Grid parity—the point at which solar electricity, without financial incentives, will cost the same as conventional electricity from your utility’s electric grid—has been the dream of the solar community for decades.
Among many efforts to speed that up is the U.S. Department of Energy’s SunShot Initiative, which has a goal of reducing the cost of solar electric systems by about 75%—to a little more than $1 per watt—by 2020.
This would make solar energy less expensive than grid energy in almost every U.S. market. However, unlike President Kennedy’s Moonshot program, which reached its goal the moment Neil Armstrong first set foot on the surface of the moon, the SunShot goal of grid parity will be achieved at different times depending on your location, roof orientation, the local cost of electricity, and several other factors. Indeed, it may have already arrived in some locations.
If you know all of the details of a particular site and solar design, determining the kilowatt-hour (kWh) price point at which PV power is equal to grid power is not too difficult. I’ve developed a grid parity calculator spreadsheet to assist (see “Web Extras” below).
Don’t be daunted by the 28 variables that go into this calculation—half of these are figured already by NREL’s PVWatts calculator (see http://rredc.nrel.gov/solar/calculators/PVWATTS/version1). And all of the variables have explanations or suggested ranges, so if you don’t have an exact figure for some, you can approximate, and update it later as needed.
There is a tendency to think that, as soon as apparent grid parity (GPA) is reached, we can do away with subsidies that are solar incentives. But this is not the case.
First, it would be unfair to the solar technologies to remove these subsidies unless fossil-fuel subsidies are also removed. This includes not only the obvious subsidies, but also the hidden subsidies, such as free disposal of wastes in the atmosphere, soil, and waterways. If the government is treating all aspects of the energy industry fairly, a residual subsidy representing the difference between GPA and true grid parity (GPT) will always be required.
Subsidies will be needed as long as there are uncertainties in any of the variables in the cost-comparison calculations. The grid parity calculator identifies three speculative variables—future inflation rate, future energy-inflation rate, and the value of modules at the end of their life cycle. Since these cannot be predicted with any degree of certainty, incentives will be needed beyond grid parity to cover the “cost” of this uncertainty—to move the solar alternative from being “probably economical” to “almost certainly economical.”
Incentives are also needed wherever PV systems are unusual or non-standard in a neighborhood. There is a natural resistance to change. Incentives beyond grid parity are needed to overcome this until rooftop PV systems become the norm.
Finally, incentives are needed to encourage people to invest their money at the beginning of the life cycle rather than buying electricity with the conventional pay-as-you-use method.
Grid Electricity. The cost of grid electricity (variable 1 in the GP spreadsheet) can be determined by reviewing your electric bill. Simply divide the total of the generation and transmission charges by the number of kilowatt-hours you’ve used.
Where utilities have time-of-use (TOU) rates in place, you’ll need to estimate how much energy your PV system will generate at various times of the day. Fortunately, PV systems often tend to hit peak production at about the same time that electricity is most expensive, so TOU customers’ PV systems will be producing a lot of the most valuable electricity. Where the utility’s rate varies depending on the amount of energy used (tiered pricing), the solar alternative will also replace the most expensive grid electricity you would have used and may also reach down into lower tiers as well. With both TOU and tiered rate schedules, you’ll need to study the system size and your usage patterns to arrive at an accurate estimate of the average cost per kWh of the electricity you’ll be saving.
The cost per kWh may sound small (ranging between 7 cents and 30 cents in 2010), but don’t let it fool you—it is a big factor in determining whether or not your PV system reaches grid parity.
Another major factor on the grid side of the equation is how fast the price of grid electricity will rise (variable 13) compared to general inflation (variable 12). Historically, energy prices have risen faster than overall inflation. But how much faster depends on which part of history you’re looking at. Typical estimates range from 0% to 6.5% per year. Just a single percentage point or two over the life cycle of a PV system can make a big difference in the grid parity calculation.
And how similar will the past be to the future? New supplies of fossil fuels are being discovered, and new methods of extracting previously uneconomic fuels are being developed. At the same time, both the worldwide population and the per-capita energy consumption are growing dramatically, especially in India and China. Here in the United States, higher gasoline prices may push people into electric cars, placing increased demand on the electric grid. Will the growth of electrical demand outstrip the growth of supply and drive up the price?
What about the cost of building additional power plants and transmission lines? How much will environmental concerns add to these costs? Will the government reduce fossil fuel subsidies? Will it add taxes to the fossil fuel industry, to reflect the true costs? Is the expansion of nuclear energy still a viable option after the crisis in Japan?
All of these factors play a role predicting the future price of electricity. While I agree that it’s likely to be more than the general inflation figure, it’s hard to say how much more. Perhaps the best approach is to try a few different estimates here, and see what the cost-per-kWh needs to be to reach grid parity at these points.
Solar Electricity. The cost-per-watt of solar-generated electricity (variable 2) is one of the biggest factors in determining when your system will reach grid parity. If you want the most accurate value, it’s best to get a quote from a local installer in your area, who can take into account any special cost factors associated with your particular site.
If you need only a general idea of the recent cost-per-watt prices in your area, you can search the National Renewable Energy Laboratory’s Open PV Project, and retrieve a list of local projects as a spreadsheet (see http://openpv.nrel.gov). Sort for recent, nearby projects about the same size as yours, and create a “cost per watt” column, which takes the total project cost on each line and divides it by the total number of watts (kW x 1,000) for that project. This will give you a good idea of local cost per watt—somewhere between $4 and $10.
Another significant variable is the PV system’s life cycle (variable 6). Most people will choose 25 years, since this is the length of a typical module warranty. It may also be about the same as the lifetime of the roofing material itself, especially if you’ve got a standard composition shingle roof. The 25-year cycle assumes a single inverter replacement midway through the cycle. You can adjust the length of the cycle if you anticipate that your modules and roof will last longer or if you feel that you’ll need to replace the modules earlier for any reason. Early replacement might also suggest that the salvage value (variable 14) will be increased as the modules will have some warranty life left.
System productivity (variable 4)—expressed in kWh per kW of system size—is also a major factor. This variable relies on NREL’s PVWatts calculator (see http://rredc.nrel.gov/solar/calculators/PVWATTS/version1). This is where the location, module orientation and tilt, weather, shading, and several other site-specific conditions are considered. Explanations and typical values are provided for most of the terms. For the GP calculator, always use 1.0 for the “DC Rating (kW),” since the system size is entered separately as variable 3. Note that “Cost of Electricity” is also already considered. Review and adjust as needed the first 10 “derate factors” according to the specifics of your system. Leave the eleventh one (age) as 1.0, since this is adjusted separately as variable 5 in the GP calculator. If parts of your array will be shaded during parts of the year, you may need to have a shading analysis done for more accuracy. After clicking the “Calculate” button in the PVWatts program, find the “AC Energy (kWh)” for the year. Results may range anywhere from 600 or less to 1,800 or so, depending on the factors.
The rate at which the modules lose productivity each year is included as variable 5. This information is usually on module spec sheets, but if you don’t have it, 1% is a reasonably conservative figure.
Although PV systems require little maintenance, it must be accounted for. The biggest cost is inverter replacement (variable 7), which is likely to be needed 10 to 15 years after the system is installed. One replacement at the midpoint of a 25-year life cycle is anticipated. If you are counting on a significantly longer system life cycle, this variable should be increased by about 8% for each year beyond the standard 25 years. The average inverter price, on a per kW basis, can be found at www.solarbuzz.com. If your system is designed already, you can plug in the actual cost of your inverter(s) divided by your system’s size in kW.
Miscellany. Tree trimming (variable 8) is included for sites where tree shade is an issue. This is simply the cost of having the trees trimmed, divided by the number of years between trimmings. Initial trimming or tree removal, or purchase of solar easements, should be included as part of the original installation costs in variable 2.
Occasional cleaning, plus monitoring and any out-of-warranty maintenance costs, are lumped into variable 9. Rainy areas usually don’t need module cleaning. System monitoring can often be set up for free, or nearly so. Many people have no out-of-warranty repair costs, but you might want to set aside a little something for the occasional oddball occurrence. If you’re spending your own time on doing any of the above work, you can estimate its value and add that into this variable.
Additional insurance costs (variable 10) should be included since it will now cost more to replace your home in the event of a disaster. Your homeowners insurance company should be able to give you a figure for the cost per $1,000 of additional equipment added.
In some states, the value that the PV system adds to your home may increase your property taxes. Fortunately, several states have eliminated these taxes (see map and www.dsireusa.org).
Finally, we need to subtract the “salvage” value of the PV system at the end of its life (variable 14). It is difficult to guess how much residual value modules will have after 25 years—they may have another 20 years of useful life. However, if module technology has improved greatly, and if their prices have come down significantly by the end of the life cycle, then perhaps rather than a net value, you will have a net cost of removing and disposing of them. Fortunately, this is not a major factor in either direction, so a guesstimate of “0” net value is often used here.
In my home state (New Jersey), 1 kWh from the grid costs 19 cents, and a 10 kW PV system can be installed for about $5 per watt. Facing anywhere from south-southeast to south-southwest, at tilts that are typical for roofs here (4:12 to 12:12), with no shading, and using equipment with reasonably low derate factors, I could expect to get 1,200 kWh per kW of PV modules. The productivity declines 1% per year and the life cycle will be 25 years. One replacement of the inverter at $715 per kW will add $7,150 in present-value dollars to the cost.
Nearby trees will need to be trimmed once every four years at $600 per trim, or $150 per year, and I’ve set aside $100 per year for maintenance costs—$25 for my time and water spent on the annual hose-down during pollen season, and $450 for an electrician’s visit once every six years (something oddball, like replacing wiring that got chewed on by critters). My insurance costs an additional $1 per year for every $1,000 of equipment I’ve added, or $50 per year. New Jersey law gives property tax breaks for added PV systems, so that factor is 0%.
I assume that general inflation will be 3% per year and that the cost of electricity will rise at 5% per year—2% faster than general inflation. I assume that at the end of the system’s life, my PV modules will be removed and sold for a residual value equal to the cost of removing the modules—i.e., the net salvage and disposal costs are zero.
After plugging the data into the GP calculator, it reports that at the end of 25 years, I would have spent $367 less on the PV system (in present-value dollars) than I would have spent for the same amount of electricity from the grid—just slightly past the apparent grid parity threshold. In this case, the tax credit and incentives available would serve as motivators to switch to solar energy, but would not actually have been needed to pay for any additional costs associated with the switch.
In most other areas, the cost of grid electricity has not yet reached 19 cents per kWh. But grid parity is still being reached in areas across the South and West despite lower grid electricity costs, since PV systems there receive more sunshine, increasing the annual kWh generated per kW installed, which offsets the grid electricity’s cost difference.
Tax credits and other subsidies help change potential solar sites from being financially marginal to financially viable. This creates demand, helping manufacturers and installers increase their economies of scale, bringing prices down further. Declining solar prices combined with increasing grid energy costs will be opening the solar/grid-parity window ever wider in the years ahead.
Jay Tyson is a solar planner/solar project manager living in central New Jersey.