Grid parity. It is the solar industry’s favorite expression/discussion topic/holy grail. You hear it at every conference, in many press releases, and from well-respected research institutes. We at GTM Research have even been guilty of using it ourselves. But grid parity shouldn’t be the focus of the industry -- and our expectations of that day when it arrives should be much more tempered, for four reasons. First, grid parity isn’t a single thing. Second, it won’t come at a single time. Third, we often won’t know when we’ve reached it. Finally, the initial impact of grid parity on demand is overestimated.
1) Grid parity isn’t a monolithic concept
Let’s start with the definition. At its core, grid parity is the point at which the cost of solar power matches that of grid electricity. There is, however, some nuance to be considered here:
Wholesale vs. Retail Parity. There is a meaningful difference between competing with retail electricity prices (which are higher) and wholesale prices (which are lower).
· Retail grid parity. If we’re comparing the cost to retail prices, are we in a location with time-of-use pricing? If so, we need to account for the time of generation of the solar system.
· Wholesale grid parity. The comparison here is even trickier. You can compare the cost of solar power against average wholesale prices, but solar is really a mid-peak resource. You can compare against average peak prices, but solar isn’t entirely a peak resource. You can compare against an alternative fuel (natural gas makes the most sense), but you should really account for the fact that fossil fuels are dispatchable (i.e., can be turned on/off on command), and solar is not.
Pre-Incentive or Post-Incentive Parity
When people talk about grid parity, they usually mean “unsubsidized” grid parity -- or the time at which solar competes favorably with grid prices without any incentives. In many places, though, this has only theoretical importance for the next few years. Take the U.S., for example. We have the Federal Investment Tax Credit -- a 30% tax credit (or, currently, cash grant) on system costs -- in place through 2016. Even ignoring state- and utility-level incentives, this means we don’t really need to hit unsubsidized grid parity until 2017 at the earliest. Instead, we need to become competitive with grid prices given available incentives. If we use this definition, we’ve already reached grid parity in many locations. Commercial (and in some cases residential) power purchase agreements (PPAs) are available for less than current retail prices.
On the other hand, if we’re talking about unsubsidized grid parity, we need to ask when it will become necessary. In other words, when will all incentives be removed? Until that point, unsubsidized grid parity is important only in theory.
The Cost of Solar Power
Let’s assume we have clarified the definition of grid prices. For argument’s sake, let’s say we are using unsubsidized residential retail prices as our comparison. The next question we have to ask is how we estimate the cost of solar power. After all, a PV system is almost all up-front capital and very little O&M. So in order to compare against grid prices, we need to estimate the levelized cost of electricity (LCOE) of the system -- or the price per kWh over the system’s lifetime.
Here, however, we run into a patch of thorns. What system lifetime are we assuming? Twenty-five years, as the module warranty might offer, or longer? What degradation rate are we assuming? In general, LCOE analysis is a product of assumptions, and variability among those assumptions only serves to further complicate the issue.
There is another way to look at this. In the U.S., the PPA model is king. If a company like SolarCity, SunEdison, Tioga Energy, or other is willing to offer you a PPA for less than current retail prices, you’ve reached grid parity, right? Well, that depends. Often, a PPA will have a built-in escalator of say, 3% per year. This escalator is often less than historical average annual retail price increases, but we all know that the future won’t necessarily mimic the past. So even if the PPA offers “grid parity” on day one, it could lose that status years later.
2) Grid parity won’t come at a single moment in time
Perhaps, you might say, the exact definition of grid parity doesn’t matter. It is more about the general concept of solar power being cost-competitive with grid electricity. Let’s assume that is true, or at least that we’ve agreed upon a clear definition. Even then, grid parity will be a slowly evolving phenomenon, not a discrete event that takes place at a single point in time. PV system costs vary widely from installation to installation, electricity costs vary from utility to utility, and incentives vary by state. This is not even taking into account differences in electricity prices across countries, which are even larger than those across states. Finally, let’s not forget that solar resource/insolation varies widely by location and drives meaningful differences in system output. As such, although one system may be at grid parity, its neighbor may not.
In essence, this means that grid parity will be a slowly developing phenomenon -- occurring over a number of years across various markets.
3) We often won’t know when we’ve reached it
The lynchpin to the complexity of grid parity is the fact that you generally cannot know whether a system has reached grid parity until the system is retired 20 to 30 years after installation. A simple example will help bear this out. Assume your current retail grid prices are $0.10/kWh and an installer offers you a 20-year fixed PPA at that exact price. If electricity prices stay flat or increase over the 20-year period, grid parity has been achieved. However, if electricity prices were to fall during that period, you might find that your solar electricity ultimately cost more than grid power. In other words, you’re always taking on some risk related to retail electricity prices.
The only way you can be sure that you’ve achieved grid parity right away is to receive a PPA that tracks grid prices -- and this is a difficult contract to find.
4) The initial impact of grid parity on demand is overestimated
There is often a belief in the solar industry that grid parity, however defined, will immediately open up vast amounts of new demand for solar. Consumer psychology, however, should never be underestimated. Even if I can put solar on my home (or business) for the same price as grid electricity, I still have to go through the trouble of getting a site assessment, allowing construction on my roof, and so on. If I’m not using the PPA model, I may be deciding between a PV installation and a kitchen remodel or other home improvement. To be sure, many people will happily expend a bit of effort in order to get clean, renewable power on their rooftop at a known cost.
However, it would be a mistake to assume that everyone will do so. After all, the most important question is not when we hit grid parity, whatever that means, but what (in terms of impact on demand growth) would be the consequences of that -- and the price elasticity of demand for solar is not well understood at all.
So what will really matter? In my opinion, there will be two achievements (both occurring over an extended period of time across various locations) that will make solar mainstream. For wholesale generation, it will be the point at which large volumes of solar become attractive to utilities even accounting for dispatchability. This could achieved either through the competitiveness of solar plus energy storage, or by making solar so cheap that supplementing solar generation with natural gas or another peaking resource beats the peaking resource alone. For retail generation, the difference will be made when consumers can see significant cost savings from a PV installation that are assured. In essence, the droves of new solar customers will be driven by cost savings, not cost parity.
The bottom line is that the solar industry is thriving even in our complicated, semi-parity world. Over time, prices will continue to fall and incentives will become increasingly unnecessary. In the meantime, let's see grid parity for what it is: an attractive idea that will just be one among many factors enabling the solar revolution.