Jump to content
Sign in to follow this  
dydo

Daqo (DQ)

Recommended Posts

massive stupidity.

Only if Wacker was selling to China, and guess what no tariffs on trade.DQ wrote of over $150M in one shot for their move of the plant. I am not 100% sure where DQ accounting stops and group's starts and vice versa. So big caution on this one.

Share this post


Link to post
Share on other sites

The weird thing about the DQ's write dwn, it was for the assets not moved (fixed structures like building) not the moved equipment, so they should not have gotten much lower depreciation cost in Xinjiang plant from this (nothing there was written down).

Share this post


Link to post
Share on other sites

The weird thing about the DQ's write dwn, it was for the assets not moved (fixed structures like building) not the moved equipment, so they should not have gotten much lower depreciation cost in Xinjiang plant from this (nothing there was written down).

or what they say they kept behind, they took a lot money out, maybe for the future costs too, who knows?

Share this post


Link to post
Share on other sites

I simply don't get how they are getting blended cost down this much. Let's hope it is like with locally sourced consumables becoming cheaper in China for other segments that helped get non-Si down to 40 cents. In that case all poly plants in (west) China benefits and poly can become cheaper (all demanded supply must have the same low cost level for the price to come down).

Share this post


Link to post
Share on other sites
The thing I don't know or understand is the end game.

 

The end game is ability to build generation capacity. There's one part in that chain that is harder to secure than the rest. Those that have it gets to build the lucrative generation capacity the coming two decades. Look at what GCL, OCi, Hanwha etc. are doing, letting Chinese fabs toll their poly/wafer and then build PV plants. It will be a couple of years until these cards are played. Building expensive poly plants to let others make money on building generation capacity from cheap poly makes little sense from an ROI perspective.

Share this post


Link to post
Share on other sites

I've gone through DQ's ER and CC a bit. Q3 depreciation was 11.8m. I would assume the small ingot and wafer capacity is only 1.8m of that, i.e. poly is 10m. So that's $6.50 on their targeted 6.15kT before adding the 6kT expansion. $14 at that would imply $7.5 cash cost, which seems too low. And they talked about $4 depreciation before and $3 after. The expansion would be at around 17 $/kg capex. What I think they are doing is lowering depreciation by extending period from 10 to 15 years to get to that $4 before and $3 after. Here's my breakdown:

 

Before expansion: 6.15 kT @ $9.7 cash + $4.3 depreciation = $14

 

After expansion: 12 kT @ $9.2 cash + $2.8 depreciation = $12

 

The difference to SOL's phase II with $14 cash + $2.5 depreciation = $16.5 is mainly the cash cost. SOL said they would do further optimization. Lowered cost target from cash cost optimization and phase III expansion (that they talked about in 2010 20-F, but deferred as the sector entered the depression) to lower depreciation cost and increase capacity to 16 kT would be a smash hit a la DQ's announcement. SOL's Sichuan province is between DQ's old Chongqing and new Xinjiang province (Qinghai is between Sichuan and Xinjiang though), so it is still hard to know how much of DQ's cost improvements are Xinjiang specific and how much is DQ proprietary specific. I find the 3% electricity rate discount sounding borderline irrelevant low.

 

Anyway it will be exciting to hear what SOL and LDK say about what's going on in the China poly scene.

 

$12 fully loaded Siemens, who would have thought?

 

Let the fossils rest in peace.

Share this post


Link to post
Share on other sites

For those familiar with China. When central government says they want to stimulate western China with support, which provinces does that typically include? Only Xinjiang or Qinghai and Gansu too or even Sichuan and some others?

Share this post


Link to post
Share on other sites

Another interesting point on the CC is that they said FBR poly has $3-4 lower ASP than Siemens poly. If REC says a new FBR plant in China could reach $11 cost and GCL says $10, then it does not sound very competitive with Siemens that now can be built with $12 cost if the FBR has $3-4 lower value due to the lower quality. DQ was also a bit sceptical that anyone else (read GCL claim) than REC could build a successful FBR plant with high certainty, since so much in FBR to get it efficient is proprietary technology.

Share this post


Link to post
Share on other sites
Guest eysteinh

Explo rec fbr b cash cost is 7.9 and total cost 11.3. Capex 50. Acording to 2012 presentation. Gcl cash cost is 9.About 15$/kg capex. The new versions will be all eg quality and since also granular should catch a premium when mono takes off. But yes siemens is very close now.

Share this post


Link to post
Share on other sites

Are you saying that quality will be the same as Siemens? Is it reasonable to expect capex to be lower than Siemens as GCL $15 suggest? So FBR is still the future, despite the low electricity cost of Siemens now?

Share this post


Link to post
Share on other sites
Guest eysteinh

Short answer: Yes. (rec fbr b) No Gcl fbr and rec fbr a (see source below)  A bit longer answer: same or some would argue even better quality since it is granulars and thus have economic advantages. Others would argue there could be dust off if not packaged well and thus slightly worse than Siemens (but imho it depends on packaging and dedusting methods used.)  But what I can say with a fair confidence is that they should be at around equal footing and for sure not 3-4$/kg if we are talking electronic grade quality they are selling. Solar grade on the other hand could have a discount until more and more mono producers starts realizing the economic benefits of using granulars.  About capex cost vs simens, in theory silane based fbr should require more costs since it has a lot more safety precautions to make. But I do not think it should be far behind simens no, especially for gcl who allready invested in the silane part and the rest could be considered upgrade to the silane infrastructure if they build it all at the same site. Anyhow a bit shocking news but here is about gcl: http://translate.google.com/translate?depth=1&nv=1&rurl=translate.google.com&sl=auto&tl=en&u=http://www.xedz.gov.cn/a/wzgg/2013/1118/3810.html

 

90% solar grade so low quality, 10% electronic grade. 54000 Mt. Around 6$/kg cost (add up for the silane part and you get around 15$ if i remember correctly. Could look into it if needed.) 

 

Current state of gcl fbr: http://www.scmp.com/business/commodities/article/1338208/polysilicon-recovery-be-slow-and-painful-gcl-polys-shu-hua-says

 

GCL has invested more than 200 million yuan (HK$254.8 million) on a 1,000 tonnes a year trial production line based on the fluidised bed reactor (FBR) method.

 

(33$/kg for the test reactor - so we allready know at small scale they can do around 30$/kg)

 

If fbr is the future the verdict is not out yet, I have not yet seen GCL finish this successful - especially in actual operation with all the explotion dangers with silane -  and for all of REC's fbr plants they have not actually implemented any of them and instead they are upgrading some of fbr a to fbr b in US. (I once suggested this to the board 1,5 year ago, but the scenario then where different - imho to me they are moving far far to slow, i even suggested to them 3 years ago to move more into systems.) For me this looks like saving capex, but will at best put them on equal footing than china due to mgs tariff.  If they could get around mgs tariff then this would play out differently. So i am interested what comes out of the us-china deals. And i think they could in theory process mgs and ship it into us then circumvent mgs tariff. Not sure it would be worth the cost but mgs is around 1,5$/kg or more expensive in us. 

Share this post


Link to post
Share on other sites

I thought the fbr process made it more difficult to aviod contaminations and thus made it difficult to attain same purity as Siemens? Is this problem solved in the b tech you talk about?

Share this post


Link to post
Share on other sites
Guest eysteinh

Yes. It is harder because most designs have the seed particles bouncing in the walls.- thus making metal containment. The easy solution now patented by rec is to cover the walls in silicon material so no contaminating happens. It gets a bit harder as you also need energy efficient heating of the walls (rec also have a new patent on this.) and also you need cost effective wall jackets that can be replaced. But it is definitively possible. Same with the silane going through the pipes to the reactor. I have previously linked a lot of patents, if you read some of the newer ones you will see this will be solved. I am fairly confident that we will see eq quality fbr. The pilot reactor of rec allready produce eg. Memc have produced eq quality fbr allready many years due to having a 3 reactor design (each granular goes through 3 reactors so much more costly but it is also pure eq quality since the reactors steps also cleans the polysilicon granulars.) Also there are more steps, like how you need to distill and make very pure silane from mgs and make this cost effective (look at newer patents on rec you will see they figured this out.) but as a very simplified explenation again, yes it is possible to produce eq quality on pair with simens with fbr.  When you see gcl posting 90% sg and 10% eq you can be clear they have not yet circumvented any patents on how to get all eq quality just yet. But I think they will get there eventually. 

 

But then again, if capex is high for eq quality (last posted number was 50$/kg in china by rec in 2012, the presentation is now removed from the site but if you need it I can email it to you.) So would it be worht it the price gain if you at the same time have high deprication? I am not sure, things move fast in the solar world. I think what gcl is doing is very smart. They are going hard for solar grade and at a very very cheap capex. this will drive down costs and will make room for both daqo simens eq and gcl cheap solar grade. I know JKS would love cheap solar grade as they allready have lots of patents to make use of solar grade and produce same quality modules as with electronic grade.

Bottom line for me currently: Simens is gaining ground for most producers since it will be on pair cost almost and eq quality. A few specialized producers and mono producers will love gcl solar grade fbr. Rec will struggle and introduce electronic grade fbr. This will in the end mean driving down costs for all polysilicon, and that is good news for overall competitiveness of c-si vs other energy sources or 2nd or 3rd generation solar modules (cdte and cigs). Especially when 3rd generation tech like nano structures will actually just boost 1st tech like c-si. Short example, one type of nanostructure will make the photon split into 2 photons in a c-si cell effectively doubling efficiency... (but that's future tech like 2020 +)

 

Anyhow for me this means c-si still is the clear clear winner going forward. So I am happy I am in some of the c-si manufacturers. Try to be in system producers (with lower module costs they will have fatter margins - csiq does come to mind but I am sure there will be others) and also spot poly-silicon players (jinkosolar) is my line of thinking currently.

Share this post


Link to post
Share on other sites
Guest eysteinh

Like I said I will follow gcl but until it is actually up there we don't know. All we know is daqo managed very cheap capex now. Why are you sorry for my sake? I have laid it out quite fair and square and even warned of the same thing you said. If gcl manage 9$/kg cash cost on fbr they will not need 20$/kg asp. But yes currently they seem to need so. Perhaps because of sg&a and interest? Even rec with sg&a and interest need above 20$/kg. I would think the same for daqo too. (and sol) currently. 2015 will be different because daqo will lower a lot at this point. If gcl fbr at 9$/kg is introduced also gcl will need less. Rec with fbr b will have higher asp (3-4 $/kg more).  What makes the capex number more belivable is that daqo is doing it currently at very low costs. That is the real story for me currently as it influences the belivability of also others to manage very low capex. That means old 100$-130$/kg capex factories who have not yet written down a lot will need to do so now as more and more cheaper and better factories come online. 

Share this post


Link to post
Share on other sites

Thanks for the explanation Eystein. It still seems like if DQ have reached cash cost in the $9's, why make the effort with FBR? I know mixing FBR granules and Siemens chunks can improved furnace feed process during ingot growth, so it would still make some sense although more effort for same cost. You're saying mono benefits more from this? But mono is also more quality sensitive which still sounds like Siemens preference to me..?

Share this post


Link to post
Share on other sites

And I agree with you, companies with much higher capex like GCL will have to take massive writedowns.

 

Why?

Share this post


Link to post
Share on other sites
Guest eysteinh

because many previously have modeled 6-7$/kg depreciation over time and this will simply not be happening if we are seeing 16-17$/kg total cost. But gcl I am not sure have higher capex (fbr will be very low capex according to the latest news) , Rec already did a big writedown of silicon factory, and daqo did so for old equipment and do not need this for future. Question then is what about sol I am holding. I honestly dont know, and I am interested to read this in the cc. Luckily sol have a lot of other good things going for it. 

Share this post


Link to post
Share on other sites

SOL will not write down. DQ wrote down the chongqing plant, because a poly plant without any poly equipment has low value. REC on the other hand wrote down extreme amounts. Even RECSOL. Imo they are almost cheating as it looks like they have better return on assets now than they actually have. It's easy to be profitable if all costs are written off and all debt is cleared, like in the case of RECSOL. If SOL did not think write down (cheating) was appropriate even when poly price was $15 last December, they won't now when it is $18.

Share this post


Link to post
Share on other sites

because many previously have modeled 6-7$/kg depreciation over time and this will simply not be happening if we are seeing 16-17$/kg total cost. But gcl I am not sure have higher capex (fbr will be very low capex according to the latest news) , Rec already did a big writedown of silicon factory, and daqo did so for old equipment and do not need this for future. Question then is what about sol I am holding. I honestly dont know, and I am interested to read this in the cc. Luckily sol have a lot of other good things going for it. 

 

Eyestein the factor which dictates the write down of the asset or the adjustment to its value comes from the accounting test, which starts with the ability of this asset to produce returns at profitable gross margin. So another words if the company is unable to produce profit, the test is automatically executed. It does not mean that the asset is rendered useless. It is actually very difficult and meticulous format and made individually for a machine to a plant.

 

I guess what I am saying is that COGS includes the depreciation as you know. Depreciation or amortization allows you to offset your revenue on the bases of the recovery the spent on capital expense or CAPEX, which is part of your asset, when spent.

At the end if ASP is higher than your COGS, impairment does not exist. 

Example GCL sold poly at $17 during the first half of 2013. They claim to reach to $17.1 cost per kg on June 30th. The company took an impairment during the period as clearly without it the amortization portion exceeded the ASP. Upon impairment they stopped.

Since you guys are seeing the ASP growth, the impairment does not apply.  If the company is quoting lower processing costs as part of its blend cost it means that CAPEX per kg, is a lot lower.  The amount would have to be significantly low to lower the blend, but writing off is not a choice, but accounting mechanism

Poly plant GCL 

 

"The recoverable amounts of the plant and machinery belonging to the production plant of polysilicon 
in the solar business with carrying amounts of HK$17,980,563,000 as at 30 June 2013 are determined 
based on a value in use calculation by the directors of the Company. The calculation uses cash flow 
projections covering the useful lives of those property, plant and equipment in relation to the production 
of polysilicon based on a 5-years financial budget approved by management at a discount rate of 
13.34% (31 December 2012: 13.43%). Cash flows beyond the five-year period are extrapolated using 
zero growth rate. Other key assumptions for the value in use calculations relate to the estimation of 
cash inflows/outflows include budgeted sales and gross margin. Such estimation is based on past 
performance and management’s expectations for the market. As the value in use is lower than the 
carrying amounts of the related assets, as a result, an impairment loss of HK$143,558,000 (year ended 
31 December 2012: HK$552,961,000) is recognised on property, plant and equipment in relation to 
the production of polysilicon accordingly."
And wafer 

"Due to deteriorating profitability and downturn of the market for wafer, primarily as a result of lower 

demand in the market, and cessation of the construction of certain wafer production plants during the 
period, the management determined to write-off the relevant assets. As a result, impairment losses of 
HK$111,812,000 (year ended 31 December 2012: HK$312,147,000) are recognised on those property, 
plant and equipment."

Share this post


Link to post
Share on other sites

SOL will not write down. DQ wrote down the chongqing plant, because a poly plant without any poly equipment has low value.

SOL will not write down the plant now, but they have in the past. Since they use the vertical from poly to module, the plant itself is just a spoke in a wheel, as they do not do external sales to trigger such a test immediately, instead of something set for once per year. Instead. SOL adjusts value of the inventory of own made poly. They took the adjustment in Q2, but it was few thousands, versus millions. I suspect another one to take place due to stoppage of the plant. 

Share this post


Link to post
Share on other sites
Guest joshchang

Cause if DQ can built a 1/5th poly plant with all in costs of $12, and GCL is wasting 25-43% margins to break even at $20 when we currently sit at $17, they can use all the help they can get. And even based on you, there is not a chance in hell that we will see over $20 in 2014. So the grand canyon is smaller than GCL ever seeing profitability on their poly operation as it currently stands. And that assumes that DQ isn't lying which is a big assumption in itself. JMHO.

GCL was producing at $17 and even if they stated $20 to start showing profit on their income statement, that's about 3/20=15% GM to break even. Required GM to break even is lower than some low cost module producers out there. With more production in 2014 and targeting cost to $14/kg, ASP required to break even on income statement should drop another $3/kg or more. So GCL will make money at $18/kg poly price and profit more if poly goes higher next year. On cash flow aspect, they will make considerably good amount of cash flow at $18/kg ASP.

 

We just have to admit, GCL and DQ are steps ahead of SOL in producing low cost poly. For DQ and GCL, they are not counting on poly ASP increase next year to make money on poly production but keep improving on cost to achieve profitability.

 

BTW, Zhu already said they started to break even at the beginning of October. 

 

保利协鑫光伏业务10月份起已全部扭亏为盈

 

 

http://news.xinhuanet.com/fortune/2013-11/13/c_125693910.htm

Share this post


Link to post
Share on other sites

SOL will not write down the plant now, but they have in the past. Since they use the vertical from poly to module, the plant itself is just a spoke in a wheel, as they do not do external sales to trigger such a test immediately, instead of something set for once per year. Instead. SOL adjusts value of the inventory of own made poly. They took the adjustment in Q2, but it was few thousands, versus millions. I suspect another one to take place due to stoppage of the plant. 

 

They have not taken any impairment on the poly plant. Only impairment SOL has taken was on some old mono furnaces last year. They took some inventory write downs in 2012, but who didn't? This year they did not have to do that (compare with HSOL who still seem to produce too much goods above their selling price, both report cash flow).

 

To be honest I do not know how to account for stoppage due to natural disasters. Stoppage for upgrades are not charged (similar to not starting depreciation until construction is done and asset is ready for use). Stoppage for not having enough customers is charged as asset is ready for use, but is still not utilized due to inability to sell produced goods. For a flood, who knows..?

Share this post


Link to post
Share on other sites
Guest joshchang

They have not taken any impairment on the poly plant. Only impairment SOL has taken was on some old mono furnaces last year. They took some inventory write downs in 2012, but who didn't? This year they did not have to do that (compare with HSOL who still seem to produce too much goods above their selling price, both report cash flow).

 

To be honest I do not know how to account for stoppage due to natural disasters. Stoppage for upgrades are not charged (similar to not starting depreciation until construction is done and asset is ready for use). Stoppage for not having enough customers is charged as asset is ready for use, but is still not utilized due to inability to sell produced goods. For a flood, who knows..?

I don't think SOL needs to do impairment test on fixed asset (poly plant) now since they use most of their poly internally. If they are selling poly into the market and are producing poly below market price in a normal operation fashion (not in optimization status), they might need a impairment test. Whether to write down in the test is determined by reasonable doubt of whether their producing cost would be consistently lower than market ASP. Right now, I think they are safe not to do a write down.

Share this post


Link to post
Share on other sites
Guest joshchang

Josh, the only thing that anyone can admit with any certainty is what GCL themselves said. That they need $20 ASP to break even. GCL is oceans away from profitability. SOL by comparison is a stone away from profitability. GCL just got in a whole lot of trouble if DQ's numbers are real. But I'll give you that is a big IF.

I was comparing GCL's poly business with SOL's poly business not with SOL's overall business. I don't care if that $20 ASP is true or not. All that I care is whether GCL can keep driving down cost and providing low cost poly/wafer to chinese partners.

Share this post


Link to post
Share on other sites

They took some inventory write downs in 2012

They took inventory writedown in Q2 2013.  I do not have a time to look this up. I thought they have done something in 2009.

I don't think SOL needs to do impairment test on fixed asset (poly plant) now since they use most of their poly internally. If they are selling poly into the market and are producing poly below market price in a normal operation fashion (not in optimization status), they might need a impairment test.

As I stated, selling to the market is different than internal production. Companies do not do tests, unless they have negative GMs. That may not even cause test as they may have the period, half a year and year tests. 

Share this post


Link to post
Share on other sites
Guest joshchang

GCL has 65,000 MT of poly operations, that if DQ's numbers are right, just became a massive worthless weight on them. By comparison SOL has only 3,500MT (their old poly plant), while their new Phase II is brand new and should have the same cost basis as DQ (plus $1/kg for electricity advantage).

Their overall poly cost is about $17/kg by the end of Q2 and with asp of $18-$19 I don't see why that's a worthless asset. And as long as their cash cost (might be in $13-$14) is under market ASP then they should keep producing poly. Don't forget capital invested in the past is sunk cost and all going forward decision matters is cash cost. Again, they are targeting at $14 all in cost next year and that would give them huge cash flow if they can achieve it.

Share this post


Link to post
Share on other sites

GCL has 65,000 MT of poly operations, that if DQ's numbers are right, just became a massive worthless weight on them.

What are you saying?

Share this post


Link to post
Share on other sites
Guest eysteinh

Odyd I am well aware of impairment tests. Used to read the basis of recs impairment tests in the annual reports before. I think everything I have written still stands. (also please note that while I have seen that it is possible for a squeeze in 2014, the long term trend of asp is down, so impairments might come. And it sure is coming down faster if capex for new equipment can be found in 15-30$/kg range and producing at 12$/kg. Someone is bound to invest more then.) Thats why I said the real story of daqo was the extremly cheap capex they are spending.   

Btw I think gcl might actually just upgrade the capacity from simens to fbr. Will be interesting to follow.  Anyhow good discussion. When it comes to SOL we just have to see the cc before we know more. I have absolutly zero news about renesola polysilicon, but would it be possible if we had the adress of the plant to just google map it to see if it was running ahead full scale? (just stuff like how many cars parked at the facility etc.) I know a bit crazy but just trying to think out of the box. 

Share this post


Link to post
Share on other sites

Odyd I am well aware of impairment tests. Used to read the basis of recs impairment tests in the annual reports before. I think everything I have written still stands. 

Reason for impairment test is a negative margin, then there is an exploratory mission to determine which asset is not creating profitability.  Asset value is taking down, reducing the amortized amount, literally removing expense from the COGS. If the group here saying that ASP for poly is claiming you will have not have impairments

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
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
      33
    • Total Posts
      92,381
  • Who's Online (See full list)



×
×
  • Create New...