PV manufacturing capacity expansion announcements top 55GW in 2015

Share on facebook
Share on twitter
Share on linkedin
Share on reddit
Share on email

PV Tech’s analysis of PV manufacturing capacity expansion announcements in 2015, which includes thin-film, solar cell, dedicated module assembly and integrated cell and module categories more than doubled (21GW:2014) year-on-year to reach 55.19GW. 

However, no one should panic that a new cycle of overcapacity is about to hit the industry in 2016, as it is important to reiterate that this figure is all inclusive. By that we mean that it accounts for actual plans that are expected to become ‘effective capacity’ over a 12-month period or extended period over several years of phased expansions, as well as memorandum of understanding (MOU) and letters of intent (LOI), that may never happen. 

Indeed, in certain cases announcements made in 2015 have already been cancelled (see below analysis of Hanergy Thin Film), companies have gone bankrupt and previously announced plans (SunPower) re-announced.

As plotted in 2014, announcements by new entrants and start-up’s have much longer lead times, primarily due to difficulty in raising finance for production plants, or are simply dropped. 

Specific to 2015 was a ridiculous number of MOUs and LOIs surrounding manufacturing across all segments, including polysilicon that related to India. The vast majority of these announcements have not been counted, however there is around 6GW of announcements that we have recorded. This is due to the specific companies involved that have a history of good execution as well as those that have finance in place and some plans have already started to be implemented. However, that still leaves several gigawatts of announcements in India that have yet to progress from paper plans. 

Also more specific to 2015 was the scale of many announcements in the 1GW plus range. However, many of these have more humble initial production capacity ramps than the headline figures.

Therefore a complete refresh and reappraisal of all announcements on a monthly basis was undertaken with the intention of providing a more measured headline figure for 2015. 

Capacity expansion announcements on a monthly basis.

Effective capacity announcements in 2015

We were therefore able to eliminate over 15GW of announcements in 2015 that failed the validity tests, at this time. As a result a more measured figure for capacity expansion announcements in 2015 is 39.87GW.

Going further and only counting initial phase capacity expansions as being valid at this time, we can eliminate a further 10.7GW, bringing the most realistic figure of announcements in 2015 that could potentially migrate to effective capacity over the next 12 to 24 months, to around 29GW. 

Capacity expansion announcements on a quarterly basis.

November reappraised

We also undertook a specific reappraisal of the record announcements made in November, 2015. We had reported preliminary figures for November of over 17GW of new planned capacity expansions, but the reappraisal after the dust had settled revealed an extra 7.7GW of capacity planned, bringing the total to over 25GW. 

However, including the validity tests and checks, November stood out for including previously announced capacity expansion plans that would have lead to duplication, as well as phantom capacity due to outsourcing and the inclusion of small-scale initial phase capacity attached to multiple gigawatt future phased expansions with uncertain timescales. 

As a result November’s figures were cut by 11GW, bringing the realistic figure down to 14GW.

November set several individual benchmarks such as setting a new record monthly and subsequently quarterly figure. 

A key reason for this change was the massive involvement of ‘Silicon Module Super League’ (SMSL) members (Trina Solar, Canadian Solar, JinkoSolar, JA Solar and Hanwha Q CELLS) announcing significant expansions as existing facilities in China, which remain non-integrated classification as well as new and further expansions of facilities outside China, namely Malaysia, Thailand and South Korea. 

Please note that clicking on the company specific links above provides an in-depth analysis of each company’s plans that were announced in November. 

Thin film

All thin-film type capacity expansions announced totalled around 4.4GW in 2015. All but 1MW of a-Si thin-film announcements came from Hanergy Thin Film, which totalled 2.7GW.

However, all the customers have since cancelled contracts, negating completely these capacity plans.

More validity exits with CIGS thin-film plans, which totalled over 1.6GW in 2015. Initial production ramps and expansions phases are only expected to be around 500MW, which includes the odd R&D pilot line. The bulk of the planned CIGS expansion relates to an eventual 1.5GW ramp over multiple years of the AVANCIS/CNBM plant in China. 

Also excluding Hanergy TF from 2014 announcements means thin-film expansions totalled just over 650MW that year, primarily due to First Solar and a lesser extent, Solar Frontier, both notable for executing on those plans and becoming effective capacity additions through 2015. 

The relatively inactive thin-film segment (excluding First Solar and Solar Frontier) marks several more years of thin-film underachievement and the growing dominance of crystalline silicon technologies.

c-Si solar cells

In that respect, dedicated c-Si solar cell expansion plans topped 20.4GW in 2015, compared to 8GW in 2014. 

Interestingly, dedicated module assembly announcements failed to exceed those for solar cells, reaching over 18.5GW in 2015, compared to 12.4GW in 2014. 

We had highlighted in monthly analysis reports last year that the lack of new cell capacity plans in 2014, compared to module assembly would need to be rebalanced, this is indeed what has happened in 2015. 

Another strong trend through the first nine months of 2015 had been the plans for integrated cell and module expansions, a trend not seen in 2014, when dedicated module assembly expansion announcements dominated. 

However, the integrated cell and module category gained no further traction in the fourth quarter of 2015, replaced by a much stronger trend of dedicated solar cell capacity and dedicated module assembly capacity expansion announcements. 

Full-year breakout highlights that integrated cell and module plans peaked at just over 6.3GW.

Multi verses mono

The year was also marked by the number and scale of announcements related to N-type and P-Type monocrystalline cell and module assembly capacity expansion plans, which reached a combined total of nearly 9.4GW, compared to 4.38GW in 2014.

N-type mono cell and module assembly (including heterojunction) announcements topped 2.23GW in 2015, compared to 2.38GW in 2014. However, applying the validity tests we can eliminate 500MW from the 2014 figures and 1.4GW from the 2015 figures, at this time, resulting in realistic figures of 1.88GW in 2014 and 830MW in 2015.

Clearly, the momentum for mono has built momentum over the last two years but N-type has been in moving at a much slower pace. 

That is not the case with multicrystalline solar cell and module assembly (includes integrated cell/module) capacity expansion announcements, which topped a combined total of over 41.5GW in 2015. In 2014, multicrystalline announcements stood at over 14GW and therefore remains the dominate technology and is outstripping monocrystalline technologies momentum. 

C-Si module assembly

The capacity expansion figures that mean the most relate to c-Si module assembly, which includes integrated and dedicated segments. 

Total c-Si module assembly capacity expansion announcements in 2015, totalled around 27.5GW. 

However, validity tests indicate almost 9GW of announcements are highly suspect, based on duplication, financing and past history of company’s execution.

This would bring the total down to around 18.6GW, still a sizable number. Simply applying analysis undertaken on 2014 announcements to actual effective capacity metrics (see below analysis) it would be prudent at this time to discount a further 30%, simply from an overall execution perspective. 

Therefore, around a further 5.58GW could be expected not to materialise from 2015 announcements in the next 24 months. Potential effective expansions of c-Si module assembly capacity from 2015 could be around 13GW. 

Global regional split of planned capacity expansion announcements in 2015.

Regional shifts

Looking at the geographical chart where capacity expansion announcements were targeted in 2015, disguises the fact that the number and scale of expansions announced in China in the first half of the year only reached 4.7GW and more than half was related to Hanergy TF.

No major China-based module manufacturer announced expansion plans in China in the first half of the year. Instead, they announced more than 6.7GW of planned capacity expansions in a number of overseas countries, including India, Malaysia, Thailand, South Korea, Brazil and the US. 

The lack of new capacity expansions in China contrasts with over 12GW of new announcements in 2015 for other regions across Asia, with Chinese producers accounting for just over half of the capacity announcement figures. 

However, the second half of 2015 went a long way to restoring the balance but Southeast Asia remains a hot destination. 

New capacity announcements in China increased by over 10GW in the fourth quarter of 2015 alone, resulting in the total reaching over 17.5GW in 2015 and remains the destination leader, but the figure is below the 19GW announced in the previous year. 

Although India is clearly the second largest destination with over 7.8GW of announced plans in 2015, and only 1.4GW in 2014, less than 1GW has moved from the MOU/LOI category but is certainly the emerging destination to watch in 2016. 

A similar situation exists in relation to Brazil, which has exceeded 1GW of announced capacity plans but again the effective capacity by the end of the year was almost non-existent. Yet, with gigawatts of PV power plant projects planned, effective capacity activity could move forward in 2016. 

One of the surprises for 2015 was South Korea, which generated just over 3GW of capacity expansion announcements. Already in 2016, LG Electronics has built on that momentum with plans to expand N-type monocrystalline cell production from the current 1GW nameplate capacity to 3GW by 2020. 

Both Germany and the US did not disappoint with both countries nudging announcements around the 1.5GW mark, with the US beating Taiwan for the second consecutive year. 

Interestingly, the figures for Malaysia (2.21GW) and Thailand (2.25GW) may not accurately represent final figures as a number of companies such as Trina Solar, JinkoSolar and SunPower have stated some of their capacity expansion by location but not yet confirmed several gigawatts of planned expansions, all potentially in either Southeast Asian country. 

The overriding trend from a location perspective is that manufacturing outside China has seen significant growth and the emergence of a larger global footprint for manufacturing is taking shape. 

By the end of 2015, the total cumulative effective c-Si module assembly capacity added from 10.7GW of 2014 announcements had reached nearly 6.4GW.

C-Si module assembly effective capacity expansion ramp analysis for 2014

As already highlighted, equating capacity expansion announcements with ‘effective’ capacity is misleading. Understandably, there is a time lag between many announcements and the time that capacity comes on stream and potentially meets its annual nameplate figure within the 12-month period from the start of the ramp.

A recent reappraisal of all 2014 c-Si module assembly capacity expansion announcements was made. This was to determine whether the plans happened and plot the effective ramp rate timelines and the overall effective capacity additions made to the global PV industry from 2014 announcements through to when the final phase of those expansions can be treated as completed effective capacity.

It should be noted that the analysis also includes and adjusts for effective capacity added but for other reasons, such as bankruptcies or sector exits, was later shutdown. 

Not surprisingly, there was a short period of inactivity from announcements made early in 2014, but an initial momentum phase started in March, 2014 and peaked in July. A plateau appeared through the remaining nine-months of 2014, due to a stall in further new planned expansions beginning to ramp.

At the end of 2014, a total of around 2.57GW of effective capacity had been added to global c-Si module assembly capacity from 2014 announcements that had totalled 10.7GW. This equated to only a 24% conversion rate. 

However, significant ramps kicked-off again in January, 2015 with the first quarter adding just around 1.4GW. By mid-year cumulative effective c-Si module assembly capacity had reached just short of 5GW from 2014 announcements, almost a 47% conversion rate. 

By the end of 2015, the total cumulative effective c-Si module assembly capacity added from 10.7GW of 2014 announcements had reached nearly 6.4GW. 

Importantly, by mid-2016 all effective c-Si module assembly capacity from 2014 announcements would be accounted for, bringing the total effective nameplate capacity to around 7.33GW from the 10.7GW of 2014 announcements. 

Therefore, around 68.5% of c-Si module assembly capacity expansion announcements from 2014 would have converted to effective nameplate capacity by mid-2016, ending incremental nameplate capacity from that year.

The analysis reaffirms that the headline figures from 2014 get diluted over time (32.5% in that year) while effective nameplate timescales can stretch much longer than many may have expected. 

Of course effective c-Si module assembly capacity from 2015 announcements has to be included to get the closest realistic picture of completed effective new nameplate capacity for that year. 

Taking the assessment of effective capacity additions from 2014 announcements by the end of 2015, a potential 3.25GW could have been added to the total. This would have meant around 10.5GW of new effective c-Si module assembly capacity was added in 2015. 

One challenge with these assumptions is that preliminary analysis indicates a larger number of companies with much larger capacity plans than those announced in early 2015 existed, indicating that the potential 3.25GW could be highly conservative. 

However, should end-market demand in 2015 prove to be in the range of various estimates of 57GW (BNEF, IHS) to 59GW (GTM), adding in effective thin-film module capacity, the actual effective supply of PV modules has simply kept pace with end-market demand since the beginning of 2014.

We plan to publish analysis of capacity expansion announcements made in January, 2016, later this week. 

PV Tech and Solar Intelligence plan to present the most comprehensive and in-depth analysis of the latest manufacturing capacity expansion developments at the inaugural PV CellTech technical conference being held in Kuala Lumpur, Malaysia between 16 – 17 March, 2016. More details on how to attend can be viewed here. 

Read Next

June 22, 2022
Trina Solar has started construction on a vast new industrial park in the central Chinese province of Qinghai that will cover almost the entire PV manufacturing chain from polysilicon production all the way down to modules.
June 22, 2022
Tongwei has announced that its subsidiaries have signed a RMB50.9 billion (US$7.58 billion) silicon material sales contract for 216,100 tons with Qinghai Gaojing Solar Energy Technology Co., Ltd.
PV Tech Premium
June 20, 2022
After years of protracted disputes and business uncertainty, the US solar industry will soon find out the precise terms of the Uyghur Forced Labor Prevention Act (UFLPA) and its implications for the supply of solar products to the country.
June 13, 2022
The Indian solar sector looks set for significant levels of disruptions over the next two years, with the government’s drive to establish a domestic manufacturing base risking deployment and investments needing to double if it is to meet its climate targets, according to industry analysts.
June 9, 2022
US conglomerate General Electric (GE) is looking to triple its solar and battery energy storage manufacturing capacity at its newly launched Renewable Hybrids factory in India by the end of 2022 to 9GW per annum. 
June 6, 2022
Solar inverter manufacturer SMA has unveiled plans to double its production capacity by 2024 at its Germany headquarters.

Subscribe to Newsletter

Upcoming Events

Solar Media Events
October 4, 2022
New York, USA
Solar Media Events
October 11, 2022
Virtual event