SunEdison (SUNE) – Deja vu all over again

SunEdison, a US based renewable energy company popped up 2 times on my radar screen. Once a year ago as one of David Einhorn’s top picks and last week as one of the very few published long investments of John Hempton at Bronte.

I try to sum up Einhorn’s 2014 thesis in four bullet points:

– Solar energy is competetive, strong growth almost guaranteed
– SUNE has a moat and will grow strongly by maintaining its margins
– extra value is created via the “YieldCo” subsidiary
– investors don’t understand the company especially the fact that most of the debt is “non-recourse”

The “Moat”

From Einhorn’s slide deck:

As an experienced project developer, SUNE’s financial, legal, and due diligence expertise gives it a competitive moat. It has opened offices in the most attractive international markets several years before anyone else, giving it a first mover edge and unique geographic diversity in an industry that faces capricious governments, currency fluctuations, sovereign risk and competition.

Well, now it is pretty easy to point out that this thesis might have some flaws after the stock cratered in the last weeks:

Let’ just look at the annual report where SUNE reports on competition:

Competition. The solar power market in general competes with conventional fossil fuels supplied by utilities and other sources of renewable energy such as wind, hydro, biomass, concentrated solar power and emerging distributed generation technologies such as micro-turbines and fuel cells. Furthermore, the market for solar electric power technologies is competitive and continually evolving. We believe our major competitors in the renewable energy services provider market include E.On, Enel, NextEra, NRG, SunPower Corporation, First Solar, Inc., JUWI Solar Gmbh and Solar City. We may also face competition from polysilicon solar wafer and module suppliers, who may develop solar energy system projects internally that compete with our product and service offerings, or who may enter into strategic relationships with or acquire other existing solar power system providers.
We also compete to obtain limited government funding, subsidies or credits. In the large-scale on-grid solar power systems market, we face direct competition from a number of companies, including some utilities and construction companies that have expanded into the renewable sector. In addition, we will occasionally compete with distributed generation equipment suppliers.
We generally compete on the basis of the price of electricity we can offer to our customers; our experience in installing high quality solar energy systems that are generally free from system interruption and that preserve the integrity of our customers’ properties; our continuing long-term solar services (operations and maintenance services) and the scope of our system monitoring and control services; quality and reliability; and our ability to serve customers in multiple jurisdictions.

If you compete mainly on price, then there is obviously not much of a moat. There are no network effects, they don’t have any patents and clients don’t care about the brand of a solar project company. In contrast, a strongly growing markets attracts many new entrants which will drive down margins especially if it is relatively easy to enter the market. or even if there would be an “econimies of scale advantage”, in a strongly growing market this is not worth much

Germany is here maybe already some years further in the experience curve and one learning here was that there wasn’t any first mover advantage. In contrast, many of the first movers made some real mistakes like contracting solar modules for fixed prices and were then wiped off by the followers who bought cheaper.

Success metrics

If you look at SunEdisons investor presentation, you don’t see any GAAP numbers, only adjusted EBITDAs and self created metrics like MW and GW delivered etc. The reason is clear: GAAP numbers look awfull, both earnings and cashflows at all levels. The company is using boatloads of money under GAAP reporting.

Overall, the accounts are pretty much incomprehensible not only on the financing side but also cash flow wise. So non-recourse debt sounds great but without earnings it will be a quite difficult investment case.

The YieldCo – TerraForm Power

TerraForm Power is a consolidated subsidiary of SUNE but has a stock listing and minority shareholders. The sole function of TerraFrom power is to buy the projects from SUNE, leverage them up ~4:1 or 5:1, hold them and pay out dividends. The stock price got hit hard along SUNE as this chart shows:

However according to Einhorn the participation is extremely valuable due to 2 reasons:

1. A Yieldco structure is value enhancing per se as Yieldco investor require much lower returns on investment as stock investors
2. Terraform and SUNE have a structure in place where SUNE retains much of the upside of the YieldCo, so the worth to SUNE is much higher than the market value of the shares

Einhorn makes some remarkable comments in his presentation, but I was struck mostly by this one:

In the recent sell‐off, Terraform’s shares declined with the oil and gas MLPs. Because most MLPs pay out cash flows from depleting oil and gas reserves that need to be replaced with new wells, these companies need continued access to cheap capital just to sustain their dividends. Terraform doesn’t face that risk because solar assets don’t deplete. So Terraform will only raise capital for growth.

Well, this is clearly wrong. Of course do Solar panels deplete. They seem to deplete clearly slower than oilwells but the problem is that there are not that many old solar panel installed to actually get statistical relevant numbers. Some studies show that there is a relatively high loss of power in the beginning (~5%) and then a depletion of capacity of around 1% per year. Additionally, most of the funding and the electricity take-off agreements have to be renewed at some point in time which includes some significant “roll over” risk ithin the YieldCos.

Another thing that struck me is the fact that both, SUNE and Einhorn assume ~8,5% p.a. unlevered return on their renewable assets going forward which then can be levered up nicely even if you have to pay 6% interest on your bonds. I don’t really know the US market, but assuming such a yield in Europe would be completely unrealistic. Unlevered yields for renewable energy projects are at 4-6% p.a. max and you can only lever them up with “low cost” leverage for instance pension or insurance liabilities, it doesn’t really work with long term more expensive “subordinated” capital as many companies have found out the hard way.

Maybe the US market is less competitive to allow such returns ? I find that hard to believe. Just by chance I have been involved in some uS wind projects and the returns are nowhere near 8% unlevered but rather similar to European yields.

Another thing which is different to European projects: In Europe, you don’t have specific credit risk in the projects as the electricity has to be taken off from the grid, which means that basically all grid user guarantee your return. SunEdison’sproject contain undisclosed credit risks because if the client default there will be no backstop.

That leads to the question: Who on earth is actually buying into those YieldCos ? In TerraForm’s case any upside is capped and equity holders are fully exposed to any problems that could show up like increasing interest rates, defaults of off-takers, debt roll risk etc. So who is prepared to take equity like risk but accepting bond like returns ? I do know but my guess is that many yield starved private investors will most likely not care about the risks as long as they get a “juicy” dividend. In Germany something similar but on a lower scale happened. a lot of the renewable companies financed themselves with “participation rights” and promises of high dividends but most big cases ended in spectacular failures. I covered some here for instance

To shorten this: Yes, at the moment the Yieldco structure could actually generate some value because for the time being there seem to be enough stupid investors out there who buy something with equity risk in exchange for bond like returns. But this could go away quickly especially if some of them blow up spectacularily. It’s the same old reason why people on Wallstreet earn so much: Pretending that repackaging an asset increases its value.

Financing structure

Although the complicated financing structure attracted me to the stock in the first place, based on what I have written above I don’t think it’s worth the time to dig deeper. One thing that John Hemption seems to have missed in his post is the fact SUNE has implemented a margin loan with TerraForm Power shares as collateral. Such a strcuture alone for me already indicats that either those guys don’t know what the are doing or that they are really desperate.

In such a case the only “safe place” in the capital structure is within the senior secured paper, everything else in my opinion is more a gamble than a value investment.


At the first glance Sun Edison looks interesting. You can buy into a (still) strongly growing company at around 1/3 of the price David Einhorn paid a year ago. From my point of view however the business relies on two fundamental assumptions to perform as planned:

– the ability to continously source renewable energy projects with really high yields (“risk free” plus 6% or so)
– enough stupid investors who buy into YieldCos with equity like risks and bond like returns to subsidize the development company

If Germany as one of the renewable power pioneer markets is any indication, both assumptions will not hold for very long. In Germany’s case, the yield for the projects went down very quickly especially after government subsidies were reduced and the “yield investors” got fleeced massively as a consequence.

Clearly, in the short run SUNE and TERP could make massive jumps up and down in price but mid- to long term I don’t think that they will be great investments.

P.S.: It might look like I want to bash David Einhorn, as this is already the third time that I strongly disaggree with him after Delta Lloyd and Aercap. But on the contrary, i do still think that he s one of the best investors in the hedge fund area, he just had some bad luck and a lot of money to manage which makes things difficult.


  • Well, that was a timely article with a very factual analysis….
    -50% in the past month!

    So, MMI:2, Einhorn: 0 ?

  • Another point is the special nature of electricity. Grid parity is simply not enough. The timing is also very important, as long as one can’t store electricity in large amounts with low cost.

    This subsidy of getting paid for the renewables without considering current market price of electricity and supply/demand is at the moment borne by grid’s costumers. I believe the producers of the rewenable electricity should pay it.

    If your neighbour produces renewable energy and sells into the grid, you have to pay more for your electricity. Is this fair?

  • Sounds like you’re just not a believer in Solar – period. Basically, you’re skeptical about everything – financing, operations, margins etc. There is also a bull case though. TerraForm is supposed to provide a long term low risk dividend income stream. Will this happen? Time will tell. But if the high quality counterparties do provide steady income $TERP would be bond like risk for bond like returns. At a basic level, the non-recourse nature of debt and warehouses are important since these allow Sunedison to earn margin workout burdening the sponsor. Also even at current prices, $SUNE’s stake in $TERP and $GLBL are worth $1.5 billion. If their valuation improves, those vehicles alone are worth $SUNE’s entire market cap.

    • Well, having seen the same “game” played here in Germany already some years ago, the answer is clearly: Yes I am skeptical about everything. But maybe it works in the US ? Who knows….

  • EInhorn has been wrong quite a bit before, just look at his longs like Delta LLoyd, MRVL and Evonik. some of the above weren’t total disaster but in all of them the initial thesis has been proven wrong. I think EInhorn sort of lost it during the last few years.

  • The reason neither Abengoa nor Sunedison have worked is pretty simple: solar still isn’t competitive in most places. The numbers just don’t add up unless you assume lots of leverage, historically low interest rates, subsidies, perpetual PPA renewals, low panel degradation, etc. The business models of Sunedison and Abengoa are predicated on all of these variables being exactly in their favor, and even then it’s still questionable.

    Or flip the argument around – if solar was economically competitive on a fundamental level, without all these nonsense assumptions that stretch decades into the future, wouldn’t these companies thrive? Instead they’re all blowing up. What does that tell you?

    I think investors got ahead of the technology, and perhaps even got a little blinded by the nobility of replacing dirty fossil fuels with clean solar.

  • I completely agree with your analysis ( I thought i was stupid for seeing this way) and I think you are pointing to emperor´s clothes,

    I did mention the case about Abengoa Yield in John´s blog, that I think is similar. I don´t know anything about a solar projects but i think Einhorn´s commentary “Terraform doesn’t face that risk because solar assets don’t deplete. So Terraform will only raise capital for growth.” is misleading/wrong, so the investment idea is flawed as it is based on a false assumption.

    I talked to Abengoa Yield´s IR, and when I asked them to provide a ROE figure they just referred to the cash yield (9%) they provided in the presentation (of course you never see this cash yield figure and its assumptions, or a reconciliation of CAFD with Ebitda figures in the 20F) , then I suggested that this figure is misleading because is including return of capital as assets depreciate, so return on equity or IRR should be at best 6% if an investor invested at book value (that is actually the same wacc figure they are using to discount their projects according to their 20F) , if I trusted their numbers, and then he answered that they expect their assets would be able to be operating after their contract (some of them after a 20-30 year contracts?!?) and continue to sell electricity at pool price…
    I asked then why they don´t supply their internal calculations for the total duration of their projects as I think is best way to proof the market is misunderstanding the economics of their business and that their dividend are sustainable, then he answered that they are thinking to do that after their project ´s ramp up… I will wait until I see it (or stock price offers enough yield to cover asset depreciation)

    The developerCo case, I think is even worst because they rely on cheap capital from the yieldco to sell their projects and earn their fees.

    • Abengoa bonds trading in the 40s. That tells you how weak the sponsor is.

      I have been trying to short Yield for 2 months but can’t find a provider…..

  • Beautifully written.

    I would just add add what we learned from the last photovoltaic waifer crash: solar is a tech investment more than a utility one (despite the fact that the industry wants to paint it as the latter).

    As such, the biggest depletion risk in my view is not the loss of efficiency but the pseudo-Moore law that applies to solar: as newer tech yields a higher conversion rate, the depreciation on your existing asset is steeper as you become less competitive (considering that you pay the same rent for the land on which you are placing your panels).

    Second risk is subsidies (off take) cuts./ It was reported last month that the UK (where I live) and Germany are now on par with fossil fuel and don’t need to be subsidised any longer…. a great thing for the environment but not such a good thing for an investor.

    Thanks again for sharing.


  • I agree with jb.

    After losing money with Q-Cells and Solarworld some years ago (although I got out before 2011) I surely agree with you, not with Einhorn. Maybe I am a little bit biased now, but I really do not see myself investing in the renewable energy sector for a while. The only exception could be hydro power. In solar, there is no moat.

  • You don’t have to say sorry for disagreeing with another investor – that is what makes markets! You were correct in the last ones you disagreed with. Thinking for one’s self is sensible. I disagreed with Einhorn on Moody’s and it went up 4 fold-ish – he was short. No shame in thinking for yourself.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.