Generally, articles that go by the title “solar myths” are pro-solar polemics that seek to assuage the concerns of the layperson about the perceived disadvantages of adopting PV (see here and here for examples). Separate from these, but equally deserving of this unflattering moniker, are a number of oft-repeated claims that have proliferated through thesolarindustry over the years and whose truth is taken to be self-evident. A few of these require a reasonably nuanced explanation to unravel their illusory veracity, while others are just blatantly untrue. Below, we examine some prime offenders.
The Oil-Solar Myth
“With oil at $100/$20 per barrel, solar is headed for explosive growth/doomed.” How many times have we heard something of the sort from the mouths or keyboards of otherwise rational human beings? Yet, for all but a fraction of the world, oil is largely a transportation fuel, while solar is an electricity generation source. They don’t compete. Comparing crude oil prices to solar equities yields no correlation whatsoever outside that related to the overall market (so a positive relationship rather than a negative one). And as of 2010, oil-powered generation made up 1.1 percent of America’s electricity consumption.
The Myth of Grid Parity
The myth of grid parity goes thusly: one day, the cost of producing electricity from solar generation will fall to levels commensurate with that of fossil fuels. This moment will be a historical inflection point for solar demand. From then onward, deployment will proceed at an explosive, exponential pace for many years.
Forget the difficulty of defining the notion of parity with the grid, which makes it problematic enough to talk about meaningfully without providing a laundry list of assumptions (subsidized or unsubsidized? Wholesale or retail grid prices? Levelized cost or payback period?). The real problem with this neat, monolithic narrative lies with two embedded premises. The first is that the onset of grid parity is a relatively singular moment in time across the history of the world. In truth, the determining variables for any calculation of this sort -- insolation levels, subsidies, grid prices, financing structures -- vary so widely across geographies that convergence between solar energy costs and grid prices is likely to be decades apart in different regions of the world.
The second problematic premise is that levelized cost is the only barrier to widespread adoption. It doesn’t matter what the LCOE of a $1/W system in rural India is if there is no means to provide financing for the upfront system cost. It doesn’t matter how high grid prices are in Hawaii if there are major transmission bottlenecks preventing deployment.
For both these reasons, anyone expecting the standard “hockey stick” conception of grid parity to bear out is likely to be very disappointed. Instead, the long-term solar deployment curve will probably have more twists in it than a Christopher Nolan film. Simply put, “grid parity” isn’t an event in time; it is a concept, one that is sometimes useful, but mostly misleading.
The Myth of Co-Located Manufacturing
Imagine, if you will, a place where manufacturing and generation exist side by side. Where modules are deployed but a few miles from where they are produced, hot off the laminator and straight into the ground, and where shipping costs cease to exist. This is the concept of co-located manufacturing, and it has been touted time and time again.
Undeniably, it is a wonderful idea in theory. It has simplicity and a certain sentimental appeal, and allows for the creation of high-tech and construction jobs within the same zip code. No wonder, then, that as many politicians as businessmen have succumbed to its charm down the years.
But step back for a moment and ask yourself whether, local content requirements aside, it really makes sense. Solar module manufacturing ideally requires availability of high-tech workers and proximity to supply of key consumables (glass, encapsulant, etc). Solar generation requires abundant insolation and access to transmission. They have nothing in common. Optimizing the equation for one set of concerns will almost always mean compromising on the other.
The trouble doesn’t end there. From the perspective of the installation, why would you want to restrict yourself to a single supplier, when less expensive and equally robust products might be available from the market? Conversely, from the module plant’s point of view, the benefits of the installation as a captive sales channel cease to exist after the plant has been built, which in the case of a 50-megawatt facility built to serve a 50-megawatt installation, is one year and change.
None of this is rocket science. And yet we have seen several announcements of co-located manufacturing projects in recent years. Examples include Isofotón’s 50-megawatt plant in Ohio for Agile Energy’s 49.9-megawatt Turning Point farm, ENN’s tandem-junction plant to go along with a planned 700-megawatt project in Nevada and Nexolon’s $100 million facility as part of OCI Solar Power’s 410-megawatt project in Texas. It’s likely that the creation of permanent jobs was necessary to garner political support for these projects, which is where manufacturing comes into play.
My guess is that these firms are secretly hoping that when push comes to shove, they won’t have to build these plants. If there is one thing our recent report on the PV supply chain makes clear, the world definitely doesn’t need more manufacturing capacity right now.
Myth Until Proven Otherwise: Modules as a Commodity
Since 2009, we have heard in ever-growing volumes the words “modules” and “commodity” spoken of in the same breath. It seems a matter of consensus now, so often it is repeated: performance and reliability do not differ significantly from one producer to another, at least in the case of crystalline silicon modules. Consequently, all that matters when choosing between different vendors is price. Sales managers at module firms talk about the dreaded “booty call,” the time when a developer calls them for an eleventh-hour quote, interested only in whether they can beat a competing supplier’s price.
Certainly, there is certainly a growing perception in the market that modules are a commodity, and this has been reflected in recent module pricing dynamics. But perception is not reality, and I certainly don’t think it’s wise to look to the developer-installer community’s behavior as an indication of the truth of this matter. Prices have to be paid right now, while the risk of under-performance or failure may not become a reality until several years down the line. As the subprime mortgage crisis showed, it is human nature to prioritize short-term gain over longer-term risk. Most module buyers understand very little about the manufacturing process or the limitations of the standard certifications. Most bankability reports are paid for by the manufacturer, which raises questions as to their objectivity, and the correlation between a firm’s bankability “tier” and product quality is questionable as it is (see this GTM webinar for more on this). Finally, warranties provide little security in the case of a pure-play firm, whose continuing existence, especially in the current environment, cannot be taken for granted.
Only data on long-term warranty claim rates for different manufacturers will reveal a conclusive answer to this question, and that will take time. But for now, I think it’s wise to treat the solar commodity mantra as an unproven assertion. Until the verdict is in, assuming that product quality across the more than 300 active suppliers in the module space is more or less equal is at best naive. At worst, it is sheer folly.
Shyam Mehta is a Senior Analyst at GTM Research and the author of the just-published report Global PV Module Manufacturers 2013: Competitive Positioning, Consolidation and the China Factor.