Jump to content
Sign in to follow this  
dydo

CEDR Posts

Recommended Posts

Hi there, I spent good part of the day on attached file. I hope you will like information contained. I hope to ask the group for notes, observations, which would help us to understand the variables to create a solid gauge of future ASPs against GADP.

Share this post


Link to post
Share on other sites

Wow, impressive file! Thanks! I added an attempt to automatically exclude outliers, which means my GADP has change slightly from previous attachment. I then added domestic sales adjustment to translate reported ASP to global ASP. I then applied a simple model where the global ASP change from Q4 to Q1 according to how the GADP change from Q4 to Q1 (revenue recognition timings are not considered yet). I'm attaching my results. Note that the Chinese ASPs are wild assumptions. Let me know if you think they're off. Due to JKS retainage sales in China the 55 cent China ASP translates to 59 cents actual China ASP. The big drops in GADP for HSOL, JASO and CSIQ cause low Q1 global ASP projection. It looks a bit suspicious. SOL had a big GADP in March. These shipments might not be recognized until Q2, so projected global ASP might be off to the high side.

Share this post


Link to post
Share on other sites

I did all this math, I am looking at your tables, we have too many factors to make relationship of actual reported ASP and GADP. Second feeling I have is that we are looking possibly at the group of companies, which declare Chinese price per watt as value for export.This group has SOL, JKS, YGE, CSUN in it. Then you have a group with tendency to declare foreign value, companies which belong to this group are, CSIQ, HSOL, JASO and TSL. STP is in this group as well. I think we can assume for time being that we are looking at the average of 4.53% reduction in ASP for Q1. I removed STP from the calculation to come to this conclusion. I have one showing of "own" % drop to Q1 value, and using an average for the group as well. last two quarters ASP was down around 10% on average and prices did fluctuate. I am going to leave those numbers as per this table and see how we fare in the coming quarterly results.

Share this post


Link to post
Share on other sites

Pop or Spin and Payback how do you guys feel about this data at this point? I think we should avoid making forced relationship with announced ASPs but watch it move for couple of months. I would like to hear from you.

Share this post


Link to post
Share on other sites

I want to share a revelation with the group here on the delivery data. I was confused when I explained shipping lead times and other aspects versus revenue recognition. As it turns out, revenue recognition of the shipment is when delivery occurred. This happens at point of when the goods have passed over the ship's rail at the named port of shipment. We all need to pay attention here to "named port of shipment". In my limitation I assumed it was destination, but no siree, it is the place where goods are being shipped from. So technically, I would say to you, unless it takes long time at customs, we could assume that pretty much everything we are looking crossing customs is revenue recognized shippment, unless of course it is destination own warehouse. Very interesting. If we only knew when and how much is shipped to themselves? Canadian Solar has subsidiaries in USA, Canada, Japan, Germany, Australia, South Africa, using simple math of holding about 50% inventory at your sub (as oppose at Chinese location) I would say that 262MW of modules from the deliveries ended up for Q1 to be revenue shipments (perhaps more).

Share this post


Link to post
Share on other sites

Odyd and explo thanks for all the time and effort you have both put into crunching the numbers and putting it into spreadsheet form. I believe once we get some earning from the Chinese solars we can really fine tune the data and make it deadly accurate. In regards to your estimated 4.53% reduction in ASP I'm very curious to see how this lines up with what's reported. Once we can compare data with what's reported I'm sure we we can really dial in future numbers before the street actually knows what they are....

Share this post


Link to post
Share on other sites

Thanks for your feedback. I am always concern about the quality of the exercise and how invaluable things can be as long as we know how to understand it. I just posted another "open eyes" observation on March data, you may find this quite interesting. Thank you

Share this post


Link to post
Share on other sites

Odyd, i think we have to wait for q1 results to get a picture of the preditive value of the ASP data. I'm thinking like this. The absolute value seems a bit all over the place for names like HSOL, but the two example trends I calculated of Q4 to Q1 and the one month adjusted Sep-Nov to Dec-Feb and their average should be searched in each companies actual GASP and COGS trends to see if relationship can be found. Either we find consistent correlations between some of these for all companies or for groups of companies. If not we have to wait for Q2 to see if their is correlation between predicted and actual of some of the example trends individually for companies.

Meantime some questions:

[*]You have different JASO ASP by one cent. I just used Klothilde's 61 estimate after receiving product mix info from JASO IR. Do you think your 62 estimate is more accurate?

[*]Just to eliminate some potential issues. Example: company exports 100 pallets. Do they declare price per pallet and SZ estimates watts in a pallet or is exact watt and value declared? Do you think there's any risk of currency use, mixes and mix ups by customs or SZ?

[*]Could CSIQ lower Japan ASP lately be because they use more updated exchange rates than others in their declarations?

Share this post


Link to post
Share on other sites

We are surprised that companies that succesfully grew Japanese sales in Q1 saw little reward in projected GASP growth. I think there could be an explanation. They had bad luck. The generous FiT was seriously devalued in USD as it is stated in Japanese Yen. http://finance.yahoo.com/q/bc?t=1y&s=JPYCNY%3DX&l=on&z=l&q=l&c=Usdcny%3Dx%2CEURCNY%3DX%2C+&ql=1

Share this post


Link to post
Share on other sites

We are surprised that companies that succesfully grew Japanese sales in Q1 saw little reward in projected GASP growth. I think there could be an explanation. They had bad luck. The generous FiT was seriously devalued in USD as it is stated in Japanese Yen. http://finance.yahoo.com/q/bc?t=1y&s=JPYCNY%3DX&l=on&z=l&q=l&c=Usdcny%3Dx%2CEURCNY%3DX%2C+&ql=1

Very Interesting observation explo, a point that may have been overlooked by almost everyone.

Share this post


Link to post
Share on other sites

Very important clarification, thanks. So different contract types like CIF and FOB only regulates who takes the cost of the sea transport. The delivery still always occurs pre sea transport regardless of contract (sometimes these deliveries are to subs instead of to third parties as you say and then that delivery is not recognized as external revenue of course)?

Share this post


Link to post
Share on other sites

Thanks, this helps. So with CIF terms (which I think dominate in this industry..?) and FOB (which is second most common) the delivery is done when goods have been cleared for exports and loaded onto the ship. Delivery is the minimum requirement for revenue recognition. Other things like collectibility probability plays a role too if you do risky shipments.

Share this post


Link to post
Share on other sites
Guest Klothilde

Hmm... very good, I learned sth new today as well. I thought revenue recognition for CIF occured at the port of destination but it does occur at the port of shipment. So I agree, the vast majority of PV volumes fall under CIF and FOB and are thus recognized at the port of shipment.

Share this post


Link to post
Share on other sites

CIF and FOB are recognizing the revenue in the same fashion at the named port of shipment. CIF shipper pays transport. If you recall our manual to GADP, this is how we explained the process of shipment in regard of declared value. So I came back to this statement again to review this in context of rev recognition. I think we can also understand that probably most of the Chinese do CIF, but SOL does FOB, they offer a bit lower price of module to compensate for it? The income statement looks a lot cleaner, We will see when volumes of modules come up for them if the price for shipping goes up with it. Magic explained.

Share this post


Link to post
Share on other sites

Just to clarify. I think the companies add CIF cost to revenue (thus ASP) and selling expenses as shipment and handling. So shipment on FOB terms should have around 3 cents lower ASP than CIF. That's my background assumption. I'm not sure how you concluded this "I think we can also understand that probably most of the Chinese do CIF, but SOL does FOB, they offer a bit lower price of module to compensate for it?". So SOL's 63 cents blended Q4 ASP, that due to 50% domestic shipments translates to around 65-66 cents global ASP would be 68 cent if they used CIF terms? Not sure what you meant by this "The income statement looks a lot cleaner"

Share this post


Link to post
Share on other sites

Explo, please simplify. SOL has no shipping costs. This is the first statement. Second is that CIF pays for marine transport. Third FOB does not. You do not know how CIF cost is accounted for but all aspects of the business tell us, it is not not accounted in COGS.

ust to clarify. I think the companies add CIF cost to revenue (thus ASP) and selling expenses as shipment and handling. So shipment on FOB terms should have around 3 cents lower ASP than CIF. That's my background assumption

Your background assumption is good for declaration not accounting purposes. How I concluded is this. I sell a module which has a operating invoice of 3 cents for shipping it. It is priced X. SOL sell is at X-0.2, without operating invoice for shipping across the see, since receiver pays It is a cleaner income statement, since does not heavy opex weight to it.

Share this post


Link to post
Share on other sites

Explo, please simplify. SOL has no shipping costs. This is the first statement.

Whose statement? I think SOL had around 2 cents freight in 2012. The low opex is due to low overhead.

Share this post


Link to post
Share on other sites

Forget low opex, we are on shipping cost. SOL has 1 cent per watt if you use modules only and nothing for shipping wafers. How is everyone having between 2 to 3 cents and they have 1 or less? It has to be due to no shipping costs. They must be shipping in the method that they do not pay for it. If that is the case, buyer who would pay for the shipping cost would get a break on ASP? Is this possible? Transport and I work in the indutry, has a cost that affects everyone the same. No efficiency to be gained. So short of miracle, transportation agreements must be different, which in turn can explain lower ASP on SOL's side.

Share this post


Link to post
Share on other sites

Forget low opex, we are on shipping cost. SOL has 1 cent per watt if you use modules only and nothing for shipping wafers.

Yes, I'm on shipping cost. They had 14m in 2012 on 710 MW module shipments of which 550 MW went on exports, so that's 2 cents per shipped module and 2.5 cents per exported module. The 14m should double in 2013.

So short of miracle, transportation agreements must be different, which in turn can explain lower ASP on SOL's side.

63 cents was in mid range. YGE was at 62, JASO 61, HSOL 60 and JKS 57/59.

Share this post


Link to post
Share on other sites

I have no access to 20-F, this is a total shipping cost, right? How much they spend to ship wafers to Taiwan, Korea, I can recall two out of three locations? Nothing? How did you know that many went to China? ASP for those in Q4 mentioned is diciated by shipments to China. I am told by Jason that SOL exports most of its modules, so it is more like CSIQ, TSL, JASO.

Share this post


Link to post
Share on other sites

I have no access to 20-F

Here's the 20-F section on shipping cost:

http://phx.corporate-ir.net/preview/phoenix.zhtml?c=210622&p=irol-SECText&TEXT=aHR0cDovL2FwaS50ZW5rd2l6YXJkLmNvbS9maWxpbmcueG1sP2lwYWdlPTg4NzcwNDEmRFNFUT0xJlNFUT0xMzUmU1FERVNDPVNFQ1RJT05fUEFHRSZleHA9JnN1YnNpZD01Nw%3d%3d

Here's the whole 20-F:

http://phx.corporate-ir.net/preview/phoenix.zhtml?c=210622&p=irol-SECText&TEXT=aHR0cDovL2FwaS50ZW5rd2l6YXJkLmNvbS9maWxpbmcueG1sP2lwYWdlPTg4NzcwNDEmRFNFUT0wJlNFUT0wJlNRREVTQz1TRUNUSU9OX0VOVElSRSZzdWJzaWQ9NTc%3d

How did you know that many went to China? ASP for those in Q4 mentioned

is diciated by shipments to China. I am told by Jason that SOL exports

most of its modules, so it is more like CSIQ, TSL, JASO.

They said the shipped close to half their modules to China in Q4 (which means they must have increased stocks in Europe) on their ER conference call:

http://seekingalpha.com/article/1273901-renesola-management-discusses-q4-2012-results-earnings-call-transcript?part=single

But in terms of the first question, in the fourth quarter, as we mentioned, a lot of the demand came out of China, I think close to half on the module side.

Share this post


Link to post
Share on other sites

Lets answer this first: 14M represent enitre shipping activity for the company or just for modules and wafers do not have shipping costs?

Share this post


Link to post
Share on other sites

The 18.5% rise in ReneSola's EU GADP from 50.8 cents in January to 60.2 cents in March is interesting against the background of what JulyWebb wrote here: EU Commission agrees provisional duties on China solar panels There's subtle patterns that provide valuable insight in this declared price data. Once we master deciphering it we'll have a huge advantage on the market.

Share this post


Link to post
Share on other sites

LOL, I think you need to get your connection working. :)

In 2012 they shipped 1500 MW wafers and 700 MW modules. Here's what the 2012 20-F says:

Shipping and handling costs incurred on sale of products and included in sales and marketing expense were $4,800,640, $9,058,534 and $14,524,853 for the years ended December 31, 2010, 2011 and 2012, respectively.

In 2009 they shipped 470 MW wafers and very little module. Here's what the 2009 20-F says:

The Company recognizes revenues when persuasive evidence of an arrangement exists, the products are delivered and title and risk of loss has passed to customers, the price to the buyer is fixed and determinable, and collectability is reasonably assured. Revenue includes reimbursement of shipping and handling costs. Shipping and handling costs incurred on sale of products and included in sales and marketing expense were $19,948, $78,705 and $360,779 for the years ended December 31, 2007, 2008 and 2009, respectively.

Share this post


Link to post
Share on other sites

I think couple of things have change since 2009, destination for wafers is no longer China. Shipment cost have gone up. last year in tune of 30%. SOL has the lowest shipment per watt, which indicates different shipment arrangements (my opinion). At this point this conversation has very little relevancy. Other things have taken a first view.

Share this post


Link to post
Share on other sites

Ok, I'm just trying to understand your view better and help out with the analysis if I can. I'm not claiming that the 14.5m freight cost is vastly dominated by module shipping cost, it's just my rough guestimate, so I'm open for this not being the case if there is good reason to suspect so, but it is the basis of your suspicition I have harder to follow. My assumption that roughly only 560 MW of the 716 MW modules shipped went on exports was based on management comments during the Q&A session of the Q4 ER conference call ("close to half of the 320 MW modules were shipped to China"). There's always a risk they say things that are not correct there. Back to the basis. You speculate on whether SOL's low ASP could be due to them including lower freight costs in the ASP than others do (due to SOL using other shipping terms) and thus SOL's freight cost excluded ASP relative to peers would actually rank higher than the freight cost included we see in ERs. The only thing I don't understand here is that, since SOL's ASP already ranks high relative to peers in Q4 ERs, why suspect that ranking for SOL is artificially lower than a freight adjusted ranking would be? In Q4 SOL had fourth highest ASP of Solar9 and probably 3rd highest on the exports. So it is the basis of the speculation I don't follow. So just to clarify, you are comparing reported ASP and not the GADP that we've calculated, right? If so were you expecting SOL to be even better than 63 cents and postion 4 on ASP?

Share this post


Link to post
Share on other sites

If I imply that lower cost of shipment is a result of different arrangement, and as the deal concept it would involved lower selling price than yes that would be indication why ASP was lower in the past, or could have been higher otherwise. Whether this was the case of Q4 or ever, I do not know. Maybe they have changed this formula on shipping and now you see better ASP going forward from Q4, but shipping costs will rise? I am obviously of the observation that shipping costs are the fixed operator for all companies.If I can see benefit in this when I strip everything down to comparable value, what else would it be? Again, the conversation was brought in by different method of shipping costs. I do not really believe this has any importance to any outcome at this point.

Share this post


Link to post
Share on other sites

Ok, it is a good point. If we could strip out the shipping costs included in all reported quarterly ASPs they would be more comparable. I think we can do this on an annual basis (assuming modules dominate the shipping cost), but it might be less interesting. So it is a good point, but we need to figure out if/how we can use it.

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Donate

    Please donate to support this community. We appreciate all donations!

    Donate Sidebar by DevFuse
  • Upcoming Events

    No upcoming events found
  • Forum Statistics

    • Total Topics
      32
    • Total Posts
      91,601
  • Who's Online (See full list)



×
×
  • Create New...