You want to be greedy when others are fearful. You want to be fearful when others are greedy. Warren Buffett
On Jan. 7, 2016 SunEdison (NYSE:SUNE), a major solar developer, announced that it will raise very expensive debt. The markets feared that the Company could not support the new debt load and sent the stock plummeting. Currently, the stock is down by more than 50.0% from $6.00 per share to $2.50 a share, which is down more than 90% from its 2015 high of $33.00 per share.
We believe that the company can support the new debt load. Additionally, we maintain that the market still does not accurately understand the company's debt profile and does not properly appreciate the company's earnings power.
We maintain that SunEdison is a strong buy. The company is priced to default; however the company has a pro-forma 2016 debt service coverage ratio of ~2.1x and a pro-forma 2016 net operating leverage ratio of ~2.8x, which implies that the company has a strong investment grade debt profile.
On this note, we will first walk through SunEdison's debt profile and debt servicing requirements. Then, we will walk through SunEdison's earnings power. Lastly, we will compare SunEdison's debt servicing requirements against the company's earnings power and defend that SunEdison has an investment grade debt profile.
SunEdison Debt Profile and Debt Servicing requirements
We believe that the capital markets do not have a fundamental understanding of SunEdison's debt profile and inappropriately believe that default is a real possibility. Most of the capital market participants believe that SunEdison has a whopping $11.7B in funded debt as of 09/30/2015 and is highly levered. One example is shown below.
SunEdison's group credit profile ... reflects a distressed debt "situation"...reflecting our assessment of its "weak" business risk profile and "highly leveraged" financial risk profile. (S&P)
We believe that SunEdison's debt profile is significantly more conservative than what is reported on its GAAP financials. SunEdison consolidates the debt of the following entities onto its balance sheet: TerraForm Power (NASDAQ:TERP), TerraForm Global (NASDAQ:GLBL), project debt, and corporate debt. Due to the consolidation of debt, SunEdison's debt profile is materially overstated. While TERP's debt, GLBL's debt, and project debt are on SunEdison's balance sheet, we do not believe that these credit facilities should be considered SunEdison's debt for the following reasons:
- TerraForm Power's debt should not be considered as SunEdison's debt. Firstly, TerraForm Power is its own publicly traded company and TerraForm's creditors do not have recourse to SunEdison. Secondly, TerraForm Power's debt does not finance SunEdison's operations and is used solely to provide a levered yield to TerraForm's shareholders. Thirdly, SunEdison does not have a majority economic interest in TerraForm Power (43.0% ownership of Class A shares), however, is required to consolidate TerraForm Power's debt onto its balance sheet as it maintains a controlling interest in Terra Form Power (91.0% ownership of Class B shares). Please note that Class B shares allow for voting rights but do not represent any economic interest.
- TerraForm Global's debt should not be considered as SunEdison's debt. Firstly, TerraForm Global is its own publicly traded company and TerraForm's creditors do not have recourse to SunEdison. Secondly, TerraForm Global's debt do not finance SunEdison's operations and is used solely to provide a levered yield to TerraForm's shareholders. Thirdly, SunEdison currently does not have a majority economic interest in TerraForm Global (35.0% ownership of Class A shares), however, is required to consolidate TerraForm Global's debt onto its balance sheet as it maintains a controlling interest in Terra Form Global through its Class B shares. Please note that Class B shares allow for voting rights but do not represent any economic interest.
- Project debt should not be considered as SunEdison's debt. Firstly, Project debt does not stay for long periods of time on SUNE's balance sheet and does not finance SunEdison's operations. As an example: SunEdison develops a $100.0MM solar energy project financed by $20.0MM equity and $80.0MM project debt. SunEdison finishes the project and sells it for $120.0MM, and the buyer of the project assumes $80.0MM in debt and pays SunEdison $40.0MM in cash. Therefore, project debt is only on SUNE's balance sheet only for as long as it is developing the project. Secondly, project debt has no recourse to SunEdison. SunEdison creates special purpose vehicles (SPV) or joint ventures (JV) or variable interest entities (VIE) for a project and the project debt lenders only have recourse to the project's asset and the project's cash flows in the special purpose vehicle.
Peeling away SunEdison's non-recourse debt, we find that SunEdison's true corporate debt is only $3.0 billion versus $11.7 billion as reported on the GAAP financials as of 09/30/2015. SunEdison's $3.0 billion in corporate debt is the sum of the following credit facilities listed in the table below. Additionally, provided below is a list outlining the important features of each portion of SunEdison's debt on the table. This information was obtained from SunEdison's 3Q2015 SEC filings and its business update dated Dec. 24th, 2015.
- 2.000%; $300.0 million convertible notes (strike price $14.62; anti-dilutive call spread overlay strike price $18.35) due 2018
- 0.250%; $600.0 million convertible notes (strike price $26.87; anti-dilutive call spread overlay strike price $37.21) due 2020
- 2.750%; $300.0 million convertible notes (strike price $14.62; anti-dilutive call spread overlay strike price $18.35) due 2021
- 2.375%; $460.0 million convertible notes (strike price $25.25, anti-dilutive capped call feature strike price $32.72) due 2022
- 2.625%; $450.0 million convertible notes (strike price $38.65, anti-dilutive capped call feature strike price $62.12) due 2023
- 3.375%; $450.0 million convertible notes (strike price $38.65, anti-dilutive capped call feature strike price $62.12) due 2023
- 6.250%; $410.0 million margin loan (LTV 40% secured by TerraForm Power Class A shares; cash collateral required for LTV shortfall) due 2017
- 3.750%; $336.0 million exchangeable notes (exchangeable for TERP Class A shares; initial strike price $34.58 per TERP share, however, initial exchange rate subject to "make-whole fundamental changes", which will increase the exchange rate for holders) due 2017.
- 4.670%; $54.0 million system pre construction debt with direct recourse to SunEdison.
- 4.470%; $388.0 million other credit facilities with direct recourse to SunEdison (including GS loan).
Source: 3q2015 sec filings starting on pg 22 note 8; credit agreement filed with the SEC - location details included in the table of contents starting on pg 90 on 3q2015 10q and table of contents starting on pg 47 on 2014 10k.
Fast-forwarding to Jan 7, 2016, SunEdison needed additional liquidity. After adverse fluctuations in working capital components, cash earmarked for constructions, and repayment of the margin loan and the Goldman Sachs (NYSE:GS) term loan, SunEdison was hard pressed for liquidity and had to raise capital in very short notice.
SunEdison raised new debt, which was priced similarly to corporate loans extended to companies on the verge of default, fueling to the irrational fear that SunEdison is nearing bankruptcy. As of January 7th 2016, SunEdison, according its 8-k filings, obtained the following new credit facilities including (1) $555.0 million second lien loan priced at LIBOR + 1,000 bps (10.0%) with an equity kicker in the form of 19.8 million warrants with a strike price of $0.01 due 2018, (2) $225.0 million second lien loan priced at LIBOR + 1,000 bps (10.0%) with an equity kicker in the form of 8.9 million warrants in exchange for $335.8 million in existing convertible bonds, and (3) $225.0 million second lien loan priced at 5.0%, with a conversion strike price of $7.50.
After adjusting for the capital raise, SunEdison's true corporate debt is $3.4 billion ($2.9B net of discounts/ balance sheet carrying value) in 01/07/2015. The company obtained three new credit facilities, however paid off (1) the margin loan as the company needed to post cash collateral against this facility (2) the exchangeable note as the company did not want to fulfill an unfavorable make whole clause with TERP Class A shares and (3) the GS term loan. After adjusting for these movements, SunEdison's new debt profile is listed below.
The transaction increased the interest payment per year by ~$41.2million from $363.0 million to $404.2 million. The interest expense increase of ~$67.0 million reported on SunEdison's transaction summary does not factor in the interest saved on the margin loan and exchangeable debt repayment but does adjust for the convertible preferred shares. We believe that the weight average effective interest rate is approximately 14.0% (vs a nominal basis of 4.4% as shown in the chart above) due to the amortization of bonds issued at a discount and due to the amortization of the costs related to the convertible features on the bonds, which had costly call spread overlays and expensive capped call features.
SunEdison's debt servicing requirements are $404.2 million for fiscal year 2016. Debt servicing requirement is typically calculated as interest payments + principal payments + lease payments. However, currently, the company's debt servicing requirements solely consist of the interest expense as there is no current maturity of long term debt and no recourse leases.
To conclude SunEdison's debt analysis, we believe the following: (1) of the $11.7B of SunEdison's consolidated debt, only $3.4B is true corporate debt ($2.9B net of discounts) with recourse to SunEdison as of 1/7/2015 (2) debt servicing requirement is ~$404.2 million, which only includes interest payments as there are no current maturities of long term debt (earliest maturity in 2018) and the leases are non-recourse to SunEdison.
SunEdison's Earnings Power
Currently, the capital markets inappropriately believe that SunEdison is a money losing business, which cannot support high debt levels. One example is included below:
SunEdison's woes illustrate how fortunes can quickly change for a money-losing company when investors grow skeptical.
The renewable energy company has failed to turn a profit for the last 2½ years, but financial performance is recent quarters has gotten worse. This year's second-quarter loss was $263 million, or 89 cents a share. (Wall Street Journal)
We believe that the capital markets do not fundamentally understand SunEdison's true earnings power. The intricacie lies in the ways that the solar energy facilities that SunEdison develops are sold and the different ways that the sales are accounted for. While sales of solar energy facilities to third parties are easily understood, sales from warehouses and to yieldco's are hard to decipher.
GAAP gross margins and GAAP operating metrics do not accurately depict SunEdison's true earning power; however GAAP metrics are regularly regurgitated by the press, such as in the WSJ note above. We list out the accounting intricacies for each individual type of sale below.
- Sale to a Third Party: In a sale to a third party, SunEdison books the expenses associated with the development of the project and also books the revenues corresponding to the sale of the project.
- Sale to Yieldco: In a sale to a yieldco, SunEdison books the expenses associated with the development of the project but does not book the revenues corresponding to the sale of the project. While counter-intuitive, the mechanisms of a transactions with a yieldco helps us understand this method of accounting. When SunEdison develops a project it contributes equity and obtains debt to finance the project. When the project is complete, it is dropped into its yieldco. However, the "drop" consists of the yieldco purchasing the equity initially put into the project by SunEdison. Therefore, SunEdison books the cost, does not book the revenue corresponding to the sale but does get a cash margin for selling the equity that it initially puts into the project, which is calculated as the cash proceeds of the sale less the initial equity put in less the assumption of project debt by the buyer.
- Sale of asset from Warehouse: In a sale of an asset from a warehouse, SunEdison books the expenses associated with the development of the project but does not book the revenues corresponding to the sale of the project. While counter-intuitive, the mechanisms of a transactions with a warehouse helps us understand this method of accounting. When SunEdison develops a project, it can either contribute its own equity or obtain equity from a warehouse facility. When the project, using warehouse equity is completed, SunEdison could drop the completed facility into its yieldco or sell it to a third party, and would collect a cash margin upon exit, which is calculated as the cash proceeds of the sale less the payback of the warehouse equity and less the assumption of project debt by the buyer. Therefore, SunEdison books the cost but does not book the revenue corresponding to the sale when assets exit from a warehouse facility to a third party or to a yieldco.
Simplifying the entire dynamic, we calculate SunEdison's pro-forma 2016 net income at $230.4 million, assuming yieldco and sales from warehouse facilities are treated similarly as third-party sales. We believe that SunEdison will sell 3.5 GW (1 GW = 1 billion watts) at $2.03 per watt (including tax equity contribution) with a gross margin of 17.5% ($0.35 per watt) and an operating margin of 8.9% ($0.18 per watt). We further assume that SunEdison will pay interest expense of $404.2 million and taxes of $12.1 million.
We believe that any projection model is only as good as its assumptions and we believe our projections are well backed and quite pessimistic. We will defend our revenue assumptions, gross margin assumptions, and operating margin assumptions. Interest expenses assumptions are described in the debt section above and minimal income taxes are assumed given that the company has an NOL of $500.0 million, which can be used to offset tax expenses (foot note of pg 16 of Dec 24 business update).
1. Revenues: We believe that our revenue assumption of $7.1 B or 3.5 GW * average selling price of $2.03 per watt is fairly achievable. We are confident of this revenue projection as SunEdison currently has 2.9 GW under construction, as shown below, 5.5GW of backlogs, 7.9 GW of pipeline projects and 49.1GW of potential leads. Additionally, historically, SunEdison has out-delivered on its guidance.
Source: SunEdison 3q15 earnings presentation
2. Gross Margins: We believe that our gross margin assumptions of 17.5% are fairly achievable.
Firstly, SunEdison's competitors gave similar gross margin guidance. As shown below, FirstSolar (NASDAQ:FSLR) gave gross margin guidance of 16.0% to 18.0%. Also shown below, SunPower (NASDAQ:SPWR) gave similar gross margins guidance of 16.0% to 18.0%.
Secondly, these gross margin guidances was established when the investment tax credit, a 30% tax subsidy for solar, were expected to expire, which gives us confidence that current gross margin guidance is on the low end. We believe that due to the renewal of the investment tax credit, a solar developer's gross margins may be significantly greater than current gross margin guidance of 16.0% to 18.0%.
Thirdly, during 2015, most solar developers delivered or provided guidance for fiscal 2015's gross margins in the low 20.0%, which is well above current guidance. As shown below, JinkoSolar achieved margins of 20.0% to 22.0%, FSLR achieved margins of 24.0% to 25.0%, and SunPower achieved margins of 23.0% to 24.0% in fiscal 2015.
Source: JinkoSolar 3q2015 earnings presentation
Source: FirstSolar 3q2015 earnings presentation
Source: SPWR 3q2015 earnings presentation
Fourthly, SunEdison's recent sale of a solar energy facility to JP Morgan's TerraNova partners were at ~21.5% gross margins, which were significantly higher than current gross margin guidance of 16.0% to 18.0%. Information on this deal was limited; however, Credit Suisse provided estimates for gross margins on this project. The margins achieved on this deal gives us confidence that SunEdison can deliver on gross margin guidance of 17.5%.
3. Operating Margins: We believe that our operating margin assumption of 8.9% is achievable as most industry players gave similar guidance, as shown above. Additionally, the operating margin assumption was given prior to the extension of the investment tax credit. Due to the extension of the investment tax credit, we believe that current guidance is understated and current industry dynamics may be more favorable than the expectation that management previously provided. Additionally, SunEdison has cut its workforce by 15.0%, which should lower operating expenses and buoy operating margins.
To conclude we believe that ( 1 ) SunEdison will achieve a pro-forma 2016 economic net income of $230.4 million and ( 2 ) we believe that our assumptions are well backed and are on the cautious side. Revenue assumptions are well backed by the company's current backlog and gross margin assumptions are well backed by current industry dynamics and gross margins achieved through recent sales.
We believe that SunEdison can service its debts and has an investment grade debt profile. From our first section, we estimated pro-forma 2016 debt servicing requirement costs of $404.2 million, which is the sum of interest payments, principal payments (currently none), and lease payments (currently no recourse lease payments). From our second section, we estimated 2016 pro-forma net income to be $230.4 million, which implies a cash available for debt service of $844.7 million (net income plus tax plus interest expense plus depreciation and amortization). Therefore, SunEdison's debt service coverage ratio is 2.1x (calculations shown below). Typically, investment grade debt issuers have a debt service coverage ratio of 1.25x and above.
Additionally, SunEdison's pro-forma 2016 net operating leverage is quite conservative at 2.8x. We believe that SunEdison's true corporate debt is $3.4B ($2.9B net of discounts) and SunEdison's pro-forma EBITDA is $844.7 million. Typically, investment grade debt issuers have an operating leverage of 3.0x to 4.0x and below.
While the capital markets currently fear that SunEdison will go bust, especially after raising expensive debt, we believe that these fears are irrational. We believe that SunEdison has a high quality business model and can adequately service its debts. We believe that this high quality business model is masked by the complexity of the way it accounts for revenues, expenses, and debt.
Additionally, SunEdison's current share price of $2.5 per share assumes that SunEdison is on the verge of default. If we back out TERP at (48 million shares * 18.0 per share in a normalized environment) or $2.4 per SUNE share and GLBL (63 million shares * $8 per share in a normalized environment) or $1.0 per SUNE share; the market is valuing SunEdison as worthless. We believe that SUNE shares should trade in the mid-teens. Our revision of the sum of the parts valuation from previous estimates are: $2.4 per SUNE share for TERP, $1.0 per SUNE share for GLBL, $2.0 per SUNE share for the Samsung JV, and $9.0 per SUNE share for the devco ($0.58 2016 pro-forma EPS * 15 P/E multiple).
Disclosure: I am/we are long SUNE.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.