FINANCIALS OF SOLAR PHOTOVOLTAIC SYSTEMS FOR CUSTOMERS, INVESTORS, AND ENTREPRENEURS
The decisions to purchase, or invest in, or find the right price point for solar photovoltaic systems is complex. The assumptions and trends drive the financial viability of each of these decisions. Feasibility of meeting your goals depends upon your location which in turn drives the incident solar radiation, the size of the system, the cost of electric power, and available tax credits. Governmental policies are also major drivers in the financials making marginal investments more attractive.
Most online solar calculators are too simplistic and are biased thereby compromising credibility. Our objective is to provide independent analyses to support purchasers, entrepreneurs, and investors to help them achieve their goals with the most accurate data for their scenario.
Solar Modeling, Assumptions, and Financials
Modeling of the financials of any solar PV system begins with assessment of how much power consumption needs to be generated. This drives the size and most of the cost of the system. The location determines the incident radiation and also directly impacts the system size and cost. All of these factors are built into our model which has been refined and validated on multiple installations. Some of the general findings from our model are summarized below.
Location, Location, Location – Works for Solar as Well as Real Estate
Incident solar radiation or “insolation” is the radiation incident on a surface. Insolation includes attenuation caused by the angle of incidence, the atmosphere, and weather and varies seasonally. All data referenced here use the average insolation over a full year and are derived from 20 years of measurement and are shown in equivalent hours/day of sunshine. The input to the model uses a 500KHW/month target which sizes the solar panels based upon the hours of sun at each location. Further this chart below assumes an array cost of $5/W plus inverters and miscellaneous items like cabling and connectors. Since we assume the system to be connected to the grid, we do not include battery cost in these calculations. Stand alone systems would need to add batteries in the cost in but would not need the added cost of the grid connection that can be expensive in rural environments. Savings are based upon the average cost of electricity in the US in 2007 which was 9.13 cents/KWH. The net cost is calculated assuming the current 30% tax credit from the Federal government but without state rebates or any feed-in tariffs as few have been implemented in the US. The resulting size and cost of the system required to meet the energy need of 500 KHW/month is shown below.
The system required for an average US location with about 4 h/day of sun is approximately a 4 KW system costing ~$20,000. This total system cost is dropping and we update our model when used for clients based upon current costs. Many locations in the southwest from Texas sweeping to the northwest including parts of Oregon have substantially higher solar radiation and therefore require a smaller and less expensive system. Conversely, locations in the northeast have less solar radiation and therefore require larger and more costly systems to generate the same amount of power.
Simple Payback Metrics Simply Overestimate Payback
Simple payback with constant cost of power makes solar PV appear less attractive than it is. Simple payback does not consider the historic cost inflation of power which shortens payback times. This is shown the chart below for a small 3-4 KW system typical for household installations. Even with Federal tax credits, without cost of power inflation, the payback time is over 26 years. However, with historic inflation applied, a more accurate payback is 15 years. Additional steps can be taken to optimize the solar installation to ensure it meets objectives.
The cumulative savings of a solar installation are derived from the total cost of the system minus the cost of power saved. These savings drive the overall return enabling one to compare this potential investment with other investments under consideration for the same money. In the case described above, the cumulative savings becomes positive in the 16th year and grows substantially thereafter as is shown below.
Cost of Power is Key to the Financials
Both payback time and return are heavily driven by the cost of power offset by the use of solar. Using an average size system for our model, we calculated the impact of the cost of power on payback and return. Power costs vary by 4X across the US such that even some locations that may lower solar radiation, have expensive power that compensates thereby yielding more favorable paybacks and returns. The results of this analysis of the impact of the cost of power are shown below.
Power costs vary from a lower value of around 5 cents/KWH in Idaho to over 20 cents/KWH in Hawaii. Further, states are beginning to enable feed in tariffs (FITs) considerably larger than the cost of local power. For example, Vermont has established a 30 cent/KWH FIT for solar power guaranteeing a high value market for solar KWH sold back to the utilities for years to come. Practically, FITs apply more to utility scale installations rather than residential roof top installation since in the latter case, the amount of power fed back into the grid is a small fraction of what is generated. Such FITs have been more common in Europe and they clearly change the payback and return economics by shortening payback time and increasing the return to make solar more competitive with historic long-term returns from traditional investments. Further the administration’s commitment to carbon cap and trade most likely will result in further increasing the cost of power which will shorten payback times and extend returns over years to come.
Array Cost is Also Key to the Financials
For a particular size array under an average scenario, we model the payback and return vs. a range of array costs. For simplicity, we only vary the array cost and not the cost of the inverters. The arrays stem from diverse materials and processes and array technologies and cost are likely to be changing much faster than those of inverters. Modeling of these results appear below.
Current array costs of roughly $5/W yield payback times of 20 years and returns of barely 4%. However, array costs are decreasing and as the technology winners become clearer in the race to lower cost, the payback times shorten to 10-15 years with returns approaching 10%.
For the above scenario, we calculated the payback times for the average annual insolation by state and the average cost of power. These two influencers provide some surprising results as we see some states with shorter payback times adjacent to states with long payback times. However, in general the range of the cost of power of a factor of 4X dominates the smaller variation in the range of insolations but not always. The map below shows the fastest payback in the states of Hawaii at 9 years, California at 14 years, and the slowest of 28-30 years in the band from Virginia through Missouri and north through the Dakotas. The high cost of power in New York and Massachusetts places their payback at 17 years which is not that different from California.
For the most part, the internal rate-of-returns follow the payback times but not exactly. Hawaii and California have IRRs of 13.9% and 7.7% with West Virginia being 0% and Alaska being 0.7%. One counter-intuitive finding is the IRR in Arizona is 4.2% whereas in New York it is 5.7%.
Summary of Solar
In some locations and some scenarios, today’s payback and returns from solar investments are not competitive with traditional investments. However, emerging global trends are driving toward more favorable financials. These trends are: increasing cost of power, increasing government rebates in renewables, increasing feed in tariffs, and decreasing cost of arrays. Investment trends tend to be leading indicators of technology implementation, and it is clear investors are already anticipating a very bright future for solar PV.
Cost of Electric Power and State-by-State Sales and Marketing Priorities
The cost of power has been continually rising at a rate averaging ~4%/year since 2000. Although the price of oil and gasoline spike up and down, the petroleum content of our electricity cost is ~1.6%. Power is generated mostly by coal, natural gas and nuclear with a smaller amount from hydro, solar, wind,and geothermal. The cost of these fuels and operational costs are more stable and increase closer to the rate of inflation, yielding the steady rise in the cost of electricity. This increase in cost occurs across all four electricity sectors (residential, commercial, industrial and transportation) per DOE accounting.
The cost of power varies by a factor of ~4X across the 50 states. This variation is caused by the widely varying types of fuel and operational costs within each state. For example, the most expensive power costs -~$0.21/kWh -- are in Hawaii, which requires the shipment of fuels long distances. The least costly is ~$0.05/kWh in Idaho, where local hydro and nuclear power dominate. The map below shows that power costs in New England, California, the Great Lakes states and the Gulf states tend to be more expensive, whereas the states stretching from Utah through West Virginia tend to have cheaper power. The latter states derive power largely from coal and as a result, currently have cheaper power. Planned future policies such as carbon cap and trade are likely to change this imbalance somewhat as global warming concerns get increasing attention in Washington, D.C. On February 24, 2009, President Obama called for Congress to generate and pass a bill on carbon cap and trade so it is likely that the US will move in this direction sooner rather than later.
Just as the current state-by-state variation in cost of power represents a priority list for sales and marketing for energy efficiency and smart power companies, tomorrow's priorities will likely be driven more by carbon intensity. The common measure of carbon intensity is the amount of carbon emitted in the generation of a unit of power and is reported in units of pounds of CO2/MWh of power generated. Carbon intensity varies by a factor of ~15 across the 50 states with North Dakota having the highest, with an intensity of 2450 lbs CO2/MWh. The least emitting state is Vermont, which has an intensity of just 167 lbs/MWh. As carbon cap and trade transitions from being voluntary today to mandatory in the future, it is likely this imbalance in the cost of power will decrease somewhat, thereby shifting the sales and marketing priorities to different states. The states shown in red below are more likely to sustain larger increases in power cost in the future thereby changing the business priorities.
Customers of energy efficiency focus on simple payback time for new projects. Projects whose energy savings payback the original cost of the project within 2 years have historically been funded. Cost savings from energy savings beyond that payback time contribute directly to the bottomline. Parameters typical of energy efficiency projects have been used to map payback times across the 50 states assuming 10% savings of energy versus baseline conditions. Results show a payback time of between roughly 5 and 20 months across the US. This indicates favorable financials for funding a project even though the assumed savings are conservative based upon published data.
Investors have a wide array of mechanisms to obtain return on their investment. The metric commonly used to evaluate candidate investments is the Internal Rate of Return (IRR). The IRR is the annualized rate of return of the overall cashflow over the life of the investment. By assessing the IRR for an energy efficiency project, one can directly compare that to historical returns from all other investments from bonds, to stocks, to private equity, and to venture capital.
Analysis of the IRR for the parameters mentioned above reveals an average IRR of ~100% annually. The range varies from a low of 48% in Idaho to 143% in California to a high of 250% in Hawaii.
The input parameters and analysis vary with each energy efficiency project. The resulting paybacks and IRR will vary accordingly and must be addressed on an individual basis to establish a credible prediction.
However, general conclusions can be drawn from the above example. This project analysis reveals a highly desirable IRR and includes all expenses needed to execute the project. Since the financials appear so promising at the project level, a potential investor in the business would likely conclude that a successful business could be established based upon these energy efficiency projects.
Electric power is an essential resource for production of all goods and services. Each state differs in what it pays for power, how much it gains from products and services sold, and how much carbon is generated in the process. Understanding these state-to-state differences provides background for optimizing marketing and sales plans.
States vary widely in their ability to exploit power efficiently toward the goal of producing goods and services. One key metric is the percentage of GDP that each state spends on electric power. The states spending the smallest fraction of their GDP on power are Utah, Colorado, Washington and California. Utah spends the least at 1.68% of GDP. The states spending the most are: Mississippi, Alabama, South Carolina, and Hawaii. Mississippi spends the most at 4.37%. States spending the most as a percentage of their GDP on power have the potential to benefit the most energy efficiency improvements.
Another metric represents how much each state derives in GDP for each KWH of power consumed. This productivity metric is calculated by ratioing the $GDP generated by the KWHs required to produce the goods and services. The average over all the 50 states is $3.50 generated for each KWH consumed and this metric varies by more than a factor of 5X. The least efficient productivity for unit of power is found in West Virginia, Montana, South Carolina, and Kentucky. The least is found in West Virginia at $0.61/KWH. The most efficient productivity is found in California, New York, Massachusetts, and New Jersey, with California having the greatest at $8.60/KWH.
Another key metric is the productivity of carbon generated. The amount of carbon generated in each state depends on local resources and choices for power plants and represents infrastructure that changes slowly over time because of the large investments and time scale for building new plants. Given the carbon generated, we calculate the $GDP in goods and services per lb of CO2 generated. This metric captures the efficiency of the use of carbon to generate GDP within each state.
The average carbon productivity is $3.81/lb CO2 over the 50 states. The most productive states are Vermont, Idaho, and California with Vermont producing $25.24/lb CO2. The least productivity states are West Virginia, North Dakota, and Kentucky with West Virginia's carbon productivity being $0.30/lb CO2. As carbon cap and trade policies are implemented, the states with the least carbon productivity will be motivated more toward greater operational efficiency. As power becomes increasingly expensive, it will become increasingly difficult to hold onto inefficient practices. Carbon productivity is a leading indicator of the drive toward efficiency. Inefficient states and cities will have increasing pressure to implement energy efficiency practices.
The Stimulus Bill
The stimulus bill, more formally known as the American Recovery and Reinvestment Act of 2009, now passed into law, provides $16.8B for the DOE Energy Efficiency and Renewables. This same organization at DOE had a budget last year of $2B. Although this turnaround in funding is welcome, the flood of money creates difficulties for DOE in using existing contract mechanisms and devising new ones under the tight deadlines. Much of this is actually a work in progress and places special demands on companies seeking stimulus funding.
The challenges are to: identify specific, relevant opportunties in the emerging chaos; identify contract mechanisms that will likely have been created to handle this volume, and satisfy the requirements of the economic analysis attached to many opportunities.
We can help in all three areas. The economic analysis is a strength as we have been performing this on many products for companies and investors over the last 5 years. The above analyses show examples of our work that have been successful in generating awards for start up companies and funding from investors.
Once we establish the alignment of our interests and respective roles in a potential teaming arrangement, we can specifically help in the following ways:
Some of the ways economic analysis is addressed in an RFP is:
How to Work with Us on the Stimulus opportunities