The Political Economy of Project Delay

Author(s): S. Hayden Lesbirel

Source: Policy Sciences, Vol. 20, No. 2 (Jun., 1987), pp. 153-171
Published by: Springer
Stable URL: http://www.jstor.org/stable/4532108
Accessed: 31-07-2016 02:16 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted
digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about
JSTOR, please contact support@jstor.org.

Springer is collaborating with JSTOR to digitize, preserve and extend access to Policy Sciences

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
Policy Sciences 20: 153-171 (1987)
© Martinus Nijhoff Publishers, Dordrecht - Printed in the Netherlands 153

The political economy of project delay

S. HAYDEN LESBIREL
Department of Political Science, Faculty of Arts, Australia-Japan Research Cen
School of Pacific Studies, Australian National University, Australia

Abstract. Uncertainty about how long it will take to secure approval for larg
projects presents difficulties for policy makers in all industrialized countries. Th
to determine the factors that influence the time required to get public acceptance fo
in Japan. It suggests the importance of distributional factors and, hence, the ne
the structure of the bargaining environment; 2) the ability of interest groups to
to or support for projects; 3) bargaining strategies employed; 4) uncertainty abo
5) changing expectations, in explaining why compensation mechanisms operate mo
resolving some social disputes than others. It concludes that an analysis of Japan
this area can contribute to an understanding of other social controversies and th
has utility to policy makers involved in developing controversial projects.

Introduction

Delay in developing a range of private and public projects is an inherent featur

of all industrialized countries. Approval to develop projects not only takes a lon
time but also an unpredictably long time. The analysis of delay is important be
cause the variation in times creates uncertainty about the speed and extent to
which objectives can be achieved. We would be better off if the factors that caus
delay could be better understood and if the time required to win approval for
projects could be known more precisely. It would inject more certainty int
project planning and the management of siting policy.
There has been a growing literature on project siting problems reflecting th
importance attached to this as a policy issue. Most of the literature on project
siting problems focuses on the origins and costs of delay in the licensing an
construction stages of project development.' What little literature there is that
analyzes delay in the public acceptance of projects has not developed a systemat
ic approach to understanding the sources of delay. There is, therefore, a need
to extend that literature and analyze the factors that contribute to the variatio
in times necessary to win agreement over the development of controvers
projects.
The purpose of this article is to develop an approach, which has not been em-
ployed in the literature, for explaining why some controversies surroundin
project development take longer to resolve than others.2 I develop a regression
model which stresses the importance of distributional influences and, hence, th

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
154

need to examine the structure of the bargaining environment and the effective-
ness of compensation mechanisms in explaining the variance in public accep-
tance times. Although I do this with reference to the siting of power plants in
Japan, I suggest that the approach and key determinants of delay are likely to
have some broader applicability to the analysis of social conflicts in pluralist
democratic regimes more generally.
The first section considers the role and effectiveness of compensation
mechanisms in the resolution of disputes involving hazardous projects. The sec-
ond presents an econometric model which aims to quantify the importance of
distributional influences on the variation in settlement times and tests the
predictive power of that model. The third section discusses through a case study
the nature of site-specific characteristics, and suggests the need for going beyond
statistical techniques in analyzing the variation in settlement times. The conclu-
sion considers the extent to which the findings are likely to have wider applicabil-
ity to understanding other social controversies and the utility of the approach
to policy makers.

Compensation and project approval

The siting of major projects confers benefits on some interests, and costs on
others. In democratic countries, where communities generally have legal and po-
litical veto power over siting decisions, the siting of projects will require negotia-
tion. This process is characterized by bargaining between project promoters,
such as private companies and public authorities, and community interests, such
as local groups and the public, over the allocation of costs and benefits expected
to be generated from the project development. Settlements will normally require
that promoters compensate community interests for losses expected to be in-
curred as a result of the projects.
Most industrialized countries use compensation and other redistributive
mechanisms in the resolution of disputes over siting major facilities.3 Compen-
sation may be paid through existing institutions or redistributive mechanisms
may have to be established and can be either explicit or implicit in nature.4 Ex-
plicit compensation may be paid in monetary form, or in non-monetary form
such as in the provision of public goods. Implicit compensation may involve
postponing the development of a project to allow affected parties more time to
adjust to a new environment or changing the size or design of the project in order
to reduce negative impacts. Developers may pay compensation directly through
either existing or established mechanisms. Alternatively, it may be paid indirect-
ly to relevant communities. For example, the national community may subsidize
project developers and those subsidies may act to increase the compensation
pool that can be redistributed to communities. The strategic objective of these

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 155

mechanisms is to increase the benefits, or at least reduce the costs, of projects

to affected parties, thereby reducing or nullifying resistance and facilitating the
settlement process.5
The success of these mechanisms varies substantially. For example, in Japan,
they were extremely effective in winning approval for the Fukushima, Oi and
Mihama nuclear plants but not for the Ashihama, Namie and Kumano
projects.6 Similarly, in the United States, the use of compensation was success-
ful in reaching a settlement for the development of the Grayrocks, Skagit and
Pilgrim projects but not in the development of the Bodego Bay, Ravenswood
and Cayaga Lake projects.7 These observations provide the context for analyz-
ing the effectiveness of compensation and other redistributive mechanisms in
the resolution of project disputes as well as for developing a framework for as-
sessing the factors that cause delay in the approval of projects more generally.
Achieving politically adequate compensation takes time, and influences the
speed at which projects are approved and the final form they take. The time re-
quired to win approval for projects will be related to the extent to which existing
or established redistributive mechanisms operate effectively. Compensation
mechanisms are likely to be more successful in situations where there are large
expected net benefits from projects, where there is a large economic surplus
available for redistribution and where those mechanisms can be used effectively
to offset adverse effects consequent upon the particular project development.
The effectiveness of redistributive mechanisms is likely, in general, to be related
to a number of structural determinants, such as the pattern of expected distribu-
tional impacts, uncertainty about outcomes, the structure of bargaining power,
the ability of groups to organize and sustain either support for or opposition
against projects, the nature of bargaining strategies employed and the stability
of expectations during the course of settlement.
It is often argued, either implicitly or explicitly, by Western as well as Japanese
observers that social choice and bargaining processes are determined culturally
in Japan,8 and that Japanese facility siting experience would not contribute to
an understanding of siting processes more generally. This line of reasoning im-
plies that compensation and negotiating processes in Japanese society are very
special because social conventions, practices and mores, population relation-
ships, and a variety of other circumstances reflecting the socio-cultural environ-
ment are so different from the rest of the world. I will challenge this conventional
wisdom by arguing that structural determinants, which are found in most indus-
trialized countries, are more important than cultural determinants in explaining
the effectiveness of compensation mechanisms and the outcome of siting dis-
putes in Japan. In addition to suggesting that structural determinants are signifi-
cant factors explaining delay, I will show that an analysis of Japanese experience
can enhance our understanding of the process of delay and the way to manage it.

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
156

Factors contributing to delay: empirical results

A bargaining framework provides a useful entry point into analyzing the factors
that contribute to delay. The resolution of disputes over the siting of projects
requires striking an acceptable bargain amongst participants to the process over
the expected gains and losses from project developments. Hence, the distribution
of costs and benefits and the way in which compensation can be affected
amongst various community interests will be an important factor influencing
settlement times. The extent of the disparity in the value of projects between
promoters and community interests will be determined by distributional effects.
These effectsinfluence the value that is placed on costs and benefits and, there-
fore, the extent to which benefits exceed costs and the economic surplus that
can be redistributed in order to win approval. Resistance to projects is likely to
be stroniger in situations where adverse effects are expected to be incurred from
project developments and where there are difficulties in organizing compensa-
tion mechanisms or satisfying competing compensation claims.

The model

Table 1 defines and explicates the variables used in the analysis. The distribu-
tional effects on promoters and community interests shape responses to develop-
and accepting projects and influence approval times. These responses will
shape the bargaining environment in which settlements take place and may be
more or less conducive to a settlement depending on the direction and strength
of those responses. The willingness of promoters to install projects will be deter-
mined by, amongst other things, the need to develop additional capacity to meet
expected demand for project output. The responses of community interests to
offered 'project sites will be a function of such factors as: the extent of social
and ec'nomic opportunities, the ability of government to supply public goods,
the importance of the rural sector in the regional economy, the political structure
of the-electorate, the perceived risks involved in accepting different projects and
attitudes towards preservation of the environment. In order to assess the relative
importance of these distributional influences, I developed a regression model
for explaining the variation in approval times.
The indentification and selection of variables was based on three major con-
siderations. The first was extensive case study analysis which identified key in-
terests involved in bargaining, such as power companies, regional and national
governments, and rural interests and sought to measure quantitatively the im-
pact of projects on these interests and their responses to either promoting or
opposing projects. The second was the development of a baseline data:set which
was large enough and homogeneous enough to provide a reliable basis for es-

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 157

Table 1. Specifying and defining distributional influences.

Distributional influence Definition Variable

Expected power company Five years predi

electricity shortages expected capacity (installed capacity plus
capacity that is under construction)

Changing expectations about Trend growth of national government R ESng

national government electricity electricity shortages defined as above
shortages

Changing expectations about Trend growth in prefecture per capita income R IN

regional social and economic
opportunities

Autonomous ability of local Local Financial Index measured by the ratio LFI
government to supply public Of tax revenues from local sources to
goods expenditure on public goods

Importance of rural sector in Primary prefectural per capita product as a PC

generating regional product proportion of total per capita product

Importance of rural employment Persons employed in primary industry as a PE

opportunities in regional economy proportion of total persons employed

Leftist political party Japan Socialist and Communist party seats as LI

representation a proportion of total regional assembly seats

Social attitudes towards preser- Time period:

vation of environment SA= 0 if 63< SA <69 SA
SA= 1 if 70<SA<79

Risk associated with different Power plant category:

technologies F=0 if nuclear F
F= if fossil-fuelled

Familiarity with risk Package number: PN

PN=0 if initial

PN = 1 if subsequent

timating the relationships between the explanatory variables and public acc
tance times. The third was data availability.
Two model variants were considered in the development of a comprehens
model for analyzing the variance in public acceptance times. The first was
development of a model which considered how much of the variation in pu
acceptance times could be explained in terms of conditions at the point in

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
158

when a power company made a decision to locate a plant at a particular site,

given information about expectations at that time. This model took the value
of variables prior to and at the time of a decision to locate a power plant. The
second was the development of a model which took into account conditions that
emerged during the settlement. The model discussed in this essay combines these
two models and takes into account conditions at the beginning of the settlement
process as well as those that emerged during the course of settlement.
The method of estimation was step-wise regression. All the variables were en-
tered into the models and those which were not significant at the 0.005 level were
eliminated. Variables that measured the energy self-sufficiency ratios of prefec-
tures where projects were sited, the risks of power plants with different techno-
logical characteristics and the number of persons leaving local communities in
search of alternative employment opportunities were not found to be statistically
significant.
The analysis covers 48 large-scale power plants developed in Japan from 1960
to 1979, the period when the majority of Japan's power plants were constructed,
and seven nuclear plants which had not received approval by 1979. I obtained
data on leadtimes by sending questionnaires to all electric power companies in
Japan. I gathered social, economic and political data illustrating community
characteristics from published sources and from the Ministry of International
Trade and Industry. The Tokyo Electric Power Company provided data on elec-
tricity supply and demand. The analysis considered both nuclear and fossil-
fuelled power plants. It also considered both initial projects, defined as projects
located at sites where no other plants are in the construction stage or operating,
and subsequent projects, defined as those projects sited at locations which have
at least one plant being constructed or operating. The sample, therefore, pro-
vides extensive coverage of the siting problem in Japan.

The importance of distributional influences

Table 2 illustrates the importance of structural distributional influences on ap-

proval times. The model explains 74.5% of the variation in public acceptance
times. The F-ratio, a test on the joint statistical significance of all the co-
efficients, provides evidence that the null hypotheses of no relationship between
public acceptance times and the explanatory variables can be rejected. These
results suggest that the model can be used in the evaluation of settlement times.
The model includes variables which represent proxies of social, economic and
political conditions at the beginning of the settlement process as well as variables
illustrating conditions which emerge during the course of negotiations. Ap-
proval times can be predicted to a certain extent at the commencement of negoti-
ations by considering the financial strength of electorates, the importance of

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 159

Table 2. Factors influencing public acceptance times.

Variable Proxy measure Estimated T-ratio

co-efficienta,b

ESPc Average expected power company electricity shortages 0.26 x 10-3 11.10
R-ESng Trend growthc in expected national government
regional electricity sphere shortages -0.02 14.08
R-IN Trend growth in prefectural per capita income - 1.23 31.27
R-IN*F Interaction of R-IN with fuel category (F) 0.91 13.40
F = 0 if nuclear, F= 1 if fossil-fuelled
P(LFI*PN)d Interaction of Local Financial Index with package
number (PN) - 0.51 5.02
PN = 0 if initial, PN = 1 if sub
P(PC*F) Interaction of primary
proportion of total per capita product with F -8.79 10.80
PE*PN Interaction of persons employed in primary industry
as a ratio of total persons employed with PN 4.80 6.90
P(LI*F) Interaction of ratio of JSP and JCP assembly seats to
total assembly seats with F -3.61 9.02
SA SA =0 if 63 < SA < 69 1.22 24.00
SA= 1 if 70< SA < 79
SA*PN Interaction of SA with PN -0.01 6.69

Constant 4.98

R- Square 0.74
Standard Error 0.56
F-Ratio 10.55
Number of observations 48

Notes: a) The constant and estimated co-eff

explanatory variable; b) All co-efficients ar
prefixed by R. indicate the trend growth
period; d) Variables prefixed by P indicate
sion to locate a power plant in a particula

primary sector product and the str

can be improved by examining expec
munity incomes, the importance of
and social attitudes toward environm
tlement. Although there is some pre
of the settlement process, there is a
ing expectations during the negot
times.

The strengths and signs of the co-e

predictability in the pattern of res
terests and the impact of those re

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
160

will be conditioned by expectations about the structure of distributional impacts

and will be related to the need for promoters to meet expected electricity short-
ages, the structure of the regional political economy and prevailing community
attitudes towards the environment.
Expected electricity shortages are likely to be important in determining ap-
proval times. As expected, the negative co-efficients on ESPC and R.ESng indi-
cate that settlement times are likely to be shorter, other things being equal, in
situations where power companies and the national government9 judge that
forecast electricity shortages will be relatively large. Promoters will place more
value on the benefits of developing projects. At the same time, they will attach
less priority to the costs, such as capital costs and excess capacity, associated
with the construction of projects. Promoters who place a high value on the im-
plementation of projects are more likely to be willing to provide necessary com-
pensation to community interests.
Even in situations where electricity demand growth is high and promoters are
capable of providing compensation requirements, the structure of distributional
impacts on community interests will influence the effectiveness of compensation
mechanisms and, therefore, the speed at which settlements can be reached. The
evaluation of approval times also requires analyzing the impacts of power plants
on community interests and examining and monitoring their responses carefully.
An interesting result of the analysis is that power plants appear to be more
readily accepted in communities where per capita incomes are rising rapidly, and
this if inferred from the negative relationship between R.IN and settlement
times. This conclusion does not support the generally accepted hypothesis in en-
vironmental economics that the demand for environmental quality is likely to
be low and, hence, that there is likely to be less resistance to hazardous projects
in areas where incomes are low or are expected to remain low. 10 A boom econo-
my mentality usually emerges under conditions of rapid income expansion, and
there is generally a lag in noticing and anticipating any adverse consequences
associated with the development of projects. Communities are likely to wish to
see continued expansion of social and economic opportunities and, consequent-
ly, are likely to offer less resistance to the siting of power plants.
As suggested by the positive co-efficient on R.IN*F, the importance attached
to environmental quality, however, appears to be a stronger influence on length-
ening approval times for fossil-fuelled compared to nuclear projects. In Japan,
as in many other industrialized countries, fossil-fuelled projects tend to be locat-
ed close to major consumption areas which, in addition to having relatively high
incomes, also have comparatively high levels of pollution. In contrast, nuclear
projects are generally located away from population centers.1 Communities
with high incomes are likely to be less enthusiastic about accepting fossil-fuelled
projects which may be seen to add to existing pollution problems. An important
implication of this result, which counters the conventional wisdom on risk, is

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 161

that the risk of nuclear plants is likely to be given less weight by communities
experiencing rapid income expansion.
The autonomous ability of local governments to supply public goods is a
stronger influence in shortening settlement times in the case of subsequent
projects relative to initial ones, and this is reflected in the negative co-efficient
on P(LFI*PN). The development of initial projects expands local tax and other
revenues which are used to develop, amongst other things, social overhead capi-
tal. These financial benefits decline after the construction of projects and, fur-
thermore, maintenance costs can impose a severe drain on financial reserves.
Communities are, therefore, more willing to accept subsequent projects in order
to facilitate the maintenance of existing social capital and the continued expan-
sion of that capital. At the same time, they are likely to become more familiar
with having power plants constructed and operating in their locality.12 Conse-
quently, approval times are likely to be shorter for subsequent projects. This is
likely to be one important factor explaining the concentration of power plants
at specific locations in most industrialized countries.13
The importance of rural product in the regional economy (PC) also appears
to be a significant factor explaining delay. As suggested by the negative sign on
P(PC*F), other things being equal, settlement times seem to be longer for nucle-
ar compared to fossil-fuelled power plants in areas where the rural sector is more
important in generating regional product. Under these conditions, there appears
to be more sensitivity amongst the rural sector to the severity of risks associated
with possible nuclear accidents and the discharge of radioactive substances.
The proportion of persons employed in the primary sector to total persons
employed (PE) will also condition the response of the rural sector to accepting
projects. There is likely to be more resistance to subsequent compared to initial
projects in situations where the rural sector generates relatively more employ-
ment opportunities, and this is indicated by the negative relationship between
PE*PN and settlement times. Anxiety within the rural sector appears to increase
with the continued development of projects and the possibility that subsequent
projects will cause further environmental degradation to factors of production,
such as land and Water, and increase the risk of accidents.
The rural sector appears to be more concerned about risky projects than the
community in general. Opposition by primary producers is an important feature
of siting disputes in both the United States and Europe. For example, there have
been many cases of opposition to energy plants by wine producers in West Ger-
many.14 The rural sector appears to be more reluctant to accept increasingly
risky projects because they generally perceive that the risks will fall directly on
them. This worry was illustrated clearly by the impact of the recent Chernobyl
nuclear accident on agricultural markets in Europe and North-east Asia.
The proportion of leftist political party assembly seats to total assembly seats
(LI) will also influence bargaining processes over the development of projects.

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
162

The strength of leftist political party representation appears to be a stronger in-

fluence in lengthening bargaining processes for nuclear compared to fossil-
fuelled plants. The negative co-efficient on P(LI*F) suggests that, in Japan, left-
ist political parties tend to place more weight on the ideological costs of nuclear
power plants and less on the social and economic benefits associated with those
developments. They usually attempt to sensitize the public to the risks of nuclear
energy and encourage antinuclear movements.
Prevailing community attitudes towards preservation of the environment and
improved social welfare (SA) also need to be considered in the evaluation of set-
tlement times. As implied by the negative sign on SA, changes in community
attitudes in the late 1960s in Japan, as in other industrialized nations, to placing
a stronger emphasis on welfare rather than on material benefits derived from
high economic growth were associated with the lengthening of settlement times.
In Japan, the effect of this transformation in attitudes appears to have been less
important in delaying approval for subsequent relative to initial projects, and
this is inferred from the negative co-efficient on SA*PN. Although accidents and
pollution problems emerged, communities tended to become more accustomed
to the risks of having power stations operating in their vicinity. Consequently,
there appears to have been relatively less resistance to accepting subsequent
projects.
An important conclusion with respect to community responses to the riskiness
of power plants is that it is not the nature of the technology itself that is the
critical determinant of settlement times. There were no statistically significant
relationships between settlement times and fuel category (F) and package num-
ber (PN). Rather it suggests that the evidence presented does not support the
commonly held belief that nuclear power plants will always take longer to site
than fossil-fuelled projects because of the inherent nature of nuclear risk, nor
the belief that initial projects take longer than subsequent projects because of
the unfamiliarity with a risky environment. The importance attached to the risk-
iness of projects with different characteristics will depend on social, economic
and political conditions. Hence, the evaluation of community responses towards
power plants requires going beyond the technological features of those plants
and examining the political and economic structure of the relevant community.

Predicting times for unlicensed nuclear plants

Although there appears to be a considerable amount of predictability in ap-

proval times, it is important to test the predictive power of the model. I used
the estimated co-efficients to predict times for seven nuclear power plants that
had not received approval by 1979, the cut-off point for the analysis, and com-
pared them with actual public acceptance times defined as the time from when

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 163

a power company made a decision to locate a plant to 1979. The magnitude and
sign of the residual, the difference between actual and predicted times, provides
information about the predictive capacity of the model. The model can be said
to be useful in making predictions if the residual is large and negative. This would
imply that the model would have predicted approval at a time substantially after
1979. (Thus, my operational definition of 'actual acceptance time', while
perhaps different from the common, conversational interpretation, is a useful
and coherent notion in the context of the statistical analysis below.)
The treatment of unapproved power plants presents difficulties for the statisti-
cal analysis. Exclusion of them from the regression model does create a sampling
bias by not considering projects that have not yet been approved. On the other
hand, inclusion of them is not possible because the public acceptance times
would actually be infinity and would also distort the results. The most appropri-
ate way of incorporating them into the analysis is to exclude them from the model
and then use the model to generate predicted times for them. This is beneficial
because it does allow for their use in testing the predictive power of the model.
Table 3 presents the results of this statistical test. The model predicted a large
and negative residual for the Suzu (-102 months) and Hikigawa (-124 months)
projects. The residuals were relatively smaller but positive for the Nachi-Katsura
(7 months) and Kumano (14 months) projects, while they were somewhat larger
for the Namie (41 months) and Noto (33 months) projects. The residual was very
large for the Ashihama project (92 months) and the model would have provided
an inadequate basis for predicting the approval time for this project. All of these
projects were more problematic than those considered in the regression sample.
An important feature of the siting process at these sites was powerful opposition
by vested rural interest groups which caused delay in the development of these
projects.

Table 3. Predicting public acceptance times for unlicensed nuclear power plants.

Plant name period Actual PATa,b Model

Predicted Residual

Namie 1967- 1979 140 99 41

Noto 1967- 1979 140 107 33
Kumano 1967- 1979 140 126 14
Ashihama 1963- 1979 179 87 92
Suzu 1975- 1979 45 147 - 102
Nachi-Katsura 1969- 1979 110 103 7
Hikigawa 1976- 1979 30 154 - 124

Notes: a. Public acceptance times (PAT) a

times are calculated using 1979 as a cut-of

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
164

An important caveat of the model is that it tends to underestimate approval

times for projects that faced negotiating difficulties in the earlier period of the
analysis. When promoters are confronted with strong resistance at particular
sites, they are likely to change the relative priority rankings of these sites. They
may either intentionally slow down negotiating processes at these sites or reas-
sign these sites to candidate site status and give priority to other less difficult
sites in order to accomplish electricity supply objectives. Although this bias ex-
ists and needs to be taken into account in evaluating the results, the model, in
general, appears to provide reasonable estimates of approval times.

Interest groups and compensation management

The accurate assessment of public acceptance times does not end with generating
estimates from an econometric model. The residual varies quite significantly,
and this implies that the model can either over- or underestimate approval times,
even though it provides a relatively good fit in general. As expected, quantitative
approaches make a number of important assumptions about the nature and im-
pact of other factors, such as uncertainty about outcomes, the structure of bar-
gaining power, the composition and structure of interest groups, the effective-
ness of bargaining strategies and changing expectations due to sudden changes
in social and economic conditions, which are not easily susceptible to statistical
measurement, yet will influence negotiating processes in a complex way, For ex-
ample, the variables used in the estimation were aggregated over time, and this
ignores dynamic factors such as sudden shifts in economic conditions which can
clearly affect the reaction of local populations to hazardous projects. Although
there are constraints to dissagregation in any statistical analysis, such site-
specific characteristics can increase the variance of the data set being used.
Hence, econometric models need to be supplemented with an analysis of such
qualitative influences in order to provide a more comprehensive framework for
evaluating public acceptance times.
A case study will assist in understanding some of the sources of the unex-
plained variation in approval times and providing insights into other social con-
troversies and project management. The study selected is the protracted dispute
over the siting of the Ashihama nuclear power plant in 1963 in Mie Prefecture
by the Chubu Electric Power Company. 15 In this case, the distribution of costs
and benefits were structured in a way which, according to the model, would have
resulted in a settlement time of 87 months, a little longer than the average for
initial nuclear power plants in Japan (82 months).16 Electricity shortages were
not expected to be high when the power company decided to locate the project
or during the attempted negotiating period. Despite community attitudes that
placed little emphasis on preserving the environment, other factors were operat-

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 165

ing to inhibit the approval process. There were not strong expectations for the
expansion of social and economic opportunities. Furthermore, the regional
economy was predominantly rural and there was concern about the risks of an
initial nuclear power plant to this primary sector on income and employment
opportunities.
An assessment of these distributional influences and the use of the model
would have provided a markedly better estimate of delay than Chubu Electric's
estimate (25 months). Although the model could have been used as a startin
point for predicting the delay that was encountered, it would have been als
necessary to consider more fully the composition of interests groups who were
expected to be affected adversely by the project, changing economic conditions
the group's ability to organize and sustain opposition to the project, their strate-
gies during the conflict and the inability of the power company to manage effec
tively compensation requirements. In addition, it would also have been essential
to monitor the increased negotiating costs of the project to the power company
that eventually forced it to shelve the project, at least temporarily.
In early 1963, Chubu Electric, after completing preliminary investigations,
decided internally to locate a nuclear project at Ashihama, a small hamlet i
Nanto town in Mie Prefecture. These investigations revealed that the physical
characteristics of the site, such as the availability of flat and stable terrain, th
adequacy of cooling water, and the accessibility of transportation routes, were
suitable for the construction of a nuclear power plant. Furthermore, they sug-
gested that the productive capacity of the land upon which the plant would be
built was extremely low and that only one small fishing co-operative to the sout
would be disadvantaged by the project. Chubu Electric judged that this co
operative could be compensated easily and that approval for the project would
be relatively quick.
After announcing plans to develop the project, the power company was sur-
prised to see the emergence of powerful opposition to the nuclear plant from
an alliance of local fishing interests. During the 1950s, there was a small fishin
industry in Nanto and the productive capacity of this industry was very low.
However, the Nanto area became famous for the production of pearls in the first
half of the 1960s and supplied the Mikimoto Pearl Company with the majority
of its pearls. During this time, Nanto fishing co-operatives were experiencing
a boom in the production of pearls and, by 1965, the value of production h
increased to $ 100 million annually. As a result, a large proportion of the popul
tion in Nanto was drawn into the industry. This sudden change in economic cir
cumstances conditioned strongly the response-of the industry to the project. Th
prospect of having a nuclear plant led to substantial concern about the discharg
of waste water from the project and the negative impacts on production and em-
ployment prospects.
The structure of economic and social relationships within the local industry

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
166

also facilitated the formation of the alliance. In Nanto, pearl cultivation was
characterized by specialization in the production process. Seven co-operatives
were engaged, some cultivating the seedlings and others growing the oyster
shells. Although the power plant was only expected to discharge waste water in
the vicinity of one of the smaller co-operatives, the economic structure of the
industry meant that the other six larger co-operatives, even though they were
located some distance from the project, would also be affected. The costs of the
project, therefore, fell equally across all the co-operatives in the industry.
A second important characteristic of the local pearl industry was a strong net-
work of inter-personal relationships which developed as a consequence of a tra-
dition of inter-co-operative marriages. This tradition created strong social rela-
tionships and was crucial in organizing and sustaining opposition in the 1800s
to rival co-operatives from other regions which attempted to poach in the Nanto
area. The structure of social relationships also facilitated the formation of the
alliance which sustained strong resistance to the project.
Confronted by this unexpected opposition, Chubu Electric, like project
promoters in other democratic countries, attempted to investigate the environ-
mental impact of the project on the local fishing industry. It persuaded the
prefecture to commission a study by a team of marine experts. The report was
favourable to the power company's position, stating that the discharge of waste
water would only increase marginally the temperature of the ocean in the vicinity
of the outlet pipes and that this would not adversely affect the fishing industry.
It concluded that the promoters should discuss the results of the report with
representatives of the fishing industry so that further investigations could take
place.
The seven fishing co-operatives rejected this proposal and issued a declaration
that contained three major criticisms of the study. First, they argued that the
experts had been personally selected by the prefectural governor on the presump-
tion that the project would go ahead. They rejected the study's results, arguing
they would only provide background information that would be used to promote
the project. Second, they argued that it was impossible to obtain even intermedi-
ate results in a two-month period. Any investigation, they concluded, should
take at least a year on the grounds that the impact of waste water may differ
according to varying ocean temperatures and currents in different seasons.
Third, they criticized the report for not considering explicitly the impact of waste
water on pearl cultivation.
The pearl industry's response put Chubu Electric in a very difficult position.
The power company did not wish to pay a large amount of compensation for
fear that a huge outlay would decrease the economic viability of the project. Fur-
thermore, it was worried that it would create a precedent for large compensation
payments and thus invite criticism from other power companies that were either
negotiating settlements or intending to enter into them. The power company was

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 167

constrained heavily in the management of its compensation policy and felt that
it could not negotiate unilaterally with the pearl industry.
It decided to ask the prefecture to provide a subsidy for regional development.
In late 1965, the prefectural government announced that it would offer a subsidy
of $90 million to Nanto and other surrounding towns for the development of
social and economic infrastructure such as roads and hospitals. Chubu Electric
hoped that this would foster economic and political pressure on the pearl indus-
try by increasing the benefits of the project to the local population and surround-
ing communities. It felt that engendering support for the project would weaken
the bargaining position of the fishing alliance.
The fishing industry declined the offer of subsidy. It argued that the subsidy,
while providing benefits to the regional community, did not offer direct benefits
to the fishing interests, who were expected to incur most of the costs of the
project development. Moreover, they were angered at the power company's
strategy of persuading the prefecture to exert pressure on them via other regional
interests in order to isolate them politically and economically.
The fishing co-operatives responded by protesting to and lobbying political
interests at both the prefectural and local levels, strategies commonly employed
in social conflicts. More specifically, they sought to politicize the issue at the
prefectural level and gain electoral access to government on the local level. Local
pearl interests appealed to prefectural assemblymen who were sensitive to their
claims. A number of assemblymen relied on the fishing sector for electoral sup-
port and were concerned about the possible loss of this support in the forthcom-
ing elections. Furthermore, the pearl industry had become an important pillar
in the economic development of the prefecture. These tactics were successful and
the prefecture subsequently voiced opposition to the project.
The mayor of Nanto, who was supported by other commercial interests, was
eager to promote the project in order to expand the tertiary sector in the town.
The fishing co-operatives appealed to local assemblymen and the public, arguing
that the nature and extent of benefits from the project, even taking into account
the subsidy, were uncertain and that there were substantial risks involved in giv-
ing their consent to the project development. Fishing interests subsequently
commenced a recall movement that ousted the mayor. Approximately ninety per
cent of the new assemblymen represented the pearl industry, and the assembly
subsequently announced a town policy of rejecting the project.
The delay encountered in developing the Ashihama project was more than the
time predicted by the model because Chubu Electric faced uniquely intense op-
position by local economic interest groups who expected to incur large losses
and, as a result, formed an alliance to oppose the project. Chubu Electric mis-
managed its negotiating policy by trying to isolate the fishing co-operatives and
exert pressure to persuade them to come to the bargaining table. Fishing interests
responded by persuading the prefectural government to oppose the project and

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
168

by gaining electoral access to local government. From the power company per-
spective, the economic and political costs of negotiating a settlement had be-
come unacceptably high, and the company decided to shelve temporarily the
project in favour of an alternative and less costly location, thus delaying the
project even more.
This dispute offers important insights into project evaluation and manage-
ment. First, developers need to consider explicitly compensation requirements
at the evaluation stage. Site selection not only requires ascertaining the physical
characteristics of the site, but also the structure of the social and political en-
vironment. In particular, developers must consider the composition of interest
groups who will be affected adversely, their ability to sustain opposition and the
possible strategies they will employ to improve their bargaining positions. Sec-
ond, when confronted with strong opposition, promoters are likely to intensify
conflict if they do not manage compensation policy effectively. The payment
of compensation needs to offset directly the costs that strong and powerful in-
terest groups are expected to incur as a result of the project. An indirect approach
to compensation management is likely to strengthen resistance to project de-
velopment.

Conclusions

The determinants of delay, as identified in this analysis, are likely to have br

applicability to the understanding of other social conflicts in pluralist dem
ic countries. Where projects confer externalities on different interests, the re
tion of conflicts will require a bargaining process which will determine who
and who loses from a particular project development. Hence, the distributio
impacts, the structure of the bargaining environment, the strategies emplo
uncertainty about outcomes, changing expectations and the effectiveness of
isting or established compensation mechanisms will be critical in determini
the outcomes of social controversies.
The structure of distributional impacts is an important factor explaining delay
in the siting of power plants in Japan. The model explains a relatively high
proportion of the variation in approval times without recourse to unique social
and cultural features of negotiating processes in Japan. These distributional in-
fluences are generally prevalent in siting disputes in democratic countries, sug-
gesting that Japanese experience in this area can assist in a more general under-
standing of the determinants of facility siting and negotiation outcomes. Socio-
cultural aspects of social choice processes appear to be less important than might
be inferred from casual empiricism in determining facility siting outcomes in
Japan.
The case study analysis highlighted the importance of strategies employed in

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 169

negotiations around compensation requirements. The strategies used in Japan

are ones commonly employed in other industrialized countries to either impede
or facilitate the resolution of conflicts. The opposition formed an alliance that
sought to improve its bargaining position more effectively by joining forces17
and used protest as a political resource to appeal to key target groups such as
the prefectural government and the local community.18 The supporters con-
ducted a marine investigation that attempted to use scientific expertise to per-
suade local interests that the costs of the nuclear project were not as high as they
had argued.19 Furthermore, they attempted to bring third parties, such as the
prefectural government, into the dispute in order to exert economic and political
pressure on the opposition and isolate them politically.20
The approach articulated in this article raises important issues regarding the
relationship between cost-benefit analysis (CBA) and project delay and its
management. Standard CBA ranks projects in terms of economic efficiency and
this is useful for identifying high return projects.21 This approach ignores the
importance of distributional influences, bargaining processes and the manage-
ment of compensation requirements and, therefore, presumes that the costs in-
volved in winning approval for projects are not relevant to benefit-cost calcula-
tions. It presumes that redistributive mechanisms are operating adequately, and
therefore implies zero implementation costs in reaching settlements over the de-
velopment of projects. Policy makers will, however, be interested in the im-
plementation costs of projects, as this will influence their ability to achieve ob-
jectives which require the development of those projects. The costs associated
with public acceptance of projects need to be taken into account, either quantita-
tively or qualitatively, and subtracted from net discounted project benefits in
the evaluation stage.
The approach provides a way for policy makers to consider more effectively
the sources of delay in reaching public acceptance for competing projects. The
development of models which can generate reasonable estimates of approval
times for complex bargaining processes with a relatively small number of varia-
bles could be used as an initial screening device for the selection of low resistance
sites. The use of other non-statistical techniques would be required in order to
make final decisions about the selection of particular sites. These more detailed
investigations about the composition of interest groups, the spread of costs and
benefits amongst them, their ability to sustain resistance to or support for
projects, their likely strategies and ways to manage compensation requirements
would be best done once the number of candidate sites was sufficiently nar-
rowed. Such an approach, which could provide more information about expect-
ed leadtimes, could assist in reducing the uncertainty that policy makers con-
front in making decisions.
Although this approach has been successful in the case of Japan, the co-
efficients of the model could not be exactly replicated in other social controver-

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
170

sies either in different time periods in Japan or in other democratic countries.

There are likely to be different institutions, different interests and different
decision-making procedures. Furthermore, the definition of social and econom-
ic statistics may vary. However, there is likely to be a set of variables which could
be identified and analyzed using the general framework developed in this essay.
Delay is a common characteristic of all collective action. The Japanese ex-
perience in this area enhances our understanding of the process, the sources of
delay and ways to manage it. It suggests that there is likely to be a set of more
global determinants which will assist in explaining why some social controversies
take longer to resolve than others. The challenge is to use the approach adopted
in this study as a starting point for the development of comparative analyzes
of this very interesting and important policy problem and its management.

Acknowledgements

The author wishes to thank Peter Drysdale, Ben Smith, Nancy Viviani, Stuart
Harris, Colin McKenzie, Peter Davis, Robert Cribb and two anonymous referees
for useful suggestions on both this and earlier drafts.

Notes and references

1. See, for example, Interview: John G. McLillan (1979). 'Delays push Alaska Highway line
up to $ 15 billion,' Pipeline Industry 51: 29- 31; 'Hand drilling delay becomes more expensiv
The Oil and Gas Journal 77 (30): 48- 51; Rad, Parviz F (1979). 'Delays in Construction of Nu
clear Power Plants,' Journal of Energy Division 105 (1): 33 - 46; Samson, Robert C. and Viole
Dan (1979). 'Refinery permit delays evaluated,' The Oil and Gas Journal 77 (17): 78- 83; Qui
Jones, and Terasawa Katsuaki (1981). 'Nuclear Regulation: An Historical Perspective,' Natur
Resources Journal 21: 833-855; and Reed, Krista S. and Young, C. Edwin (1983). 'Impact
Regulatory Delays on the Cost of Wastewater Treatment Plants,' LandEconomics 59 (1): 35 -
For a good review of the literature on project siting problems, see Hamilton, Michael S. (197
'Power Plant Siting: A Literature Review,' Natural Resources Journal 19 (1): 75-95.
2. For a detailed analysis, see Lesbirel, S. Hayden (1985). 'The Political Economy of Delay: Sitin
Energy Power Plants in Japan,' unpublished PhD dissertation, Australian National Universit
Canberra.

3. For a brief review of compensation and financial schemes operating in industrialized countries,
see IEA, Electricity in IEA Countries: Issues and Outlook (1985). Paris: OECD-IEA; see espe-
cially chapter VII.
4. See Cordes, Joseph J. and Weisbrod, Burton A. (1985). 'When Government Programs Create
Inequities: A Guide to Compensation Policies,' Journal of Policy Analysis and Management
4 (2): 181-183.
5. O'Hare, Michael (1977). 'Not on My Block You Don't... Facility Siting and the Strategic Impor-
tance of Compensation,' Public Policy 24(4): 428-429.
6. See Lesbirel, op. cit., chapters 4 and 5.
7. See Baccow, Lawrence S., and Sanderson, Debra R. (1980). 'Facility Siting and Compensation:
A Handbook for Communities and Developers,' MIT Energy Laboratory Working Paper (MIT-

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms
 171

EL, 80-037WD): 99-100, and Jopling, David G., Gage, Stephen J. and Schoeman, Milton E.F.
(1973). 'Forecasting Public Resistance to Technology: The Example of Nuclear Power Reactor
Siting', in: Bright, James R. and Schoeman, Milton E.F. (eds). A Guide to Practical Technolog-
ical Forecasting. New Jersey: Prentice-Hall Inc., pp. 53-66.
8. For an analysis of this view, see Stockwin, Arthur (1982). Japan: Divided Politics in a Growth
Economy. London: Weidenfeld and Nicholson, pp. 25-41.
9. There are nine power companies in Japan. Each of these companies has a monopoly over supply
to a particular electricity sphere and is required by law to supply electricity to meet demand.
From the perspective of the national government, Japan is divided into eastern, central and west-
ern spheres. Three power companies belong to each of these spheres. The national government
is concerned with balancing electricity supply and demand in each of these broader regional
spheres.
10. Bamuol, William J., and Oates, Wallace E. (1975). The Theory of Environmental Policy: Exter-
nalities, Public Outlays, and Environmental Policy. New Jersey: Prentice-Hall Inc., p. 191.
1. See IAEA (1974). The Siting of Nuclear Facilities-Proceedings of a Symposium. Vienna: OECD.
12. See Lawrence, William, W. (1970). Of Acceptable Risk: Science and the Determination of Safe-
ty. Los Altos: William Kaufman Inc., for an excellent analytic discussion of risk.
13. See Japan Atomic Industrial Forum (JAIF) (1978). Japan's Nuclear Power Plants for informa-
tion on the location of nuclear and other power plants in countries developing nuclear power.
14. Franzen, F.S. (1976). 'Socio-economic Issues for Nuclear Plants: The German Situation,' Trans
American Nuclear Society 24: 99.
15. This discussion in this section draws heavily on Lesbirel, op. cit., chapter 5.
16. See Lesbirel, S. Hayden (1985). 'Energy Power Plant Leadtimes in Japan,' Pacific Economic
Papers 118: 27. Canberra: Australia-Japan Research Centre; reprinted as 'Nihon ni okeru hat-
suden sho no ridotaimu,' Enervgii keizai [The Journal of Energy Economics]: 120-137.
17. Groennings, Sven, Kelley, E.W. and Leiserson, Michael (eds) (1970). The Study of Coalition Be-
haviour: Theoretical Perspectives and Cases from Four Continents. New York: Holt, Rinehart
and Winston p. 450; and Kelley, E.W. (1968). 'Techniques for Studying Coalition Formation,'
Midwest Journal of Political Science 12(1): 62-84.
18. See Lipsky, Michael (1968). 'Protest as a Political Resource,' American Political Science
Review 6(2): 1144-1158.
19. Nelkin, Dorothy (1975). 'The Political Impact of Technical Expertise,' Social Studies of Science
5: 35-54.

20. Schattschneider, Edwin, E. (1960). The Semi-Sovereign People. New York: Holt, R
Winston.

21. For example, see Pearce, David W. (1971). Cost-Benefit Analysis. London: MacMillan; and Mi-
shan, Edward J (1972). Cost-Benefit Analysis: An Informal Introduction. London: Allen and
Unwin. For a survey of the literature on CBA, see Prest, Alan R., and Turvey, Ralph (1965).
'Cost-Benefit Analysis: A Survey,' Economic Journal 75: 683-735.

This content downloaded from 136.159.235.223 on Sun, 31 Jul 2016 02:16:07 UTC
All use subject to http://about.jstor.org/terms