203 replies to this topic

[odyd]

Hi Sunny, I am only looking at the revenue dollars, indifferent to the source of it. I put no sales at all, to be honest, on TSL. This is where TSL could be a better idea. TSL has regained a lot of good attention today. I like a plenty, their balance sheet management and I am thankful for an English speaking CFO. I did not like they kept the revenue on the plant undisclosed and the permit question avoidance. I assumed that they would not ship to not permitted plants, but apparently that is not a stopper, so maybe I should apologize to Heiken.

[sunnysky]

I back-fitted your revenue figure to arrive at 300 MW worth of project revenue for CSIQ. For TSL, your figures are in line with mine without plant sales. So from project revenue perspective, CSIQ seems to have some upside if they execute well. But given, analysts have not taken into account what plant sales could do to TSL's bottom line, and along with the momentum of TSL's progress and name recognition, who knows,  maybe there is upside in TSL valuation metrics as well.

 

I need to finish the transcript to assess the other point you made. 

[odyd]

I cut the 2.8 to 2.7B. I also put only 19% and I know there was a lot of 20% calculations. I will be closely watching TSL's execution.

CSIQ has very explicit way to deliver information on their solar projects, but their timelines are unpleasantly associated with seasonality of solar. 

They need to put some South American projects or Southern hemisphere overall into a docket, or we will never get out of the grind of that winter blues

[sunnysky]

My blended GM is 19.5% with 300 MW of projects and 20% with 400 MW. GM for module is just shy of 15% and GM for projects is at 23%, quite reasonable I think.

 

If their revenue guidance is used, then it would mean they recognize less than 300 MW in H2 or ASP would be lower (I used $3.5/W). So I tend to believe they will smash revenue guidance in H2.

 

It's good to hear from Josh and see where he places the numbers. We know explo and eysteinh place CSIQ behind TSL.

[odyd]

Have you broken down CSIQ revenue? It looks like 49MW of modules were sold at $128M, that is $2.61 per watt. I do not think this $53M sale is in this quote? If I take this amount out the rest of modules are about 0.66 per on some 441MW am I doing this correctly? If do not, outside of the solution, module left over is becoming $0.75 which is not the case.  So plant is $5.3, solution is $2.6 and pure module is $0.655 per watt.

[odyd]

I think I am getting a bit of understanding here. Technically they can say whatever they want for average ASP on modules as they can use the figure from their solution stuff and solar plant sales plus third party sales, I used 0.67 and 0.57 and it did not make any sense to me
So this is what I did, I took $128 and added $53M; I extracted 441MW (49+10) of modules as a third party. I divided the rest of the revenue and came up to be about $0.65 per watt. so if they did 0.57 cost this looks like 12% (ouch), take the rest of the revenue and rest of the cost, as a plant and the it looks like 19.1%,
So at $0.55 processing as they show in the presentation they could have been at 16.6%. If you look at total module and revenue for Q1 and Q2, they are quoting precisely the same revenue level of some $0.93 cents per watt. I am looking at the table what come off on Q2 and I see  maybe Rutley will clear it, Penn and Val Caron, those guys should sell at 20%, so let's say they save the 2 cents from processing and sell at 0.67 and they have 17% module margin, and 20% plant sale. If their margin becomes 19%, I see plants selling at 23% but module got to be 17%

So believe it or not the biggest improvement is the module cost here. 

[sunnysky]

I think I posted my breakdown some time back. Our approaches would be similar but differ in details. Basically, there is some other revenue besides module direct sales and plant revenue. Q1 was especially distorted by this because in earlier quarters you can backfit with some reasonable ASPs without "other" revenue.

 

Anyway, what I came up with for Q1 is the following:

 

Project Rev:        $127,773

Module Rev:        $302,170

Other Rev:          $  36,381

 

Module cost:        0.5848

Module ASP:        0.67

Module GM:          12.72%   (I definitely posted this and said I could not understand how come this low

                                              - not explained by the 1% from the fire at all) 

 

My $0.68 Q2 EPS estimate is obtained after Q1 backfitting and then extending the model for Q2 based on guidance. So I'm pretty comfortable with it. This is all in GAAP. But as usual, it could be off due to various adjustment items here and there, especially tax and FX.

 

Oh btw, CB gain/loss is not included. I think they'll have to include in Q2 but the value depends on where the stock closes on 6/30. Again, not sure why they did not do it for Q1. It could have saved the stock from losing a few bucks.

[explo]

3. is $51M that is a dollar alone in EPS when fixed.

They said they might account for forex differently in future where it gets embedded already in top lines accounting. I think others like JA who hardly reports any forex use this effective currency accounting. If CSIQ had applied that in 2013, they would not have that forex charge, but instead had lower gross profit, since bottom line is the same irrespective of forex accounting method.

[Makan]

They might want to apply hedge accounting rules where any hedge ineffectiveness go to COGS. The derivatives valuation would have gone direct to equity in Q1 under these rules and be recycled to match whenever the cash flow occurs the derivative is supposed for in the future. I think last year they didnt hedge at all, so COGS was free of it and when they apply the proper hedge accounting rules in the future and have no ineffectivness (there will always be some) it is still eliminated.

[explo]

Lots of intangibles in this comparison.
TSL stated 80-150MW of projects that will be for sale...some of which have already been built.

Thanks. This is critical information that I missed. Can you paste exact quote and source, please?

Or your deduction if it was not explicitly stated.. I still haven't digested enough to arrive at the critical info on how many MW project revenue recognition they expect for 2H.

[explo]

Sunny, what makes TSL a lot more profitable is the scale. shipping 800MW to third party buyers at 17% is a huge boost versus 550MW.

If CSIQ had 71M shares it would be priced around 5 dollars more than TSL. If we use criteria of $2 per share and that way created PE, CSIQ is trading at expectation of $2.74 right now.

 

This is a good point. Let's say as an example that Trina made 75m GP on panels in Q1 based on 14 cents GP per watt and 534 MW external panel shipments and CSIQ made 45m GP on panels based on 10 cents GP per watt and 451 MW external panel shipments. Now if that GP per watt gap closes in Q2 so both makes 10 cents GP per watt then if Trina make 820 MW external shipments and CSIQ 560 MW we still got Trina having bigger contribution from panel business to GP (82m vs 56m, i.e. GP gap is 4m less than in Q1) despite them now making around the same money per panel.

[Xeloris]

nks. This is critical information that I missed. Can you paste exact quote and source, please?

From CC
 
we of course for the UK project and also for the overseas projects, we have in our plans, we are going to be eventually sell those projects.
And for the China projects, we mostly are expecting to hold them although we do evaluate each project one at a time to make sure that we use a criteria that we have formulated to make sure that whether we are going to end up with holding the target or selling the project.
As far as the China project, we have about 400 MW to 500 MW in the pipeline for this year to be completed and about 70% to 80% of those would be in China and about the rest of those in either UK markets or Japan markets.

[Xeloris]

I also liked this quote,  
 
And so, I will see that in some few respects, number one, we do believe that there is a good attempt for the total installation to be very close to 14, in 2014.
Secondly, we believe that, from our perspective it doesn’t really matter that the 11, or 12 or 13 or 14, from Trina’s perspective being the leader in providing module and downstream projects, we believe that as we are gaining market share and as we are – it’s our brand and quality of service that being recognized, the demand for our product is very much evide"
 
 
Basically they are confident in their shipment numbers because of the quality of product, all this extra demand is only feeding the lower quality producers.

[odyd]

Thanks. This is critical information that I missed. Can you paste exact quote and source, please?

 

I am aware of 40MW ready today and they say they are for sale but will not be sold, Not 80 to a 150MW.

[explo]

Thanks Xeloris and odyd.

[pg6solar]

Jefferies "reiterates on Trina" and increases PT from $23.70 to...$23.90!

[spiritcraft]

Jefferies "reiterates on Trina" and increases PT from $23.70 to...$23.90!

They stepped out on a limb with that $0.20 increase!

[sunnysky]

Funny, but a symbolic gesture. $23.70 is good target already. But what's with the fraction - how precise this can be?

[Scsnospam]

Roth maintains $25 PT

 

http://www.streetins...lick=2014052215

 

upside risk to poly? Then why is DG going down?

[pg6solar]

Its a shame that outfit like Axiom is considered as "Wall Street" as Roth, Nomura, DB, etc. Maybe guys who calling the shots at Axiom are frequent Colorado and/or Oregon guests?