When Will Grid Parity Come?

I have a confession to make: I’m not actually going to answer the above question.

But I am going to tell you it will be soon—probably within five years—and probably even sooner in many locations where “conventionally-made” electricity is more expensive than the 10 cents per KWH that is frequently bandied about as total cost for electricity (power production plus transmission and distribution costs). Indeed, in some cases, PV is already at grid parity. For example, a friend who bought a piece of land 1/4 mile from existing power lines is already at grid-parity because his whole solar system—unsubsidized—cost less than the $100,000+ the power company wanted to charge him to extend their power lines to his to-be-built house.

Although this is an anecdotal example, it illustrates a point rarely made, which is that in many places, even unsubsidized solar is already at grid parity, and that of the locations that are not at grid parity today, some places will reach “grid parity” before others. For example, many places in developing countries rely on diesel-fired generators to provide electricity a few hours per day. There is little doubt that even today, unsubsidized PV (photovoltaic) electricity would be cheaper than electricity provided by a diesel-powered generator.

It is true that today, probably less than 10% of the world’s electricity-consuming population lives in places where conventional electricity is more expensive than unsubsidized PV would be, but as conventional electricity increases in cost and as PV decreases in cost, the percentage of the world’s population who can buy PV-made electricity cheaper than conventionally-made electricity will increase.

So, it appears that in order to answer the question presented by the title of this article, we really need to answer two separate questions, as follows:

Will Conventionally-Produced Electricity Cost More In The Future?

Everybody knows that the answer to this question is “yes.” But what is somewhat less clear is, “Why?” and “By How Much”? The answer to the second question is “By A Lot”, and here is why:

1) The vast majority of the electricity in the world is produced today either by the burning of fossil fuels (mostly coal and natural gas) or via nuclear reactors. Both of these methods have been highly subsidized by the public sector, although this has traditionally been neither recognized nor quantified. To put this a different way, both conventional methods have significant societal costs that neither the producers of the electricity nor the consumers have paid out of pocket (we’ve all paid these costs in terms of air pollution and global warming, but not in actual dollars out of our pockets) This has begun to change and as these costs become appreciated and incorporated into the total cost of conventionally-produced electricity, its cost will rise substantially. For fossil fuels, for example, the cost of CO2 sequestration has started to be added into the cost of building coal-fired or natural gas fired power plants, and for nuclear, people are starting to put a cost on the issue of dealing with the spent fuel from nuclear reactors. And even if you sequester every pound of CO2 (which would be prohibitively expensive), you still have to deal with the mercury, sulfur dioxide and numerous other pollutants put out by conventional power plants, not to mention the thousands of people who die annually mining coal, transporting the various fossil fuels around the world, etc. For nuclear, even if you take all costs into account, that still doesn’t eliminate the chance of a nuclear meltdown. Yes, I know about all the “fail-safe” mechanisms in existence today, but, without argument, we can all acknowledge that nothing human is “fail-safe.”

To put it in a more personal way—If you had a choice of living 5 miles from a PV power plant or a nuclear power plant, who wouldn’t choose the former?

In fact, the best measure of this risk is and the problem of spent nuclear fuel is to note that today, without governmental financial guarantees, the private sector in the United States CANNOT competitively build and operate nuclear power plants. In other words, if a power company wants to build a nuclear power plant today—and governmental guarantees were unavailable and the company must pay for safely disposing of the nuclear waste—the cost of that nuclear power would be greater than the cost of unsubsidized PV-generated electricity. In other words, if one adds in ALL the costs of unsubsidized fossil-fuel-generated or nuclear-generated electricity, solar electricity is ALREADY AT PARITY TODAY.

2) But that isn’t all. In addition to considering the “unfunded” costs (discussed above) that have never been taken into account in pricing conventional electricity at 10 cents-per-KWH, one must examine what will happen over the next few decades to the “funded” costs that have always constituted a large percentage of the cost of conventional electricity, namely, the cost of the fossil and uranium fuels and the capital costs of building the plants themselves. Again, nobody doubts that as demand for electricity increases due to the emerging global middle-class, the costs of both fossil fuels and uranium will increase (both have increased approximately ten-fold in the last decade), and that capital costs of building the cleaner fossil-fuel plants and the safer nuclear plants will increase significantly.

Now that we know that conventionally-produced electricity will rise substantially going forward, we must answer the second question:

Will PV-Produced Electricity Cost Less In The Future?

This is an easy one, too--Everybody knows that the answer to this question is “yes.” But again, what is somewhat less clear is, “By How Much” and “Why?”

The answer to the first question is “By A Lot,” and here’s the “Why”:

1) Everbody knows what has happened to the cost of polysilicon in the past few years, and everyone expects poly to go back to where it started (and probably lower) within a few years. Although people argue about the timing, most people agree that poly will drop by 50-75%, and perhaps even more, in the next few years. Since poly accounts for about 50% of the cost of building panels, a 75% decrease in its cost will have a large impact on the cost of the panels.

2) Balance-of-system and installation costs are going to drop by at least 50% over the next few years as well. For example, 3 years ago, I paid in excess of $5,000 for a 6KW inverter that is probably less complicated than the flat-screen TV I bought last month for $499. I would bet, however, that in 2005, SMA only made a few hundred inverters of that size, whereas Vizio probably made a million of the flat-screen TV’s like the one I bought. I would be shocked if 6KW inverters are not selling for less than $1000 (of cheaper dollars) in 5 years. I suspect this will also apply to the disconnects, combiners, etc., that are used in PV systems. As quantities produced increase from hundreds or thousands to hundreds of thousands, and as the industry becomes more competitive, costs will drop significantly.

3) Finally, greater efficiencies of present PV technologies, and discovery of new technologies, have the potential to drop PV cost by an additional 50%. For example, just in the last few years, panel efficiencies have increased from about 15% to 22%, and wafer thicknesses have dropped, even though they are achieving greater efficiencies. My guess is that the watts obtained per kilo of silicon at the best fabs today is double what it was 10 years ago. It seems not unreasonable that with improved process technologies and executions, a further improvement to 30% efficiency may be achieved, and wafer thicknesses decreased yet again. Thus, a doubling of the watts that can be made per kilo of silicon over the next five years seems reasonable, especially given that the pace of R & D in this field now is probably 10 times greater than what it was 10 years ago.

In addition, it seems reasonably likely that existing non-silicon technologies will be improved and that other, perhaps game-changing, non-silicon technologies will emerge and will also yield significant cost decreases in dollars-per-watt. Several of these are already in production and indeed, First Solar (FSLR) has reported manufacturing costs approaching $1/watt, and it is likely that 2008 will see costs drop under $1/watt.

In summary, it is indisputable that going forward, conventional electric generation will increase in cost and that solar will decrease. Even if these changes only amount to a few percent per year (and on the recent conference call, Sunpower’s (SPWR) executives suggested after 2008, production costs will see decreases amounting to upper single-digit percentages), closing the gap to achieve grid-parity (to the 10-cent cost) will happen within 5 years, and will happen much sooner in places where conventional electricity is more expensive than 10 cents per KWH.

Related reading here and here. 

Disclosure: The author owns a substantial long position in CSIQ, but has no position in the other stocks mentioned in this article.

Comments

[tessant:]

grid parity is coming...when silicon valley puts their collective heads together, they come up with answers (think computers and internet)- solar and alt energy is their current project.

scott w
growthportfolio.ning.c...
"where social networking meets investing"

2008 Jan 28 08:04 PM

[alpha24seven:]

It won't come from the current Si based solar industry. The future is in solar 2.0. That is the companies like Nanosolar et al.

One item that people tend to miss. The price can and should be lower for solar today. It is kept artificially high to insure that there is no disruption in subsidies. This is a fact that every solar insider knows to be true. They will defend the subsidy with extreme prejudice. To break this paradigm, it will have to come from external sources when they bring new technologies not needing a subsidy. This will happen quickly once the politicians realize a particular technology is much less expensive and shipping. The subsidies then will be a sticky political situation and be quickly shelved leaving the entire current Si based solar industry massively exposed. It's happening already. Witness the recent lack of support for the tax subsidies in the energy bill. Only the beginning and with companies like Nanosolar shipping product to industrial MW level partners -- the writing is on the wall.

2008 Jan 29 12:38 PM

[canb888:]

alpha's comment is pathetic, it is almost saying today's solars are hopeless, let's wait for the new technology which may happen in five to ten years. I bet that is what he is doing with his own dollars - avoid or even short the silicon-based solar companies. Instead, I would argue that we need actions today just like what the article already said very well. The actual social cost of fosil fuel generated electricity is much higher than what we pay in cash, therefore, grid parity is already here in many places from a macro point of view. Petty politicians cannot see macro impact. Solar will continue to grow regardless of people like alpha or the big oil or the petty politicians because there are more people like Al Gore and progressive forces on this earth we call home that can think in macro terms. Jack, keep up with the good work and continue to contribute for the green cause. I enjoy many of your articles.

2008 Jul 05 12:48 AM