As moderator of the last panel, Mr. Strunk turned to a topic of both German and American expertise, the solar energy industry and its policies and programs. He introduced the speakers, Dr. Neuhoff and Dr. Le, noting their own considerable expertise from which to describe the German and American experiences in solar energy.
Dr. Neuhoff said he had recently joined the Climate Policy Initiative, a new global network with offices in Berlin, Beijing, and San Francisco, that has the goal of assessing the implementation of climate policies in different regions of the world.
He said he would discuss photovoltaics as a central strategy in climate policy, and look more closely at policy differences in German, China, and the United States. With current crystalline silicon technology, he said, rooftop-scale PV could produce about one-fourth of all German electricity, and about ¼ of all Chinese electricity by 2020, assuming demand continues its current growth. With more efficient cells, or solar collections at larger than rooftop scale, that amount could be increased.
42Dr. Neuhoff presented his work in partnership with Molin Huo and Thilo Grau.
Currently, he said, three PV technologies are competing for market share. The crystalline path has traditionally led in market share, although thin film technology seemed more promising technology for the last two or three years. Over the last six to nine months, however, crystalline technology has regained its lead in terms of cost competitiveness and is “probably going to be back up to 80 to 90 percent of global demand.” The third area is multi-junction technology, which is more efficient but more expensive, and also needs a structure that tracks the sun’s position, further increasing cost.
Therefore, Dr. Neuhoff said, he would focus on crystalline technology. The cost of producing PV electricity depends first on insolation. Some eastern provinces and southern provinces receive up to 1100 kilowatt-hours of sun averaged over the year and can produce electricity for about 30 euro cents per kilowatt hour. A major factor in cost, however, is the cost of capital, which needs to be near 4 percent or less. For large-scale application, he said, overall costs need to be reduced by a factor of two before solar can effectively compete with current electricity prices in Germany. Obviously, however, once the PV penetration is very high, storage facilities must be built to stabilize electricity supply, which brings additional cost increases.
In China, which has more sunshine, especially in the southern provinces, solar technology looks more attractive. To some extent, deployment costs are also high, but the break-even point against which the technology is competing is lower, especially because the costs of carbon are not built into electricity pricing.
Dr. Neuhoff turned to the 15-year trend in the price of roof-top systems in Germany. After 2000 came a gradual reduction until about 2006 to 2007. The reduction has accelerated since then, as a result of three drivers. One was the opening of the German market to international competition. “As long as we had a German-only market on both demand and supply sides, there wasn’t much pressure on companies to innovate or improve competitiveness. You could actually observe all the actors in the value chain reaching their margins.” Second were several certain production scarcities, such as a sudden drop in the silicon supply, which brought a high price for a time. Finally, several innovations contributed to recent cost reductions.
For this symposium, he said, he would look at five kinds of innovations that could bring such cost reductions. One of the biggest kinds of innovation was that of the U.S. company, First Solar, which introduced a new commercial thin film cell of cadmium telluride in 2003. Building the cell required about 15 years of public support through universities and national laboratories. The company made all its decisions internally and maintained ownership of their technology longer than do most companies. It remains one of the biggest producers globally, with much of its production capacity in China. What drove much of the company’s growth, including private investment, was the expectation of a future market, and indeed, the market in Germany, and then briefly Spain and Italy, were major causes of growing global demand. All three were driven by
government incentives, especially feed-in tariffs that guaranteed utility buy-backs of PV-generated electricity.
The second innovation was a set of technological advances that increased PV efficiency from 17 to about 19 percent in crystalline cells. Behind this progress, he said, was companies’ understanding of the production process of the entire cell, and good links or even internal lines between the module components and wafer components. “You have to make adjustments across this entire process when you introduce a big innovation into the technology. If you look at the industry structure, both in Germany and China, you see quite a few companies that were able to do this.”
A third behavior that has been helpful is that many of these companies were competing “in parallel,” or seeking to innovate in the same areas. When one company succeeded in perfecting a technology, the others were able to adopt it quickly, benefiting the industry as a whole.
The fourth set of innovations was the arrival of new expertise in the PV industry. For example, Dr. Neuhoff said, Applied Materials bought Varian Semiconductor Equipment Associates in May 2011 for $4.5 billion. One of the stated goals for Applied Materials, a major equipment supplier in PV, was to gain access to the new technology of ion implantation, which was not yet used in the PV area. Such buying of companies to gain new technology increased the efficiency of the final products, and therefore of PV.
Finally, over the past two years, equipment supply companies had entered the PV arena, which had traditionally been dominated by the solar cell manufacturers. For example, cell producers had been experimenting with the use of selective emitter cells in solar arrays without notable success. Suddenly, two large global equipment suppliers, Gebrüder Schmidt, followed quickly by Centrothem, developed equipment of their own to produce selective emitter cells. They soon found an Asian partner willing to take some risk in adjusting the production process for this change, and in exchange the Asians—a and the industry—has an innovative technology a few percent more efficient.
Dr. Neuhoff said that some cost reductions over the last two years were also generated by moving production to China, where input materials are cheaper. At the same time, innovations to the input materials, especially the pads used to fasten the conductors to the wafer cell, have brought higher quality. “They have become better,” he said, “so you can make the cells thinner, you lose less light, and you have better contact between materials.”
Some challenges have emerged as well, he said. Previously, demand effectively created incentives to select different technologies and invest in those with highest efficiencies and lowest cell costs. Now, for both the United States and Germany, the cost of modules has come down to the extent that the installation and system integration costs of PV account for less than half the total cost of the PV. And yet, he said, the costs of putting a PV panel on the roof are still significantly higher in the United States than in Germany.
One implication of this changing cost structure, Dr. Neuhoff said, it that it is very difficult to anticipate what it will cost to install a PV panel. “If you
don’t know what the cost will be,” he said, “it is difficult to anticipate demand by offering a fixed price.” Germany now has a lot of capacity to install PV panels, he said, both on the roof and on also a large scale, and as a result, PV demand has picked up more quickly than anticipated. This is enhanced by the global fall in PV prices. As a result, the PV support level in Germany was reduced over the last year. Weekly deployment volumes in Germany rose in anticipation of this price reduction as many households decided to deploy panels quickly. Thus, deployment has become more sensitive to the level of the feed-in tariff because the consumers respond more quickly.
Italy experienced this even more dramatically, he said, because its tariff was rather generous, and when the tariff rose, many German installing companies were driving to Italy to build PV panels. “Using the feed-in tariff offers you low capital costs,” he said, “but the technology is now small scale and established enough that we need to be quicker in adjusting it and careful in the implementation. It’s an effective but tricky instrument.”
One large benefit established in Europe, Dr. Neuhoff said, is the renewable energy trajectories, by which countries commit to the EC to follow a certain trajectory for the technology they want to develop in coming years. This provides a guideline against which countries can adjust their tariffs over time. “This year we are discussing whether to adjust the feed-in tariff according to the market price almost week by week.”
A second big challenge, he said, is the balance between supply and demand. The global supply, or production volume, was 15 Gw in 2010, while installed capacity is now between 30 and 40 Gw. This is a setting, he said, “where, for a capital intensive industry, you might be concerned about boom and bust.”
To some extent, Dr. Neuhoff said, the 40 Gw number is high, because it includes some old capacity and some that could operate only at peaking load. “But the number is still an important indicator over the next 12 months. If the market senses a lot of excess capacity, and investors start to feel uncomfortable, we need to know how that is going to impact the confidence of investors and innovators. If cash flows are restricted, what happens to innovation? Where are companies going to cut first? On the other hand, some excess supply creates incentives to reduce costs and helps choose the best technologies.”
Another challenge, he said, was to apportion public support among three needs: early-stage R&D, where much innovation arises; manufacturing, which is desirable in generating employment and stimulating R&D; and support for demand and “putting panels on the roof.” One goal is to allow the private sector to identify the most promising innovative activities. He noted the earlier discussion in favor of supporting battery manufacturers, saying that too much support makes it difficult to design incentives for innovating and selecting the best technologies. He said that over the last two years, countries provided much higher subsidies for deployment than for manufacturing. “I think there will be increasing understanding among countries that it is unfair to support only manufacturers, and that this must be balanced by support for demand.”
Dr. Neuhoff summarized by saying that many innovations, certainly for crystalline PV, had been possible only because of strong innovation networks, which allowed ideas to be integrated into existing production processes. This led to gradual learning by doing, innovations in efficiency, and cost effectiveness. “We don’t know which innovations over the next five years will help us make a technology twice as good as it is today,” he said. “But we certainly should create an environment that maintains opportunities.” One important part of the environment, he said, is transparency, especially as more government programs are created. This is necessary both to protect government against supporting the wrong technologies, thereby forgoing new opportunities, and to let countries learn from one another which technologies are most effective.
Dr. Neuhoff concluded that “it’s the market pull that has driven this technology, and has attracted a lot of talent and technology companies into this field. Let’s maintain this ‘pull policy’ by having more commitments of government to advanced deployment policies. That strengthens confidence, and is going to be crucial in this imbalance of supply and demand. Government has a role to reassure investors that there’s going to be growing demand over the next three to four years. This demand is still going to be policy driven before the technology is cost-competitive at the wholesale level.”
Dr. Le began with a familiar comparison of the solar energy available to the United States and to Germany. The United States, he said, receives about as much insolation as Spain, a notoriously sunny nation. The U.S. state receiving the lowest insolation is Alaska—which receives approximately the same amount of solar energy as Germany. And yet Germany, dark though it is, generates about half of world demand for solar energy.
There are grounds for hope, however, in U.S. demand. In 2010, he said, the United States installed approximately 900 Mw worth of photovoltaics (PV), about a 100 percent increase over the prior year. Projections suggest that that growth will continue, and by 2013 or 2014 may actually exceed demand in Germany.
The United States employs 65,000 to 95,000 people in the solar sector, while Germany employs about 120,000. In the amount of solar installed unit per land area, or solar per capita, however, the United States lags Germany by a factor of 20 to 30.
Dr. Le reviewed the policy tools that influence demand and supply. On the technology push side, he said, the United States uses R&D funding for new technologies, R&D tax credits, and manufacturing tax credits. One example is the solar incubator program run by the National Renewable Energy Laboratory
(NREL), which since 2007 had funded 24 PV startups with an aggregate value of about $59 million. Those 24 companies had leveraged about $1.3 billion in private capital, and employed more than 1,200 people. “We funded some of these when they were as small as two people,” he recalled some now have 300 to 500 employees. Against these successes were other companies that did not succeed, although he said that shutting a firm down can be seen as a positive in freeing up capital, both human and financial, from a technology that was not able to mature.
Tools for the deployment of PV, he said, include feed-in tariffs, renewable portfolio standards (RPS), investment tax credits, loans, loan guarantees, and other financial mechanisms to lower the cost of capital, as well as depreciation tools that are modified for tax purposes.
Unlike Germany, he said, the United States relies heavily on the R&D investment tax credit of 30 percent. Germany relies primarily on the national feed-in tariff. Many secondary support mechanisms for both manufacturing and demand are fairly similar in the two nations.
The German feed-in tariff, Dr. Le said, had proven to be a “tremendous market pool mechanism that has been able to increase demand in the marketplace. It can also create some high peaks in demand, and is expensive to the ratepayer, though not to the taxpayer. The German FIT cost roughly 4.6 billion Euros in 2009, and a little more in 2010, representing 0.2 percent of GDP. If we deployed capital at that level in the United States, it would total about $30 billion.”43
While there is no national FIT in the United States, a number of states already have them, or are studying them; so are some municipalities, including Sacramento, San Antonio, and Gainesville, Florida. “A FIT at the national level would be difficult,” he said, “because of regulatory hurdles at state and local levels.”
The favored mechanism to incentivize demand in the United States is the renewable portfolio standards (RPS), which have been adopted or planned by 30 states. A significant fraction have solar carve-outs, or solar-specific renewable standards. This creates markets for renewable energy credits, or solar renewable energy credits, which in turn stimulates demand.
The primary national incentive structure is the investment tax credit (ITC), which automatically adjusts as price declines. On the negative side, one needs a “tax appetite” to take advantage of the ITC, as well as up-front capital. The lack of long-term clarity can be a problem, as the ITC is set to expire in
43Germany launched the era of the feed-in tariff (FIT) in 1990. The FIT requires utilities to purchase electricity generated from renewable sources at prices a percentage of the prevailing retail price of electricity. The law was revised in 2000 to guarantee purchase prices for 20 years, allow utilities to participate, and reduce costs annually based on expected cost reductions. Because the FIT sets rates based on cost of production, wind power producers receive lower rates, for example, than PV producers.
2016. In Germany, the FIT is both stable and simple—two features that encourage investors and allow capital to deploy widely. The negative aspect is the cost to the ratepayer, but many are willing to bear it.
The two nations are similar in government funding for solar R&D as a percentage of GDP. For the United States, the ARRA, passed in 2009, was a very important policy tool to spur deployment of manufacturing and renewable energy technologies. One such tool was the grant in lieu of tax credits. Unlike the ITC, which requires a tax appetite and does not provide payment until taxes have been paid, the grant in lieu of tax credit provides the money up front.44Wind energy, in particular, has been able to benefit, receiving $5.2 billion that enabled it to expand in the United States. Wind represented some 40 percent of new generation capacity in the United States in 2010. Solar received about $600 million from that program.
The more significant stimulus for solar has been the advanced manufacturing tax credit.45 Manufacturers of polysilicon thin film receive about half of the $2.3 billion in the solar program, with other technologies receiving less. Similarly, the loan guarantee program has been a crucial part of the U.S. portfolio for the deployment of solar energy. Of the $10.7 billion conditionally committed or finalized in loans, $8.3 billion went to the solar sector in 2010 for deployment and manufacturing. Some of the larger grants went to utility-scale solar farms, PV or solar manufacturers, and concentrating solar power (CSP) research.
While these programs are helpful, Dr. Le said, they are not sustainable. “The private sector needs to take over if this sector is to expand,” he said. The total amount of VC and private equity financing deals in 2010, he said, totaled about $2.3 billion. Roughly ¾ of those deals were done in the United States, which has a very strong VC community that has been an engine for innovation. This is an important first link of that value chain, he said, but it is not sufficient. “Debt financing and asset financing is what will be required for this industry to expand much more widely than it is today.” Of the $44 billion in solar debt financing world-wide in 2010, the United States held only a 9 percent market share. “That isn’t much,” he said.
Dr. Le then showed a graph that reflected “a sad state of affairs.” In 1995, the United States produced 43 percent of PV cells and modules worldwide. “Admittedly the industry was a lot smaller. But over the course of the next 15 years, U.S. market share eroded, and in 2010 the United States held only 7 percent of world market share. “Most stark is the growth of PV cells and modules produced in China and Taiwan. Rising from almost nothing in five years, they have captured almost 60 percent of the worldwide market share in a short period. This speaks to the erosion of our manufacturing competitiveness.”
44The grant in lieu of tax is described in section 1603 of the National Tax Code.
45The ITC is found in Section 48C of the National Tax Code.
He referred to a point made by Andrew Grove, a co-founded of Intel, who said that abandoning today’s manufacturing can “lock you out of tomorrow’s emerging industries.” To the degree that a country’s manufacturing sector declines, so does its ability to innovate. This was one perspective that helped shaped the innovation policies of the Obama administration, which was trying to use the interplay between solar research and solar production to make the technology cost-competitive with fossil fuels—and to do that without subsidies. “If we can,” he said, “solar will be deployed across the United States—not just the West, but across Maine, Michigan, North Dakota. These are exciting and audacious challenges. But please know they are not easy: they mean a 75 percent reduction in cost.”
Dr. Le said that the United States can draw inspiration from other countries. In the United States, the cost of residential solar installation was approximately $6.50 a watt in 2010. In Germany, it was approximately $3.80 a watt. “Given that the cost of PV modules is the same worldwide,” he said, “what’s really different is the balance of systems, especially installation and permitting costs.” These can be widely different in two cities in same state. Lowering such high costs and removing cost inequities are administration priorities, he said.
In conclusion, Dr. Le said, government policies can accelerate clean energy “at speed and at scale.” The German FIT had been a very effective tool to help spur the solar industry, and had resulted in global solar leadership on the demand side. The United States has a hybrid mixture of federal, state, and local incentives to spur not just demand, but also supply. “This is an exciting time to work in solar. If we can reduce solar costs by a factor of three or four by the end of the decade we can play an important part in meeting president Obama’s clean electricity standard of 80 percent by 2035.”
Mr. Strunk noted that the U.S. picture had long been characterized by the availability of “cheap oil and many natural resources.” In Germany, he said, the picture was quite different, and the need for alternatives much greater. He asked, “Does this in your opinion highly influence the ambitious program to advance solar energy in Germany, with the result that we are far ahead?”
Dr. Le said that with trillions of dollars of assets already deployed in the U.S. energy sector, change “does not happen rapidly; there’s also an incumbency you have to overcome. It will take time. But if we don’t start, we’ll never get there. So we have to start working on all those challenges.”
Dr. Neuhoff agreed, adding that while “energy security” is commonly emphasized in the United States, “in the end it’s the political economy of incumbent groups that want to maintain their position investment. So it’s about creating a dynamic where new technologies have bigger expectations and representation in the political process, and then gather support from the voters.”
A questioner asked why Germany, despite its large demand pool, was not spending more on R&D. Dr. Neuhoff agreed that until about three years ago the solar industry did have a low level of R&D spending. Recently, however, he said that the pace of R&D seems to have increased, in both public and private programs, because “international competition really created the moment when they realized they had to increase cost competitiveness.”
Dr. Wessner asked about the common view that “the United States is doing the research, the Germans are installing, and the Chinese are producing.” Dr. Le said that view no longer seemed to be justified, with function in the United States coming into balance. “In the United States, by about a year ago, we installed about as much solar panel capacity as we produced domestically. So we have a fairly even one-to-one production vs. installation.”
Dr. Neuhoff said that indeed the picture is mixed. Much of the research in Germany is dedicated to improving crystalline technologies, while much public R&D money in the United States goes to very new concepts, or “long shots.” Also, much of the equipment used globally to produce PV cells is still produced in Germany, “so while you usually see how many PV modules we import from China, you rarely see how much equipment is being sold to China to build those modules. China has a lot of interest in manufacturing more of its own equipment, but at the moment but it appears to be almost a balanced picture.”
A questioner asked about the roles of transmission and storage. Dr. Le said that they would both become more important as solar and other renewables provide a greater fraction of electricity to the grid. “In the United States now, however, solar represents roughly 2 percent of our electric generation capacity, while in Germany it’s about 10 times as much. We are investing in solar thermal technologies, which have inherent storage capabilities, and that will be an important portion of our grid.”
A questioner asked how a government in an era of fiscal stringency could justify putting public dollars in an industry where capacity is twice as great as demand.
Dr. Le replied that “if you don’t make the investments today, you will certainly not be in the game tomorrow. As a taxpayer I believe you need to make certain investments that will enable a prosperous future.” Dr. Neuhoff agree that it was “quite a challenge to build up capacity,” but that if Germany were to stop its support programs, financial participants would also stop their support, “and within a few months we would have cash flow issues across the sector. We would lose a lot of the capacity that was carefully built up and that might contribute both to global PV development and to German economic success. You need to maintain the momentum if you want to play in the game.”
A participant who worked in Adlershof Science Park in Berlin addressed the wisdom of innovation subsidies, which can be both “right and wrong, and a paradox. Sometimes it’s a necessity to build up new technologies with subsidies in the process of long-term planning. It took us about 15 years to
develop this solar industry. Meanwhile we have a company producing thin-film solar panels and exporting them to China—not the other way around.”
Dr. Prabhakar agreed that Germany had been “extremely effective” in translating R&D into deployment, and in making Germany a global leader. She asked whether the progress of solar had been “out of proportion to what Germany would like to accomplish with other renewable energy sources, and whether the country had paid too high a price to achieve its huge solar leap forward. Would you do it again or are there better ways to get to this point?”
Dr. Neuhoff said that the technology costs had fallen more quickly than most people had expected, and that was the important goal. When he began working on PV in 2005, he said, the “perspective was that wind was close to cost-effective, so why would we want to go into solar?” He said that in terms of storage, “it’s going to be valuable for the German system to have a set of different renewable technologies. Storage requirements decline faster as the portfolio of technologies becomes more diverse.” He conceded that if one now took a long view of the renewable system, solar could have been built “a bit slower, but at the end, it won’t really matter by 2030.” He said that the investment costs have been high, but “in the end we’ve succeeded in developing the technology. He said that the German progress had also inspired China to raise its goals for solar deployment from 5 Gw by 2015 to about 20 Gw by 2020, which created a general good. “So yes, sometimes you have a make a first step; you might benefit yourself, but also contributed quite a bit to development elsewhere.”
A questioner asked whether thin film, a hot technology a year previously, was “a little bit of a has-been” with the drop in price of crystalline technology. Dr. Le said that for the U.S. Government, “the official answer is that we’re technology agnostic. We set goals for technology to achieve in terms of performance and cost, and technologies that achieve those goals will be deployed widely. We’re not trying to pick winners.” At the same time, he said, the DoE follows every energy market, and makes investments in all promising technologies, including crystalline silicon and thin films. He said he was personally optimistic about thin film, “and there are good reasons to be optimistic about some of these new technologies.”
A questioner asked whether Germany had invested too much money in a technology that was not very efficient in generating electricity and “which seems to be a niche technology at best?” Dr. Neuhoff admitted that the PV technology did require more improvement in efficiency and cost, but that “if we want to decarbonize our power sector, which is a stated goal across Europe, it’s difficult to do this with individual technologies. It really takes a portfolio of them. Why not use the opportunities to have PV across our roof spaces? I think it is in the cost range where it’s going to be viable, and it provides about a quarter of the electrical power for Germany.”
Dr. Le added that from a technology standpoint, “there is a still a significant gap between what our ‘hero cells’ can achieve in the laboratory, and what those cells could theoretically achieve from basic physics’ first principles,
and a bigger gap between those cells and what our manufacturing lines can achieve day in and day out. There are opportunities in both places to improve performance.”
Mr. Strunk asked what would be next step in dealing with alternative energies—”not just solar or wind, but integrating systems, including storage. Would storage allow the renewable energies to have a final breakthrough?” Dr. Le said that a number of breakthroughs will be required—not just technology, transmission, and storage of electricity—but also financial. “This is a capital-intensive industry,” he said. “We need finance mechanisms that will enable industry to expand.”
Dr. Neuhoff added that “we are in the middle of a transformation. It is reassuring that we don’t have to wait for one big decision from government, because these issues are difficult to deal with in commercial environments. It’s really about gradually evolving the system, and I think we are in a world where we can learn from each other in this process.”
“In Germany now,” he continued, “we have the revision of the renewable energy law, which happens every two years, and we anticipate a move from the feed-in system to a system that offers a premium to sell the electricity in the market. I think the discussion behind this decision was a bit too narrow, because we want to open the electricity market to different technologies. I don’t think we need to adjust the feed-in tariff for this purpose; that would make it more complicated and therefore risk misbalancing it. What we need is to change the power market so it is open to other technologies. The United States has experimented for years, but now has established the renewable portfolio standard that has spread across the liberalized markets and is demonstrated to be effective. So let’s learn from each other and adopt another system which is established, well practiced, and tried out.”