consistent execution and growth their valuations should increase and long-term investors would benefit.

Valuation for Shareholders: Dividend Discount Model

To estimate the value of the yieldcos as ongoing investment vehicles using a DDM, we prepared a forecast of each yieldco’s dividends per share for the next 20 years and then estimated the terminal value. To estimate the pre-tax CAFD generated by existing assets and debt, we used the same assumptions as for the DCF valuation above. Then we estimated the ability of yieldco to increase the dividend over time by growing CAFD per share through project acquisition.

We assumed that the projects acquired would be financed with common equity up to ~30-40% of the total project value. We assumed that the vast majority of acquisitions will be new wind and solar assets and that tax equity for half of the value of projects would be available for wind projects through 2020 and for solar through 2024.[1] Any remaining project cost was assumed to be financed with non-recourse project debt. The yieldco’s 30-40% equity investment was forecasted to be funded from either retained cash, holdco debt and, in the case of NEP, equity issuance. Although some recent deals have been completed at 10% CAFD yields, we forecast the purchase price of new projects to generate a 9% CAFD yield on the equity investment by the yieldco. To simplify forecasts, we assumed that the CAFD generated by projects was constant for 20 years, corresponding to the assume term of project PPAs, and that the project debt amortized in mortgage style payments, keeping debt servicing costs flat.

For most yieldcos, the ability to acquire new projects was limited in our analysis. Retained cash accumulated slowly as we assumed 85-90% of CAFD was paid out each year in dividends. We limited holdco debt to a total of less than 3.5x pre-tax CAFD plus the interest on the holdco debt, but by the 20th year we limited it to less than 3.0x. The exception was NEP, which we assumed issued equity to acquire new projects. We assumed that new shares were issued at the current yield on NEP stock, but applied a 5% discount to estimate the proceeds to NEP. We also did an additional analysis of NYLD and PEGI that assumed, contrary to the current policy of either company, that they began to issue equity more aggressively to fund acquisitions. Due to our expectation of a negative initial reaction from the market, we applied a 10% discount to the current valuation for share issuance during the first few years before reverting to a 5% discount similar to NEP.

Project acquisitions create a new basis for tax depreciation that extends the existing tax shield for each yieldco. Consequently, yieldcos with the capacity for sustained, large acquisitions should be able to shelter their taxable income with accelerated depreciation well beyond the expiration of their initial tax shield.

We incorporated all of this analysis into a single dividend forecast for each company. For all of the yieldcos except ABY and NEP we assumed an 85% payout ratio of CAFD for dividends, with some variation to allow for consistent growth or maintenance of the dividend over time. (See Exhibit 4) At ABY, based on the initial low payout ratio, declining debt service expense and long duration contracts, we forecasted an increase in the dividend to $1.50 per share, with 4% annual growth, increasing the payout ratio to 85% by 2033, at which time the dividend was cut due to taxes. At NEP, we took the midpoint of management’s dividend growth guidance of 12-15% per year through 2022 and then grew the dividend by 4% per year thereafter, using equity issuance to fund the growth. We discuss the drivers of the individual company dividend forecasts below.

Exhibit 4: 20 Year Dividend Forecasts[2]

___________________________________________________________________________________________________________________________________________________________________________________________

Source: Company reports, SSR analysis

Finally, we estimated the terminal value based on some additional analysis of the cash flows beyond 2036. This resulted in a similar calculation of terminal value to the DCF analysis, or ½ of the final year dividend divided by the discount rate, for all of the companies. For CAFD, NYLD and PEGI, the absence of significant growth resulted in a portfolio value similar to the no growth scenario of the DCF analysis. For NEP, with over 90% of the remaining cash flows coming from projects acquired over the next 20 years, and a tax shield in 2036 large enough to last another 10 years, we estimated their 2036 dividend could be maintained for another 10 years before significant declines. ABY, was a mix of the two extremes, with significant new project acquisitions in the next 20 years, thereby ensuring longer lived cash flows than the yieldcos with limited acquisitions, but the impact of taxes was already in the dividend by year 20.

When a dividend discount model is applied to these forecasts, NEP shows the greatest upside due to its capacity to fund accretive acquisitions through continued equity issuance and its consequent strong dividend growth. ABY also shows some upside at reasonable discount rates.

Exhibit 5: Dividend Discount Model Valuations[3]

___________________________________________________________________________________________________________________________________________________________________________________________

Source: Company reports, SSR analysis

Atlantica Yield (ABY)

ABY owns a diverse portfolio of contracted international assets, including solar, wind, transmission lines, conventional gas fired generation and water treatment facilities. Due to the bankruptcy of their former sponsor, Abengoa, they have not raised their dividend for several quarters, in spite of increasing CAFD, as they are waiting for the final waivers from all of their project debtholders. This means that they are starting with a very low payout ratio, increasing cash on hand, and significant capacity for additional holdco debt. Looking forward, there will be a 10% drop in CAFD in 2020 due to the repricing of their Spanish solar PPAs to reflect lower Spanish bond yields. We also estimate more than half of their current CAFD will roll-off between 2032-2036. However, we expect ABY to be able to raise their dividend to $1.50 in by 2018, and then grow their dividend by 4% p.a. through 2032, as it continues to acquire projects at a pace of $10-20 million of CAFD per year. To finance these acquisitions, we assume they use retained cash, conserved by only gradually growing the dividend into an 85% payout ratio, plus the holdco debt capacity. Starting in 2033 we expect the dividend to drop by ~50% as the tax shield will have been consumed, exposing earnings to taxes.

8point3 Energy (CAFD)

8point3 Energy is the most dramatically overvalued yieldco in the DDM and significantly overvalued in the DCF. 8point3 has the largest percentage of its contracts repricing to market within the next 20 years, starts with a payout ratio over 80%, and its holding company is over-levered. As a result, it has no ability to grow its dividend and, indeed, will have to cut its dividend in 2020 by at least 35%, when its $664 million of holdco debt and $50 million of project debt matures. Even if it refinances the debt, which we assume will occur, it will have to start paying it down over time, reducing available cash flows. It will have to cut its dividend again by 50% in 2029/30 when it starts paying taxes and then, after contracts are repriced to market in 2032-35, 8point3 will be unable to pay any dividend. As a result, even on a nominal basis our forecasted dividends total less than the current share price. We are therefore adding CAFD to our least favorite list of yieldcos.

NextEra Energy Partners (NEP)

We view NEP as the best positioned of the yieldcos, reflecting its capacity to fund accretive acquisitions through continued equity issuance and the strength and support of its sponsor, Nextera Energy (NEE). NEE is the largest U.S. developer of renewable power and its current plans should bring online enough new projects to support NEP’s growth plans. In addition, NEP has acquired contracted intrastate gas pipelines in Texas and would likely have the opportunity to acquire similar assets that NEE might acquire or construct, including large pipelines in the Southeast U.S.

Critically, from NEE’s perspective, at a drop down CAFD yield of 9%, the projects it sells to NEP are valued much higher than they are as part of NEE, encouraging NEE to continue to support its yieldco. NEE’s actions since 2014 suggest that they view NEP as an important tool for their growth: they have reduced their incentive distribution rights to allow for continued growth at NEP, they have purchased large amounts of NEP equity to fund acquisitions by NEP when NEP’s valuation was at its lowest levels, and they have continued to drop down projects at valuations that allow for material accretion at NEP even when that was lower than the valuations of acquisitions made by other yieldcos.

As a result, we believe NEP will be able to achieve its distribution growth guidance of 12-15% through 2022 and we expect it will be able to continue to grow at a moderate pace afterwards. This forecast assumes continued access by NEP to the capital market risks, as NEP would need to issue ~$850 million of equity in 2018 and ~$1.0-1.3 billion of equity annually thereafter in order to fund this growth, resulting in a tripling of shares to ~450 million over the next 20 years. It would also entail acquiring projects generating ~$175 million of CAFD annually, driving an increase in project level debt of ~$28 billion and holding company debt of ~$11 billion over the same period, or ~13% annually. Absent the creation of new yieldcos or more aggressive growth by the existing yieldcos during that time, NEP would comprise 90% of the yieldco sector. Nevertheless, we believe this growth is still achievable.

NRG Yield (NYLD)

Without equity issuance NYLD is well positioned to grow its dividend modestly through 2023, using cash on hand and holdco debt to finance acquisitions. In 2024, however, the expiration of PPAs and repricing to market of 1750 MW of gas fired generation in California will result in a 20% decline in CAFD and likely a similar decline in dividend. Following this, we forecast continued small acquisitions funded by retained cash as we do not expect NYLD to have additional holdco debt capacity once the California PPAs are repriced. The acquisitions, including the $644 million acquisition already made in the first quarter of 2017, should allow NYLD to defer cash taxpayer status until 2032, at which point we expect the dividend to be cut by 60% to ~$0.50.

We also analyzed NYLD’s valuation if they were to issue equity to make acquisitions. We found that, at their current dividend yield, they could improve their value through acquisitions using equity. Acquisitions of ~$30-40 million of CAFD/year or ~300-600MW of projects, which we think is achievable, even after they have exhausted the assets available from NRG, their current sponsor, would achieve 2-3% dividend growth annually over the next 20 years and defer taxation for several years beyond. However, even with this growth it would still be somewhat overvalued on a DDM valuation. (See the row labeled “NYLD – Shares” in Exhibit 5.)

Pattern Energy (PEGI)

We view PEGI as similarly positioned to NYLD. PEGI does not face the same large drop in CAFD in 2024, but it faces the roll-off of hedges and PPAs on nearly 75% of their portfolio over the next 20 years. Nevertheless, holdco debt capacity plus retained cash will allow them to grow their dividend by 25% over the next 10 years and defer taxation until 2031/32, at which point they would have to cut their dividend by almost 55%.

Unlike other yieldcos, PEGI’s sponsor is not publicly traded, but is a private developer, which has raised concerns about the ability to continue to source new projects for acquisition. We are less concerned, as we think they should be able to source enough projects both from their sister company and third parties. Yesterday’s announcement of a series of transactions and investments with Pattern Development, Riverstone and the Public Service Pension Investment Board is positive and helps to address concerns about PEGI’s ability to source new projects over the longer term, but does not address the need for PEGI to issue equity to growth more rapidly.

PEGI could grow their dividend for longer if they pursued a more aggressive acquisition policy. We found that a plan based on issuing $75-100 million of equity the first few years and growing to $200 million per year after executing successfully would result in an attractive DDM valuation for PEGI. This plan would allow acquisitions of $25-35 million of CAFD annually, or ~250-500 MW of projects. This could be viewed negatively initially, but it is likely that after a few years of consistently achieving this growth PEGI could achieve a higher valuation, allowing for less equity to be issued or yet higher growth to be achieved. (See the row labeled “PEGI – Shares” in Exhibit 5.)

We recognize that a growth plan requiring continual equity issuance does leave PEGI exposed to the capital markets, but if applied flexibly, with more equity issued in years when the stock valuation is high and less equity when the markets are weak, we think PEGI would be a more attractive investment. However, we cannot recommend PEGI as an investment until they embrace such a strategy because their underlying dividend trajectory is not enough to justify the current share price.

Limitations of Our Analysis

In this analysis we have made a number of simplifying assumptions to facilitate the analysis and the comparison among companies and to provide some estimate of the value of the companies. Most importantly, as income vehicles, the valuation of yieldcos is going to closely tied to interest rates. We believe that the range of discount rates we applied in our valuation should accommodate a moderate rise in interest rates from current levels, but a sharp rise of 150 basis points should cause a significant dislocation in yieldco valuations. However, a gradual rise interest rates would have a more moderate impact on yieldcos as valuations of project acquisitions would also decline making them more accretive.

We have also simplified the amortization of debt and the treatment of tax equity. Regarding debt amortization, which can have a significant impact on CAFD, after the first few years when debt amortization was disclosed by the companies, we amortized the project debt as a portfolio so as to eliminate the debt before the PPAs expired. Because all of the project debt matures before the PPAs expire, the difference in cash flows should only be minor variations in timing. We assumed that existing holdco debt could be refinanced upon maturity, but would need to amortize by the end of 20 years.

Finally, we assumed that tax equity would have de minimus impact on cash flows. In truth, the impact of tax equity on cash flows varies depending on the structure, but the vast majority of the return for tax equity investors are the tax benefits of the projects, which do not affect project cash flow. While it is possible some projects could be seriously impacted by tax equity, that is generally not the case and we determined that impact on the analysis was not significant enough to justify the effort needed to identify the few projects where it would matter, especially in light of the limited disclosure on tax equity.

©2017, SSR LLC, 225 High Ridge Road, Stamford, CT 06905. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results.

  1. This assumes that the IRS treatment of start of construction for wind, which allows the full value of the PTC if construction was started by 2016 and completed by 2020, will be maintained and that similar treatment will be given to solar, allowing most solar projects completed by 2024 to receive the full ITC.
  2. “NYLD – Shares” and “PEGI – Shares” refer to scenarios we modelled for NYLD and PEGI that incorporate more aggressive equity issuance and project acquisition than either company’s current strategy. Please see our discussion of these companies on pages 10 and 11 below.
  3. “NYLD – Shares” and “PEGI – Shares” refer to scenarios we modelled for NYLD and PEGI that incorporate more aggressive equity issuance and project acquisition than either company’s current strategy. Please see our discussion of these companies on pages 10 and 11 below.
Print Friendly, PDF & Email