The Economics of Distributed Energy Generation: A Literature Review 1

Grant Allana, Igor Eromenkoa, Michelle Gilmartinb*, Ivana Kockarc and Peter McGregord

Paper published in Renewable and Sustainable Energy Reviews

Abstract
The UK electricity system is likely to face dramatic technical and institutional changes in the near future.
Current UK energy policy focuses on the need for a clean, affordable and secure energy supply. Decentralisation
of the electricity system is recognised as one means of achieving efficient and renewable energy provision, as
well as addressing concerns over ageing electricity infrastructure and capacity constraints. In this paper we
provide a critical literature review of the economics of increased penetration of distributed energy generation.
We find that there exists a large volume of research considering the financial viability of individual distributed
generation technologies (and we are necessarily selective in our review of these studies, given the wide variety
of technologies that the definition of distributed generation encompasses). However, there are few studies that
focus on the pure economics of individual or groups of distributed energy generators, and even fewer still based
on the economy-wide aspects of distributed generation. In view of this gap in the literature, we provide
suggestions for future research which are likely to be necessary in order adequately to inform public policy on
distributed generation and its role in the future of UK energy supply.

a
Fraser of Allander Institute, Department of Economics, University of Strathclyde
b
Division of Economics, University of Stirling
c
Department of Electronic and Electrical Engineering, University of Strathclyde
d
Department of Economics and Strathclyde International Public Policy Institute, University of Strathclyde

*Corresponding Author: Division of Economics, University of Stirling, Stirling, FK9 4LA. tel: +44 (0) 1786
467 478; fax: +44 (0)1786 467 469; email: michelle.gilmartin@stir.ac.uk.

JEL classification: H54; L94; P48; Q42; Q43

Keywords: Energy and the macroeconomy; public investment and capital stock; electric utilities; non renewable
resources and conservation; alternative energy sources.

1
The authors are very grateful to two anonymous referees for their comments and acknowledge the support
of the EPSRC’s Highly Distributed Energy Futures consortium.
1. Introduction and Overview

Driven by various technological advances, regulatory issues and emissions reduction policies,
the UK electricity supply framework, and its associated transmission and distribution
networks, has been undergoing significant change in recent years. The development of
renewable electricity generation technologies, the growth of competition in the electricity
industry, concerns over ageing infrastructure and capacity constraints have stimulated
increasing interest in the potential for distributed electricity generation to address such issues.
Distributed generation2 (DG) encompasses a broad range of typically (though not always)
‘low carbon’ or ‘efficient’ technologies which are small-scale in comparison to conventional
generation, and located closer to the end user. Such technologies may give rise to benefits in
terms of transmission and distribution savings, as well as their potential to remove the need
for costly infrastructure and capacity upgrades.

Moreover, the UK Government sets out three key priorities in its Energy Review: to reduce
greenhouse gas (GHG) emissions; to secure its future energy supply; and to reduce fuel
poverty (DTI, 2007a). Whilst some changes to the current UK energy system may lead to
trade-offs among these goals (such as the potential for high-cost renewable energy
installations to reduce emissions but increase fuel poverty), increased penetration of
distributed energy technologies may contribute towards the achievement of all three goals
simultaneously. There are potential emissions savings associated with the low carbon output
(on average) of DG technologies; whilst increased diversification in the range of the type of
energy supply technologies and resources associated with DG could mean reduced reliance
on energy imports and increased security of supply of UK energy3; and the ‘efficient’ nature
of DG technologies such as CHP, combined with possible savings relating to reduced system
transmission and distribution costs, could contribute towards lower-cost energy supply than

2
The term ‘distributed generation’ is commonly referred to in the literature, but the terms ‘dispersed’,
‘embedded’, ‘decentralised’ and ‘on-site’ are also used with reference to the same concept.
3
Though the link between reduced energy imports and increased security of supply of domestic energy is far
from clear. Stirling (1994) provides a useful discussion of the factors affecting energy security of supply.
that associated with conventional centralised generation. Whilst in its report the Government
acknowledges that the existing centralised system of energy production and delivery provide
‘economies of scale, safety and reliability’, it also states that a ‘combination of new and
existing technologies are making it possible to generate energy efficiently near to where we
use it, potentially delivering lower emissions, increased diversity of supply and in some cases
lower cost’. Thus DG has the potential to achieve a ‘triple dividend’ in terms of meeting
energy policy objectives4.

In this review we acknowledge the potential for distributed energy resources fundamentally
to alter the way in which UK energy requirements are met. Conventionally, the UK
electricity framework is characterised by large-scale, centralised electricity generation plants.
Electricity is delivered to a huge number of consumers located across a large area via a
complex transmission and distribution network. In the past, this system is widely understood
to have worked well, providing the advantage of economies of scale, reliable, secure and
relatively low-cost electricity to consumers. In contrast, DG technologies are located close to
the demand source. A greater number of smaller, modular energy generation devices are
required, each producing much smaller amounts of energy. DG systems can either be stand-
alone or grid-connected. In the former case the DG technology produces power
independently of the grid, and the operational capacity is matched to the demand. In the
latter, the main purpose is for the device to service the electricity needs in the local area. Any
surplus generation is fed into the grid, whilst any shortage of electricity is drawn from the
grid (see Figures 1a and 1b). In such a system, both demand and generation are directly
connected to the distribution network, close to the point of end use. Consequently, the
electricity losses and inefficiencies, which occur as centrally-generated electricity is
transported across the network, are potentially reduced, and the electricity supply system as a
whole is more flexible. Such developments may avoid (or certainly delay) the need for the
widely anticipated and costly investments in the existing centralised electricity network,
which would otherwise be required to address capacity constraints and ageing infrastructure.
Furthermore, the government’s Energy Review (DTI, 2007a), suggests that a ‘community-

4
The Scottish Government regards economic growth stimulated by the energy sector (with a particular
emphasis on renewable technologies) as a further ambition for Scottish energy policy. If DG stimulates
economic activity, there exists the potential for an additional dividend. (See e.g. Allan et al, 2008)
based energy system could lead to a greater awareness of energy issues, driving a change in
social attitudes and, in turn, [could lead to] more efficient use of our energy resources’.

Despite these potential theoretical benefits of distributed energy generation, there are also a
number of complexities and constraints involved in its further penetration into the energy
mix. The integration of distributed generation technologies within the existing network is
likely to create significant issues relating to the costs of energy provision and price of
electricity, power quality, infrastructure requirements, and technical performance. DG
necessitates a more active distribution network than that which currently exists in the UK. In
particular, there is a need for electricity to flow in two directions, both from the network to
the consumer for use at home or in industry, and also from the distributed generation source
to the network when exporting excess generation (Figures 1a and 1b). Furthermore, there are
considerable uncertainties regarding the financial viability of individual and wide-spread DG
applications, as well as the social costs and benefits attached to the increased penetration of
distributed generation in the UK, not to mention the macroeconomic effects of such a
fundamental change in energy provision.

Figure 1a: Conventional Electricity Distribution Network

Figure 1b: Electricity Distribution Network with Distributed Generation

Source: Ofgem, 2002

In this paper we provide a critical review of the literature on the economics of distributed
generation. This is with a view to informing the wider energy policy community and
identifying important informational and research gaps, as policy makers seek to make
decisions towards developing an efficient, secure and financially and environmentally viable
electricity network for future energy needs.

Evaluating the economics of the increased penetration of distributed energy generation is not
straightforward. Distributed energy technologies vary widely in terms of their technological
design and generation capacity, as do their capital, maintenance and fuel costs. For example,
there are the potential costs of electricity infrastructure adjustments that may be required in
order to make widespread use of distributed energy. These potential costs should sensibly be
compared to the alternatives, such as the network upgrades that would be necessary to
increase the capacity of conventional, centralised generation. Furthermore, there are
uncertainties regarding the characteristics and extent of future policy support mechanisms, as
well as the likely regulatory and institutional arrangements for distributed electricity
generation (for example generators’ obligations and costs for connecting to the grid). Data
on the financial costs and benefits of distributed energy generation tend to be highly project-
specific and estimates of the social costs or benefits of such generation (for example potential
reduction in carbon emissions), are necessarily assumption-driven and subject to
uncertainties. As such, there are no standard models or tools for analysing the economics of
distributed generation. In this paper we consider the findings from a range of research that
we believe to be informative about the key issues regarding the economics of distributed
generation. These include studies of: the financial viability of a number of individual
distributed energy generation systems; the social costs and benefits of distributed generation,
including environmental costs; and the wider macroeconomic impacts of increased
penetration of distributed energy generation.

The remainder of this paper proceeds as follows. In Section 2 we discuss a suitable definition
for DG, for which no precise consensus exists in the literature, and we also describe the key
features of some distributed energy technologies. In Section 3 we comment on the current
penetration of distributed energy in the UK and the current policy support mechanisms in the
UK relevant to DG, and we briefly compare this with other countries where DG has a greater
presence. In Section 4 we discuss some important aspects of estimating the cost of
distributed versus centralised generation. In Section 5 we review the literature on the
economics of distributed generation, focusing on: both the financial and wider social costs
and benefits of distributed energy systems; and the economy-wide impacts of DG in the UK.
In Section 6 we provide brief conclusions and identify opportunities for further research to
address the key issues raised in earlier sections.

2. Distributed Generation: A Definition

In general terms, DG refers to the use of stand-alone or grid-connected small, modular

electric generation devices which are located close to the point of consumption (Arthur D.
Little, 1999). The key defining characteristics of DG technologies include the size of the
power production of the technology and the location and application of the device. DG
systems are generally located close to the power demand, on the customer side of the meter
or on the distribution network, rather than on the transmission network5. The systems mostly
produce between 1kW and 5MW of power supply6 (Carley, 2009). Some systems include:
stand-alone rural or remote applications (for example where there are grid access constraints);
grid-connected devices for the purpose of exporting electricity to the grid; utility-owned
devices (for the purposes of improving power quality and reducing power losses in certain
areas); and combined heat and power (CHP) devices.

In terms of a precise definition of DG, there is no consensus in the literature. Dondi et al

(2002) and Chambers (2001) concur with the definition of Arthur D. Little (1999) in defining
DG as small-scale electric power generation that is located close to customer needs. Ofgem
defines DG as ‘electricity generation which is connected to the distribution network rather
than the high voltage transmission network’ (Ofgem, 2002). The questionnaire-based
international survey by CIRED (1999) found that for respondents in some countries, the
definition of DG is based on the voltage level of the system, whilst for respondents in other
countries the classification was based on whether the system was stand-alone or not.

5
The electricity ‘transmission network’ is used for the bulk transfer of high-voltage electrical energy from
generating stations (power plants) to power substations which are located close to areas of high demand. The
electricity ‘distribution network’ is used to transfer low-voltage electrical energy from power substations to
the end consumer via a local wiring system.
6
Larger DG applications can produce between 5 and 300 MW of power, though there remains disagreement
about whether such devices can truly be considered to be DG technologies (Ackermann et al., 2001).
Ackermann et al (2001) examine different characteristics of DG with a view to arriving at a
very specific definition of distributed power generation. In that paper, the authors suggest
that whether or not a system is considered to be DG depends on the location of the device
(the authors suggest that a DG device should be close to the distribution network or on the
customer side of the network); the type of service supplied by the system (active supply of
power is required for a device to be defined as distributed generation; reactive power supply
is not); and generation capacity (with micro distributed generation assumed to be less than
5kW; small to medium distributed generation to be greater than 5kW and less than 50MW;
and large distributed generation to be greater than 50MW and less than approximately
300MW).

DG therefore encompasses a broad range of technology devices, and can include both
renewable (e.g. solar photovoltaic; wind; biomass and marine) and non-renewable energy, as
well as ‘efficient’ technologies such as CHP. Table 1 provides some examples of low carbon
distributed energy technologies.
Table 1 Examples of Low Carbon Distributed Energy Generation Technologies

Technology Description
Distributed Heat Technologies
Solar water heating Uses the heat of the sun to produce hot water

Heat pumps Uses the warmth stored in the ground or air, via a
cycle similar to that used in refrigerators, to heat
water for space heating
Biomass Small-scale biomass installations from approx 10kW
to 2MW that provide space and water heating by
combustion of wood, energy crops or waste
Distributed Electricity Generation technologies
Solar Photovoltaics (PV) Panels, often roof-mounted, that generate electricity
from daylight
Wind Large wind turbines that convert wind energy directly
to electricity
Micro-wind (<100kW) Small wind turbines that generate electricity - can now
be roof-mounted as well as attached to tall masts
Micro-hydro Devices that capture the power of flowing water and
convert it to electricity
Biomass/waste Installations range from landfill gas generation to
large power-only facilities approaching 40MW
Combined Heat & Power Technologies
Biomass/waste Installations range from 100kW biomass CHP to
around 85MWth/20MWe
Micro-CHP and CHP up to 1MW Small devices, usually gas-fired, that produce
electricity and capture the waste heat produced as a
by-product. CHP used on this scale tends to be for
heat and power for a single house or on a community
or commercial scale (e.g. a housing estate or office
block)
CHP from 1MWe-10MWe CHP on this scale tends to be large community
projects or small industrial applications
CHP over 10MWe CHP on this scale tends to be large gas turbine
industrial applications that require a substantial heat
load on a continuous basis

Source: adapted from DTI, 2007b.

3. The Penetration of Distributed Generation Technologies in the UK Energy System

The current electricity system in the UK is dominated by conventional, centralised

generation, with energy supplied through a nationwide network. DG systems produced less
than 10% of total electricity supply in the UK in 2006 (DTI, 2007b). Other European
countries have a much greater share of distributed energy technologies contributing to the
overall electricity supply than in the UK: in Denmark more than 50% of generating capacity
is sourced from distributed energy – mostly wind and small-scale CHP technologies (Lund et
al, 2006; Sørensen et al, 2006); whilst in the Netherlands around 25% of electricity supply
comes from DG (Foote et al., 2005). In Sections 3.1 and 3.2 we briefly consider respectively,
possible barriers to the adoption of DG in the UK, and identify key policies that have been
deployed in an effort to overcome these and encourage wider deployment. The contents of
Section 3.1 and 3.2 are summarised in Table 2.

Table 2 Summary of potential barriers to DG and policies to stimulate DG development

Potential barriers to DG Selected Policies to stimulate Selected

references DG references

Institutional (e.g. grid e.g. Pepermans Tariffs/subsidies for e.g. Bergman et al

access, regulation, et al (2005), DTI DG technologies (2009), Watson et
licensing and planning) (2007b), Watson al (2008), Finney et
et al (2008). al (2012)

Uncertainty (e.g. e.g. Uyterlinde et

certainty of policy al (2002),
support, technology Balcombe et al
performance) (2013).

Consumer attitudes and e.g. Bergman et

social acceptance (e.g. al (2009), Devine-
inconvenience, value on Wright (2007),
housing asset) Sauter and
Watson (2007),
Balcombe et al
(2013)
3.1 Barriers to the Adoption of DG

Despite the existence of some policy support measures (see Section 3.2 below), there exist a
number of institutional barriers to the adoption of DG in electricity systems where large,
centralised generators dominate. Pepermans et al (2005) note that such issues include the
potential for discriminatory access to the grid, while Uyterlinde et al (2002) suggest issues
relating to uncertainty over future policy support and planning and installation constraints. A
joint report by the UK government and Ofgem (DTI, 2007b) identified four key barriers
considered to be important, all of which were judged to continue to apply, to varying degrees,
in the more recent assessment of Balcombe et al (2013). Firstly, DG technologies are
typically less commercially attractive than alternatives since they tend to have: higher capital
costs; longer payback periods; and the payments for exporting excess electricity to the grid
are inadequate. Balcombe et al (2013) identify that for some technologies, potential
“adopters” of DG systems, would accept payback periods of around ten years, while the
payback periods for current technologies combined with existing support mechanisms were
often considerably longer. Secondly, potential users cannot easily access information about
DG, and the incentives available are not easily understood. Thirdly, aspects of the electricity
industry structure in the UK make it difficult for small generators to connect and operate
within it. These could include the complex system of licensing applicable for the generation
and supply of electricity to the network. Such regulations – while enforcing system stability
and safety – are more costly for smaller generators7. Additionally, Watson et al (2008) note
that the fiscal system appeared “biased towards business investments in central power
stations” (p. 3100). These include the existing system of capital allowances in place for
businesses, but not private actors, and the operation of the settlement system favouring
centralised generation8. Finally, regulatory barriers exist in the form of the planning process,
inhibiting community developments and initiatives associated with new housing.

7
As DTI (2007b) notes, a system of exemptions does exist for those DG schemes with net capacity below a
specific minimum amount.
8
The impact of micro-DG technologies on the balancing market is specifically examined in Van der Veen and
De Vries (2009).
A further barrier to the wider adoption of DG is consumers’ apparent resistance. The
potential importance of what appears to be less than entirely rational consumer behaviour has
been emphasised by many, including Bergman et al (2009), Devine-Wright (2007), Keirstead
(2007) and Rogers (1995). For example, Bergman et al (2009) argue that people generally
view government or industry as being responsible for environmental change and expect them
to take the initiative, whereas it may be argued that significant behavioural change requires
action at the collective, social level. Attitudes towards adoption of innovations differ among
heterogeneous consumers, from ‘early adopters’ to ‘laggards’, and there is some evidence to
suggest that these attitudes differ systematically by age, income, class and political belief (see
for example, Balcombe et al (2013) for a review and Claudy et al (2010), Karytsas and
Theodoropouloul (2014) and Claudy et al (2011) for more recent evidence). To the extent
that this evidence is accepted, successful policy action would depend on more than simply
addressing the financial barriers to adoption9, though it may be that incentives of sufficient
scale can induce changes in attitudes. Balcombe et al (2013) additionally identify a potential
barrier to adoption from the anticipated impact of a technology on the property to which
domestic-scale technologies would be connected. They provide evidence that technologies
that most closely resembled known energy technologies would be more favourably regarded
by householders. While these barriers to the adoption of DG continue to exert an impact,
there have recently been important policy initiatives aimed at mitigating their effects (notably
in respect of reform of the planning process and the introduction of feed in tariffs (FiTs)),
which we now consider.

3.2 Policies to Encourage the Adoption of DG

The UK government has implemented a number of policies that serve to promote the
adoption of a range of DG technologies. The Renewables Obligation (RO) is the most
important UK policy instrument directed at (larger scale) renewables initiatives. Under the
RO scheme, operators of accredited renewable electricity facilities receive Renewables
Obligation Certificates (ROCs) for each MWh of electricity they produce. The introduction

9
Sauter and Watson (2007) provide an analysis of the social acceptance of DG, which they regard as a pre-
requisite for the adoption of DG, drawing on a range of surveys.
of ‘banding’ within the RO system (Renewables Obligation Order, 2009; Renewables
Obligation Amendment Order, 2010), is intended to provide additional incentives for
investment in emerging, and thus generally more expensive, renewable technologies, and this
has resulted in increased support for some DG technologies (see Table 3). Technologies are
presently grouped into five ‘bands’, with each band receiving multiples (or fractions) of
ROCs for their electricity generation. Among the technologies assumed to be ‘emerging’ and
in receipt of additional ROCs support are solar photovoltaics, some CHP applications, wave,
tidal, offshore wind and biomass generation. Each of these generation types are entitled to
two ROCs per MWh, compared to one ROC/MWh for onshore wind and hydro-electric
generation. This effectively lowers the cost to developers of some DG facilities in the UK10.
In the year 2006-7, one ROC was worth £49.28 (Ofgem, 2008) to an accredited renewable
electricity generator.

10
These ROC ‘bands’ are applicable across the UK. However, the Scottish Government has indicated its
intention to introduce significantly higher levels of ROC support to electricity generation from Wave and Tidal
sources in Scottish waters. It is proposed that Wave and Tidal technologies might receive, in total (i.e. including
the UK “banded” support) 5 ROCs and 3 ROCs for each MWh of electricity generated (Scottish Government,
2008).
Table 3. Bands, technologies and level of Renewables Obligation support

Source: Renewables Obligation Order 2009; Renewables Obligation (Amendment) Order

2010

A number of other policies impact, at least potentially, on private sector incentives to adopt
DG. These include: the Climate Change Levy (CCL); the Carbon-Emissions Reduction
Target (CERT); the Climate Change and Sustainable Energy Act (2006); The Low Carbon
Building Programme (LCBP); VAT relief for ‘energy savings’ items; income tax exemption
for revenue from microgeneration export; time-limited stamp duty exemptions applied to sale
of zero-carbon dwellings and new building regulations. We consider each of these very
briefly. The Climate Change Levy (CCL) was introduced in April 2001. It is effectively a tax
on the use of energy in industry, commerce and the public sector, with revenue recycling to
reduce employers’ NI costs. As stated at the Department of Energy and Climate Change
(DECC) website (www.decc.gov.uk), the aim of the CCL is to encourage businesses to
become more energy efficient and reduce their greenhouse gas emissions. Good quality CHP
systems are exempted from CCL.

The Carbon-Emissions Reduction Target (CERT) was adopted in 2008 for three years and
replaced the Energy Efficiency Commitment (EEC), which was in operation from 2001. The
CERT is an obligation for energy suppliers to reduce the CO2 emissions of their residential
customers. Hawkes and Leach (2008) stress that CERT includes both energy efficiency and
microgeneration measures. The Climate Change and Sustainable Energy Act (2006) aimed to
promote micro-generation and required the Secretary of the State to set one or more national
micro-generation targets. The Microgeneration Strategy was adopted in 2006, aiming to
promote easier access to ROCs and to motivate local authorities to be more proactive in
developing microgeneration through the use of planning policies. The strategy provides grant
support for residential adoption through the Low Carbon Building Programme (LCBP),
which we now consider.

The LCBP, reflecting government recognition that ROCs are insufficient to support
technologies which are smaller-scale and further from market, supports microgeneration
installation through direct grants and initially had £86m of grant funding for microgeneration
installations in homes, communities, public and private sector to 2009 (DTI, 2007b). Allen et
al (2008b) note that there are two phases of this programme. Phase 1 grants were available
for households and for public, non-for-profit and commercial organisations. Demand from
households was much higher than the programme anticipated and some adjustments had to be
made. Phase 2 made funds available for the installation of microgeneration units by public
sector and charitable bodies, but not for households and commercial companies. Specific
technologies are supported: solar PV, solar thermal, wind, ground source heat pumps, and
biomass. Also, purchase and installation of technologies is limited to a short-list of seven
suppliers. This later development was criticised for excluding a large number of suppliers
and installers.
Bergman et al (2009) regard VAT relief for ‘energy savings’ items as another important
policy directed at the development of microgeneration. This policy instrument was first used
in 1997 and cuts the VAT rate from the standard of 17.5% to 5%. Energy-efficient measures
include installation of wind turbines, solar PV, water turbines, micro-CHP and some other
technologies. Bergman et al (2009) also mention two other incentives for microgeneration:
income tax exemption for revenue from microgeneration export and a time-limited stamp
duty exemption applied to the sale of zero-carbon dwellings. The authors cast doubt that that
these two incentives will have any significant effect, since the costs of installing
microgeneration units considerably exceed any benefits received from these measures.
However, they do believe that new building regulations may have a significant impact.
Currently these regulations are applied only in England and Wales and require all buildings to
be built to a truly zero-carbon standard from 2016. This is potentially important since the
only way to achieve the status of zero-carbon dwelling is to adopt microgeneration within the
dwelling and export electricity to the grid.

Of course, broader energy policies also impact on DG. For example, the EU’s Emission
Trading Scheme (EU ETS), which establishes a carbon price for major covered industries in
the UK (including large scale generators), serves to improve the relative cost-competitiveness
of DG, though for a variety of reasons the price has fluctuated and has rarely reached levels
that many would regard as an appropriate long-run price of carbon. The Climate Change Bill
came into effect in 2007 and set a target for reducing UK carbon dioxide emissions by at least
26-32% by 2020 and at least 60% by 2050 (subsequently increased to 80% on the
recommendation of the Climate Change Committee) compared to 1990 levels. While targets
are legally binding, it is not yet clear what impact these targets will have on DG, or more
widely, since, for example, unlike the Monetary Policy Committee (MPC) the CCC has not
been delegated any policy instrument with which to achieve the targets (McGregor et al,
2012).

However, undoubtedly the most significant recent policy initiative in this area, targeted at
small-scale distributed energy systems, is the Feed-in Tariff (FiT) scheme (DECC, 2010),
which replaces the LCBP and RO for installations under 5MW. This policy was intended to
increase the installation of small-scale renewable and low carbon non-renewable generation
technologies. The scheme requires licensed electricity suppliers to pay a generation tariff to
small scale low-carbon generators for electricity generation11, and an export tariff when the
electricity is exported to the grid, for the operational lifetime of the device. By obliging
electricity suppliers to purchase renewable energy from suppliers at a favourable price, the
FiTs policy provides emerging renewable technologies an opportunity to compete in the
electricity market. The policy is intended to increase the uptake of small-scale low carbon
technologies, and thus many technologies falling under the DG band, by increasing their
costs effectiveness for households and communities.

The FiTs are scaled according to technology, and payments are scheduled to gradually fall
over time, so as to incentivise cost-cutting and efficiency measures in renewable electricity
industries. The idea behind the gradual tariff reductions is that as demand for small-scale
renewables devices grows, manufacturers can take advantage of economies of scale, price
reductions are passed on to the consumer and the industry becomes competitive on its own.
In some European countries, however, FiT rates have recently been cut more sharply than
planned due to the perceived success of the FiTs scheme. In Germany, for example, FiT
rates have been cut by up to 16% for solar installations (compared with planned gradual
reductions of 1-6.5%), in response to considerable growth in the solar heating sector and a
steady fall in the price of solar panels (see e-parliament, 2010; euractiv, 2010). Recent
empirical evidence on the success of the UK FiT scheme in encouraging significant new
small-scale renewable facilities appears to be overwhelmingly positive. Bush et al (2014)
describe the development of small-scale PV since FiTs were introduced in April 2010 as a
“revolution” (p. 86), with under 1MW of capacity installed prior to its introduction and
1.5GW by the end of May 2013. However, reservations have been expressed about the
inequitable impact of FiTs12. With the success of the policy, significant and faster
reductions in the tariff levels, in particular for PV technology have been introduced ahead of
schedule to ensure the anticipated costs of the policy were not exceeded (see for example,
Cherrington et al (2013) and Muhammad-Sukki et al (2013) for reviews of changes to the PV
FiT schedule). Additionally, Finney et al (2012) discuss the implications of the global

11
Regardless of whether the electricity generated is exported to the national grid.
12
See Jardine (2010) for an ex ante analysis of the likely impact of FiTs on PV and a critique of the impact on
equity, and Morri et al. (2010). We return to this in our discussion of social costs and benefits in Section 5
below.
economic downturn and funding for decentralised energy in the UK post-2010, and the
implications for the range of FiT, RO, LCBP and CERT schemes supporting DG
technologies.

4. Estimating the Cost of Distributed versus Centralised Energy Generation

A sustainable future power system for the UK will likely comprise a diverse portfolio of
generation techniques and plants, including both DG and conventional generation. In the
UK, large-scale centralised electricity generation has been associated with economies of
scale and high reliability, and the future penetration of DG systems will be determined by the
costs and benefits of DG vis a vis the current centralised setup. A distributed electricity
network does have the theoretical potential to offer cost reductions. However, since the
existing electricity framework has been designed to support the requirements of large-scale
conventional transmission-connected generation, some aspects of the system may prevent a
level playing field for the introduction of DG technologies. This could act as a barrier to
entry for DG, and preclude the development of an economically efficient electricity system.
However, it does seem inappropriate to assess DG simply on the basis of standalone cost
estimates, since it will inevitably form part of a generation portfolio of energy resources. In
this context, distributed generation may offer reduced overall risk for any given (levelised)
cost, though this will vary by technology: in particular if gas generation is involved the link
to fuel prices will not be entirely broken13.

By connecting electricity generators closer to the point of use, the extent of the infrastructure
needed to transport the electricity is much reduced, as are the costs associated with
transmission and distribution. Approximately 6.5% of generated electricity is lost as it is
transmitted and distributed to consumers (DTI, 2007b)14, representing significant potential for

13
See e.g. Awerbuch (2008) and Allan et al (2011) for applications of portfolio theory to renewable electricity
generation.
14
However, in some cases a distributed electricity system could necessitate reinforcements to the
transmission network: for example in Scotland, the output from DG may exceed local demand at times, and
excess supply may be exported to the grid on large scale.
savings with DG. In many cases, power generated from DG technologies does cost more
than that from conventional electricity networks, though some authors suggest that methods
for making truly cost reflective comparisons should explicitly take into account the use of the
transmission and distribution systems. The cost of electricity produced by a centralised
system is estimated to be around 2-3 p/kWh. This compares with a much higher value of
electricity of 4-10 p/kWh for DG technologies (see, for example Strbac et al. (2007) and
DECC (2011)). Since DG systems are located closer to the consumer than electricity from a
centralised system, however, DG electricity has a much lower requirement for the transport
services provided by the transmission and distribution networks, and thus avoids the costs
associated with their use (Watson et al, 2008). These ‘avoided costs’ however are not taken
into account when comparing the p/kWh of DG versus centrally-generated electricity. Strbac
et al (2007) argue that excluding such potential avoided costs results in non-cost reflective
network systems, and could lead to unnecessary network reinforcement and inefficient
integration of DG into the wider electricity system.

Ayres et al (2007) also suggest that centralised generation is not necessarily optimal. The
authors suggest that whilst the capital cost of installing a large electricity plant is around
$500-1500/kW, the ‘true’ capital cost can be much higher. In addition to the costs of the
central plant, there needs to be investment in associated transmission and distribution
capacity increases to accommodate the extra load, as well as to accommodate line losses and
to provide reserve margins. The authors suggest that this could drive up the ‘real’ capital cost
of a new central plant to over 5.5 times the assumed minimum capital costs of $500/kW, and
almost 3 times the assumed maximum of $1500/kW. In contrast, installing DG systems
involves no costs associated with transmission and distribution capacity or line losses, and
needs relatively small costs associated with distribution. Additionally, there are potential fuel
savings associated with utilising waste heat in decentralised CHP systems.

From an emissions perspective, many (though not all) DG technologies are associated with
lower carbon emissions than conventional technologies. Some technologies are renewable
(such as solar PV, biomass), and others bring about efficiency savings (such as CHP, via the
recycling of waste heat that is produced as a by-product of electricity generation). The true
benefit of the emissions savings is not represented in the market cost of DG technologies,
however. Although society as whole values emissions savings, the benefits of emissions
savings associated with DG technologies are not fully reflected in its price. For large
electricity generators, however, the European Union Emissions Trading Scheme (ETS) does
impose a cost of carbon on generators. Under the ETS, EU member states agree on national
emissions caps, and then allocate emissions allowances to industrial operators. Operators
may reassign or trade their allowances, treating it like a financial instrument. This results in
an incentive towards low carbon distributed energy for large suppliers. However, in some
cases emissions caps have been insufficiently tight to bring about a reduction in emissions
(Climate Change Committee, 2008), resulting in a carbon price that is ‘too low’, reflecting
the difficulties inherent in estimating the true cost of carbon emissions.

Additionally, distributed generation may lead to potential cost reductions in terms of the
postponement of required investments and upgrades associated with the infrastructure and
plants of centralised generation (Hoff, 1996; Hoff et al., 1996). Ayers et al (2007) argue that
economies of scale associated with large centralised plants are coming to an end due to
capacity constraints, whilst small and renewable generators are benefiting from rapid
technological advancements. Furthermore, the authors suggest that real fuel price rises are
likely to continue, in line with the long-term decline in the discovery rate of fossil fuels.
Additionally, it is well known that new technologies generally begin with high unit costs,
which tend to fall with cumulative capacity installed. The incorporation of learning curve
effects, though of course subject to uncertainty, may prove significant (Gross et al, 2013).
These factors may increase the competitiveness of distributed energy systems compared with
centralised generation15.

5. Review of the Literature on the Economics of Distributed Generation

In this section, we review the literature on the economics of distributed energy generation. In
doing so, we focus on research in a number of key areas: calculating the financial viability of
individual or groups of DG plants; estimating the social cost and benefit of DG technologies,
including environmental benefits; and modelling the macroeconomic impact of an increase in
the uptake of distributed generation technologies.

15
Other analyses of the possible impact of DG on the electricity system include: Willis (2000), Energy Saving
Trust, E-Connect and Element Energy (2005), Infield and Li (2008), Kelly et al (2008), Pudjianto et al (2008),
Alacron-Rodriguez et al (2009) and Haesen et al (2009).
5.1 The Financial Viability of Individual DG Systems

The vast majority of the literature concerned with modelling the impact of DG focuses on
microeconomic analyses: a wide variety of studies assess the financial feasibility of
individual grid connected or stand-alone DG systems. However, because of the heterogeneity
of DG schemes, there is no generalised methodology in the literature for assessing the
financial or economic viability of decentralised energy systems. Further, we find a varied
range of results across the literature, suggesting that the outcome of the financial calculations
is highly project-, location-, and technology-specific. In assessing financial/commercial
feasibility, most authors make use of standard investment appraisal methods such as Net
Present Value (NPV) and Pay Back Period (PBP)16. These analyses focus virtually
exclusively on the private costs and benefits associated with DG, that is those cost and
benefits that accrue directly to the investor, and so conventional discounted cash flow
methods are appropriate. In this section we summarise a sample of financial viability studies
of DG technologies; whilst in Section 5.2 we review some studies that focus on the wider
economic – as opposed to purely commercial - viability of DG plants, by also incorporating
the social costs and benefits associated with DG. These are the costs and benefits that are
generated by DG but not borne by the private investors. (For example, social costs could
increase by adopting DG if there are any local disamenity effects, while any reductions in
CO2 emissions would reduce social costs.)

Stand-alone DG technologies are particularly well-suited to remote or inaccessible

geographic locations and developing countries, and individual assessments of the viability of
DG plants in the literature reflect the impact of these characteristics. Such stand-alone DG
systems are often financially viable in remote areas compared with conventional systems
because of the high costs of setting up conventional systems including connectivity.
Accordingly, there are many studies showing the economic efficiency of stand-alone DG
systems, in particular PV systems for remote applications, and for developing countries,
where the cost of other alternatives, such as extending utility power lines or transporting fuel,
are very high. Kolhe et al (2002) compare the economic efficiency of a stand-alone solar PV

16
The NPV of a project is the discounted value of the investment’s income, minus the discounted value of its
outflows. An investment should have a NPV of greater than zero to be financially viable. The PBP measures
the length of time taken to recover the initial investment for a project, ignoring the time value of money. NPV
is therefore, in general, a more appropriate method of investment appraisal.
system with a conventional diesel-powered system employing life-cycle costs comparisons
using parameters for India. The authors find that the solar PV system is competitive up to an
output of 68kWh per day. Their sensitivity analysis showed that even under an unfavourable
economic scenario, the solar PV system still performs better for an energy output of up to
15kWh per day. Bernal-Augustin and Dufo-Lopez (2006) conduct a financial analysis of grid
connected photovoltaic (PV) systems in Spain. The authors consider the profitability of the
system by assessing the NPV and PBP of the system. The results clearly suggest that the
project is financially viable, but with very long pay back periods, suggesting a potentially
important disincentive for investors, given that capital markets may be imperfect. Kaldellis
(2003) conducts a feasibility assessment of a number of widely dispersed wind energy
systems at various locations across Greece, and calculates the PBP and benefit to cost ratio of
all wind power installations. The author finds that, in general, the wind power program leads
to substantial financial losses, due to the low energy production of most of the wind power
plants, as well as technological failures. Bakos and Tsagas (2003) calculate the PBP of a
hybrid solar/wind installation that is designed to provide thermal and electric power to urban
residences in Greece. The authors find that the system is associated with a twelve-year PBP,
but note that the PBP could be reduced if the externalities of conventional power generation
technologies were to be taken into account.

Audenaert et al (2010) conduct a financial evaluation of photovoltaic grid connected system

(PVGCS) for firms located in Belgium, to consider whether such investments represent a
good financial decision for firms in Belgium. The authors calculate key financial criteria
such as NPV and PBP, taking the tax deductions applicable to PV in Belgium into
consideration. The authors also conduct sensitivity analysis to determine the key factors
influencing the financial variables. The costs include those for components, O&M, financing
and insurance. The revenue stream includes saved energy expenses, and tax deductions and
subsidies associated with green energy installations. However, for the example studied, the
investment yields a negative NPV and that the payback period is long (between 8.3-10.2
years). Kahn and Iqbal (2005) consider potential remedies to the problems of stand-alone
decentralised energy systems which sometimes make them non-viable options - such as low
capacity factors and excess battery costs. The authors suggest that stand-alone systems could
be used as a hybrid with other sources of energy carrier (both renewable and non-renewable)
so as to increase their cost effectiveness. They use HOMER software to find the optimal
combination of energy technologies in Newfoundland. The results suggest that some hybrid
systems (such as a wind-diesel-battery hybrid system) are commercially feasible, but that
other systems (such as an environmentally friendly hydrogen-based hybrid system) are too
costly to be commercially viable17.

Cherrington et al (2013) show how recent changes to the FiT tariff for domestic-scale PV
systems in the UK have changed the private viability of such technologies. They find that
even major reductions in the applicable FiT rate from 43.3p/kWh to 21.0 p/kWh or 16.0
p/kWh delays the PBP by two years (12 years, rather than 10 years). While the return on
investment falls, it remains “healthy” (Cherrington et al, 2013, p. 421) at 7%. Muhammad-
Sukki et al (2013) show how the payback period for solar PV projects differs across Europe
with tariff levels, and that reductions in tariffs could have implications for the future
deployment of such technologies.

5.2 The Social Costs and Benefits of DG Penetration

In practice, however, there are additional costs and benefits associated with the use of DG
technologies over and above the financial costs and revenues. In particular, the potential
environmental benefit of distributed generation systems is one of the main drivers of the
current enthusiasm for DG. DG can play a role in helping environmental obligations be met
in two key ways. Firstly, CHP applications (associated with DG technologies such as fuel
cells, gas turbines and microturbines) allow for emissions savings and optimal energy
consumption for firms or communities where there is a simultaneous demand for heat and
electricity. Secondly, most renewable energy technologies (with the exception of large hydro
stations) are decentralised because of their nature. Various studies attempt to capture such
externalities (social costs and benefits) associated with DG energy systems.

There are a number of techniques that can be used for appraising and valuing the social costs
and benefits associated with distributed energy generation. Life Cycle Assessment (LCA) is
an important method of assessing the environmental impact of a technology over its entire
life. In a LCA, all energy and materials use, including waste or pollutants associated with an

17
Other relevant studies include: Hawkes et al (2009a,b) who formulate and apply a techno-economic
modelling framework to study fuel cell micro-CHP system design and control; and Carley (2009) who develops
an econometric model of utilities’ decisions on whether to adopt DG and, if so, how much capacity to deploy.
activity or products, are quantified during the full life cycle of that activity or product.
Consequently, the contribution of a product towards a predefined environmental impact over
its lifetime is calculated18. All environmental effects associated with an activity are
computed, including geographically diverse effects, such as material inputs that are
imported19. However, a LCA, ‘only’ identifies the environmental impact – though this is
often a huge undertaking if genuinely comprehensive. Cost Benefit Analysis (CBA) is yet
more ambitious in that it seeks to assign monetary values to all of the costs and benefits of a
project, even where no market price exists to facilitate valuation (which is, of course, a major
challenge). CBA is the public policy counterpart to private sector investment appraisal
methods. In principle, LCAs could constitute one of the inputs into an overall CBA, though a
monetary value would have to be attached to the environmental impacts, and it would then be
included along with all other costs and benefits into the overall welfare assessment20.

Chakrabati and Chakrabati (2002) consider an existing stand-alone solar PV system for the
electrification of a remote area in India. In addition to illustrating that diseconomies of scale
are associated with conventional power generation for such a remote application, the authors
also demonstrate the social viability of such a system via an observed improvement in
education, trade, commerce and increased participation of women in non-household
activities, though these results are based on (ex ante and ex post) frequency distributions
collated from household samples, rather than by estimating the value of such social impacts
via, for example, CBA21. The authors also note the zero emissions costs from the solar
system, and quantify comparative emissions costs associated with various fossil fuel
alternatives. Ravindranath et al (2006) consider the carbon abatement opportunities
associated with substituting bioenergy technologies (BETs) for centralised fossil fuel energy
systems in India. They compare the costs per tonne of carbon abatement of ten BET projects

18
The LCA methodology is standardised according to the Society of Environmental Toxicology and Chemistry
guidelines: ISO standards 14040 and 14044.
19
See e.g. Allen et al (2008a), McManus et al (1999), Molloy (2003), Hammond and Winnett (2006) and Bush
et al (2014) for an exposition and critique. Among the problems are: the data requirements of LCA; the typical
neglect of local environmental impacts and the reluctance of private companies to share environmental data
that they may regard as sensitive.
20
See Pearce (1998) and Ackerman (2008) for an exposition and critique.
21
So that the ceteris paribus assumption is important here.
with that of conventional alternatives, and find that six of the BET projects represent more
cost-effective carbon mitigation opportunities compared with the conventional generation.

Walker (2008) assesses the link between distributed energy systems and fuel poverty in the
UK. The author notes that the wide range of potential DG technologies, as well as issues
such as the operation, ownership, installation, network requirements and maintenance of the
individual systems will determine the effect on fuel poverty. The author suggests that
increased penetration of distributed energy systems could improve energy affordability for
low income households in the UK. However, he notes that the upfront costs required to over
the installation of residential distributed energy generation systems are an important barrier
for low income households. Existing evidence suggests that the early adopters of
microgeneration systems are higher income groups (Caird and Roy, 2007). If policies
designed to encourage DG are focused on households installing and paying for small-scale
generation technologies (as is the case for the current FIT scheme, and the Low Carbon
Buildings Programme22), Walker (2008) argues that there is the risk of middle classes
actively investing in such technologies, whilst the low-income groups rely on traditional
electricity and gas supplies. If technology advances mean that investors in microgeneration
benefit from falling energy costs, then the problem of fuel poverty will be exacerbated.
Walker (2008) instead argues that national and local governments, housing associations
and/or energy providers should actively pursue the provision of microgeneration technologies
in alternative ways, for example via fuel poverty programmes that provide grant funding for
low-income groups.

Gulli (2006) implements a social cost-benefit analysis of the decentralisation of energy

supply, focusing on both residential and service sector applications (CHP installations in both
a residential building and hospital). In doing so, the author calculates both the financial costs
and benefits (including the price per unit consumed of energy from a centralised versus a
decentralised system), as well as wider social costs and benefits (including estimates of the
cost of energy-related externalities such as pollution emissions). This exercise is conducted
for different hypothetical DG systems in Italy, compared with comparable conventional

22
A capital subsidy policy for low carbon generation installations; under this scheme households still pay a
minimum of 70% of the cost of the installation costs.
generation techniques. The author finds that, in terms of private costs, DG systems are, in
general, uncompetitive both in the residential and service sector applications. To calculate
the social costs and benefits, Gulli (2006) takes into account effects such as the impact on
public health, agriculture, the ecosystem in general and pollution emissions. The value of
such effects are calculated in monetary terms, measured via a ‘willingness to pay or accept’
measure. The results suggest that even when such social externalities are taken into account,
the DG project, in general, is still more costly than the conventional centralised systems (with
the exception of two cases: a gas engine CHP system in Palermo (for the hospital case), and a
gas turbine CHP system in Milan (also for the hospital case). The author finds that the
efficiency benefits associated with CHP use and the avoidance of transmission costs are not
sufficient to compensate for higher investment costs of the DG applications. However, the
author notes that the methodologies used to evaluate the external costs are imperfect, and also
that technological developments in the efficiency of dispersed energy systems (with specific
reference to the development of fuel cells) could increase DG performance. Additionally, in
this paper the author attaches a higher environmental impact coefficient to emissions that
occur in urban areas than to emissions that occur in non urban areas. The DG systems are
located in urban areas, necessarily close to demand, and therefore, in relative terms, the CHP
systems are associated with higher emission costs. The appropriateness of such an
assumption should perhaps be considered, and sensitivity analysis around this assumption
would be informative.

Hawkes and Leach (2008) consider the environmental impact of energy use in the residential
sector for three different types of micro-CHP installations in the UK, and five different types
of residential dwelling, as well as three different electricity demand values. The authors
calculate the ‘equivalent annual cost’ and CO2 emissions using the CODEGen model, a
generalised model of heat and power provision that minimises the present value lifetime cost
of meeting a given energy demand. The authors find that the micro-CHP system can reduce
CO2 emissions by between 10-20% of current CO2 emissions for the residential sector, and
generate annual cost savings of between approximately £100-£500 per tonne of CO2. The
authors also note that the cost of the CO2 savings are, in the majority of cases, such that
micro-CHP can be an economically efficient instrument for reducing carbon emissions (given
the carbon price in the EU ETS).
In Allen et al (2008a) the authors use LCA to evaluate the environmental impact associated
with the installation and operation of a micro-wind turbine for domestic electricity
generation. The authors collect data for the manufacture of the micro-wind generator, which
includes: materials and components; transportation of materials and components to the
turbine factory; transportation of assembled turbine to the customer; and materials and
production of a mounting station for the turbine23. During the lifetime operation of the
turbine, the energy produced by the device is assumed to offset the conventional energy that
would otherwise be required to be obtained from the centralised UK electricity system. The
authors assume a specific ‘energy mix’ that is typical of the UK grid 24. They consider a
number of ‘wind condition’ scenarios, and find that, over the life cycle of the turbine, the
device has a positive environmental impact for all scenarios except for the poorest of wind
conditions. Aside from the available wind resource, the authors find that the geographical
positioning of the turbine and the use of recycled materials or not for the manufacture of the
device are particularly important factors in determining the environmental impact and overall
environmental benefits of the device.

In Allen et al (2008b) the authors conduct a more comprehensive appraisal of the

environmental impacts of distributed generation technologies. The authors use both a LCA
and CBA to evaluate the environmental performance of three microgenerators: a micro-wind
turbine; a solar photovoltaic array; and a solar hot water system. For the LCA, the authors
find that all three devices are associated with positive environmental benefits, but that the use
of aluminium as an input mitigates the environmental benefits for the micro-wind turbine and
the solar hot water system. For the CBA, the authors identify the financial costs and benefits
associated with the devices, and also incorporate a quantification of the present value of the
environmental externalities associated with the use of the devices, and determine the net
benefits of the projects. The authors find that none of the devices are commercially viable
and that, even when avoided environmental externalities are included, the overall cost of the
devices still outweighs the benefits. The authors suggest that although the micro-generators
are not currently competitive, future technological changes, operational efficiencies, and the

23
There is heating and lighting energy use associated with the assembly period in the factory; though the
authors omit this factor due to a lack of data, and note that this aspect is likely to make little overall impact.
24
Though the authors note that if the grid ‘energy mix’ were to change - for example via an increase in the
non-fossil fuel energy share of the mix - then this could have an important impact for such analyses.
use of, for example, recycled aluminium in the production of devices, could significantly alter
the results of such studies25.

Bush et al (2014) undertake an LCA assessment of micro-wind and solar PV technologies in

the UK. In their hybrid input-output approach, they combine technological data (i.e. a
“bottom-up” assessment) with an input-output system of interactions between production
sectors, to account for upstream carbon emissions in the production process. Further, they are
able to then demonstrate the importance of wind speed and irradiance on the carbon payback
period for both technologies.

5.3 The Overall Economic, Environmental and Social Impacts of DG Penetration

We next consider the overall/ system-wide/ macro-economic, environmental and social

impacts of DG penetration. Here our concern is with the likely system-wide26 consequences
of significant penetrations of DG. Clearly the intention of policy-makers in introducing
measures to encourage the adoption of DG is that this will assist in the achievement of some
or all of their energy policy goals. To determine whether this is so we need to assess the
likely impact of significant DG penetration on economic activity as a whole. This is likely to
be of interest in its own right (since economic activity is typically one of the wider goals of
government policy, and in Scotland – for example - is one of the goals of energy policy per
se), but also because this is a key determinant of the level of GHG emissions. Furthermore,
we are also likely to be concerned with the sectoral composition of any changes in economic
activity because we know that energy demands (and the emissions intensity of output) vary
significantly across sectors. To assess the system-wide environmental impacts of significant
DG penetration we have to understand its effect on sectoral and aggregate economic activity
In addition, such changes invariably have uneven impacts across household groups, and if we
wish to track effects on fuel poverty, for example, we again need to adopt a system-wide

25
Other relevant studies include: Mendez et al (2006), Pehnt (2007), Strbac et al (2007), Newborough (2004)
and Vogel (2009).
26
Here ‘system-wide’ refers to the economy as a whole, not the electricity system. (However, for the latter
see, for example, Viral and Khatod (2012) and Tan et al (2013)).
perspective. We consider the system-wide economic, environmental and social impacts of
significant DG penetration in turn.

The system-wide impacts on aggregate and sectoral economic activity

Macroeconomic analyses of DG are few, possibly reflecting a number of difficulties and

uncertainties associated with modelling the economy-wide impacts of increasing penetration
of DG technologies. DG encompasses a wide variety of generation technologies, and the
costs associated with DG differ widely according to technology type, geographic location,
infrastructure requirements and so forth. Additionally, there are many unknowns regarding
feasible penetration scenarios for DG; policy support mechanisms, and potential cost
requirements for network adjustments, which could be important factors to consider in a
macroeconomic analysis of DG.

There have been some attempts to quantify the costs (including the social cost of emissions)
associated with the UK moving towards an electricity system that is based on DG as an
alternative to centralised generation, so these studies go well beyond cost benefit analyses of
individual distributed energy plants. BERR and WADE (2007) quantify the costs and
benefits (including the social cost of carbon emissions) of using a decentralised electricity
generation system (that encompasses a range of different technologies) to meet electricity
demand needs for the whole of the UK for the next 20 years, and compare this with the
relative costs and benefits of equivalent centralised generation. They compare alternative
‘scenarios’ (exogenously determined “bundles” of electricity generating technologies) and a
wide range of model input assumptions by the user for a DG compared with a CG system
(regarding, for example: transmission and distribution infrastructure costs; electricity output
losses associated with transmission and distribution; fuel use; electricity demand growth over
time). Given the complexities of modelling DG, the exercise is necessarily rather
assumption-driven. As a result, there are significant uncertainties associated with the ‘rule of
thumb’-type assumptions made in the study. For example, the authors note that the WADE
framework incorporates a single cost to reflect the cost of infrastructure updates required for
transmission and distribution under each of the DG and CG scenarios. In practice, however,
such costs vary significantly from project to project. Furthermore, the model adopts a very
simplistic treatment of CHP: although the generation sector is explicitly modelled in WADE,
heat is not explicitly identified and so instead the authors attempt to quantify the benefits of
better fuel efficiency associated with CHP simplistically.

The WADE model is used to calculate the costs associated with generating a given quantity
of electrical output for different technologies. For CHP, the costs are associated not only
with the production of electricity, but also heat generation – an efficiency benefit associated
with CHP. Appropriate measurement of the efficiency benefit of CHP requires a comparison
of the costs of CHP generation with that of the separate centralised generation of (power
station) electricity and (gas boiler) heat. However, since heat is not explicitly modelled in
WADE, this is not possible. In order to capture the efficiency benefit of CHP in the WADE
model, the input electrical efficiency data is increased as an approximation for the heat
benefit associated with CHP, taking account of different sizes of CHP plant and their
associated efficiencies. This an important limitation of the WADE model (and is explicitly
recognised by the modellers): CHP is likely to play a crucial role in a future DG scenario (in
this case the modellers assume over 50% of new capacity is accounted for by CHP
technologies), so the simplified treatment of CHP may well have important implications for
model results.

In a study commissioned by the DTI, Cambridge Econometrics (2003), the authors use an
energy-economy-environment framework that is designed to model the growth of CHP
capacity in the UK to 2010, and the resulting impacts on energy demand and environmental
emissions. They use a macro-econometric model that is based on a set of input-output (IO)
coefficients which are updated with a series of econometric time series relationships.
Embedded within the main model is a series of sub-models (energy, electricity supply and
environmental emissions model). These sub-models update specific prices and demands,
which then feed back into the main model and are used to update the IO data. Further, the
electricity supply and energy sub models are integrated with a CHP sub-model, designed to
allow examination of the factors that are important in influencing CHP installation decisions.
Operation of the model requires a large number of assumptions concerning the future energy
and economic environment. These assumptions relate to macroeconomic conditions over
time (including forecasts of the economic growth rate, domestic and trading partners’
inflation rates, exchange rates, interest rates, domestic tax rates, and government
expenditures); energy prices (including forecast of electricity and gas prices); and the extent
and characteristics of policy support mechanisms for the energy industry. For the CHP sub-
model, the authors also input statistics on the use of CHP in the UK (using historical data on
the uptake of CHP); cost estimates of CHP installations; and the avoided costs of alternative
energy sources. The simulation methodology of estimating the uptake of CHP involves
inputting the exogenous assumptions and data sources described above to the main and sub-
models, with the model then calculating the altered investment decisions. The historical data
on the uptake of CHP is used to estimate how the share of the associated CHP will change.
This is done by separating the individual demand for CHP into sub-sectors, calculating a
NPV for each scheme, and then allowing the model to select a proportion of the plants which
are financially feasible according to their NPV. The main output of the MDM-E3 model is
the technological capacity, rather than the impact on wider economic variables

In their simulations, the authors consider the contribution of selected policy support
mechanisms in affecting the growth rate of CHP uptake in the UK, and, given the significant
degree of uncertainty surrounding the input data and assumptions, the authors conduct
significant sensitivity analysis around fuel price and other modelling assumptions. The
extent of the sensitivity analysis means that there are a wide range of possible CHP capacity
outcomes. The authors find that fuel price assumptions, in particular, have substantial effects
on results, in line with intuition: for example, the results suggest that under a ‘more
favourable’ price scenario for CHP (where electricity prices are assumed to be 40% higher
and gas prices 40% lower than baseline assumptions), a higher CHP capacity is installed
(13.6GWe by 2010), whilst for a ‘less favourable’ price scenario for CHP (where electricity
prices are lower and gas prices are higher than in the baseline), there is downward pressure
on installed CHP capacity, with build estimates of 7.3GWe by 2010. The authors also
estimate the importance of key support mechanisms in influencing the installation of CHP
capacity, and find that exemption from the Climate Change Levy (CCL) for certain CHP
power exports has a particularly strong impact on CHP capacity installed by 2010. The
authors also find that CO2 emissions fall as a result of the CHP installations, due to CHP
being more efficient and using fuels with a much lower carbon content, on average, compared
with conventional generation. While the analysis is indicative, the results are, of course,
heavily dependent upon the specific assumptions made, although the sensitivity analysis is a
recognition of this.

There are demanding informational requirements related to the DG sector associated with
carrying out detailed macroeconomic analyses of its impact. One important requirement in
estimating the macroeconomic impacts of DG is reliable and realistic estimates of the extent
of DG generation. Government targets for the uptake of renewable energy provide some
indication of the potential penetration of DG. Furthermore, there are government and other
organisations’ forecasts/ projections at a national level of the likely system-wide uptake of
DG technologies. However, in practice, it is likely that one or two DG technologies become
dominant, and such projections do not provide information on how the overall level of
penetration is shared across technologies. Due to the heterogeneous nature of distributed
energy technologies - each technology differs greatly in terms of their financial costs (thereby
influencing overall expenditures, for example, and corresponding macroeconomic impacts),
as well as their emissions performance, for example - more detail is required not only on the
likely overall DG capacity, but also on its likely composition.

In Foote et al (2005), the authors develop penetration scenarios for low-voltage DG that
provide an indication of not only the overall system-wide potential of (low-voltage) DG
uptake in selected European countries (including the UK), but also an indication of the
technologies that are most likely to be prevalent. The authors collate government and other
organisations’ forecasts of the uptake of DG for each country, and also collect survey data
from industry experts to indicate which of the DG technologies are likely to be most
prevalent, ranking the DG technologies with respect to which are most likely to be installed
in 2010 and 2020. For the UK, the authors suggest a DG capacity of 8.24-17.30% as a
percentage of total capacity in 2010, with photovoltaics, micro gas turbines and reciprocating
engines amongst those technologies expected to be the most prevalent by 2020. This provides
a starting point for more macro-based analysis. For example, macroeconomic analysis could
be conducted by assuming an overall capacity of DG penetration, assigning generic (and
necessarily uncertain) costs associated with this capacity installation, and assigning these
expenditures to the top three (for example) dominant technologies. Scenario analysis around
benchmark values could yield significant insight into understanding the significance of key
assumptions. Such work provides insight into the expected level of DG penetration, and
informs the modelling, simulation and analysis of its impact.

Although it is difficult to do so, estimating the macroeconomic impacts of DG has clear

benefits. DG is becoming increasingly widespread and its uptake is fundamentally changing
the structure and operation of electricity networks, and increased expenditures on renewable
energy installations, for example, is likely to have wider economic interactions and feedback
effects. A number of multi-sectoral modelling approaches allow an analysis of the likely
aggregate and sectoral economic impacts (which in turn can be linked to environmental and
social impacts – see below). First, input-output (IO) analysis can be usefully employed to
assess regional (and if data permit) local economic and employment effects of, for example,
the introduction of significant DG penetration27. Such analyses can be employed, for
example, to assess the importance of supply-chain development in governing the scale of
economic impacts: the greater the degree of local embeddedness, the greater the impact on
the host economy28.

Although undoubtedly useful, IO studies have limitations. For example, while onshore wind
developments currently typically have little in the way of backward linkage into the host
region (since the technology is imported), they are often associated with significant income
flows in the form of community benefits and, increasingly, returns to local ownership. These
income flows are not captured appropriately in IO systems, but are in social accounting
matrices (SAMs), which can identify the potentially significant impacts of such income flows
on the local economy29. Such models are capable of exploring the impacts of different levels
and forms of community benefit payments and of alternative ownership models and so could
be applied to a range of DG initiatives.

Another limitation of IO models (shared by SAM models) is that they are predicated upon an
assumption of entirely passive supply: they are completely demand-driven. In circumstances
where there are important supply constraints in the host region, the supply side of the
economy has to be modelled explicitly. In principle, computable general equilibrium models
(CGEs) permit a complete, theory-consistent model of the demand and supply sides of all
markets. In CGEs prices are endogenous, and typically adjust to equate demand and supply in
each market. However, for demand-side disturbances they replicate the comparable IO
systems where supply is passive (e.g. where there is significant unemployment and spare
capacity). These studies have modelled, for example, the macroeconomic impact of increased
demand connected with marine expenditures, the potential legacy effects of such

27
See Miller and Blair (2009) for an exposition of IO and SAM models.
28
See Allan et al (2007a) for a study of the impact of alternative electricity generating technologies.
29
Allan et al (2011) adopt a SAM approach to illustrate the potential impacts of the Viking windfarm proposal
in the Shetland Islands.
expenditures, and the creation of new export opportunities (Allan et al 2008; Allan et al
2010b; Allan et al 2014a).

Economy-wide environmental impacts

Given the existence of fuel emissions coefficients and knowledge of the fuel-use of
industries, it is possible to construct integrated IO and SAM accounting systems that include
energy demands and emissions. These integrated databases can be used to calibrate energy-
economy-environment IO, SAM and CGE models, which can be used to track the impact of
disturbances on GHGs for example. The resultant databases can be used to compute carbon
footprints in a rigorous and transparent way that reflects system-wide impacts on emissions
(including those that arise indirectly through intermediate purchases and those that are
induced by consumers’ expenditure responding to changes in income). In the context of DG,
this would allow an assessment of its impact on a particular region’s carbon footprint.
However, in the context of ‘open’ regions or localities (where trade flows are relatively
large), it would be important to distinguish between production-oriented measures of
emissions (as emphasised by Kyoto) and consumption-oriented measures. In two (or more)
region context it would be possible to identify the CO2 ‘trade balance’ between regions, and
the impact of DG on this30. Clearly, small, open regions may initially import electricity from
beyond its geographic boundaries, whereas CHP development may bring generation within
those boundaries but result in a cut in imports (and possible electricity generation) elsewhere.
Production-oriented measures would clearly indicate an increase in emissions, whereas
consumption-oriented measures may indicate the reverse.

CGEs can and have been used to identify the environmental impacts of various policy and
other disturbances. This approach is essential if relative price changes are a key element of
any disturbance, as is the case, for example if a carbon tax is imposed (e.g. Allan et al,
2014b) or energy efficiency is improved. In the latter case CGEs can be used to analyse the
‘rebound’ and ‘backfire’ effects associated with energy efficiency improvements (e.g. Allan
et al 2007a; Lecca et al, 2014; Hanley et al, 2009). The approach may similarly be applied to
explore the economic and environmental impact of significant penetration of DG.

30
See e.g. McGregor et al (2008). Allan et al (2010a) is an illustrative application to the introduction of CHP to
Glasgow.
Economy-wide social impacts

In principle, the energy-economy-environment IO and SAM databases can be augmented to

include a degree of household disaggregation by income group. Again these databases can be
calibrated to yield corresponding models in which impacts on households by broad income
group are automatically tracked. Clearly this is of interest given concerns about fuel poverty,
and indeed about income distribution in general. Again such models can be usefully applied
to DG31.

6. Conclusions

In recent years, there have been significant changes to the institutional, technical, and policy
environment within which the UK energy supply system operates. The current energy
resource mix has expanded, with the increased operation of ‘clean’ and ‘efficient’ supply
sources, and a shift towards distributed energy technologies. In this paper we consider the
economics of distributed generation, and review the literature on the financial viability, social
costs and benefits, and macroeconomics of DG. This is with a view to contributing to the
knowledge base on DG, and to identifying any gaps in the literature that future research
might address to improve the evidence base for policy formulation.

Distributed energy encompasses a wide variety of technologies, and we find that the
economics of DG tends to be highly sensitive to the type of technology and deployment
context. The results of financial viability studies are mixed, although stand-alone projects in
isolated areas in developing countries seem currently to be the most attractive. As is the case
for most emerging technologies, DG technologies are typically not yet commercially viable
without support, and tend to have greater costs than conventional technologies, though many
authors acknowledge the potential for future cost reductions through technological
advancements. We believe there is scope for wider application here of the ‘learning curve’
analyses that have been applied to other emerging technologies. Furthermore, recognition that
DG will inevitably form only part of a portfolio of generating technologies suggests that
levelised cost comparisons reflect a limited perspective (even if these do incorporate wider

31
The illustrative application of Allan et al (2010a) includes a degree of household disaggregation.
electricity system impacts), since DG may reduce risks for any given cost level because it is
likely to reduce correlations with fuel prices, and to enhance security of supply32. This
perspective also suggests that there is scope for the application of portfolio theory to
incorporate DG. At the aggregate level this would focus on the benefits of diversified
technologies; at the spatially disaggregated level account should be taken of the benefits of
geographic diversification.

In terms of social costs and benefits, the limited attempts at comprehensive analysis exhibit
considerable variation in results, with DG proving less attractive than some alternatives.
However, whether that will continue to be so if learning is explicitly incorporated and
portfolio and other benefits fully valued, remains to be seen. Certainly, difficult though it
undoubtedly is, we judge that further cost benefit analysis (CBA) of DG is likely to prove
productive, with some of the uncertainties diminishing with time. CBA continues to provide,
in our view, the most comprehensive and systematic framework for evaluating individual
projects, or groups of projects, from the perspective of society as a whole. However, the
approach can, and should, be extended to accommodate general equilibrium impacts/
macroeconomic effects.

We find very few examples of macroeconomic analyses of DG. Of course, the very
heterogeneous nature of DG renders such analyses especially difficult. The appropriate
numerical representation of DG is clearly complex. Nevertheless, in order to assess the
contribution of DG to energy (and other) policy goals, we need to understand the system-
wide economic, environmental and social impacts of increased penetration of DG. Increased
decentralisation of the UK energy system via DG technologies would likely have substantial
impacts through construction expenditures, infrastructure adaptations, employment changes
in ‘green’ industries and environmental impacts. Such effects could have important
implications for economies at all spatial scales, and are of interest to local, regional and
national governments. There are also potentially important impacts of DG policy on fuel
poverty in particular and equity in general, as well as issues relating to the potential
displacement of emissions across local areas, from centralised ‘out of town’ plants to urban
locations. This highlights the complexity of, but also the need for, emissions attribution
analyses that recognise geographic boundaries (and accommodate trade flows) appropriately.

32
However, given liberalised markets levelised costs remain a part of private investment decisions. This can be
regarded as a further argument in favour of public sector support for DG.
System-wide economic-energy-environment modelling can, we believe, enhance the evidence
base for DG policy formation and implementation.
References

Ackermann T, Andersson G and Soder L (2001). ‘Distributed generation: a definition’,

Electric Power Systems Research, 57, pp 195-204.

Ackerman, F (2008). ‘Critique of Cost-Benefit Analysis, and alternative approaches to

decision-making’. A report to friends of the earth England, Wales and Northern Ireland.
Retrieved 19 July, 2010 from http://www.foe.co.uk/resource/reports/policy_appraisal.pdf

Alarcon-Rodriguez A, Haesen E, Ault G, Driesen J and Belmans R (2009). ‘Multi-objective

planning framework for stochastic and controllable distributed energy resources’, IET
Renewable Power Generation, 3, Issue 2.

Awerbuch S (2008). ‘The role of wind generation in enhancing Scotland’s energy diversity
and security’. In: M Bazilian and F Roques (Eds). Analytical methods for energy diversity
and security, Elsevier, Amsterdam, 2008.

Allan GJ, Bryden I, McGregor PG, Stallard T, Swales JK, Turner K and Wallace R (2008).
'Concurrent and legacy economic and environmental impacts from establishing a marine
energy sector in Scotland', Energy Policy, 36, Issue 7, pp 2734-2753.

Allan G, Eromenko I, Gilmartin M, McGregor P, Swales K and Turner K (2010a). ‘The

economic and environmental impacts of Distributed Generation on the city and region: A two
region input-output analysis of Glasgow and Scotland using illustrative data’. SuperGen
HiDef Discussion Paper.

Allan G, Eromenko I, McGregor P, Swales K (2011). ‘The regional electricity generation mix
in Scotland: A portfolio selection approach incorporating marine technologies’. Energy
Policy, Vol. 39, p. 6-22.

Allan G, Gilmartin M, McGregor P, Swales JK (2010b). ‘The formation of a domestic and

international market for UK tidal energy devices and technologies: Assessing the demand
side implications for the UK economy’. European Association of Energy Economics
Conference Paper, forthcoming.

Allan G, Hanley N, McGregor P, Swales K and Turner K (2007a). ‘The impact of increased
efficiency in the industrial use of energy: A Computable General Equilibrium analysis for the
United Kingdom’, Energy Economics, 29, Issue 4, July, pp 779-798.
Allan, G, Lecca, P., McGregor, P. G. and Swales, J. K. (2014a). ‘Marine Energy Impacts: A
Case Study from Scotland’, Marine Policy, doi:
http://dx.doi.org/10.1016/j.marpol.2013.05.003

Allan, G, Lecca, P, McGregor, P. G. and J. K. Swales (2014b). ‘The Economic and

Environmental Impact of a Carbon Tax for Scotland: A Computable General Equilibrium
Analysis’, Ecological Economics, Vol. 100, pp 40-50.

Allan G, McGregor P, Swales JK (2011). ‘The importance of revenue sharing for the local
economic impacts of a renewable energy project: A Social Accounting Matrix approach’,
Regional Studies, Vol. 45 (9), pp. 1171-1186.

Allan G, McGregor P, Swales K, Turner K (2007b). ‘The Impact of alternative electricity

generation technologies on the Scottish economy: An illustrative input—output analysis’,
Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy,
221, Number 2, pp 243-254.

Allen SR, Hammond GP, and McManus MC (2008a). ‘Energy analysis and environmental
life cycle assessment of a micro-wind turbine’, Proceedings of the Institution of Mechanical
Engineers, Part A: Journal of Power and Energy, 22, Number 7, pp 669-684.

Allen SR, Hammond GP and McManus MC (2008b). ‘Prospects for and barriers to domestic
micro-generation: A United Kingdom perspective’, Applied Energy, 85, Issue 6, June, pp528-
544.

Arthur D. Little (1999). ‘Distributed generation: Understanding the economics’. An Arthur

D. Little White Paper.

Audenaert A, De Boek L, De Cleyn S, Lizin S and Adam JF (2010). ‘An economic

evaluation of photovoltaic grid connected systems (PVGCS) in Flanders for companies: A
generic model’, Renweable Energy, 35, Issue 12, December, pp 2674-2682.

Ayers RU, Turton H, Casten T (2007). ‘Energy efficiency, sustainability and economic
growth’, Energy, 32, pp 634-648.

Bakos GC and Tsagas NF (2003). ‘Technoeconomic assessment of a hybrid solar/wind

installation for electrical energy saving’, Energy and Buildings, 35, Issue 2, pp 139–145.

Balcombe, P., Rigby, D. and Azapagic, A. (2013). ‘Motivations and barriers associated with
adopting microgeneration energy technologies in the UK’, Renewable and Sustainable
Energy Reviews, Vol. 22, p. 655-666.
Bergman N, Hawkes A, Brett DJL, Baker P, Barton J, Blanchard R, Brandon NP, Infield D,
Jardine C, Kelly N, Leach M, Matian M, Peacock AD, Staffell I, Sudtharalingam S and
Woodman B (2009). ‘UK microgeneration part I: Policy and behavioural aspects’,
Proceedings of the Institution of Civil Engineers, 162, Issue EN1, February.

Bernal-Agustin JL and Dufo-Lopez R (2006). ‘Economical and environmental analysis of

grid connected photovoltaic systems in Spain’, Renewable Energy, 31, Issue 8, pp 1107–
1128.

BERR, Department for Business, Enterprise and Regulatory Reform (2007). ‘Review of
distributed generation report: May 2007’. Retrieved 12 July, 2010 from http://webarchive.
nationalarchives.gov.uk/20081230223129/http://www.berr.gov.uk/files/file39025.pdf

BERR, Department for Business, Enterprise and Regulatory Reform (2008). ‘The growth
potential for on-site renewable electricity generation in the non domestic sector in England,
Scotland and Wales’. Report prepared with Element Energy Limited. Retrieved 5 July, 2010
from http://www.berr .gov.uk/files/file48018.pdf

BERR, Department for Business, Enterprise and Regulatory Reform and WADE (2007).
‘Using the WADE model to investigate the relative costs of Distributed Generation (DG)’.
Retrieved 19 July, 2010 from http://www.berr.gov.uk/files/file39385.pdf

Boedecker E, Cymbalsky J, and Wade S (2002). ‘Modeling distributed generation in the

NEMS building models’. US Energy Information Administration. Retrieved 12 July, 2010
from http://www.eia.doe.gov/oiaf/analysispaper/electricity_generation.html.

Bush, R., Jacques, D.A., Scott, K. and Barrett, J. (2014). ‘The carbon payback of micro-
generation: An integrated hybrid input-output approach’, Applied Energy, p. 85-98.

Caird S and Roy R (2007). ‘Consumer adoption and use of household renewable energy
technologies’. Design Innovation Group, Open University, Milton Keynes.

Cambridge Econometrics (2003). ‘Modelling good quality combined heat and power capacity
to 2010: Revised projections’. A final report submitted to Department of Trade and Industry.

Carley, S (2009). ‘Distributed generation: An empirical analysis of primary motivators’,

Energy Policy, 37, Issue 5, May, pp 1648-1659.

CCC, Committee on Climate Change (2008). ‘Building a low-carbon economy – The UK’s
contribution to tackling climate change’. The First Report of the Committee on Climate
Change, December. The Stationary Office.
Chakrabarti S and Chakrabarti S (2002). ‘Rural electrification programme with solar energy
in remote region—a case study in an island’, Energy Policy, 30, Issue 1, pp 33–42.

Chambers A (2001). ‘Distributed generation: a nontechnical guide’. PennWell, Oklahoma.

Cherrington, R., Goodship, V., Longfield, A. and Kirwan, K. (2013), ‘The feed-in tariff in the
UK: A case study focus on domestic photovoltaic systems’, Renewable Energy, 50, p. 421-
426.

CIRED (1999). ‘Dispersed generation’. Preliminary report of CIRED working group, WG04,
June.

Claudy, M.C., Michelsen, C., O’Driscoll, A. and Mullen, M. (2010), ‘Consumer awareness in
the adoption of microgeneration technologies: An empirical investigation ion the Republic of
Ireland’, Renewable and Sustainable Energy Reviews, Vol. 14, p. 2154-2160.

Claudy, M.C., Michelsen, C. and O’Driscoll, A. (2011). ‘The diffusion of microgeneration

technologies – assessing the influence of perceived product characerisistics on home owners
willingness to pay’, Energy Policy, Vol. 39, p. 1459-1469.

DECC, Department of Energy and Climate Change (2010). ‘Feed-in Tariffs: A response to
the summer 2009 consultation’.

DECC (2011). ‘Review of the generation costs and deployment potential of renewable
electricity technologies in the UK: Study report’, available online at
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/147863/3237-
cons-ro-banding-arup-report.pdf

Devine-Wright P (2007). ‘Reconsidering public attitudes and public acceptance of renewable

energy technologies: A critical review’. School of Environment and Development,
University of Manchester, Working paper 1.4 of the research project ‘Beyond nimbyism: A
multi-disciplinary approach investigation of public engagement with renewable energy
technologies’.

Dondi P, Bayoumi D, Haederli C, Julian D and Suter M. (2002). ‘Network integration of

distributed power generation’, Journal of Power Sources, 106, Issues 1-2, April, pp1-9.

DTI, Department for Trade and Industry (2003). ‘Modelling good quality combined heat and
power capacity to 2010: revised projections’. Report prepared by Cambridge Econometrics,
November.
DTI, Department of Trade and Industry (2006). ‘Using the WADE model to investigate the
relative costs of distributed generation’. Retrieved 13 July, 2010 from
http://www.berr.gov.uk/files/ file39385.pdf

DTI, Department of Trade and Industry (2007a). ‘Meeting the energy challenge: A white
paper on energy’. Retrieved 14 July, 2010 from http://www.berr.gov.uk/files/file39387.pdf.

DTI, Department of Trade and Industry (2007b). ‘Review of distributed generation’. A Joint
Government/Ofgem Report.

Energy Information Administration (2005). ‘Commercial sector demand module of the

national energy modeling System’. Model Documentation, EIA/DOE M066, February.
Washington D.C.

Energy Saving Trust, E-Connect and Element Energy (2005). ‘Potential for microgeneration
study and analysis’. Retrieved 21 July, 2010 from http://www.dti.gov.uk/files/file27558.pdf

E-Parl (12 July 2010). ‘Feed-in tariffs support renewable energy in Germany’. Retrieved 29
July, 2010 from http://www.e-parl.net/eparlimages/general/pdf/080603%20FIT%20toolkit.
Pdf

Euractiv (05 July 2010). ‘Oettinger backs EU-wide renewables feed-in tariff’. Retrieved 29
July, 2010 from http://www.euractiv.com/en/energy/oettinger-backs-eu-renewables-feed-
tariff-news-496148

Finney, K.N., Sharifi, V.N. and Swithenbank, J. (2012), ‘The negative impacts of the global
economic downturn on funding decentralised energy in the UK’, Energy Policy, Vol. 51, p.
290-500.

Foote CET, Burt GM, Elders IM, and Ault GW (2005). ‘Developing distributed generation
penetration scenarios’. IEEE International Conference on Future Power Systems, 16-18 Nov.

Gross, R., Heptonstall, P., Greenacre, P., Candelise, C., Jones, F. And Castillo, A.C. (2013),
‘Presenting the future: An assessment of future costs estimation methodologies in the
electricity generation sector’, A report by the UK Energy Research Centre Technology and
Policy Assessment Function, November 2013.

Gulli F (2006). ‘Small distributed generation versus centralised supply: A social cost-benefit
analysis in the residential and service sectors’, Energy Policy, 34, Issue 7, pp 804–832.
Haesen E, Alarcon-Rodriguez AD, Driesen J, Belmans R and Ault G. (2009). ‘Opportunities
for active DER management in deferral of distribution system reinforcements’. SuperGen-3
HDPS Discussion Paper.

Hanley, N., McGregor, P.G., Swales, J.K. and Turner, K. (2009), ‘Do increases in energy
efficiency improve environmental quality and sustainability?’, Ecological Economics, 68 (3),
p. 692-709.

Hammond GP, Winnett AB (2006). ‘Interdisciplinary perspectives on environmental

appraisal and valuation techniques’, Proceedings of the Institution of Civil Engineers Waste
and Resource Management, 159, Issue WR3.

Hawkes AD and Leach MA (2008). ‘On policy instruments for support of micro combined
heat and power’, Energy Policy, 36, pp 2973-2982.

Hawkes AD, Brett DJL, and Brandon NP (2009a). ‘Fuel cell micro-CHP techno-economics:
Part 1 – model concept and formulation’, International Journal of Hydrogen Energy, 34, Issue
23, December, pp 9545-9557.

Hawkes AD, Brett DJL, Brandon NP (2009b). ‘Fuel cell micro-CHP techno-economics: Part
2 – Model application to consider the economic and environmental impact of stack
degradation’, International Journal of Hydrogen Energy, 34, Issue 23, December, pp 9558-
9569.

Hoff TE (1996). ‘Identifying distributed generation and demand side management investment
opportunities’, The Energy Journal, 17, Issue 4, pp 89–106.

Hoff TE, Wenger HJ and Farmer BK (1996). ‘Distributed generation. An alternative to

electric utility investment in system capacity’, Energy Policy, 24, Issue 2, pp 137–147.

Infield D and Li F (2008). ‘Integrating micro-generation into distribution systems – a review

of recent research’. Retrieved 22 July, 2010 from http://ieeexplore.ieee.org/stamp/
stamp.jsp?arnumber=04596874

Jardine C (2009). ‘An analysis of the forthcoming Feed-in tariffs for solar photovoltaic
electricity generation in the UK’. Discussion Paper, Environmental Change Institute,
University of Oxford.

Kaldellis, JK (2003). ‘Feasibility evaluation of Greek State 1990–2001 wind energy

program’, Energy, 28, Issue 14, pp 1375–1394.
Karytsas, S. and Theodoropoulou, H. (2014). ‘Public awareness and willingness to adopt
ground source heat pumps for domestic heating and cooling’, Renewable and Sustainable
Energy Reviews, Vol. 34, p. 49-57.

Keirstead J (2007). ‘Behavioural responses to photovoltaic systems in the UK domestic

sector’, Energy Policy, 35, Issue 8, August, pp 4128-4141

Kelly N, Galloway S, Elders I, Tumilty R and Burt Graeme (2008). ‘Modelling the impact of
micro generation on the electrical distribution system’. Report Paper, Institute for Energy and
Environment, University of Strathclyde.

Khan, M and Iqbal M (2005). ‘Pre-feasibility study of stand-alone hybrid energy systems for
applications in Newfoundland’, Renewable Energy, 30, Issue 6, pp 835–854.

Koeppel G (2003). ‘Distributed generation—literature review and outline of the Swiss

situation’. EEH Power Systems Laboratory, Zurich.

Kolhe M, Kolhe S and Joshi JC (2002). ‘Economic viability of stand-alone solar photovoltaic
system in comparison with diesel-powered system for India’, Energy Economics, 24, Issue 2,
pp 155–165.

Lecca, P., McGregor, P.G., Swales, J.K. and Turner, K. (2014), ‘The added value from a
general equilibrium analysis of increased efficiency in household energy use’, Ecological
Economics, 100, p. 51-62.

Lehtonen M and Nye S (2009). ‘History of electricity network control and distributed
generation in the UK and Western Denmark’, Energy Policy, 37, pp 2338–2345.

Lund P, Cherian S and Ackerman T (2006). ‘A cell controller for autonomous operation of a
60 kV distribution area’, International Journal of Distributed Energy Resources, 2, pp 83–
100.

McGregor P.G, Swales J.K and Turner K (2008). ‘The CO2 trade balance between Scotland
and the rest of the UK: Performing a multi-regional environmental input-output analysis with
limited data’, Ecological Economics, 66, pp 662-673.

McGregor, P.G., Swales, J.K. and Winning, M. (2012), ‘A review of the role and remit of the
committee on climate change’, Energy Policy, 41, p. 466-473.
Méndez VH, Rivier J, De la Fuente, JI, Gómez T, Arceluz J, Marín J and Madurga A (2006).
‘Impact of distributed generation on distribution investment deferral’, International Journal of
Electrical Power & Energy Systems, 28, Issue 4, May, pp 244-252.

Miller R and Blair P (2009). ‘Input-output analysis: Foundations and extensions’. Cambridge
University Press, Cambridge.

McManus MC, Hammond GP and Burrows CR (1999). ‘Life Cycle Assessment’. Written
Evidence on Non-Food Crops, HL Paper 5-I, House of Lords Select Committee on Science
and Technology, London.

Molloy E (2003). ‘When is a spade not (only) a spade? When it’s an environmental
management tool’. In: A Winnet and A Warhurst (Eds). Towards an Environment Research
Agenda: A Second Selection of Papers. Palgrave Macmillan, Basingstoke.

Mori C, Hawkes A and Cockerill T (2010). ‘A review of UK Energy Policy at Multi-levels of

Governance’. SuperGen HiDef Discussion Paper.

Morthorst PE, Petersen AH, Nissen F and Cronin A (2006). ‘Denmark in a European market.
The future energy system–distributed production and use’. Risø Energy Report 4, pp 13–19.

Muhammad-Sukki, F., Ramirez-Iniguez, R., Munir, A.B., Mohd Yasin, S.H., AbuBakr, S.H.,
McMeekin, S.G. and Stewart, B.G. (2013). ‘Revised feed-in tariff for solar photovoltaic in
the United Kingdom: A cloudy future ahead?’, Energy Policy, 52, p. 532-838.

Newborough, M. (2004). ‘Assessing the benefits of implementing micro-CHP systems in the

UK’ Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and
Energy Volume 218, Number 4, pp 203-218.

Ofgem (2002). ‘Distributed generation: the way forward’ Ofgem factsheet 15, March.
Retrieved 19 July, 2010 from
http://www.ofgem.gov.uk/MEDIA/FACTSHEETS/Documents1 /1102-factsheet0602
_27feb.pdf

Ofgem (2008). ‘Renewables Obligation: Annual Report 2006-2007’. Retrieved 2 July, 2010
from http://www.ofgem.gov.uk/Sustainability/Environmnt/RenewablObl/Documents1/Ann
ual%20report%202006-07.pdf

Pearce D (1998). ‘Cost-Benefit Analysis and environmental policy’, Oxford Review of

Economic Policy, 14, Number 4, pp 84-100
Pehnt M (2008). ‘Environmental impacts of distributed energy systems — The case of micro
cogeneration’, Environmental Science and Policy, 11, Issue 1, 25-37.

Pepermans G, Driesen J, Haeseldonckx D, D’haeseleer W and Belmans, R (2005).

‘Distributed generation: Definition, benefits and issues’, Energy Policy, 33, Issue 6, April,
787-798.

Pudjianto D, Shakoor AA, Djapic P, Ramsay C and Strbac G (2008). ‘UK system level
implications of HDPS’. HDPS Report Paper, presented at First International Conference and
Workshop on Micro-Cogeneration Technologies and Applications National Arts Centre,
Ottawa, Ontario, April – May.

Ravindranath NH, Balachandra P, Dasappa S and Usha Rao K (2006). ‘Bioenergy

technologies for carbon abatement’, Biomass and Bioenergy, 30, Issue 10, pp 826–837.

Ravindranath NH and Hall D.O (1995). ‘Biomass, energy and environment—a developing
country perspective from India’. Oxford University press.

Renewables Obligation (Amendment) Order (2010). ‘Statutory Instrument No.829’. The

Stationary Office.

Renewables Obligation Order (2009). ‘Statutory Instrument No.785’. The Stationary Office.

Rogers E M (1995). ‘Diffusion of Innovations’. Free Press, New York

Sauter R and Watson J (2007). ‘Strategies for the deployment of micro-generation:

Implications for social acceptance’, Energy Policy, 35, Issue 5, May, pp 2770-2779.

Scottish Government (2008). ‘Renewables Obligation (Scotland) Introduction of Banding,

Statutory Consultation’. September 2008. Retrieved 9 July, 2010 from
http://www.scotland.gov.uk/Resource/Doc/917/0065773.pdf

Solar Power Portal (2010). ‘British Gas starts £15 million free solar installations scheme for
schools’. Retrieved 7 July, 2010 from
http://www.solarpowerportal.co.uk/news/british_gas_starts
_15_million_free_solar_457installations879_scheme/ 29 July 2010

Sørensen P, Meibom P and Nørga P (2006). ‘Flexibility, stability and security of supply. The
future energy system, distributed production and use’. Risø Energy Report 4.
Stirling, A (1994). ‘Diversity and ignorance in electricity supply investment: Addressing the
solution rather than the problem’ Energy Policy, 22, Issue 3, pp 195-216

Strbac G, Ramsay C, Pudjianto D (2007). ‘Integration of distributed generation into the UK

power system’. Summary Report, DTI Centre for Distributed Generation and Sustainable
Electrical Energy.

Tan, W-S., Hassan, M.Y., Majid, M.S. and Rahman, H.A. (2013). ‘Optimal distributed
renewable generation planning: A review of different approaches’, Renewable and
Sustainable Energy Reviews, 18, p. 626-645.

Uyterlinde MA, Sambeek EJW, Cross ED, Jorb W, Loffer P, Morthorst PE, Jørgensen BH
(2002). ‘Decentralised generation: Development of EU Policy’. ECN, Petten.

Van der Veen, R. and De Vries, L. (2009), ‘The impact of microgeneration upon the Dutch
balancing market’, Energy Policy, 37, p. 2788-2797.

Van der Vleuten E and Raven R (2006). ‘Lock-in and change: Distributed generation in
Denmark in a long-term perspective’, Energy Policy, 34, 3739–3748.

Varming, S and Nielsen JE (2004). ‘Review of the role of ICT (information and
communication technology in network management and market operations)’. SUSTELNET
project report. Retrieved 16 July, 2010 from
http://www.ecn.nl/fileadmin/ecn/units/bs/SUSTELNET/wp3-ict.pdfS

Viral, R. and Khatod, D.K. (2012), ‘Optimal planning of distributed generation systems in
distribution system: A review’, Renewable and Sustainable Energy Reviews, 16 (7), p. 5146-
5165.

Vogel P (2009). ‘Efficient investment signals for distributed generation’, Energy Policy, 37,
Issue 9, September, pp 3665-3672.

Walker, G (2008). ‘Decentralised systems and fuel poverty: Are there any links or risks?’,
Energy Policy, 36, Issue 12, pp 4514–4517.

Watson, J., Sauter, R., Bahaj, B., James, P., Myers, L. and Wing, R. (2008), ‘Domestic
micro-generation: Economic, regulatory and policy issues for the UK’, Energy Policy, Vol.
36, p. 3095-3106.

Willis HL, Scott W. (2000). ‘Distributed power generation. Planning and evaluation’. First
Edition, New York: Marcel Dekker.