Finlay Colville
Over the past few decades – from when solar PV moved into mass production status – the polysilicon sector was often the key part of the manufacturing value-chain to track from a market research perspective.
This was certainly the case between 2020 to 2024. First, when polysilicon shortages led to price increases that ended up dominating the overall cost in manufacturing a module; then as oversupply saw polysilicon ASPs collapse, allowing module suppliers to cushion the rate of margin compression while module ASPs themselves fell to record lows today.
From the second half of 2024, the cost of glass has returned to being the largest part of module costs (or at least comparable to polysilicon costs depending on cyclical monthly changes), despite glass manufacturing costs themselves seeing strong erosion.
A more complex and moving landscape than simply looking at grams-per-Watt for polysilicon consumption, glass metrics have been influenced strongly by the move from 3.2mm to 2mm thickness and the percentage of bifacial (glass/glass) modules being produced today.
Factoring this in reveals that glass costs are now the key contribution to overall module costs (or at a minimum comparable to polysilicon), reflected by the inclusion of glass suppliers now routinely among the Top 5 suppliers (by revenues) for leading c-Si manufacturers – previously a domain largely reserved for polysilicon suppliers.
The increasing importance of the solar glass sector today
It is not simply the percentage of sales costs (or COGS) from glass that is promoting the importance of solar glass today – there are other factors at play.
First, glass supply is by nature far more closely correlated with module supply than polysilicon which can be, and has been over the past 12 months, subject to chronic inventory build-up and uncertainty on precise China polysilicon production and shipment volumes (as plants are taken offline and then brought back into production).
Second, glass manufacturing can be set up relatively quickly outside China as was seen in recent years across Southeast Asia (to assist module supply to the U.S. market) from the likes of Malaysia before to Indonesia today; and characterized by the current flurry of activity in the Middle-East & Africa and eastern Europe (for example, Turkey), again very much with a view to serving c-Si supply-chains making modules targeted for the U.S. market.
Tracking the metrics shown in the graph above
To fully understand the market trends over the past five years, several metrics are shown in the graphic above and are now discussed below.
On the primary axis of the graph (to the left), metrics are shown for eight of the leading glass suppliers to the PV industry today. The upper (solid) line shows the increased market share contribution to levels above 90% today. The lower (solid) line tracks the consolidated gross margin from this grouping, summing up the individual solar PV glass specific revenues and production costs from the different companies. Therefore, this is a blended gross margin for the market-leaders – but is heavily weighted to the profits from the top 2-3 solar glass suppliers today.
Like other segments of the PV manufacturing and materials supply- and value-chains, there are companies operating in the glass supply stage that are seeing heavy losses today (and indeed were loss making prior to the current manufacturing downturn starting). However, some of these producers are shielded by parent company state-owned operations and appear to have mandates to expand overseas when pure economics would suggest that only greater losses would be incurred by such actions.
For the market leaders today, thoughts of returning to 30-40% gross margins are likely well off the table, but increasing sales volumes while squeezing overall operating costs over the next 12-18 months may create an ideal business model as annual module production moves to the terawatt level in coming years.
The second ‘part’ of the graph above relates to the two dotted lines (plotted on the secondary, or right-hand, axis). This shows the percentage of module costs (based on global best-in-class module production costs, characteristic of the leading Chinese companies) blended annually from 2020 to 2024 (actual) and forecast for 2025.
The rise-and-fall of polysilicon pricing (and subsequent percentage share of overall module COGS) between 2020 and 2024 was alluded to above and – short of an ‘act of God’ – is unlikely to happen again anytime soon. This is shown by the upper (2021 to 2023) dotted line.
The ‘lower’ dotted line (marginally higher in 2025) is the module COGS contribution from solar glass. Interestingly, this looks to be somewhat flat over the whole five year period, but this masks factors that have helped reduce costs (3.2mm to 2mm thickness, lower energy input costs, compression of glass sales margins) while others that have increased costs (in particular, the move to bifaciality in the form of glass/glass module design).
Correlation between glass shipments & market demand should be keenly watched now
In a few weeks from now, once results are released for Q3’25, a much clearer picture will emerge on what the year as a whole may look like, in terms of module production volumes (and margins).
While module production levels today are well above annual deployment volumes for the industry (owing to increased levels of warranty replacements from a larger installed base and growing repowering of existing solar farms), glass production and shipment is for now the best leading indicator for all PV market analysts.
Beyond this, and looking into 2026, solar glass supply is likely to see far greater scrutiny than in the past; noting that when glass supply was previously the dominant part of module COGS, the PV sector was not being subjected to the ESG and traceability demands that characterize supply-chains today.
In part, greater scrutiny on PV glass is coming from the changing geographic landscape for regional manufacturing (still new additions in Southeast Asia in the likes of Indonesia, increased production in India, the new facilities coming online in the MEA region and eastern Europe including Turkey, and domestic production ramp-up in the U.S. for First Solar and new c-Si module assembly plants). The role of recycling is also going be become even more important, as glass usage rates and the current market-share dominance from China becomes more widely known.
However, as noted above, increased scrutiny is also likely to be seen as module buyers are required to understand glass supply chains, in a similar way to when polysilicon companies’ business operations were highlighted in 2020.
Either way, expect 2026 to see global solar glass supply emerge as one of the most topical aspects of PV manufacturing coverage; and a greater understanding of the companies behind this increasingly important part of the solar industry, as the sector navigates out of the current manufacturing downturn.
