Why Don’t We Have Abundant Solar Power? Blame Financing, and Industry, not Science

13,600 9 Loading
Italy PV system

The Montalto di Castro Photovoltaic Power Station in Italy is one location where solar electricity is already as cheap an investment as traditional fossil fuel sources.

In the world of renewable energy resources, solar power is the epitome - abundant, reliable, and green. For decades scientists and engineers have been working tirelessly to improve the efficiency at which photovoltaic cells convert sunlight to electricity. If they can simply make PV cells efficient enough, then they would be cheaper than fossil fuels and most of the world could switch over to the better technology. Getting solar power to be as cost-effective as traditional electricity sources is known as reaching “grid parity”, and several locations around the world have already succeeded in making it a reality. How can the rest of the world join in? While scientists and engineers will be crucial in that development, loan officers and bankers may hold the key to bringing solar power to the masses.

Researchers at Queen's University in Canada have compiled a comprehensive study of solar power plants and found that improvements in financing, and industrial streamlining could allow for the technology to become as cost-effective as traditional electrical sources. In other words, a much larger part of the globe could currently reach grid parity if cheap loans and better supply chains were available.

“The major generation cost for solar PV [photovoltaics] is the upfront cost and the cost of financing the initial investment, which means the LCOE [levelized cost of electricity] is very dependent on the financing methods available and manufacturing cost reductions”
---Branker, Pathak, and Pearce, A Review of Solar Photovoltaic Levelized Cost of Electricity, 2011

The study was led by Joshua Pearce, along with K. Branker and M.J.M. Pathak of Queen's University. It found that variations in financing interest rates could alter the cost per kilowatt-hour by ten cents (Canadian) for typical solar plant configurations. Considering that the retail cost of electricity in Canada is just around ten cents, this financial-dependence is a large consideration for building new solar power plants.

Pearce and his colleagues also found that many previous studies were too conservative in their evaluations of photovoltaic cells. Most analysts assume solar cells will lose 1% in peak output every year, but PV systems from the 1980s showed considerably less loss in the past two to three decades. The Queen's University team found 0.5% loss to be a better estimate and typical losses might be expected to be 0.2 to 0.5%. As such losses compound every year, that small difference in calculation can lead to big changes in the projected cost-effectiveness of a solar power plant. In other words, such plants are much better investments than they've previously been thought. Especially when one considers that most other levelized cost of electricity (LCOE) evaluations peg solar cells at having effective lifespans of 20 years, when Peace and colleagues found 30 years to be a better estimate. As even older technologies from the 80s are going longer and stronger than most analysts assumed, the return on investment for building solar plants is higher than generally accepted.

PV parity Nikkei

Which means many more solar power plants could reach grid parity with current levels of technology. As the Queen's University team pointed out, British Petroleum test systems in California and Hawaii have already succeeded in reaching grid parity. They are not alone in that success. Nikkei Electronics Asia declared Italy had reached grid parity last year. The Montalto di Castro Photovoltaic Power Station competes with traditional sources in the nation. In fact, Italy's solar industry has grown remarkably in the past three years, up to one gigawatt of produced power. Nikkei and other industry analysts believe Southern Europe is primed for solar grid parity.

The rest of the world is quickly catching up. In the past decade, total solar power production has increased from less than 1 GW to around 16 GW. Since 2008, the cost of photovoltaic modules has dropped by a remarkable 60%! This exponential growth in total global solar electricity output lines up well with the projections of futurist Ray Kurzweil. His much discussed predictions say that the world will continue to double solar power production every two years. In 20 years the world would produce around 15,000 GW – the total amount of energy humanity consumes in all forms.

Kurzweil Graph

Ray Kurzweil's logarithmic graph for solar power production shows a clear growth that, if continued, will have solar providing enough electricity to power the world in just 20 years.

A world almost completely powered by the sun in just two decades seems like science fiction, but the work of Pearce and his colleagues seemingly corroborates some of Kurzweil's predictions, at least in the near term. Current investments in solar power are not as large as they could be, not because of a lack of government mandates, nor because photovoltaic efficiency isn't increasing quickly enough. Instead high interest rates in loans, overly conservative estimates on return in investments, and preferences for short term returns in energy markets have undervalued solar power plants. In 2010 new investments in solar power were only around $26 billion (USD), about one-fourth of that put into wind.

While solar grid parity hasn't been reached universally, it has been well documented in a few locations. If Pearce's LCOE study was to be applied, it's likely that many more such locations would be found to already be cost-effective for photovoltaic cells. Improving the solar cell industry to take advantage of economies of scale, more efficient supply lines, and more experience in installation/maintenance will only increase the number of solar-ready locations.

And, of course, the biggest factor, the actual quality of the photovoltaic technology is only going to keep improving. At 8% efficiency the entire world could obtain all the electricity it needs from solar cells covering land equivalent to the size of Colorado (~ 100,000 square miles). Some of the most recent, and experimental, PV cells have surpassed 40% efficiency.

The current solar industry is small, providing just one-thousandth the electricity consumed by humanity. Yet with the right investments, that fraction is ready to grow exponentially in the years ahead. Solar grid parity is already here for some of us, and it's much nearer than the rest of us may believe.

[image credits: 'Anna' via Wikicommons, Enel, BP, Nikkei Electronics Asia, KurzweilAI.net]
[sources: Branker, Pathak, and Pearce 2011 (preprint)]

Discussion — 9 Responses

  • buybuydandavis January 21, 2012 on 2:05 pm

    This looks bogus to me.

    ” In other words, a much larger part of the globe could currently reach grid parity if cheap loans and better supply chains were available. ”

    Yes, indeed, if only know nothing investors and banks would give you loans on terms that you want, then you could make bazillions and save the world. Looks like the people paid to analyze business plans aren’t buying it. I’m not either.

    I’ve pointed this out in other venues, but Kurzweil’s logarithmic chart for solar power is fundamentally different than his usual charts. It’s total output, instead of output per dollar. When he shows exponential growth in watts per dollar, give me a call.

    • Michal Strojnowski buybuydandavis January 22, 2012 on 2:03 am

      > When he shows exponential growth in watts
      > per dollar, give me a call.

      I do not know about Kurzweil, but these guys make extensive surveys of photovoltaic modules each month:

      And currently there is clear exponential trend, about 25% more watts per dollar each year. But it started in 2008, so it is unclear how stable this trend is.
      However, many people expect it to continue:

      • Homer Michal Strojnowski January 22, 2012 on 12:50 pm

        Thanks for the info Michal. It’s clear to me that both total output and output per dollar are following exponential trends.

      • buybuydandavis Michal Strojnowski January 22, 2012 on 1:00 pm

        3 years isn’t much time to judge a cost trend, particularly when it started right at the beginning of a global depression. in the 5 years prior, the curve is flat, if not increasing.

        Also, that’s a linear plot, not a log plot, which is what you need to see exponential growth.

        Going to the Scientific American plots, the values look flat from 2000-2007.

        And it’s not just the watts/dollar of a cell that counts. It’s the cost of an installation that delivers that power. I think half the cost of home solar power is installation. Even if the cells were free, the total cost doesn’t come down exponentially until the cost for the installation starts coming down exponentially.

        Depending on where you look, we only need a factor of 2-4 more. It’s getting close. Thanks for the info.

      • albeit Michal Strojnowski January 22, 2012 on 5:23 pm

        Solar cell manufacturers are currently receiving large government subsidies throughout the world. So how much of this “exponential” decline is due to their competing on price armed with these subsidies?

        It seems to me that if this progress is real, we just need to wait until it delivers. I suspect if we ended subsidies, this boom would quickly collapse.

        • Michal Strojnowski albeit January 23, 2012 on 1:12 am

          The whole point of subsides was to start this trend. Many people seems to think that subsides should pay for solar energy – this is nonsense, because we would need hundreds of trillions of dollars to achieve that. The idea was only to create artificial demand for several years, allowing solar energy companies to grow, compete and optimize. This should trigger real drop in prices, which would result in real market demand.

          And it seems to be working as predicted. Solar energy already reached grid parity in Italy, and soon it will reach grid parity in several other sunny regions like Spain, Australia and California. Then subsides will no longer be necessary.

          The main idea behind that is that there is no technological reason why solar power should be expensive. Even without any further advancement in technology, economies of scale alone might lower prices to competitive level.

          • arpad Michal Strojnowski January 23, 2012 on 5:01 am

            Actually, there are some very good reasons why solar will never be competitive not the least of which is its necessarily episodic nature along with the fact that, ultimately, power is a function of area covered with solar cells.

            If you proponents of solar power were honest you’d do the arithmetic to calculate the amount of land area that’s necessary to generate a worthwhile amount of power. But you know better.

            Similarly, you also shy away from *delivered* power costs. The “coupla hours a day” nature of solar means either conventional power generation for non-productive hours a day or energy storage schemes which are hugely expensive. The default position is to simply ignore the problem, hope no one notices and attack anyone who ignores the taboo on the discussion.

            *That’s* why solar will never amount to anything. *That’s* why some of the European countries that are in the worst shape, and were formerly touted as exciting leaders in the area – Spain and Portugal – have ceased to be of much interest. Their economic problems are due in part to their pursuit of this chimera and their economic problems make it impossible to pursue these bad ideas.

            • Michal Strojnowski arpad January 23, 2012 on 6:13 am

              It looks like you have not read anything about solar power, because what you claim is simply false. Every problem that you pointed out as “not addressed” and “shy out from” is plainly discussed by solar energy proponents.

              Arithmetic is really simple. USA (except Alaska) receives about 1000-2000 kWh/m^2/year of solar radiation. This includes day cycle, annual cycle and weather conditions. USA consumes about 4000 TWh of electricity a year, which, assuming 10% efficiency of solar cells, would be met by no more than 40000 km^2 of panels. This is less than 0.5% of USA land area, and less than total area covered by roads in USA. So this is only a matter of price. If asphalt could generate electricity, you would already have enough power.

              As for energy storage, solarbuzz provides cost analysis of a complete systems, including battery backup. This is expensive, but not hugely expensive. Effective costs are doubled, giving final price of about 30 cents/kWh.

              *This* is why it is better to read about some topic, before trying to be smart about it.

  • NeilJefferyRW January 25, 2012 on 9:50 am

    Excellent article, demonstrating the potential of solar power to meet a significant proportion of our energy needs, if it is given sufficient financial support. At Renewable World we facilitate this through the provision of capital to kick-start renewable energy initiatives for poor, off-grid consumers in the developing world. Our programmes in East Africa and Central America, in particular, have successfully introduced solar cells to produce electricity for communities previously without access to modern energy. With the help of this up-front financial support, communities can develop further methods of income generation, stimulating innovation and enterprise. Widespread adoption of solar is thus both feasible and economically viable.