Valuing the Attributes of Renewable Energy Investments

Paper Menu >>

Journal Menu >>

Smart Grid and Renewable Energy, 2009, 43–53

Published Online September 2009 (http://www.SciRP.org/journal/sgre/).

Valuing the Attributes of Renewable Energy Investments

ABSTRACT

Increasing the proportion of power derived from renewable energy sources is becoming an increasingly important

part of many countries’s strategies to achieve reductions in greenhouse gas emissions. However, renewable energy in-

vestments can often have external costs and beneﬁts, which need to be taken into account if socially optimal investments

are to be made. This paper attempts to estimate the magnitude of these external costs and beneﬁts for the case of re-

newable technologies in Scotland, a country which has set particularly ambitious targets for expanding renewable en-

ergy. The external effects we consider are those on landscape quality, wildlife and air quality. We also consider the

welfare implications of different investment strategies for employment and electricity prices. The methodology used to

do this is the choice experiment technique. Renewable technologies considered include hydro, on-shore and off-shore

wind power and biomass. Welfare changes for different combinations of impacts associated with different investment

strategies are estimated. We also test for differences in preferences towards these impacts between urban and rural

communities, and between high-and low-income households.

Keywords: Renewable Energy; External Costs and Beneﬁts; Choice Experiments

1. Introduction

Increasing the proportion of power derived from renew-

able energy sources is becoming an increasingly impor-

tant part of many country’s strategies to achieve reduc-

tions in greenhouse gas (GHG) emissions. However, re-

newable energy investments can often have external

costs and beneﬁts, which need to be taken into account if

socially optimal investments are to be made. This paper

attempts to estimate the magnitude of these external costs

and beneﬁts for the case of renewable technologies in

Scotland, a country which has set particularly ambitious

targets for expanding renewable energy. The external

effects we consider are those on landscape quality, wild-

life and air quality. Unlike other papers in the literature,

we do not restrict our investigation to the effects of par-

ticular technologies (such as hydro or wind, Alvarez

Farizo and Hanley, 2002; Hanley and Nevin, 1999), but

consider impacts applicable to a wide range of renewable

technologies. We also consider the welfare implications

of alternative investment strategies for employment and

electricity prices. The methodology used to do this is the

choice experiment (CE) technique. Renewable technolo-

gies considered include hydro, on-shore and off-shore

wind power and biomass. Welfare changes for different

combinations of impacts associated with different in-

vestment strategies are estimated. We also test for dif-

ferences in preferences towards these impacts between

urban and rural communities, and between high and

low-income households.

In what follows, Section 2 sets out some background

detail on energy policy in Scotland. Section 3 provides a

brief overview of the CE method, whilst in Section 4 we

outline the design and conduct of our empirical study.

Section 5 presents results from data analysis, including

the conditional logit models estimated from the CE data.

Section 6 evaluates the welfare effects of alternative in-

vestment strategies in renewables, whilst the ﬁnal section

presents some conclusions.

2. Scotland as a Case Study

Scotland has recently started down a new path in how it

generates electricity (ROS, 2002). The Scottish Execu-

tive has set two challenging targets for use of renewable

pow- er sources in the next 20 years:

Corresponding author.

Tel.:

+44

–

141

–

330

–

5982;

fax:

+44

–

141

–

330

–4940.

E-mail address: arielbergmann@yahoo.com (A. Bergmann).

doi:10.1016/j.enpol.2004.08.035

 by 2010, 18% of electricity consumed should come

from renewable generation,

 by 2020, that portion should rise to 40%.

Valuing the Attributes of Renewable Energy Investments

Currently only 10% of the electric energy produced in

Scotland comes from renewable sources such as wind

energy, hydro and waste-to-energy plants. These targets

are thus ambitious, as the potential for large-scale hydro

is close to fully developed, and as wave energy technol-

ogy has not yet been adequately tested on a commercial

scale. The 400% increase in clean electricity production

will have to come from new energy projects and new

technologies. The 2020 goal is three times larger than the

negotiated portion Scotland has committed to as its con-

tribution the UK’s reduced GHG obligation. Denmark,

with a 79% renewable electricity goal by 2030, is the

only European Union member with comparable volun-

tary goals (de Vries et al., 2003).

The major political reasons for promoting renewable

energy are external to Scotland. The UK has accepted a

legally binding target of reducing emissions of a bundle

of GHGs by 12.5% below 1990 emission levels by

2008–2012, as its share of the European Union negoti-

ated target of an 8% reduction in GHGs under the Kyoto

Protocol. The Energy White Paper “Our energy futurec-

reating a low carbon economy”, published in February

2003 by the British Government, includes an even more

ambitious path of reducing CO2 emissions by some 60%

of current levels by 2060 (DTI, 2003). Currently, the UK

Renewables Obligation-a requirement on power supply

companies to meet certain minimum fractions of total

supply from renewables-has set a target of 10.4% by

2010–2011.

The economic reasons for Scotland developing re-

new-ables are multifaceted. The ﬁrst reason is that re-

newable energy projects by their very nature should be

highly sustainable. There is minimal or no resource de-

pletion due to the use of renewables technologies, as

compared to gas-, oil- and coal-based energy. Renewable

energy projects, as with traditional fossil fuel projects,

tend to be capital intensive, so the opportunity to develop

and manufacture renewable energy equipment for do-

mestic use and international export exists. In 2001, Ves-

tas, a major manufacturer of wind turbines, announced a

manufacturing facility would be opened in Scotland

(Scottish Executive, 2001), although most capital equip-

ment is currently imported. There is the potential to

transfer some of the job skills learned in the North Sea

oil industry to the marine renewables sector, which in-

cludes tidal, wave, and ocean current power generation

technologies, as the off-shore oil industry declines (a

European Marine Energy Centre (EMEC, 2004) was

opened on Orkney in 2004 to assist in the advancement

of marine energy). Off-shore wind farm development

may need this skilled labour pool. England and Wales

will have a more difﬁcult time building sufﬁcient re-

newables capacity to provide adequate non-fossil fuel

energy that their populations will need to meet domestic

targets (OXERA, 2002). Scotland, on the other hand, has

some of the greatest renewables potential in all of Europe,

and therefore may have sufﬁcient excess supplies to trade

south of the border. Finally, rural areas of Scotland, with

some of the greatest needs for economic development,

will be the location of most all land-based renewable

energy projects (Hassan et al., 2001). These rural com-

munities may well reap beneﬁts from these long-term

projects.

A fundamental restructuring of the power industry will

need to be undertaken to achieve these renewable targets.

Since the 1880s in Scotland, as in the rest of the world,

the power industry has been organised with a centralised

hierarchical technological and management structure.

Ever-larger generating facilities based on fossil fuels and

nuclear power, and a uniﬁed transmission network to

distribute the electricity over hundreds of miles, were the

model of development. The nature of land-based renew-

able energy projects makes this development style tech-

nically impossible at this time. Current scale economies

dictate that projects like wind farms and biomass genera-

tion plants be 3–5% the size of a traditional 1200 MW

coal-ﬁred plant. Even the largest wind farms being plann-

ed today are only 20% of this size. Also, because of the

intermittency problems of renewable sources, greater

quantities (measured by megawatt capacity) of generat-

ing assets are needed because of the lower average usage

of this capacity. Renewable energy projects normally

require large amounts of space to capture the energy in

wind, water or solar radiation in sufﬁcient quantity to be

commercially viable. Dozens of communities in Scotland

will therefore likely be impacted by renewable energy

projects that will need to be constructed to meet the Scot-

tish Executive’s clean energy goals. As noted above,

only 10% of electricity in Scotland is currently generated

from renewable sources, mainly from hydro power sta-

tions (Scottish Executive, 2002b). The biggest current

share of generation (44%) is from nuclear stations: how-

ever, all of these are planned to close by 2023 (NIA,

2003), and no new nuclear plants are likely to be com-

missioned.

The primary policy instrument being utilised by the

Scottish Executive to motivate what is thus a very large

expansion of renewable energy sources is the Renew-

ables Obligation (Scotland) (ROS). The ROS has com-

bined a demand-push legal requirement for renewable

power usage with a supply pull ﬁnancial incentive pro-

gramme to reward private industry for constructing and

investing in new renewable energy generation projects.

The ROS compels licensed electricity suppliers to source

speciﬁc quantities of eligible renewable energy for sale

to all customers (residential, commercial and industrial),

or face ﬁnancial penalties for the shortfall. The associ-

ated increase in electricity costs is shared by all consum-

ers and avoids the free-rider problem (Rose et al., 2002).

The original minimum supply of renewable power by

Valuing the Attributes of Renewable Energy Investments45

retailers, by quantity, was set at 3% for2002–2003, rose

to 4.3% for 2003–2004, and will rise annually to 10.4%

in 2010–2011. During the 18 months since the ROS was

implemented in April 2002, signiﬁcant increases in re-

newable generation projects have applied for or are mov-

ing towards application and planning consent. In all,

1500 MW of capacity has sought consent and another

2500 MW of capacity is near requesting consent (BWEA,

2003).

The ﬁnancial incentives for private investment in re-

newable power facilities are created by the use of the

renewable obligation certiﬁcates (ROCs). Electricity sup-

pliers use these certiﬁcates as evidence that the required

percentage of sales is matched with eligible green power

production. The ROCs are traded separately from the

actual electricity being generated and had a market price

of £45–50 per megawatt during the ﬁrst year of the ROS.

This money is earned by the renewable power generating

company and represents revenue above the value of the

electricity being sold to the power market. Renewable

power generators earned £63–75 per megawatt of pro-

duction during the 2002–2003 period as compared to

£17–25 per megawatt paid for fossil fuel-powered pro-

duction.

3. The Choice Experiment Method

Renewable energy investments in Scotland are thus ex-

pected to grow rapidly in the near future. These invest-

ments will produce a series of potential impacts on the

environment, price of electricity and employment. Envi-

ronmental impacts will include landscape effects, effects

on wildlife and changes in air pollution (e.g. waste to

energy plants emit air pollution). Exactly what environ-

mental impacts occur, what happens to electricity prices

through changes in cost, and any changes in employment,

will depend on the exact investment mix (e.g. the balance

between on-and off-shore wind farms; the extent of hy-

dro developments). Taken together, environmental ef-

fects, price effects and employment effects can be

thought of as the attributes of a renewable energy strat-

egy. Knowing something about the relative economic

values of these attributes is important if we wish a re-

newables strategy to (i) take some account of public

preferences and (ii) take some account of economic

efﬁciency (beneﬁt-cost) concerns. Choice Experiments

are an economic valuation method which enables this

kind of information to be produced.

3.1 The Characteristics Theory of Value

and Random Utility Theory

CEs are based on two fundamental building blocks:

Lancaster’s characteristics theory of value, and random

utility theory (RUT). Lancaster (1966) asserted that the

utility derived from a good comes from the characteris-

tics of that good, not from consumption of the good

itself. Goods normally possess more than one charac-

teristic and these characteristics (or attributes) will be

shared with many other goods (Lancaster, 1966). The

value of a good is then given by the sum of the value of

its characteristics.

RUT is the second building block. RUT says that not

all of the determinants of utility derived by individuals

from their choices is directly observable to the research-

er, but that an indirect determination of preferences is

possible (McFadden, 1974; Manski, 1977). The utility

function for a representative consumer can be decom-

posed into observable and stochastic sections:

anan an

UVe



 (1)

where Uan is the latent, unobservable utility held by

consumer n for choice alternative a, Van is the systemic

or observable portion of utility that consumer n has for

choice alternative a, and ean is the random or unobserv-

able portion of the utility that consumer n has for choice

alternative a. Research is focussed on a probability func-

tion, deﬁned over the alternatives which an individual

faces, assuming that the individual will try to maximise

their utility (Bennett and Blamey, 2001; Louviere et al.,

2000). This probability is expressed as

(\)[() ()]

nananjn

PaCP VeVe



 (2) aj

for all j options in choice set Cn, a and n are as previ-

ously described, or

(\)[() ()]

nanjnjna

PaCP VVee



 (3) aj

To empirically estimate (3), and thus to estimate the

observable parameters of the utility function, assump-

tions are made about the random component of the model.

A typical assumption is that these stochastic components

are independently and identically distributed (IID) with a

Gumbel or Weibull distribution. This leads to the use of

multinomial (sometimes called conditional) logit (MNL)

models to determine the probabilities of choosing a over j

options (Hanley et al., 2001):

exp( )

()

exp( )

an jn

PU UjV





 (4) aj

Here, µ is a scale parameter, inversely related to the

standard deviation of the error term and not separately

identiﬁable in a single data set. The implications of this

are that the estimated β values cannot be directly inter-

preted as to their contribution to utility, since they are

confounded with the scale parameter. When using the

MNL model choices must satisfy the independence from

irrelevant alternatives (IIA) property, which means that

the addition or subtraction of any option from the choice

set will not affect relative probability of individual n

choosing any other option (Louviere et al., 2000). Mod-

elling constants known as alternative speciﬁc constants

(ASCs) are typically included in the MNL model. The

Valuing the Attributes of Renewable Energy Investments

ASC accounts for variations in choices that are not ex-

plained by the attributes or socio-economic variables,

and sometimes for a status quo bias (Ben-Akiva and

Lerman, 1985).

The estimated coefﬁcients of the attributes can be used

to estimate the tradeoffs between the attributes that re-

spondents would be willing to make. The price attribute

can be used in conjunction with the other attributes to

determine the willingness-to-pay of respondents for gains

or losses of attribute levels. This monetary value is call

the “implicit price” or part-worth of the attribute:

nonmarket attribute

part worthmonetaryattribute



















(5)

The scaling problem noted above is resolved when

one attribute coefﬁcient is dividing by another, as in

the part-worth equation, since the scale parameter

cancels out.

4. Study Design and Implementation

To meet Scottish Executive targets, hundreds of renew-

able energy projects of all sizes and types of technology

have been proposed. These range from large wind farms

and new hydroelectric schemes that have signiﬁcant im-

pacts on the countryside and local communities, to small

changes like the addition of solar panels to rooftops and

district heating plans with impacts that may only be felt

by the immediate residents. This paper’s objective is to

estimate the value of positive and negative impacts aris-

ing from the kind of renewable energy projects that will

be developed over the coming years.

4.1 Designing the Choice Experiment

In any CE, attributes must be chosen which meet a num-

ber of requirements. These are that they are:

 relevant to the problem being analysed,

 capable of being understood by the sample popula-

tion, and

 of applicability to policy analysis.

Identifying the set of attributes and the levels these

take is a key phase in CE design. To this effect, focus

groups were conducted with members of the general pub-

lic (Dewar, 2003). The objective set to each group was to

identify the ‘characteristics’ of ‘green’ electric energy

production that were regarded as ‘good’ or ‘bad’. The

facilitator had each group identify all the types of re-

newable power technologies that they could, and then

discuss the good or bad characteristics of each type of

energy project. Technologies that were identiﬁed were:

windmills, hydro schemes (run of river and reservoir),

tidal and wave power, solar (photovoltaic and hot water

panels) geothermal, various types of biomass or waste

combustion like burning municipal solid waste, wood

burning, animal and organic waste, natural gas from

landﬁlls and fermentation of organics. After

identiﬁcation of the attributes of each technology, the

groups were requested to separate into smaller sections

of two or three persons, and rank these attributes by im-

portance to them. After that exercise, individuals were

asked to indicate their personal choices for which char-

acteristics were most important or of concern to them.

Three characteristics that dominated all others were re-

vealed by the focus groups. One was that renewable en-

ergy projects have a low environmental impact, and

should reduce how we change or pollute the environment.

Another was that the projects be aesthe- tically pleasing.

This characteristic was a little more contentious because

some group members felt that both windmills and reser-

voirs are pleasing to observe, while other members felt

that large man-made structures took away from nature’s

scenic beauty. The ﬁnal dominant characteristic was that

wildlife should not be harmed any further and that pro-

jects that improved wildlife should be supported. Other

less signiﬁcant characteristics mentioned by individuals

or groups were the creation of jobs, the effect on electric-

ity prices, the abundance and sustainability of the re-

sources, and more localised control and responsibility for

the project.

Five key attributes were then chosen, based on these

focus groups responses, and also on government state-

ments (e.g. Scottish Executive, 2002b) and the literature.

The attributes chosen were:

 impacts on the landscape,

 impacts on wildlife,

 impacts on pollution levels, in particular, air pollu-

tion,

 creation of long-term employment opportunities, and

 potential increases in electric prices to pay for renew-

able sources.

Once these attributes were determined, a question-

naire was constructed that presented the context of re-

newable energy development in Scotland. The national

commitment by the UK to reduce production of GHGs

was explained. Survey participants were told that the

survey was not concerned with any speciﬁc type of re-

newables technology, but with the impacts that could

result from development of any renewable energy re-

source. The ﬁve attributes noted above were described,

with examples being given to clarify each type of impact.

SPSS (VERSION 10.0) was used to select a set of op-

timal choice proﬁles, which were then combined to make

up the choice sets used in the experiment. Table 1 shows

the attributes and levels as used in the ﬁnal design. Given

the ﬁve attributes and 17 associated levels, 1800 possible

proﬁles exist, which was an unfeasible number to employ

in the survey. We thus used a fractional factorial design

to reduce this full factorial to 24 proﬁles that could be

used to estimate main effects. This smaller set was also

Valuing the Attributes of Renewable Energy Investments47

Table 1. Attributes and attribute levels

designed for orthogonality, which is a desirable, but not a

required, mathematical characteristic of choice set con-

struction. Twenty different choice sets were thus de-

signed and used in the questionnaire, blocked into sets of

four (that is, each respondent worked there way through

four choice tasks). Combined plans were alternated in the

order in which they appeared as a choice set and the or-

der of the individual plans were alternated between the

ﬁrst or second within the choice set. This was done to

avoid any bias from the ordering of choices.

Choice sets were then presented, and the survey par-

ticipant was requested to indicate their preference. Each

set contained three options. Plans A and B were possible

renewable energy projects, each with different attribute

levels. A third option of choosing neither was given. This

‘neither’ option, commonly called the opt-out or status

quo option, stated that there would be no increase in re-

newable energy, that alternative programmes would be

implemented to avoid climate change, and that North Sea

natural gas usage would be expanded to provide for fu-

ture electricity generation. Figure 1 gives an example

choice set. The ﬁnal page of the questionnaire was con-

cerned with collecting standard socioeconomic informa-

tion about the participant. Information was requested

about location of household, number of children, em-

ployment in the energy sector, membership in a conser-

vation group, age, household income, education attain-

ment and amount of last electric bill.

Because of budgetary constraints, the design was se-

lected to estimate principal effects only. No secondary

cross-effects can be determined from the choice design

being used. The sample size requirements grow too rap-

idly when cross-effects are to be studied. The question-

naire and accompanying cover letter were then submitted

to a small pretest with regard to their clarity and useful-

ness of the information contained. Feedback from this

process lead to a revised and shortened version of the

cover letter, clariﬁcation of some terminology and chan-

ges in how the socio-economic information was re-

quested in the questionnaire. The questionnaire and other

survey materials can be found in Hanley et al. (2004).

4.2 Sample Selection

The sampling frame for this project was the Scottish gen-

eral public. Our sample population was randomly se-

lected from the list of registered voters in eight council

districts of Scotland. The districts are Aberdeenshire,

Highlands and Islands, Western Isles, Edinburgh, Glas-

gow, Stirling, Borders and Dumfries and Galloway. Ap-

proximately 250 names were from Glasgow and Edin-

burgh, 80 from Aberdeenshire and 30–45 names from

each of the other districts.

A mail survey was used due to constraints imposed by

project funding. Some 547 names were selected and

mailed survey packages with a cover letter during the ﬁrst

week of September 2003. As an incentive to participate a

£20 prize draw was offered. Three weeks later a follow-up

postcard was mailed to encourage the completion and re-

turn of the survey. By October 2003, 219 households had

returned surveys, a 43% response rate after undeliverables

are considered. Two hundred and eleven surveys were

received in time to be part of the sample set. Eight surveys

were returned too late to be included. Two hundred and

eighty-seven households did not respond. This response

rate is acceptable, and comparable to other studies (Ek,

2002; Hanley et al., 2001) that had response rates ranging

from 44% to 56%, for a survey mailed to the general

population. Mail surveys tend to have the lower response

rates than telephone or face-to-face interviews (Bateman et

al., 2002). The sample group of respondents should also be

tested for any self-selection bias, which could result in a

biased MNL estimates and WTP amounts.

5. Data Analysis

To model the information collected from the question-

Attribute

Description Levels

Landscape

The visual impact of a project is dependent on a

combination of

None, low, moderate,

high

impact

both the size and

location

Wildlife

impact Change in habitat can inﬂuence the amount and diversity of species Slight

improvement,

no impact, slight

harm

Air

pollution

Many types of renewable energy projects create no

additional air

pollution, but some projects do

burn non-fossil fuels. These

projects

produce a very small amount of pollution when compared

electricity generated from coal or

natural gas

None, slight

increase

Jobs

All renewable energy projects will create new local

long-term

employment to operate and

maintain the projects.

Temporary

employment increases during the

construction

phase are not

being

considered

1–3, 8–12, 20–25

Price Annual increase in household electric bill resulting from

expansion

of renewable energy projects.

An average household pays £270

year (£68 per quarter) for

electricity

£0, £7, £16, £29, £45

Alternate specific

constants

ASC-A Takes value of 1 for Plan A, 0 otherwise. Acts to

represent

variations that cannot be explained by

the

attributes

socio-

economic

variables

ASC-B Takes value of 1 for Plan B, 0 otherwise. Acts to

represent

variations that cannot be explained by

the

attributes

socio-

economic

variables

Valuing the Attributes of Renewable Energy Investments

Figure 1. Example choice set

naire, each choice set has three lines of code that com-

bines the attribute levels, ASCs and socio-economic

variables (Bennett and Blamey, 2001). The data matrix

appeared in the form:

alternative plan A :

aaattributessociecon

VASCX Y





 

alternative plan B :

bbattributessoci econ

VASCX Y





 

no renewables option : attributessoci econ

Vn XY





, (the

neither=opt-out plan)

where V is the

conditional

indirect utility,

ASCa,b are

the alternative speciﬁc constants for each choice

plan,

attributes

is a vector of coefﬁcients associated with

the

attributes

X and levels, and

socio-econ

is a vector

coefﬁcients associated with the socio-economics

descry-

tors Y of the

respond

ents.

NLOGIT 3.0/LIMDEP 8.0 econometric software was

used to estimate the MNL model. Attributes were effect

coded, rather than being coded using dummy variables, as

this will provide estimates that are uncorrelated to the in-

tercept of the model (Louviere et al., 2000). Effect coding

means that at least one level of each attribute is not in-

cluded as an identiﬁed variable: thus a three-level attribute

generates two variables. The excluded level is coded as

negative one. The attributes levels chosen for exclusion

were the ones hypothesised to have the most negative ef-

fect on environmental amenities. Therefore, the estimated

coefﬁcients for each of the remaining levels indicate the

value respondents placed on the change from the lowest

valued (omitted) level to the level of greater utility. The

omitted levels were: high landscape impact, slight wildlife

harm and slight increase in air pollution. The effect of

these omitted levels on utility is given by the negative of

the sum of the coefﬁcients on all the included levels.

5.1 Descriptive Statistics

Any mail survey has the risk of self-selection bias. Com-

paring the socio-economic information collected on the

211 respondents used in the CE against the statistical

proﬁle of the Scottish population is one test for such a

bias: the null hypothesis that the experiment population

is equal to the national population must be rejected for

bias to be suspected. In our sample, respondent’s income

and location of residence are different from the national

distribution at 10% level. Our sample is thus lower in-

come than the national average, and more rural. These

two descriptors are in fact correlated with each other.

Rejection of the null hypothesis means that the estimated

coefﬁcients and the calculated WTP values may not be

statistically valid representations of the whole Scottish

population.

5.2 Model estimation and results

Results for all 211 respondents from the MNL model are

shown in Table 2. The ‘‘simple’’ model shows results

when only the CE attributes are included in the regres-

sion. The coefﬁcients are interpreted as the parameters of

the indirect utility function, although the fact that they

are confounded with a scale parameter means that one

cannot directly interpret their numerical value (the scale

parameter cancels out when calculating implicit prices or

welfare measures). Coefﬁcient signs show the inﬂuence

of attributes on choice probabilities: here, all attribute

coefﬁcients have the expected signs. The signs of all but

the price attribute are positive, as consumer preference

theory predicts, since these attributes are coded to show

an increase in environmental quality, which should lead

to increased utility. Price is negative and therefore also in

accord with standard economic theory. All of the envi-

ronmental attributes are signiﬁcant determinants of utility

at some level: changes in air pollution, landscape effects

and wildlife effects. However, employment creation is

not a signiﬁcant attribute.

A series of socio-economic variables were proposed

for inclusion in an ‘‘expanded’’ model based on standard

consumer theory. The Student’s t-test and log-likelihood

tests were then used to determine acceptance or rejection

of each variable. The rejected descriptive variables were:

does the respondent have children, employment in the

energy sector, membership in a conservation group,

amount of last electric bill, age by ﬁve categories and

education by three categories. The covariates used in the

“expanded” model in Table 2 thus show either statistical

signiﬁcance or are included on theoretical grounds. Edu-

cation and age are in the former class, while income is

the latter case.

A likelihood ratio test was used to compare the “sim-

ple” and “expanded” models, and rejected the null hy-

pothesis that the parameter values of the two models are

equal at the 95% signiﬁcance level. Implicit prices de-

rived from the two models are not statistically different.

Simple visual examination of this is conﬁrmed by the

large overlap of the conﬁdence intervals (95% level) of

implicit prices of both models (the delta method was

used to calculate the conﬁdence intervals). The adjusted

Valuing the Attributes of Renewable Energy Investments49

Table 2. Multinomial model results

Model Model: expanded model w/covariates Simple model: attributes only

Descriptor Coefﬁcient Implicit

price (£)

(std.

error)

(95% conﬁdence

interval) Coefﬁcient Implicit

price (£)

(std.

error)

(95% conﬁdence

interval)

Moderate

Landscape

Low 0.29 5.58 (2.99) (0.28–11.44) 0.20 4.07 (2.99) (–1.79–9.93)

Landscape 0.15 2.82 (3.56) (–4.16–9.79) 0.16 3.21 (3.56) (–3.77–10.19)

None

Landscape 0.42* 8.10* (1.94) (4.30–11.90) 0.39* 7.88* (1.94) (4.08–11.68)

None

Wildlife 0.22** 4.24** (2.18) (–0.03–8.51) 0.27* 5.51* (2.18) (1.24–9.78)

Improved

Wildlife 0.63* 11.98* (1.88) (8.30–15.66) 0.50* 10.11* (1.88) (6.43–13.79)

None

Air pollution 0.74* 14.13* (1.88) (10.45–17.81) 0.71* 14.40* (1.88) (10.72–18.08)

Employment

Price

ASCA

0.02

–0.05*

2.80*

0.32 (0.22)

(–0.11–0.66)

(0.0065)

0.01

–0.05*

2.96*

0.23 (0.22) (–0.20–0.66)

(0.0058)

ASCB 2.73* 2.80*

IncomeA –0.01

IncomeB2

Higher

educationA

–0.01

0.99*

Higher

educationB2 0.85*

Under age 40-A 1.06**

Under age 40-B 0.88***

Log-likelihood –434 –509

No. of

observations 739 836

Pseudo-R2 0.31 0.29

*Signiﬁcant at 1% level, **signiﬁcant at 5% level and ***signiﬁcant at 10% level. Bold indicates monetary implicit

values.

Table 3. Implicit prices from the model with covariates

Landscape

impact Households are WTP £8.10 to decrease high impact landscape changes to having no landscape impact

Wildlife impact WTP of £4.24 to change a slight increase in harm to wildlife from renewable projects to a level that has no harm.

However, households would be WTP £11.98 per annum to change a slight increase in harm to wildlife from renewable pro-

jects to a level that wildlife is improved from the current level

Airpollution

impact

Households are WTP £14.13 to have renewable energy projects that have no increase in air pollution, compared to a pro-

gramme which results in a slight increase in pollution

McFadden pseudo-R2 is also improved with the addition

of the covariates. Louviere et al. (2000) state that a

McFadden statistic in the 0.20–0.30 range is comparable

to an ordinary least square (OLS) adjusted-R2 of 0.70–

0.90. Therefore, the expanded model with covariates is

deemed the superior model, and implicit prices from this

are used in the following discussion. Implicit prices

(“part-worths”) are interpreted as the incremental will-

ingness-to-pay through an increase in electricity charges

per annum per household for a change in any of the at-

tributes. They reveal the following (Table 3).

Looking closer at landscape impacts, moderate and

low landscape impacts were not statistically signiﬁcant

compared with a high impact. Respondents thus only

seem WTP to reduce high landscape impacts, but not to

reduce low or moderate impacts. One very interesting

ﬁnding is that employment effects are not statistically

signiﬁcant determinants of choices or of utility: respon-

dents as a whole did not seem to care about employment

effects to a signiﬁcant degree.

An internal validation question was included in the

questionnaire to test for consistency of these results. Re-

spondents were asked to indicate which single attribute

was most important to them. The ordering of the attrib-

utes by votes from respondents was: air pollution, wild-

life, electricity price, landscape and employment. This

shows consistency with the preference results shown in

Table 2. Also, there is inferred consistency of the indirect

utility measurement of individuals as the implicit prices

are in the same rank order. Consistency with preference

theory is also demonstrated by the estimated willing-

ness-to-pay increasing with increased improvement of

the qualitative attributes (for instance, with regard to

wildlife effects).

5.3 Heterogeneous Preferences

One important factor that may determine one’s attitudes

to renewable energy projects is where one lives, in par-

Valuing the Attributes of Renewable Energy Investments

ticular whether one lives in the countryside or not. A way

of testing this in our survey is to examine whether there

is a statistical difference between rural and urban esti-

mated MNL coefﬁcients and implicit prices. To do this,

the sample was partitioned according to place of resi-

dence as disclosed in the questionnaire. The sample

population was thus segregated into two groups, those

located in villages or the countryside and those who re-

side in towns and cities. Separate MNL models were then

run for each group (Table 4). A likelihood ratio test re-

jected the null hypothesis that the segregated subsets

were equal at the 5% level. Moderate landscape impacts

now register as signiﬁcant in the rural model, as do jobs.

Jobs remain insigniﬁcant in the urban sample, but be-

come strongly signiﬁcant in the rural model: this is per-

haps unsurprising given most peoples’ likely expecta-

tions about where jobs would be created. Note that the

McFadden pseudo-R2 for the rural subset has increased

to 0.34 from the 0.29 level with the complete sample.

Another reason why attitudes towards renewable en-

ergy investments might vary across people is their in-

come: either because environmental concern is a “lux-

ury” (Hokby and Soderqvist, 2003), or because rising

energy prices hit poorer households disproportionately

hard. To test this hypothesis, the sample was split by

annual household income level into two subsamples: low

income (£16,000 or less per year) and higher income

(greater than £16,000 per year). The log-likelihood ratio

test failed to reject the null hypothesis that the two sub-

sets were equivalent to the complete sample set: there are

no signiﬁcant differences in preferences therefore be-

tween these two income groups.

6. Welfare Analysis for Alternative

Investment Plans

One of the strengths of CEs is that estimated coefﬁcients

of the attributes maybe used to estimate the economic

value of different ways in which the attributes can be

combined. In the context of this paper, alternative re-

newable energy investments may be compared in terms

of the welfare changes that they are associated with. To

determine the change in economic surplus from possible

alternative projects in a multi-attribute MNL model, a

“utility difference” is calculated as

(1 /)(),

monetary ii

enonomics surplusVV



 (6)

where

i is the conditional indirect utility associated

alternative i. The superscript 1 is the base

ase-here

deﬁned

as an expansion of an existing fossil fuel

plant-and

superscript 2 is the alternative

ene

energy

case (Bennett and Blamey, 2001). Four

different

energy

project scenarios were

con

ide

ed:

 Large off-shore windmill farm: a 200 MW, 100 tur-

bines each at 80 m nacelle hub height, 6–10 km

from shore.

 Large on-shore windmill farm: 160 MW, 80 turbines

each at 80 m nacelle hub height.

 Moderate windmill farm: 50 MW, 30 turbines each at

60 m nacelle hub height.

 Biomass power plant: 25 MW, emissions stack height

up to 40 m, portions of building up to 30 m, fuelled by

energy crops.

Table 5 gives results for the welfare change of each in-

vestment scenario relative to the case, using Equation (6).

The above values can be interpreted as the price that

households are willing-to-pay (or would have to be of-

fered in compensation in case B) on an annual basis to

have renewable energy projects with the indicated attrib-

ute levels, rather than the base case expansion of fossil

fuel power. These Scottish households place the greatest

value on off-shore wind farms, with the major determi-

nant of the prospective welfare change being the absence

of landscape impacts. This result matches with prior pub-

lic opinion research in Scotland (Scottish Executive,

2003). The next most valued type of energy project is

biomass generation. The major determinant for this type

of project being highly valued is the amount of employ-

ment that is associated with plant operation and agricul-

tural production of the energy crops; and with the wild-

life beneﬁts associated with this expansion in biomass

production. Most signiﬁcant of the results presented in

Table 5 are the large and substantial negative value

placed on large on-shore wind farms by our sample

households in Scotland. The high level of landscape in-

trusion and the low amount of other beneﬁcial attributes

make the promotion and construction of large-scale

windmill farms a poor social welfare choice, other things

being equal. The negative monetary value indicates that

households would have to be compensated in excess of

£19 per year for the construction of this type of project, if

their utility is to be held constant.

It is important, however, to note several caveats. These

alternative projects do not produce the same amount of

electric power. As an example, two moderately sized

wind farms would have to be constructed to generate

similar quantities of electricity each year as a biomass

generating plant. Quantifying the cumulative effect of

numerous projects being constructed in a region is, how-

ever, beyond the scope of this paper.

7. Conclusions

Renewable energy offers a partial solution to the problem

of reducing greenhouse gas emissions whilst meeting

future energy needs. Yet different renewable energy pro-

jects can have varying external costs in terms of impacts

on the landscape, wildlife and air pollution. In addition,

strategies vary in their likely impacts on jobs and elec-

tricity prices. The choice experiment method used in this

paper enables these effects to be jointly evaluated in wel-

Valuing the Attributes of Renewable Energy Investments51

Table 4. Implicit prices of

attributes

comparing rural, urban and all

responden

Model: attributes only (standard error and 95% conﬁdence intervals)

Full sample set Rural subset Urban subset

Descriptor Implicit price (£) Implicit price (£) Implicit price (£)

Moderate

Landscape 4.07(2.99) 12.15**(6.3) 0.50(3.31)

(–1.79–9.93) (–0.196–24.5) (–5.99–6.98)

Low

Landscape 3.21 (3.56) –5.68 (7.09) 7.15 (4.03)

(–3.77–10.19) (–19.58–8.20) (–0.74–15.04)

None

Landscape 7.88* (1.94) 5.32 (3.32) 8.73* (2.41)

(4.08–11.68) (–1.18 to –11.83) (4.01–13.45)

None

Wildlife 5.51* (2.18) 6.18 (3.71) 4.43 (2.69)

(1.24–9.78) (–1.08–13.45) (–0.83–9.70)

Improved

Wildlife 10.11* (1.88) 15.23* (3.16) 7.62* (2.42)

(6.43–13.79) (9.04–21.49) (2.87–12.36)

None

Air pollution 14.40* (1.88) 19.08* (3.73) 11.77* (2.08)

(10.72–18.08) (11.77–26.39) (7.70–15.85)

Employment

0.23 (0.22) 1.08* (0.44) –0.19 (0.26)

Log-likelihood (–0.20–0.66) (0.20–1.95) (–0.69–0.32)

–509 –200 –290

No. of observations 836 349 475

Pseudo-R2 0.29 0.34 0.27

*Signiﬁcant at 1% level and **signiﬁcant at 5% level. Bold indicates monetary implicit values.

Table 5. Welfare change for alternative energy projects

Scenario

Base case

Fossil fuel power station

expansion

Large off-shore wind

farm

Large on-shore wind

farm

Small on-shore wind

farm

Biomass power plant

Attribute levels

Landscape Low None High Moderate Moderate

Wildlife None None None None Improve

Air pollution Increase None None None Increase

Employment +2 +5 +4 +1 +70

Welfare change

(£/hsld/yr) +£6.60 –£19.40 +£2.40 +£4.60

fare-consistent terms. This enables conclusions to be

drawn about the net social beneﬁts of different renew-

ables investment strategies.

Reviewing our main results, we found a substantial

sensitivity to the creation of projects that will have a high

impact on landscapes. Conversely, there seems to be no

sensitivity, or at least no positive mean willingness-to-

pay, to reduce landscape impacts if the projects are de-

signed to have moderate or low levels of landscape ef-

fects. Wildlife is highly valued by our sample group and

avoiding impacts on wildlife comes out as being as im-

portant as avoiding impacts on landscape. The implicit

price to maintain a neutral impact on wildlife is 75% of

the price households would pay to reduce landscape im-

pacts from high to none. Any project that creates the po-

tential to harm wildlife thus needs to have large offset-

ting beneﬁts. The converse of this is the growing of cop-

piced willow as biomass for use in energy production is

expected to create greater bio-diversity on farmland. Our

results show that such increases in wildlife attract a high

economic value. We have not included beneﬁts related to

the carbon sequestration function of biomass growth, but

this might be an important part of the overall case for

promoting biomass generation. Conversely, avoiding air

pollution from renewable energy investments was highly

valued by our respondents. This would add to the case

Valuing the Attributes of Renewable Energy Investments

against burning biomass for power.

Investing in renewable energy might well result, at

least over the short to medium term, in an increase in

electricity prices. Our results show that, unsurprisingly,

increases in prices reduce consumer utility, since the

coefﬁcient on price in all of our models is negative and

signiﬁcant. However, we do not ﬁnd in the survey sample

that income groups differ in their preferences towards

renewable energy. However, this study did not have a

sufﬁciently large sample to test for those households near

the ‘energy poverty’ level. This is an issue for further

research.

Turning to spatial issues, there are important differ-

ences between urban and rural responses in this choice

experiment. There is some evidence that accepting nega-

tive environmental impacts from the development of

projects (e.g. landscape impacts) is more acceptable to

the rural population: the rural sample show no willing-

ness-to-pay for reducing landscape impacts from high to

none. Conversely, rural people value wildlife beneﬁts

and reductions in air pollution more highly than their

urban cousins (this latter may be due to a perception that

biomass combustion was more likely in rural areas, i.e.,

close to the supply of such material). Finally, we found

that employment creation is a statistically and economi-

cally signiﬁcant attribute to the rural sample, but not to

the urban sample. Rural respondents would be will-

ing-to-pay an additional £1.08 per year from each hous-

ehold for each additional full time job created by the re-

newable projects.

8. Acknowledgements

We thank the Scottish Economic Policy Network for

funding the research on which this paper is based, and an

anonymous referee for many useful comments.

REFERENCES

[1] Alvarez Farizo, B., Hanley, N., 2002. Using conjoint

analysis to quantify public preferences over the environ-

mental impacts of wind farms. Energy Policy 30 (2),

107–116.

[2] Bateman, I.J., Carson, R.T., Day, B., Hanemann, M.,

Hanley, N., Hett, I., Jones-Lee, M., Loomes, G., Mourato,

S., Ozdemiroglu, E., Pearce, D., Sugden, R., Swanson, J.,

2002. Economic Valuation with Stated Preference Tech-

niques: A Manual. Edward Elgar, Cheltenham.

[3] Ben-Akiva, M.E., Lerman, S., 1985. Discrete Choice

Analysis: Theory and Application Travel Demand. MIT

Press, Cambridge, MA.

[4] Bennett, J., Blamey, R. (Eds.). 2001. The Choice Model-

ling Approach to Environmental Valuation. Edward Elgar,

Cheltenham, UK.

[5] British Wind Energy Association (BWEA), 2003. Avail-

able from/http://www.bwea.com/index.htmlS (Accessed 1

December 2003).

[6] de Vries, H.J., Roos, C.J., Beurskens, L.W.M., Kooi-

jman-van Dijk, A.L., Uyterlinde, M.A., 2003. Renewable

Electricity Policies in Europe: Country Fact Sheets 2003.

Energy Research Centre of the Netherlands, Amsterdam,

The Netherlands.

[7] Dewar, M., 2003. Report for the University of Glasgow

Renewable Energy Project: Synopsis of Focus Group

Responses. Focus Ecosse, New Lanark.

[8] DTI, 2003. Energy White Paper ‘‘Our Energy Future-

Creating a Low Carbon Economy’’. HMSO, London.

[9] Ek, K., 2002. Valuing the environmental impacts of wind

power: a choice experiment approach. Licentiate Thesis,

Lulea University of Technology, Sweden, p. 40. http://

epubl.luth.se/1402-1757/2002/40/index-en.html (Accessd

1 December 2003).

[10] European Marine Energy Centre (EMEC), 2004. http://

www.emec. org.uk/ (Accessed 1 August 2004).

[11] Hanley, N., MacMillan, D., Wright, R., Bullock, C.,

Simpson, I., Parrisson, D., Crabtree, B., 1998. Contingent

valuation versus choice experiments, estimating the

beneﬁts of environmentally sensitive areas of Scotland.

Journal of Agricultural Economics 49(1), 1–15.

[12] Hanley, N., Nevin, C., 1999. Appraising renewable en-

ergy developments in remote communities: the case of the

North Assynt Estate, Scotland. Energy Policy 27,

527–547.

[13] Hanley, N., Mourato, S., Wright, R., 2001. Choice mod-

elling approaches: a superior alternative for environ-

mental valuation? Journal of Economic Surveys 15 (3),

435–462.

[14] Hanley, N., Bergmann, A., Wright, R., 2004. Valuing the

Environmental and Employment Impacts of Renewable

Energy Investment in Scotland. Scotecon, Stirling.

[15] Hassan, Gerald and Partners, Ltd., 2001. Scotland’s Renew-

ables Resources 2001, Vols. 1 and 2. Research commis-

sioned by Scottish Executive. Document No. 2850/G R/03.

[16] Hokby, S., Soderqvist, T., 2003. Elasticities of demand

and willingness to pay for environmental services in

Sweden. Environmental and Resource Economics 26 (3),

361–383.

[17] Lancaster, K., 1966. Anew approach to consumer theory.

Journal of Political Economy 74, 132–157.

[18] Louviere, J.J., Hensher, D., Swait, J., 2000. Stated Choice

Methods: Analysis and Application. University Press,

Cambridge.

[19] Manski, C., 1977. The Structure of Random Utility Mod-

els. Theory and Decision 8, 229–254.

[20] McFadden, D., 1974. Conditional logit analysis of quali-

tative choice behaviour. In: Zarembka, P. (Ed.), Frontiers

in Econometrics. Academic Press, New York.

[21] Nuclear Industry Association (NIA), 2003. Available

from /http:// www.niauk.org/article_33.shtmlS (Accessed

1 December 2003).

[22] OXERA, 2002. A Report to the DTI and the DTLR: Re-

gional Renewable Energy Assessments. Oxera Environ-

Valuing the Attributes of Renewable Energy Investments 53

mental, Oxford, UK.

[23] Rose, S.K., et al., 2002. The private provision of public

goods: test of a provision point mechanism for funding

green power programs. Resource and Energy Economics

24, 131–155.

[24] Scottish Executive, 2001. Press Release SE 1472/2001,

http://www.scotland.gov.uk/news/2001/06/se1472.asp

(Accessed 1 August 2004).

[25] Scottish Executive, 2002a. Key Scottish Environmental

Statistics, 2002. A Scottish Executive National Statistics

Publication, Edinburgh.

[26] Scottish Executive, 2002b. Securing a Renewable Future:

Scotland’s Renewable Energy. HMSO, Edinburgh.

[27] Scottish Executive, 2003. Public Attitudes to Wind Farms:

A Survey of Local Residents in Scotland. HMSO, Edin-

burgh.

[28] The Renewables Obligation (Scotland) Order (ROS),

2002. Scottish Statutory Instrument 2002, No. 163.

HMSO, Edinburgh.