Renewable Energy 57 (2013) 137e143

Contents lists available at SciVerse ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

A comparison of the cost and nancial returns for solar photovoltaic

systems installed by businesses in different locations across the United
States
Kenton D. Swift*
School of Business Administration, The University of Montana, 32 Campus Drive, Missoula, MT 59812, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 3 July 2012
Accepted 8 January 2013
Available online 24 February 2013

Depending on what one reads, the cost of solar photovoltaic energy systems (solar PV systems) has
already reached parity with grid produced electricity or, at the other extreme is far from parity. This
leaves potential end users who are considering the installation of solar PV systems confused as to
whether such an investment makes economic sense. These contrasting views are not surprising for
several reasons. First, many analyses of the costs and nancial returns of solar PV systems ignore
important factors impacting the potential costs of such systems. Second, the factors impacting costs and
nancial returns such as levels of solar insolation, government and utility incentives, and the cost of grid
produced electricity vary dramatically based on location. Finally, the cost of solar PV systems is declining
rapidly over time. This study compares the costs and nancial returns of solar PV systems installed by
businesses, to grid produced electricity, in four specic locations across the US for 2012. The amounts are
calculated after taking into account local solar insolation, and all available tax incentives and other
available incentives and rebates. The results demonstrate that costs and nancial returns of solar PV
systems vary dramatically depending on the location where they are installed. The differences are primarily the result of variations in state incentives, electricity prices, and the level of solar insolation. This
conrms that detailed, site specic, information about incentives, grid produced electricity rates, and
levels of solar insolation is needed to determine whether the installation of a solar PV system makes
economic sense in a particular location.
2013 Elsevier Ltd. All rights reserved.

Keywords:
Photovoltaics
LCOE
Solar nance
Grid parity
Solar economics

1. Introduction
Parity with grid produced electricity is the economic holy grail
for solar photovoltaic energy systems (solar PV systems). Depending on what one reads, the cost of solar PV systems has already
reached parity with grid produced electricity [1] or, at the other
extreme the cost is far from parity with grid produced electricity
[2e4]. This leaves potential end users who are considering the
installation of solar PV systems confused as to whether such an
investment makes economic sense.
These contrasting views are not surprising for several reasons.
First, many analyses of the costs and nancial returns of solar PV
systems ignore important factors impacting the potential costs of
such systems. Second, the factors impacting costs and nancial
returns such as levels of solar insolation, government and utility

* Tel.: 1 406 243 4182; fax: 1 406 243 6925.

E-mail address: kent.swift@business.umt.edu.
0960-1481/$ e see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.renene.2013.01.011

incentives, and the cost of grid produced electricity vary dramatically based on location. Finally, the cost of solar PV systems is
declining rapidly over time. The result is that solar PV systems are
likely to become cost-effective in more and more locations as costs
continue to decline.
The purpose of this study is to compare the costs and nancial
returns of solar PV systems installed by businesses, to grid produced
electricity, in specic locations across the US for 2012. The costs and
nancial returns are calculated after taking into account local solar
insolation, and all available tax incentives and other available incentives and rebates. The results are valuable in identifying the
factors that are likely to make installed solar PV systems economically viable in some locations, but more expensive than grid produced electricity in other locations. In addition, in locations where
solar PV systems are not currently economically viable, it is possible
to estimate the cost at which a system is likely to achieve grid parity.
The method used to complete this analysis is capital budgeting,
which is a standard method used for analyzing potential capital
expenditures by organizations. Capital budgeting can be used by

138

K.D. Swift / Renewable Energy 57 (2013) 137e143

businesses to determine whether the installation of a solar PV

system makes sense in the rms specic set of circumstances.
2. How companies make capital budgeting decisions
In making a solar PV system purchase decision, businesses have
little interest in generalized discussions of potential parity between
such systems and grid produced electricity. Rather, the organization
wants to make a specic comparison between its present and future
grid produced electricity costs, and the net cost of a solar PV system
over its useful life. Simple parity with grid produced electricity is not
enough. This is because the organization will have to commit capital
to purchase the solar PV system. Since capital is limited companies
need to invest in those projects that produce the greatest return.
Thus, an investment in a solar PV system must provide a nancial
return in comparison to grid produced electricity that compares
favorably with other investment opportunities.
2.1. Selecting a method for analyzing capital investments
A variety of methods are available for analyzing and comparing
potential capital investments and, in aggregate, these are normally
referred to as capital budgeting techniques. The net present value
(NPV) method discounts future cash ows to their present value
using the organizations cost of capital as the discount rate [5]. The
present value of the future cash ows is then compared to the
initial investment. A positive NPV occurs when the present value of
the future cash inows exceeds the initial investment. When the
present value of the future cash inows equals the initial investment, the NPV is zero. A zero NPV means that the projects future
cash ows are exactly sufcient to repay the invested capital and to
provide the required rate of return on the capital. Thus, a positive
NPV means the projects future cash ows exceed the amount
necessary to repay the invested capital and to provide the required
rate of return on the capital.
A common way of reporting the present value of the cost of
a solar PV project is the levelized cost of electricity (LCOE). LCOE is
the estimated present value of the cost of 1 kWh of electricity.
Rather than comparing the present value of the cost directly to the
benets of the solar PV system, it is compared to the cost of grid
produced electricity. A review of the methodology for calculating
LCOE is provided in Branker et al. [1]. Making such a comparison is
critical in an organizations decision process because they are
interested in identifying the least expensive source of electricity.
The LCOE of the solar PV system is often compared to the current
cost of grid produced electricity, rather than the LCOE of grid produced electricity. This type of comparison is really a comparison of
apples and oranges because the current cost of grid produced
electricity ignores future ination in energy prices. A better alternative is to calculate the after-tax LCOE of grid produced electricity,
taking into account expected ination, and compare this to the
LCOE of the solar PV system.
The internal rate of return (IRR) method is closely related to the
NPV method. The IRR method provides a rate of return from an
investment, rather than using a cost of capital in the calculation,
and it is dened as the rate of return, or discount rate, that forces
the NPV of the investment to equal zero. Thus, it is not necessary to
select a discount rate, as it is in NPV calculations.
For the capital budgeting cases presented in this article two
types of results are presented for comparing the lifetime cost of
a solar PV system to the cost of grid produced electricity. The results
provide both a comparison of the LCOE of the solar PV system to the
LCOE of grid produced electricity, and a presentation of the IRR of
the installed solar PV system. The IRR calculation includes the
future savings from not having to pay for grid produced electricity.

2.2. Considering loans in the capital budgeting analysis

In the literature on the nancial performance of solar PV systems, the cash ows related to debt nancing are often included in
the analysis [1,6]. In capital budgeting, however, debt nancing is
normally separated from the analysis of the nancial return on the
asset being purchased, except for the determination of the weighted average cost of capital used in the discounting of future cash
ows. So, for instance, the availability of low-interest debt nancing
for a solar PV project impacts the potential return of the project
indirectly, by changing the rms cost of capital. Bierman and Smidt
[7] state:
Cash disbursed for interest is normally excluded from the cash
ow computation used in analyzing investments. The interest
factor is taken into consideration by the use of the present value
procedures. To include the cash disbursements for interest
would result in double counting (p. 64).
In the uncommon circumstance in which debt nancing is
included in a capital budgeting analysis there is another important
issue to consider. An analysis that includes debt cash ows must
use a discount rate that reects the expected return to the rms
owners, or equity holders, not the overall cost of capital to the rm.
This is because the return being calculated is really the return to
equity holders after considering debt. The expected rate of return to
equity holders is, by denition, higher than the rms weighted
average cost of capital, which is normally used in capital budgeting.
This additional expected return creates the positive leverage that
provides the incentive for rms to borrow money. Thus, in such an
analysis the discount rate used in the LCOE calculations must be
increased to reect the expected return to the rms owners.
The cost of capital normally used in capital budgeting is the
weighted average cost of each type of capital including both debt
and equity. The cost of each type of capital is multiplied by the ratio
of the market value of the securities (debt or equity) representing
that source of capital to the market value of all securities issued by
the company [7]. Thus, if 60% of a companys total capital is debt
with an after-tax rate of 6.5%, and 40% of the companys total capital
is equity with a required return to investors of 15%, the weighted
average cost of capital for the company is 9.9%
((0.6  0.065) (0.4  0.15)). This is the rate the company would
use in its capital budgeting analyses. Projects that provide a rate of
return equal to, or exceeding, the weighted average cost of capital
are acceptable, and those that have a rate of return that is less than
the weighted average cost of capital are not acceptable.
For the LCOE calculations in this article a weighted average cost of
capital is used. The rate used is 7.0%, which is the weighted average
cost of capital across US public companies as of January 2012. The
rate is taken from Damodaran Online, which provides the weighted
average cost of capital across 5891 US public companies [8].

3. Factors to consider in the solar PV system analysis

The information needed for analyzing a solar PV system investment decision can be divided into four general categories.
These are:
 Cost of electricity: The electricity costs, both present and
future, that will be saved by installing a solar PV system.
 Sunlight: The amount of sunlight, or solar insolation, available
to provide power for the solar PV system.
 PV system costs and performance: The cost and performance
of photovoltaic panels and related components of the solar
energy system.

K.D. Swift / Renewable Energy 57 (2013) 137e143

 Financial incentives: The tax and other incentives provided by

federal, state, and local governments, and by utility companies
[9].
Each of the factors listed above can vary dramatically based on
location. That is, to accurately estimate the cost and rate of return
on a solar PV system the devil really is in the details of the circumstances in each location. To illustrate the impact of these factors on the cost and nancial return from the installation of a solar
PV system, case studies are provided for 2012 installations in four
different locations across the US. All signicant factors impacting
the cost and nancial return, including taxes and other incentives,
are included in the analysis.
The rst step is to calculate the net initial project cost of the PV
system using the following formula. Table 1 provides a key to the
formula nomenclature.
Net initial project cost:

NIPC IPC  FTC  STC  UR  1  MTR

(1)

In this study the following formula is used to calculate the solar

PV system LCOE. The formula includes tax benets and all other
incentives available for the installation of solar PV systems.
Levelized cost of electricity (LCOE) of solar PV system:

PT
LCOE NIPC

t 1

MAINT INVERTER  DEP  PBI  SREC

1 rt
t

PT

t 1

E  1  d
1 rt

(2)
The LCOE for the solar PV system is compared to the after-tax
LCOE of grid produced electricity in the location where the system is installed. The after-tax LCOE of grid produced electricity is
calculated using the following formula.
Levelized cost of electricity (LCOE) of grid produced electricity:


PT
LCOE


E  1  dt  PRICE  1  MTR

t1

1 r
PT

t1

1 r

4. Case studies e general information

Case studies are performed for the commercial installation of
a 50-kW solar PV system in the following locations:





Honolulu, Hawaii
Newark, New Jersey
Phoenix, Arizona
Minneapolis, Minnesota

For each case it is assumed that the system is installed by

a corporation with a marginal US federal income tax rate of 34%,
plus a marginal state income tax rate equal to the maximum state
corporate income tax rate where the system is installed. The National Renewable Energy Laboratory (NREL) gathers data on the
installed cost of solar PV systems through its Open PV Project. For
the period January 1, 2011 through December 31, 2011 the average
installed cost of systems between 40 and 60 kW was $5.81 per DC
watt before incentives. This average is based on 119 different installations with a total capacity of 5.9 MW [10]. For the 2012 case
studies an installed cost of $5.25 per DC watt is used, which is
approximately a 10% decrease in costs over 2011. For a 50-kW
system this is a total installed cost of $262,500 before incentives.
Other assumptions used in the case studies are described below. A
complete summary of assumptions is also provided in Table 2.
4.1. Cost of electricity
The Energy Information Administration (EIA; www.eia.doe.gov)
provides statistics on the average retail price of electricity in each
US state. The information is updated monthly. The current price of
electricity used in each case study is the average retail electricity
price for commercial and industrial customers for 2011 in the state
where the solar PV system is installed [11].

Table 2
General assumptions used in IRR and LCOE calculations for case studies. These assumptions are used in the case studies unless otherwise noted in the discussion.
System:

E  1  dt

139

(3)

- The solar PV system is pointed due south and tilted at an angle equal to

the latitude of its geographic location.

- The system is tied to the utility grid and net metering is permitted.
- The performance of the system degrades 20% over its 25-year useful life.

Degradation takes place in equal increments each year.

- The system inverters must be replaced after 13 years, and the replacement

cost is 9.5% of the initial cost of the PV system, adjusted for ination.
Table 1
LCOE calculation nomenclature.
Nomenclature
NIPC
IPC
FTC
STC
UR
MTR
T
t
MAINT
INVERTER
DEP
PBI
SREC
E
d
PRICE

- The DC-to-AC derate factor is 77%. That is, the actual AC output of the

PV system is 77% of its DC power rating.

- The system has a zero scrap value at the end of its 25-year useful life.

General:
Net initial project cost ($)
Installed cost of solar PV system (including sales taxes,
if any) before tax benets and rebates
Federal income tax credits
State income tax credits
Upfront utility rebates
Marginal combined federal and state income tax rate
Life of the project (years)
Year t
Maintenance cost for t net of income tax savings ($)
Cost of replacement inverter for t ($)
Income tax benet of depreciation for t ($)
Performance-based incentive for t, net of income tax cost ($)
Solar renewable energy certicate revenue for t, net of
income tax cost ($)
Energy output of system (kWh)
Degradation rate (%)
Price of one kWh of electricity for t

- The cost of electricity in year 1 is the average retail price of electricity

for commercial and industrial users for the state in which the system is
installed, for 2011.
- The ination rate for grid produced electricity is 1.6% per year.
- The annual maintenance and insurance cost is 1.0% of the initial cost of
the PV system, adjusted each year for ination.
- There is no incremental cost for the space occupied by the system.
Taxes:
- All grants and rebates are fully taxable in the year received.
- The federal income tax rate for the taxpayer is 34%.
- The PV system is depreciated for income tax purposes using normal

MACRS depreciation over ve years. In the rst year 50% bonus

depreciation is taken as allowed by IRC Sec. 168(k). No IRC Sec. 179
expensing election is made.
- SRECs received by the taxpayer are sold as they are received and are
taxable at ordinary income rates.

140

K.D. Swift / Renewable Energy 57 (2013) 137e143

Since solar PV systems have useful lives of 20e30 years it is also

necessary to consider future changes in the price of electricity. The
EIA forecasts the expected price of electricity in the US for 25 years
into the future. In its Annual Energy Outlook 2011, the EIA predicts
a nominal average annual increase of 1.6% per year through
2035 [12].

4.2. Solar insolation

The fundamental requirement for a successful solar PV system is
solar insolation. The reality is that different parts of the US receive
dramatically different levels of sunlight. For instance, in Anchorage,
Alaska the average daily hours of solar insolation is 3.0 kWh/m2/
day for a at-plate collector pointed south with a xed tilt angle
equal to the sites latitude. This compares to Phoenix, Arizona
where the average daily hours of solar insolation are 6.5, more than
double the level in Anchorage [13].
The NREL PVWatts calculator is used to determine the annual
level of solar insolation in each case study location. The calculation
assumes the use of a at-plate collector pointed south with a xed
tilt angle. In addition, A 77% DC-to-AC derate factor is assumed [14].1

4.3. Solar photovoltaic system costs

Solar panels require little maintenance and should last a long
time. A manufacturers guarantee typically provides that the panels
will produce at least 80% of its original capacity for 25 years. The
other critical piece of hardware in a solar electric system is the
inverter. It is a relatively small part of the entire system, making up
approximately 9.5% of the total cost [15]. Inverters do not last as
long as solar panels and the case studies assume that they must be
replaced after 13 years.

4.4. Financial incentives

There are a variety of nancial incentives available to businesses
investing in solar PV systems. These include federal tax incentives,
state tax incentives, state and local government rebates, and utility
rebates.
4.4.1. Federal income tax incentives
One of the biggest incentives for installing solar energy systems
in the US is the federal business energy credit described in IRC Sec.
48 and which is part of the investment tax credit. The amount of the
credit is 30% of the cost of property placed in service during the
year, and the credit is available through December 31, 2016.
Businesses also receive a tax benet from deducting the
depreciation on the solar energy property. US tax law requires that
when the business energy credit is taken for this property, the
propertys basis for depreciation must rst be reduced by one-half
the amount of the credit. Solar energy property described in IRC
Sec. 48 is ve-year property for purposes of federal tax (MACRS)
depreciation [16]. For 2012, solar energy property is also eligible for
50% bonus depreciation in the year of purchase under IRC Sec.
168(k).

1
The PVWatts calculator determines the solar radiation incident of the PV array
and the PV cell temperature for each hour of the year. The DC energy for each hour
is calculated from the PV system DC rating and the incident solar radiation, and
then corrected for the PV cell temperature. The AC energy for each hour is calculated by multiplying the DC energy by the overall DC-to-AC derate factor and
adjusting for inverter efciency as a function of load. Hourly values of AC energy are
then summed to calculate monthly and annual AC energy production.

4.4.2. State, local government and utility incentives

State and local government tax incentives for solar energy system installations generally consist of income tax credits, property
tax exemptions, and sales and use tax exemptions. For state income
tax credits the percentage of the credit varies widely by state, and
can make a dramatic difference in the net cost of an installed solar
PV system.
States, local governments, and utilities also offer a myriad of
grants and rebates to promote the installation of renewable energy
systems. The form of the grant or rebate is typically either an upfront
payment to reduce the initial cost of the system, or a performancebased incentive (PBI), in which the producer is paid based on electricity production. Solar energy producers participating in a PBI
program receive a double benet: the incentive payment plus the
electricity costs saved by using a solar energy system. Fortunately,
there is an online comprehensive database of state and local incentives called the Database of State Incentives for Renewables &
Efciency (DSIRE), at www.dsireusa.org. Funded by the US Department of Energy, DSIRE is an ongoing project of the North Carolina
Solar Center and the Interstate Renewable Energy Council.
Policymakers in several states, including New Jersey, have also
created Solar Renewable Energy Certicates (SRECs) to ensure that
a certain amount of solar energy capacity is installed in a designated area. States with SREC programs have a renewable portfolio
standard (RPS) that requires utilities to secure a portion of their
electricity from solar generators. In order to demonstrate that they
meet this requirement, the utility is required to earn or purchase
SRECs. One SREC represents 1 MWh of production. The SREC is
separate from the value of the electricity itself. In states with these
programs, SRECs provide an additional incentive to invest in solar
PV systems over and above the electricity savings. This is because
SRECs generated by a solar PV system can be sold to a utility
needing to meet its solar RPS requirement.
In order to produce SRECs, a solar PV system must rst be certied by state regulatory agencies. Once the system is registered,
SRECs can be issued based on electricity production. SRECs are not
assigned a monetary value. Instead, prices are set in the marketplace and are a function of supply and demand. Utilities can purchase SRECs in the marketplace or pay a penalty instead. The
penalty is referred to as a solar alternative compliance payment
(SACP), and the amount of the penalty sets an effective cap on the
market price of a SREC.
5. Case studies e results
Using the assumptions discussed above, case studies for a solar
PV system installation were created for Honolulu, Newark, Phoenix,
and Minneapolis. The results demonstrate the dramatic variation in
the cost and nancial return from the installation of a solar PV
system depending on location.
5.1. Honolulu, Hawaii
There are a number of factors that make Hawaii an ideal location
for solar PV systems. The average annual solar insolation is high, the
cost of electricity is high, and the state offers substantial income tax
credits for installing solar PV systems. Honolulu gets an annual
average of 5.7 h of solar radiation per day. The average retail price of
electricity in Hawaii for 2011 for commercial and industrial business sectors was $0.304/kWh, and the state offers a 35% income tax
credit on the cost of commercial solar PV systems, up to a maximum
of $500,000 [17].
The maximum state corporate income tax rate in Hawaii is 6.4%
[18]. Solar PV systems are also subject to an excise tax of 4.0% at the
time of purchase, and in Honolulu there is an additional excise tax

K.D. Swift / Renewable Energy 57 (2013) 137e143

141

Table 3
Returns for case studies inclusive of all benets.
Honolulu $5.25/DC
watt

Newark $5.25/DC
watt

Phoenix $5.25/DC
watt

Minneapolis $5.25/DC
watt

Inputs:
Installed system cost per DC watt
Initial system cost before rebates and credits
Federal income tax credit e 30%
State income tax credit
Maximum state corporate income tax rate
Utility rebates/performance-based payments
Average annual hours of solar insolation per day
Initial price of electricity per kilowatt-hour

$5.49a
$274,313
$82,294
$96,010
6.4%
$0
5.7
$0.304

$5.25
$262,500
$78,750
$0
9.0%
SRECb
4.5
$0.125

$5.25
$262,500
$78,750
$25,000
6.968%
PBIb
6.5
$0.080

$5.25
$262,500
$78,750
$0
9.8%
$0
4.6
$0.076

Outputs:
LCOE of PV system after tax benets and all other incentives
LCOE of grid produced electricity after tax benets
Internal rate of return (IRR) over 25-year life

$0.055
$0.208
31.60%

$0.073
$0.082
7.71%

$0.081
$0.054
3.34%

$0.180
$0.049
8.27%

The installed cost includes the 4.5% excise tax on the purchase of solar PV systems in Honolulu.
See the case study discussions for details of the specic benets of the performance-based incentives (PBIs) in Phoenix, and the solar renewable energy certicates (SRECs)
in New Jersey.
b

of 0.5% for a total of 4.5% [19]. The 4.5% excise tax must be added to
the cost of a solar PV system. In Honolulu, solar PV systems are
exempted from property taxes [20].
Consider an installation of a 50-kW PV system in Honolulu. The
30% federal tax credit on a $274,313 ($262,500 $11,813 excise tax)
solar PV system amounts to $82,294 and the 35% state income tax
credit is $96,010. Thus, the net cost of the system after tax credits is
only $96,009. As indicated in Table 3 this combination of incentives
results in a 31.60% IRR over the assumed 25-year life of the system!
The LCOE of the system is $0.055/kWh versus an after-tax grid
produced LCOE of $0.208/kWh. Clearly, the installation of a solar
electric system in Honolulu is likely to provide a substantial economic return.
The federal and state income tax credits have a signicant impact
on the return of the solar PV system. If the $96,010 state income tax
credit were not available in the example just described, the IRR
would be reduced to 9.10%, and the LCOE of the system would
increase to $0.176/kWh, as indicated in Table 4. But even without the
state tax credit, the solar PV system has a better rate of return than
grid produced electricity. If both the federal and state income tax
credits were not available the IRR is still positive at 4.44%. The LCOE
of the solar PV system, however, increases to $0.262/kWh which is
greater than the LCOE of grid produced electricity of $0.208/kWh.
5.2. Newark, New Jersey
At rst glance, Newark may seem like a poor location for commercial solar PV systems. The average annual solar insolation is only
4.5 h per day and there are no state income tax credits. But, because
Table 4
Returns for Honolulu, Hawaii solar PV system case study e with and without federal
and state income tax credits.
Returns with both Returns without
federal and state
35% state income
income tax
tax credit
credits available

LCOE of solar PV
$0.055
system ($/kWh)
LCOE of grid produced $0.208
electricity ($/kWh)
Internal rate of return 31.60%
(IRR) over
25-year life

Returns
without 35%
state income
tax credit and
without 30%
federal income
tax credit

$0.176

$0.262

$0.208

$0.208

9.1%

4.44%

of a robust market for SRECs in New Jersey, the city of Newark could
be a good location for the installation of solar PV systems.
Businesses installing solar PV systems in New Jersey receive
SRECs that can be resold in the marketplace. Under New Jersey law,
solar PV systems can qualify to generate SRECs for 15 years from the
date of installation. The amounts received from the sale of SRECs
are in addition to the electricity cost savings from installing a PV
system. The State of New Jersey Board of Public Utilities sets the
SACP payment which, in turn, sets the effective cap on the market
price of a SREC. In a document dated September 21, 2011, the Board
recommended a declining set of SACP payments for the next 15
years. These amounts set the theoretical cap on the sales price of
SRECs. Unfortunately, the availability of SRECs in New Jersey has
created an oversupply of solar PV installations in relation to New
Jerseys mandates for solar power. Thus, resale prices for SRECs
have become volatile and for the rst three months of 2012 New
Jersey SRECs sold for an average of 54.6% of the SACP. In this case
study 54.6% of the recommended SACP for each of the next 15 years
is used as the resale price for SRECs.2
Consider the installation of a 50-kW PV system in Newark at
a total installed cost of $262,500. In New Jersey the maximum state
corporate income tax rate is 9.0% [21]. Solar PV systems are exempt
from state sales taxes and state property taxes [22]. The federal
income tax credit amounts to $78,750. As indicated above there are
no state income tax credits but the benets from selling the SRECs
are substantial. For the system described in this case study the
SRECs sold in the rst year would amount to revenue of approximately $20,702, or about $0.35 per AC kilowatt-hour. The result is
that the IRR for a solar PV system installed in Newark, using the
assumptions described here, is 7.71%, which is slightly greater than
the 7% weighted average cost of capital used in the LCOE calculation. Thus, the LCOE of the system is $0.073/kWh versus a grid
produced LCOE of $0.082/kWh. That is, installing a 50-kW solar PV
system, as described, should have a slight advantage over the
purchase grid produced electricity.
5.3. Phoenix, Arizona
Located in the southwestern part of the US, Phoenix, Arizona
receives signicantly more sunlight than most other parts of the

2
Data on New Jersey SREC pricing is available at http://www.njcleanenergy.com/
renewable-energy/project-activity-reports/srec-pricing/srec-pricing, 2012. Also, see
New Jersey Statute Sec. 48:3-51 et seq. and New Jersey Administrative Code Sec. 14:
8-1.1 et seq.

142

K.D. Swift / Renewable Energy 57 (2013) 137e143

Table 5
Installed cost per DC watt to achieve parity with grid produced electricity.

Installed cost to achieve grid parity

with federal and state Incentives
Installed cost to achieve grid parity
with federal incentives, but no state incentives
Installed cost to achieve grid parity
with no federal or state incentives
a
b

Honolulu
Installed cost per DC watt

Newark
Installed cost per DC watt

Phoenix
Installed cost per DC watt

Minneapolis
Installed cost per DC watt

$4.30

$1.43

$2.16

$1.89

$1.43b

$4.36

$1.44

$1.27

$0.95

Returns were greater than parity with grid produced electricity.

Minneapolis provides no state incentives.

country. The average annual solar insolation per day is 6.5 h. While
this high level of sunlight would seem to make solar PV systems
a good bet, there are other factors in Phoenix that reduce the potential nancial benets. For instance, the average 2011 retail price
of electricity in Arizona for commercial and industrial businesses
was just $0.080/kWh.
Arizona provides a 10% state income tax credit for nonresidential solar PV system purchases, up to $25,000 in any one
year [23]. This is a great benet, but it is much less than the income
tax credit Hawaii offers. Utility companies in Arizona, however, also
have rebate programs. APS, one of the utilities serving the Phoenix
area, offers both upfront grants and performance-based incentives
(PBIs). The PBI offered for 50-kW solar PV systems is $0.08/kWh up
to a total incentive of $105,000.3 The maximum state corporate
income tax rate in Arizona is 6.968% [24], and solar PV systems are
exempted from sales taxes and property taxes [25].
The IRR from a 50-kW system installed in Phoenix over the 25year life of the system is 3.34%. This is a positive IRR, but it is less
than the weighted average cost of capital used in these case studies
of 7.0%. The result is that the LCOE of the solar PV system is $0.081/
kWh while the LCOE of grid based electricity is $0.054. Thus, even
with the high level of sunlight in Phoenix the cost of such a system
does not beat the cost of grid based electricity. This is because grid
based electricity in Phoenix is inexpensive, and state income tax
incentives are limited.
All may not be lost however. If the installed cost of a 50-kW PV
system were to drop by $1.00 per watt to $4.25 per DC watt, all else
being equal, the IRR would rise to 7.23%. The LCOE of such a system
would be $0.053/kWh compared to a $0.054/kWh LCOE for grid
based electricity. At this price the installation of a solar PV system
begins to make economic sense in Phoenix.
5.4. Minneapolis, Minnesota
Making a solar PV system pay is difcult, if not impossible, in
Minnesota. The average annual solar insolation per day in Minneapolis is 4.6 h, and the average cost of electricity in 2011 for commercial and industrial users was only $0.076/kWh. Consider the
installation of a 50-kW PV system in Minneapolis. In Minnesota the
maximum corporate income tax rate is 9.8% [26], and solar PV
systems are exempt from state sales taxes and state property
taxes [27].
The federal income tax credit amount is $78,750. But the state
provides no state income tax credits or rebates for installing solar
PV systems. Combining this with the low cost of electricity and the
limited sunlight, results in an IRR for a solar PV system installed in
Minneapolis of 8.27%. The LCOE of the solar PV system is $0.18/
kWh and the LCOE of grid provided electricity is only $0.049.

3
The rebate amount offered by APS for a particular solar PV system can be
estimated using the worksheets available at www.aps.com/main/green/choice/
solar/Business/default.html, 2012.

To make matters worse, a 50-kW solar PV system is not eligible

for net metering in Minnesota. In Minnesota only solar PV systems
with a capacity of less than 40-kW are eligible for net metering [28].
Thus, if an organization installing such a system cannot use all of
the electricity generated from the system, some of the savings
would be lost.

6. Achieving parity with grid produced electricity

The case study information in Section 5 can also be used to
predict the installed cost of solar PV systems that would be needed
to achieve parity with grid produced electricity. As discussed earlier, solar PV systems, installed in Honolulu and Newark, with all
federal and state incentives available and using the assumptions in
the case studies, already have a lower LCOE than grid produced
electricity (see Table 3). In addition, in Hawaii, even if the state
incentives are removed, the LCOE of a solar PV system is less than
that of grid produced electricity (see Table 4).
We can also predict the installed cost at which a solar PV
system would achieve parity with grid produced electricity for all
locations, with and without, federal and state incentives. This is
the cost at which the LCOE for the solar PV system equals the LCOE
for grid produced electricity. Table 5 provides a summary of the
installed cost, per DC watt, that would be needed to achieve grid
parity in each of the case study locations, using different
assumptions.
In Phoenix and Minneapolis, grid parity, with all federal and
state incentives available, would occur at $4.30 and $1.43 per DC
watt, respectively. A low cost is required to achieve grid party in
Minneapolis because solar radiation is relatively low (average of
4.6 h/day), electricity in Minnesota is relatively inexpensive
($0.076/kWh), and the state provides no incentives for installing
solar PV systems.
As discussed above, a solar PV system in Honolulu has a lower
LCOE than grid produced electricity, even without state incentives.
This is because of the high cost of electricity in Hawaii. In Newark,
Phoenix, and Minneapolis, parity without state incentives is achieved at installed costs of $2.16, $1.89, and $1.43, per DC watt,
respectively. Newark achieves parity at the highest installed cost
because electricity is relatively expensive in Newark ($0.125 per
kWh) and the top state corporate income tax rate is a relatively high
9%. Minneapolis provides no state incentives so the cost to achieve
grid parity is the same as that with all incentives available ($1.43
per DC watt).
If both federal and state incentives are removed, even lower
installed costs are required to achieve grid parity. In Honolulu,
Newark, Phoenix, and Minneapolis, parity without federal or state
incentives is achieved at installed costs of $4.36, $1.44, $1.27, and
$0.95, per DC watt, respectively. Based on these results, solar PV
systems are likely to achieve grid parity, or be less expensive than
grid produced electricity, across the US as installed costs approach
$1 per DC watt.

K.D. Swift / Renewable Energy 57 (2013) 137e143

143

7. Summary

References

The results from the four case studies demonstrate that costs
and nancial returns of solar PV systems vary dramatically
depending on the location where they are installed. At one extreme,
the case study for Honolulu has an IRR of 31.60%, while at the other
extreme the case study for Minneapolis has an IRR of 8.27%. The
differences in the returns from the solar PV systems case studies are
primarily the result of variations in state incentives, electricity
prices, and the level of solar insolation. This demonstrates that
detailed, site specic information, about incentives, grid produced
electricity rates, and levels of solar insolation are needed to
determine whether the installation of a solar PV system makes
economic sense in a particular location.
In addition, by varying the installed cost of solar PV systems in
specic locations it is possible to estimate the installed cost at
which such a system becomes economically attractive. For instance,
if the installed cost of a solar PV system in Phoenix, Arizona decreases by $1.00 per DC watt from $5.25 to $4.25, the IRR on the
system increases from 3.34% to 7.23%, and begins to make economic
sense.
Finally, it is possible to predict the installed cost at which solar
PV systems reach parity with grid produced electricity, both with
and without federal and state incentives. Without incentives the
installed cost of solar PV systems would need to drop substantially
to achieve grid parity. The cost at which this is likely to occur depends on the level of solar radiation and top state income tax rates
in a given location. It is certain, however, that as the price of solar
PV systems continues to drop they will become economically viable
in more and more locations.

[1] Branker K, Pathak MJM, Pearce JM. A review of solar photovoltaic levelized cost
of electricity. Renewable and Sustainable Energy Reviews 2011;15:4470e82.
[2] Peters M, Schmidt TS, Wiederkehr D, Schneider M. Shedding light on solar
technologies e a techno-economic assessment and its policy implications.
Energy Policy 2011;39:6422e39.
[3] McHenry MP. Are small-scale grid-connected photovoltaic systems a costeffective policy for lowering electricity bills and reducing carbon emissions?
A technical, economic, and carbon emission analysis. Energy Policy 2012;45:
64e72.
[4] Yang C. Reconsidering solar grid parity. Energy Policy 2010;38:3270e3.
[5] Brigham EF, Daves PR. Intermediate nancial management. 9th ed. Mason,
Ohio: Thomson/South-Western; 2007.
[6] Singh PP, Singh S. Realistic generation cost of solar photovoltaic electricity.
Renewable Energy 2010;35:563e9.
[7] Bierman Jr H, Smidt S. The capital budgeting decision. 9th ed. New York:
Routledge; 2007.
[8] Damodaran Online. Cost of capital by sector. Stern Business School, New York
University, http://people.stern.nyu.edu/adamodar/; January 2012.
[9] Swift KD. Is a solar energy system right for your organization? Management
Accounting Quarterly 2011;12:38e47.
[10] http://openpv.nrel.gov.
[11] U.S. Energy Information Administration. Electric Power Monthly February
2012.
[12] U.S. Energy Information Administration. Total energy supply and disposition,
and price summary. Annual Energy Outlook 2011.
[13] National Renewable Energy Laboratory. Solar radiation data manual for atplate and concentrating collectors; 1994.
[14] http://www.nrel.gov/rredc/pvwatts/.
[15] Galen B, Darghouth N, Wiser R, Seel J. Tracking the sun IV: an historical summary
of the installed cost of photovoltaics in the U.S. from 1998 to 2010. Environmental Energy Technology Division, Lawrence Berkeley National Laboratory.
[16] U.S. Internal Revenue Code Sec. 168(e)(3)(B)(vi)(1).
[17] Hawaii Revised Statute Sec. 235-12.5.
[18] Hawaii Revised Statute Sec. 235-71.
[19] Hawaii Revised Statute Sec. 237-13.
[20] Ordinance of Honolulu 8-10.15.
[21] New Jersey Statute Sec. 54:10A-5.
[22] New Jersey Statute Sec. 54:4-3.113a et seq.; Sec. 54:32B-8.33 and New Jersey
Administrative Code Sec. 18:24-26.1 et seq.
[23] Arizona Revised Statute Sec. 43-1085 and Sec. 43-1164.
[24] Arizona Revised Statute Sec. 43-1111.
[25] Arizona Revised Statute Sec. 42-5061(N); Sec. 42-5575(A)(14); and Sec.
42-11054(C)(2).
[26] Minnesota Statute Sec. 290.06.
[27] Minnesota under Minnesota Statute Sec. 272.02, and exempt from sales taxes
under Minnesota Statute Sec. 297A.67.
[28] Minnesota Statute Sec. 216B.164.

Acknowledgments
An early version of this study was presented at the World
Renewable Energy Forum 2012 in Denver, Colorado. The author
wants to thank the participants in the session for their helpful
comments.