China added approximately 317 GW of new solar capacity in 2025, 14% more than the year before. By the end of the year, the country had roughly 1.2 TW of installed solar capacity, while electricity generation from solar increased by 40%.
A few years ago, looking at numbers like these would have led to a fairly simple conclusion for silver: more solar panels meant more silver.
But that equation is starting to change.
Solar manufacturers are becoming increasingly efficient in their use of silver in solar panels. The amount of metal required per watt is falling, and higher silver prices have accelerated the process. In 2025, this trend reached the point where record growth in solar capacity was no longer enough to offset the decline in silver use per unit of power.
At first glance, this is one of the strongest arguments against the long-term bullish case for silver.
But I think stopping there would be a mistake.
The Solar Boom Continues. Silver Consumption Doesn’t
Until recently, photovoltaics were one of the most reliable engines of industrial silver demand.
In 2024, total industrial demand reached a record 679.0 Moz. It was the fourth consecutive record year, driven in large part by photovoltaics, vehicle electrification, investment in power grids, and new applications related to AI.
Just one year later, the picture changed.
In 2025, total industrial silver demand fell 3% to 657.4 Moz. The Silver Institute specifically identified weaker photovoltaic demand as one of the main reasons, as higher silver prices and intense competition among manufacturers accelerated thrifting and substitution of silver in solar panels.
What makes this particularly worth watching is that the solar industry itself did not shrink.
China added 317 GW of new photovoltaic capacity during the same year, while its total installed capacity increased by 35% year over year.
Indicator | Value | Context |
| Global industrial silver demand (2024) | 679.0 Moz | Record high |
| Global industrial silver demand (2025) | 657.4 Moz | -3% YoY |
| New Chinese PV capacity (2025) | 317 GW | +14% YoY |
| Total Chinese PV capacity (end-2025) | ~1.2 TW | +35% YoY |
To me, this matters far more than the headline number of gigawatts being installed.
The number of watts continues to rise rapidly. But the number of milligrams of silver required for each watt is falling even faster.
The future of silver demand from the solar industry will increasingly be determined by the race between those two curves.
Silver Has Become Too Expensive a Component to Ignore
Thrifting is nothing new. Manufacturers have always tried to reduce their use of expensive materials.
What has changed is the economic incentive.
According to the World Silver Survey 2026, silver represented approximately 8–10% of solar cell manufacturing costs at the beginning of 2025. As silver prices rose, that share climbed to more than 20% during the year.
At that point, it is no longer a minor detail in the manufacturing process.
When a single raw material accounts for more than one-fifth of the cost of a cell in an industry where manufacturers are fighting for margins, the incentive to reduce its use becomes enormous.
The result is a combination of several technologies.
Manufacturers are using finer conductive lines, improving printing techniques, changing cell designs, and increasingly experimenting with pastes in which copper replaces part of the silver.
By the fourth quarter of 2025, according to the World Silver Survey, silver-coated copper pastes containing only around 10% silver had entered high-volume production.
The next step includes zero-busbar technologies and advanced printing methods. Compared with some current designs, these could reduce silver consumption by another 10–20%.
The Five-Milligram Threshold Is Approaching Fast
What happens next is even more interesting.
The World Silver Survey 2026 expects average silver loadings in mainstream photovoltaic cells to fall below 5 mg per watt by 2027.
There is even a technological path that could take silver usage much lower.
Copper electroplating could theoretically reduce silver use to below 1 mg/W. Compared with earlier generations of solar cells, that would represent a fundamental change.
But there is a large gap between what is technically possible and what can reliably be manufactured at a scale of hundreds of gigawatts.
Pure copper pastes still face challenges including oxidation, adhesion, manufacturing yields, and long-term reliability. The Silver Institute therefore does not expect widespread adoption in 2026.
Silver has one inconvenient characteristic — at least from the perspective of a manufacturer trying to replace it.
It is exceptionally good at what the solar industry needs it to do.
Silver has the highest electrical conductivity of any metal and also offers excellent physical properties for contacts and conductive pastes. That makes it economically attractive to reduce silver use dramatically, while making the final portion much harder to eliminate completely.
To me, that distinction matters.
Thrifting is no longer a hypothetical future development. It is happening now, and its impact on demand is already visible in the data.
Complete substitution is a much higher technological hurdle.
For Silver, the Multiplication Matters More Than Either Number Alone
This is where we get to the part that I think both simple bearish and bullish arguments often overlook.
Silver consumption in photovoltaics can be reduced to a very simple equation:
installed watts × silver per watt.
If silver loading falls 20% while the number of installed watts increases 30%, total silver requirements can still rise.
If loading falls 30% while installations increase only 10%, silver demand declines.
The year 2025 showed that the second scenario is not merely theoretical.
And according to the Silver Institute, 2026 is expected to continue in the same direction. Global solar installations are expected to keep growing, but thrifting and substitution are forecast to drive another decline in PV silver consumption. Total industrial silver demand is now forecast to fall another 3% to 639.6 Moz this year, its lowest level in four years.
That is a real headwind.
But silver is not only a solar metal.
While the PV segment faces pressure, other areas are benefiting from the very same electrification of the global economy:
- expansion of transmission and distribution grids,
- electric and hybrid vehicles,
- charging infrastructure,
- power electronics,
- data centers and AI infrastructure.
Growth in AI, the automotive sector, and investment in power grids helped offset part of the decline in photovoltaic demand during 2025.
I would therefore be careful with the claim that solar no longer matters for silver. It does. But we can no longer take a forecast for new gigawatts and automatically translate it into an equally rapid increase in silver consumption.
The Strongest Bearish Argument Still Hasn’t Eliminated the Deficit
And this is the part of the equation that interests me most.
Photovoltaics were one of the strongest sources of growth in silver consumption for several years. Now the technological response is arriving: the industry is aggressively reducing the amount of silver used per watt and beginning to replace part of it with copper.
That is exactly the mechanism that should gradually relieve pressure on a tight market.
So far, however, it has not been enough.
The year 2025 ended with the fifth consecutive structural silver market deficit. For 2026, the Silver Institute expects another deficit — the sixth year in a row.
And that is despite industrial demand being expected to decline.
That changes the way I look at solar thrifting.
It is not evidence that silver has no supply-and-demand problem. Rather, it shows that one of the metal’s largest consumers is already actively responding to higher prices and tighter conditions — yet the market is still expected to remain in deficit.
Supply also does not respond at the same speed as the technology sector.
For 2026, the Silver Institute expects global silver mine production to remain broadly flat, with mine supply forecast to edge lower by approximately 2.5 Moz as gains in some producing regions are offset by declines elsewhere.
That is hardly the kind of supply response one might expect after such a strong rise in silver prices.
A large share of the world’s silver is produced as a by-product of lead, zinc, copper, and gold mining. A higher silver price therefore does not automatically mean miners can simply turn on the taps and deliver hundreds of millions of additional ounces within the next year.
If demand continues to exceed new supply, the difference has to come from somewhere — primarily above-ground inventories.
And to me, that matters more over the long term than whether an individual solar panel uses a few milligrams less silver next year.
What I’ll Be Watching Instead of the Number of Solar Panels
After 2025, I would no longer use growth in global photovoltaic installations as a simple bullish indicator for silver.
The equation has become more complicated.
Over the next few years, I will primarily be watching the rate at which silver loading declines relative to the growth rate of new installations. In my view, that relationship will determine the actual amount of silver consumed by the PV industry.
The second thing to watch is the commercialization of copper. If pure copper pastes or copper electroplating can move into mass production without major compromises in reliability and manufacturing yields, that would represent a much more significant change for long-term silver demand than the gradual thrifting we are seeing today.
Third is what happens outside the solar sector. If PV uses less silver while demand from power grids, vehicles, power electronics, and AI infrastructure continues to grow, the decline in PV consumption alone will tell us less and less about the overall industrial market.
And finally, I will be watching something even simpler: whether the structural deficit actually begins to close.
That is where the story meets the data.
The solar industry is already doing exactly what we would expect in response to high silver prices. It is using the metal more efficiently. It is replacing silver where possible. And it is investing in technologies that require less silver for every watt produced.
That is not a detail I would want to hide as a silver investor.
It is one of the biggest bearish factors on the industrial-demand side of the market.
And yet 2026 is still expected to be the sixth consecutive year of a silver market deficit.
So for me, the most interesting question is no longer whether the next solar panel will contain less silver. It probably will.
The more interesting question is this:
How little silver can the world use in each individual device if it keeps manufacturing more devices, electrifying more of the economy — and the market still fails to generate enough new supply to fully meet demand?
Leave a Reply