SUNE - Sunedison Inc.

[KJP]

Regarding slide 19, it appears SUNE gets their development equity back once they drop it to the warehouse but don't receive any more cash until after it is sold out of the Construction warehouse.

 

Just a very rough liquidity calculation: If I assume $250M in total interest cost over next 6 quarters. 2nd lien obligation of $169M in Q3 2016, Firstwind earnout of $510M, Renova obligations of $100M, $404M margin loan gets to $1433M in cash needs.  Add in 17c/w for 875MW per quarter in OpEx (slightly more for Q4 2015 and Q1 2016) for $930M more in OpEx.

 

So all in Cash needs are $2.4B over next 6 quarters, subtract 850MW at $0.35W cash margin each quarter = $1.785B and your left with about $600M shortfall of cash.

 

You have a 3Q 2015 ending cash balance of $1.380B so you have enough to fund any shortfall. Or at least that's how it appears.

 

I don't think SunEdison is going to run out of liquidity in 2016.  The real question I have is whether the company actually makes enough money to justify the current enterprise value. 

[xtreeq]

SunEdison - Now with $739 Million in Extra Recourse Debt?

http://www.bloomberg.com/news/articles/2015-11-16/sunedison-now-with-739-million-in-extra-recourse-debt-

 

[KJP]

I just bought shares. Curious to know who is getting margin called here.

 

Altai Capital must be crapping their pants.

 

Picasso, do you have a fair value estimate, or is your investment based more on the belief that there is significant, non-fundamental driven selling?

[dbuch]

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

[Picasso]

I think there's significant selling which will lessen shortly.  A lot of investors (including a bunch of hedge funds) bought a lot higher without really knowing what they owned and underestimating how bad management is.  It's never a good feeling to be on the verge of a blow up when you don't trust management.  Those situations almost always lead to forced selling.

 

I think TERP is probably trading much closer to a reasonable discount under fair value.  There's probably a good chance the CEO gets replaced and we'll see the market get more comfortable with the story.  This isn't a terrible business model, but you need to manage the growth better and use more realistic cost of capital assumptions.

 

I think at this price it's a "eh" business at a silly price.  I don't think it's far fetched to see a 3x return from this level if you get better headlines or replace the CEO.  Hell, just a tweet from Icahn will send this back to $6.  It won't be a zero for at least several months either so you've got some pricing support here like an option gets priced.

[Palantir]

I don't think mgmt is bad, but they've handled this very poorly. Who will be first to go Wuebbels or Chatila? Conversely mgmt could say that they don't need equity capital so the price drop is immaterial....

[Picasso]

Management is terrible.  I'd rather have Steve Ballmer or Marissa Meyer running this operation. 

 

They've put the shareholders of SUNE in a position of being insolvent by crafting a stupid deal with Vivint.  There are other stupid things going on, but that's probably by far the worst. 

[jrallen81]

That's a little unfair to Ballmer :)

[jrallen81]

bad but not like Mayer or SUNE

[KJP]

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

[dbuch]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

 

 

 

 

[KJP]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

[KJP]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

 

cmlbr:

 

Do you agree with dbuch's back-of-the-envelope numbers?  Have different ones?

[arcube]

I don't think mgmt is bad, but they've handled this very poorly. Who will be first to go Wuebbels or Chatila? Conversely mgmt could say that they don't need equity capital so the price drop is immaterial....

 

Palantir - You, Sir are a very forgiving and generous shareholder. Me, I want Ahmad's ass booted out of there. Such bad deals that he has gone after and created unnecessary complexity in this whole process. I know the jury is still out there but given what is happening, I don't know what is going to stop another 20%-30% slide in the next few days.

[cmlber]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

 

cmlbr:

 

Do you agree with dbuch's back-of-the-envelope numbers?  Have different ones?

 

I don't quite agree with them. 

 

Here's my back of the napkin valuation (as of late last night, before the blood bath today):

 

Total DevCo Debt = $4,667 million (excludes non-recourse debt with the exception of project level non-recourse debt)

(+) Vehicle Interest Support = $276 million (undiscounted sum)

(+) First Wind Earn Out = $500 million

(+) Vivint Value Destruction = $907 million (non stock consideration)

(-) Total SUNE Cash = $1,380 million

(-) Value of TERP/GLBL = $1,270 million (10x 2016 Dividends + IDRs)

"Net Debt" = $3,700 million

(+) Market Cap = $1,558 million

"Enterprise Value" = $5,258 million

 

DevCo EBIT = $695 million (Einhorn calculation that I agree with)

 

So, when it's all said and done, you're effectively buying the DevCo for 7.5x EV/EBIT and getting the call options on warehoused projects for free and the technology that they claim is worth $1 billion for free.  Plus you're getting many years of ridiculously low interest debt from the convertibles, so the debt arguably has a much lower FMV then I'm adding to EV.  And I think I excluded about $500 million in equity value of projects on the balance sheet.

 

How long will the DevCo be sustainable, and what are those options worth, are the key questions. 

 

Disclaimer:  I sold out most of my position this morning, thankfully in the 4s.  Just had enough with the constant creep of non-recourse debt into recourse debt, terrible disclosures, misleading statement from management, etc.  Also, I have no idea what's going on with TERP, the dividend yield is now like 13-14%.  I should have hedged that out when I got in but was lazy.  I'll probably get back in at some point and am going to dig deeper on TERP.

 

KJP, what do you think it's worth?

 

[Palantir]

SunEdison - Now with $739 Million in Extra Recourse Debt?

http://www.bloomberg.com/news/articles/2015-11-16/sunedison-now-with-739-million-in-extra-recourse-debt-

 

I don't understand the case here, this debt was listed as recourse in the October presentation as well as this one...

[Guest Grey512]

Check out the new piece from John Hempton on the Bronte Capital blog. (i know, i know; COBF hates Hempton).

 

The interesting part is where SunEdison management stopped responding to communications and info requests from CreditSights.

 

This, by the way, during a time where SunEdison miraculously still has time to meet with analysts from investment banks (e.g. Deutsche Bank) that have Buy ratings on the stock.

 

The wonderful things about the markets is that they never cease to amaze. Becoming even more of a skeptic and misanthrope when looking at this (slow-moving) train crash.

 

 

Disclosure: No position in the stock.

 

[KJP]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

 

cmlbr:

 

Do you agree with dbuch's back-of-the-envelope numbers?  Have different ones?

 

I don't quite agree with them. 

 

Here's my back of the napkin valuation (as of late last night, before the blood bath today):

 

Total DevCo Debt = $4,667 million (excludes non-recourse debt with the exception of project level non-recourse debt)

(+) Vehicle Interest Support = $276 million (undiscounted sum)

(+) First Wind Earn Out = $500 million

(+) Vivint Value Destruction = $907 million (non stock consideration)

(-) Total SUNE Cash = $1,380 million

(-) Value of TERP/GLBL = $1,270 million (10x 2016 Dividends + IDRs)

"Net Debt" = $3,700 million

(+) Market Cap = $1,558 million

"Enterprise Value" = $5,258 million

 

DevCo EBIT = $695 million (Einhorn calculation that I agree with)

 

So, when it's all said and done, you're effectively buying the DevCo for 7.5x EV/EBIT and getting the call options on warehoused projects for free and the technology that they claim is worth $1 billion for free.  Plus you're getting many years of ridiculously low interest debt from the convertibles, so the debt arguably has a much lower FMV then I'm adding to EV.  And I think I excluded about $500 million in equity value of projects on the balance sheet.

 

How long will the DevCo be sustainable, and what are those options worth, are the key questions. 

 

Disclaimer:  I sold out most of my position this morning, thankfully in the 4s.  Just had enough with the constant creep of non-recourse debt into recourse debt, terrible disclosures, misleading statement from management, etc.  Also, I have no idea what's going on with TERP, the dividend yield is now like 13-14%.  I should have hedged that out when I got in but was lazy.  I'll probably get back in at some point and am going to dig deeper on TERP.

 

KJP, what do you think it's worth?

 

I agree with most of your numbers, but would adjust DevCo EBIT down to account for interest payments on non-recourse construction financing, then compare that number to your enterprise value to get a EV/EBIT number.  It's not cheap if you look at it that way.  But that also gives it no credit for the upside optionality. 

 

If you play around with margins and assumptions a bit on 2016 and then model some margin compression going forward, I think you can see a situation in which the equity is a zero.  So, you're potential downside here is 100%, or close to it.  On the other hand, your potential upside is multiples of that, particularly in the blue-sky scenario in which TERP can start issuing equity again.  So, I can understand people who say it's uninvestable (Rule No. 1: Never lose money), and I can understand people who think the optionality and very high upside are worth the risk (5x/1x upside-downside bet). 

 

Moreover, significant share price volatility should be expected at this point it's becoming an equity stub, so even relatively small changes in enterprise value estimates translate into big swings in equity value.

 

Another question to ask:  Is TERP a better investment at current prices?  At current yields, you'll be getting nearly 3.5% of your purchase price back every quarter (pre-tax) as a dividend, and dividends should be increasing given the upcoming purchases.

[dbuch]

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

« Last Edit: November 17, 2015, 02:38:00 PM by KJP »

 

I heard sell side apparently talked to management and they have $1.35B of cash with $700m of cash earmarked for working capital. Also apparently construction and preconstruction debt interest expense is capitalized into COGS.

 

So the $700M working capital reduces my rough estimate of floor value to $7/share

 

 

 

 

 

[Guest longinvestor]

Check out the new piece from John Hempton on the Bronte Capital blog. (i know, i know; COBF hates Hempton).

 

The interesting part is where SunEdison management stopped responding to communications and info requests from CreditSights.

 

This, by the way, during a time where SunEdison miraculously still has time to meet with analysts from investment banks (e.g. Deutsche Bank) that have Buy ratings on the stock.

 

The wonderful things about the markets is that they never cease to amaze. Becoming even more of a skeptic and misanthrope when looking at this (slow-moving) train crash.

 

 

Disclosure: No position in the stock.

 

Yes, old timers around here hate him. But we are also a relatively happy lot. Unlike him, he was miserable then, spreading lies, hiding in the shadows, trying to throw dirt etc. That was over 10 years ago with Fairfax, he is still seeped in misery. That's what you get to be while constantly looking for short ideas. None of us hate him for that. We understand, happiness needs to be balanced with misery, that's his role. We're fine here.

[cmlber]

Quote from: dbuch on Today at 12:01:54 PM

 

Just my rought estimate...If we can assume they make $0.35/watt gross and pay $0.17/watt OpEx they would make $630M operating income. Less the $150M interest costs and taxes equals call it $1/share So say DevCo is worth $10

 

TERP I think should be worth at least an 8% yield. Lets say $1.6 DPS/.08 = $20/share x 60M shares =$1.2B/350M shares =$3.5 a share to SUNE

 

Recourse Debt of $3B - $1.4B cash =$1.6B of net debt which equals a negative $4.6/share

 

$0 value for GLBL and IDRs

 

Total value of $8.90 a share. This assumes GLBL is worth nothing and IDRs are a donut. But it also assume they actually can sell their projects to third parties and they earn $.35/watt

 

 

 

 

I appreciate you sharing your estimates.

 

I think you have another embedded assumption:  by putting a 10x multiple on DevCo "earnings," you're also assuming those margins are sustainable beyond 2016. 

 

Also, a significant portion of the current $1.4 billion in cash is going to be used to fund the construction of MWs upon which you rely, so aren't you double-counting?

 

Finally, where does $250MM in interest for the next 1.5 years come from?  Isn't the interest on construction financing alone going to exceed that number?

 

1) 10x multiple - Yes that assumes the devco is sustainable. If this isn't the case then SUNE is probably worth zero.

 

2) I'm assuming they construct 3500MW a year and sell all of it to third parties every year so they are essentially paid back for whatever construction costs they incur. What is left is the amount of debt they have and the cash already on the balance sheet.

 

3) Interest costs is tough to figure out but here's what I calculated based on just the recourse debt.

 

Convertible Senior Notes $1.8B x 2.05% = $40M

Margin Loan $404M x 6.25% = $25M

Exchangeable notes $336M x 3.75% = $13M

Construction Term Debt $54M (recourse portion) x 4.67% = $3M

Leasback $32m (recourse portion) x 4.59% = $2M

Credit Facilities $388M (recourse portion) x 4.47% = $18M

Preferred Stock $650M x 6.75% = $44M

 

Total of $145M/year in interest. I rounded up to $250M from $220M over 6 quarters.

 

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

 

cmlbr:

 

Do you agree with dbuch's back-of-the-envelope numbers?  Have different ones?

 

I don't quite agree with them. 

 

Here's my back of the napkin valuation (as of late last night, before the blood bath today):

 

Total DevCo Debt = $4,667 million (excludes non-recourse debt with the exception of project level non-recourse debt)

(+) Vehicle Interest Support = $276 million (undiscounted sum)

(+) First Wind Earn Out = $500 million

(+) Vivint Value Destruction = $907 million (non stock consideration)

(-) Total SUNE Cash = $1,380 million

(-) Value of TERP/GLBL = $1,270 million (10x 2016 Dividends + IDRs)

"Net Debt" = $3,700 million

(+) Market Cap = $1,558 million

"Enterprise Value" = $5,258 million

 

DevCo EBIT = $695 million (Einhorn calculation that I agree with)

 

So, when it's all said and done, you're effectively buying the DevCo for 7.5x EV/EBIT and getting the call options on warehoused projects for free and the technology that they claim is worth $1 billion for free.  Plus you're getting many years of ridiculously low interest debt from the convertibles, so the debt arguably has a much lower FMV then I'm adding to EV.  And I think I excluded about $500 million in equity value of projects on the balance sheet.

 

How long will the DevCo be sustainable, and what are those options worth, are the key questions. 

 

Disclaimer:  I sold out most of my position this morning, thankfully in the 4s.  Just had enough with the constant creep of non-recourse debt into recourse debt, terrible disclosures, misleading statement from management, etc.  Also, I have no idea what's going on with TERP, the dividend yield is now like 13-14%.  I should have hedged that out when I got in but was lazy.  I'll probably get back in at some point and am going to dig deeper on TERP.

 

KJP, what do you think it's worth?

 

I agree with most of your numbers, but would adjust DevCo EBIT down to account for interest payments on non-recourse construction financing, then compare that number to your enterprise value to get a EV/EBIT number.  It's not cheap if you look at it that way.  But that also gives it no credit for the upside optionality. 

 

If you play around with margins and assumptions a bit on 2016 and then model some margin compression going forward, I think you can see a situation in which the equity is a zero.  So, you're potential downside here is 100%, or close to it.  On the other hand, your potential upside is multiples of that, particularly in the blue-sky scenario in which TERP can start issuing equity again.  So, I can understand people who say it's uninvestable (Rule No. 1: Never lose money), and I can understand people who think the optionality and very high upside are worth the risk (5x/1x upside-downside bet). 

 

Moreover, significant share price volatility should be expected at this point it's becoming an equity stub, so even relatively small changes in enterprise value estimates translate into big swings in equity value.

 

Another question to ask:  Is TERP a better investment at current prices?  At current yields, you'll be getting nearly 3.5% of your purchase price back every quarter (pre-tax) as a dividend, and dividends should be increasing given the upcoming purchases.

 

I'm wondering the same thing regarding TERP.  No idea why it's so cheap.  Runoff value must be close to or in excess of this price, and you get the optionality free...  Haven't done enough work on the TERP side, but plan to spend some time on it. 

[KJP]

 

Also apparently construction and preconstruction debt interest expense is capitalized into COGS.

 

 

Capitalizing the interest is actually consistent with SunEdison's presentations.  I had thought that was a possibility, but couldn't find any disclosure to confirm it.  Can you point to a disclosure somewhere that says the COGS number implied by the $.35/watt gross margin guidance includes capitalized interest on construction financing? 

[CONeal]

What about interest on the non-recourse construction financing?  I don't think it's included in the $.35/watt gross margin or the $.17/watt OpEx forecast.

 

Also, won't ramping up to 3500MW per year require a permanent increase in working capital that will permanently eat part of the $1.4 billion in cash?  The 10x multiple on DevCo earnings assumes the step up in working capital is permanent.  That's the double-counting I was referring to.

« Last Edit: November 17, 2015, 02:38:00 PM by KJP »

 

I heard sell side apparently talked to management and they have $1.35B of cash with $700m of cash earmarked for working capital. Also apparently construction and preconstruction debt interest expense is capitalized into COGS.

 

So the $700M working capital reduces my rough estimate of floor value to $7/share

 

Are you taking into account that 900 million of the 1.3 billion is earmarked if the vivint solar deal is approved.

[dbuch]

Capitalizing the interest is actually consistent with SunEdison's presentations.  I had thought that was a possibility, but couldn't find any disclosure to confirm it.  Can you point to a disclosure somewhere that says the COGS number implied by the $.35/watt gross margin guidance includes capitalized interest on construction financing? 

 

I'm not sure if it is disclosed anywhere but sell side confirmed it with management.

 

Are you taking into account that 900 million of the 1.3 billion is earmarked if the vivint solar deal is approved.

 

SUNE's portion of the acquisition was $777M. Stock was $370M, Convertible Note of $350M, cash of $57M. I'm not sure if the convertible note is included with the other convertible notes in the debt balance. If not I need to subtract $407M more. If so then it will use just $57M of cash so good catch on that.

 

Also I'm not including any value for GLBL. In reality they will do $1.1 DPS and with a 15% required yield that's $7.33/share x 60M shares = $1.25

 

So subtract another $1.15 for Vivint debt and cash and add $1.25 for GLBL basically it evens out. So rough valuation of $7.00 for no growth no IDRs.

 

[Artha158]

What do you think about Axiom Capital's view on SUNE?

• Credit Risk Appears Worse-than-Forebode. After some pressure from a number of SUNE pundits following our downgrade of the shares last week (given SUNE’s shrs had already moderated -75% vs. +1% for the S&P 500 over the same timeframe), we decided to do another “scrub” of the company’s 10-Q published 11/9. Following this exercise, we are even more resolute in our subdued outlook, SELL rating, and yr-end C16 PT of $2/shr (34% downside). Why? Five reasons, namely:

o (1) when excl. cash committed for construction projects, SUNE has just ~$600mn in cash for general corporate purposes (which we now blv may not be enough to sustain SUNE through 2Q16) – Ex. 2,

o (2) SUNE’s decision to borrow $169mn in 1yr paper from Goldman Sachs (GS; NR) in 3Q15 at a 15.4% interest rate (incl. $9mn prepayment) to put up collateral, we blv, for the 8/11/15 $152mn margin call on its $410mn Deutsche Bank (DB; NC) loan (an addtl. $91mn of collateral was required from SUNE 10/15 [Ex. 3], and we surmise more since then with the fall in Terraform Power’s shrs [TERP; NC]), pointing to emergency cash needs as recently as 3Q15 – who borrows 1yr paper at 15.4%?, (Ex. 4), other than a distressed company,

o (3) Renova’s right to put 7mn of GLBL shares to SUNE at a price of $15/shr 3/31/16 (a $105mn liability) – Ex. 5,

o (4) SUNE’s potential obligation to buy ~16% of Renova for $250mn using its own shares (i.e., 83mn shrs), suggesting sig. dilution to equity holders in the offing – Ex. 6, &

o (5) TERP’s recent revelation that it put up ~27% (i.e., $388mn) of the capital necessary to fund the Invenergy Warehouse, implying SUNE’s aspirations for ~$6bn in Warehouse funds to “house” its ~3GW in projects being developed may require a ~$1.65bn cash infusion (Ex. 7). Barring unforeseen incremental cheap funding in the offing, we see a credit event as likely before 3Q16.

• Valuation. Our yr-end C16 PT remains $2/shr (34% downside). While more valuation detail is below, with acute stress on its core biz at present, & shortcomings selling huge amounts of projects into the secondary mrkt over a short period of time thus far, we blv SUNE will miss ~3.5GW developed in C16. Using ([1.42GW × $0.18 (op. prof.) - $150mn interest × 70% (tax)] ÷ 316mn shrs) = $0.23/shr in dev. co. EPS, & applying a 9x P/E multiple (assumes 15% DevCo GM into perpetuity; likely high given 9.6% 2Q DevCo GM), SUNE is worth $2.