Solar panels are now cheap. Thank you government policy.

From an economic point of view, the main challenge of climate change is that the usual way of proceeding – the dirty, carbon-intensive method – is generally less expensive than the new, less polluting alternatives.

Solving the problem means reducing the cost of these solutions. Simple, no?

But in practice, it is not so simple. In fact, we still do not quite understand what drives innovation and technological improvements. Is it basic scientific research? Early R & D? Learning by doing? Economies of scale?

If we want to make clean technologies cheaper, we need to better understand how the process works. Among other things, Silicon Valley guys are spending billions on "moonshot" startup initiatives – it would be nice to spend that money effectively.

There is a voluminous academic literature on these topics, but a new article in the journal Energy policy helps cut through the fog. It focuses on a specific technology and seeks to identify and quantify the different forces that have resulted in reduced costs.

This technology: good old photovoltaic (PV) solar panels, the cost of which has decreased by 99 percent in recent decades.

Shutterstock

The authors are Jessika Trancik, associate professor at MIT, Goksin Kavlak, postdoc, and James McNerney, researcher. They are part of a team that, in collaboration with the Department of Energy's SEEDS program, is attempting to develop a comprehensive theory of technological innovation. , focusing on photovoltaic solar energy.

"Evaluating the Reasons for Reducing Photovoltaic Costs" outlines the results – which has resulted in such a rapid decline in PV costs, and when.

The details deserve to be examined, but the big lesson is quite simple: it did not happen. At each stage he was guided by intelligent public policy.

Photovoltaic solar has become cheaper at a derisory pace

First, as a context, it is important to understand the remarkable evolution of solar photovoltaics. Again, the costs of solar modules have dropped nearly 99 percent in the last 40 years.

Trancik et al.

In other words, these declines have continued since 2015, and market experts are expecting them to accelerate in the foreseeable future.

Photovoltaic solar energy has defied all expectations, continuing to become cheaper and deploying faster, even though experts predict, over and over again, that it will stabilize.

This rapid decline in costs is a puzzling and surprising phenomenon. This requires explanation.

Of course, there have been many studies on the subject, but most have relied on a "correlational analysis", linking the decline in photovoltaic costs to other ongoing trends. For example, it is common to note, partly on the basis of this article, that PV costs decrease by about 20% for each doubling of cumulative capacity (both trends are correlated).

There are also device-level studies that examine the components of PV systems and their contribution to costs in an instant.

"The absence of these studies," writes the MIT team, "is a method of accurately quantifying how each change in a feature of the technology or manufacturing process contributes to cost reduction when many changes occur. occur simultaneously ". create – a dynamic model that can distinguish and quantify the main causes of price declines over time.

The drivers of lower PV costs have changed over time

The team distinguishes two basic engine types for cost reduction, low level and high level. The former are "measurable, technology-specific cost drivers" such as the area of ​​the unit, the efficiency of the modules and the size of the manufacturing plant. These are "processes like R & D, learning by doing, and economies of scale that involve minimal cost reductions."

The idea is to bring together bottom-up and top-down approaches to understand the evolution of technology. There is a long explanation of the modeling methodology, filled with equations, in the document, if you are interested in this kind of thing.

The team examined overall PV costs from 1980 to 2012, distinguishing the different factors. In their results, they first examine the role of low-level mechanisms; then they link these mechanisms to high-level mechanisms.

Here is a visual representation of the low-level mechanisms that resulted in lower costs, by how much and when:

Trancik et al.

A quick look at the graph reveals the shape of the results. In the early years of the photovoltaic industry, gains were relatively homogeneously distributed across several low-level mechanisms, mainly due to module efficiencies (24%) and lower non-silicon costs ( 22%) and silicon (18%). panel components.

In other words, the first improvements were mainly in appliances, basic science and panel engineering.

In the more recent period of the evolution of the sector, the factors have evolved considerably. Efficiency costs, excluding silicon and silicon, fell to 12, 15 and 3%, respectively. And one driver exceeded all others: the size of the factory. Photovoltaic solar energy has become a big business. Bulk manufacturing in large factories has resulted in a rapid decrease in costs.

The document then goes to the high-level mechanisms. Here is the similar distribution:

Trancik et al.

This serves as a more abstract representation of the same dynamics in the first graph. In the early years of PV, device-level R & D played a major role in reducing costs. People spent time and money to improve the signs.

In the years that followed, R & D declined slightly and economies of scale increased. People started to cut costs by making tons of photovoltaic panels. (Note, however, that R & D has continued to play an important role.)

So, what does all this tell us about politics?

It is possible to intentionally make cheap clean energy technologies

Several interesting policy implications are built into these results.

For example, as mentioned earlier, the causes of lower costs have been fairly evenly distributed over low-level mechanisms in the early years. The components of the supply chain at the device level and the engineering challenges were many and varied. There were many directions to attack the problem. In an email, Trancik calls this "having several buttons to turn," which allowed a variety of simultaneous solutions.

Perhaps the most interesting implications, however, have to do with the timing politics.

As Hal Harvey, an expert on clean energy policies, said in our recent interview, technology is taking a fairly predictable path in the learning curve, and different types of policies can move them forward at different stages.

Energy innovation

When things work, as technologies become cheaper and closer to commercialization, R & D gives way to performance standards. And when the industry is mature, price signals (like a price on carbon) take over.

As a first approximation, this is what happened with solar PV, except that the market is still dominated by performance standards (such as renewable energy mandates), while that price signals (such as carbon taxes and cap-and-trade programs) are having a little trouble getting increased

Nevertheless, there are lessons to be learned for other clean energy technologies that we know will be needed for decarbonisation – batteries, better electric cars, advanced nuclear reactors, hydrogen fuels, fuel-based from algae, microgrids, carbon removal, etc. It is possible to target public policy on a technology based on its position on the learning curve and consciously accelerate its development.

For some early technologies, such as algae, it will be essentially research and development. For something like solar photovoltaic, an active sector, future cost reductions will most often result in accelerated deployment (ongoing cost reduction of existing technology) and ongoing research and development (protection against corruption). technology could stabilize prematurely).

Prism Solar

Finally, in terms of public policy, the general role of politics in PV development needs to be emphasized.

The document refers to two basic types of policies: firstly, publicly funded R & D and second, "market-stimulating policies", which create legal or economic incentives for private actors to seek, develop and invest in in technologies.

When people think of technological innovation, they tend to think of the first, scientists and engineers of government labs. But it is the latter who do most of the work.

"Market stimulation policies have played a central role in reducing the costs of PV modules," notes the team, "with private R & D, economies of scale and on-the-job learning contributing, according to the estimates, to 60% of the decrease in costs in Photovoltaic Modules between 1980 and 2012. "

It's worth repeating: policies that encourage private investors to develop and deploy solar panels are responsible for well over half of the cost reduction of solar PV. Most of the rest is public R & D.

Trancik et al.

Recently, the rich guys in Silicon Valley have shown a keen interest in financing private R & D in the field of clean energy technologies. There is an almost libertarian pretext among this crowd that the government is slow and inefficient.

But this research shows that public policies can be extremely effective in creating the market conditions in which people can innovate. The technology entrepreneur who hides behind various (laudable) clean energy investment projects would do well to study it.

The real lesson of photovoltaic solar energy is simple: we know how to make clean energy cheap. We did it. We can do it again if we want.