WPS4331


Policy ReseaRch WoRking PaPeR                     4331




    Infrastructure Governance and Corruption:

                               Where Next?

                                  Charles Kenny




The World Bank
Sustainable Development Network
Finance, Economics and Urban Division
August 2007

Policy ReseaRch WoRking PaPeR 4331


 Abstract

 Governance is central to development outcomes in                                      provides some recent examples of reform efforts and
 infrastructure, not least because corruption (a symptom                               project approaches, and suggests an agenda for greater
 of failed governance) can have significantly negative                                 engagement--primarily at the sector level--to improve
 impact on returns to infrastructure investment. This                                  governance and reduce the development impact of
 conclusion holds whether infrastructure is in private                                 corruption. The discussion covers market structure,
 or public hands. This paper looks at what has been                                    regulation, state-owned enterprise reform, planning and
 learned about the role of governance in infrastructure,                               budgeting, and project design.




 Thispaper--aproductoftheFinance,EconomicsandUrbanDivisionoftheSustainableDevelopmentNetwork--ispartof
 a larger effort in the department to research issues connected with governance and corruption in sustainable development.
 Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted
 at ckenny@worldbank.org.




          The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development
          issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the
          names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those
          of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and
          its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.


                                                          Produced by the Research Support Team

                Infrastructure Governance and Corruption: Where Next?

                                           Charles Kenny1




1Senior Economist, FEU, ckenny@worldbank.org. Thanks to Jon Halpern for two rounds of detailed
comments and Maria Vagliasindi for extensive comments and suggestions on the SOE section. The views
and remaining errors are those of the author, not necessarily of the World Bank, its Executive Directors, or
the Countries that they represent.

Introduction

Governance of infrastructure is hardly a new topic for developing countries or donors.
The 1994 World Bank World Development Report on infrastructure tackled a range of
issues including public and private roles in provision, regulation and management of state
owned enterprises. It highlighted the vital importance of governance and suggested many
approaches repeated in this paper --most of the differences between the two are a matter
of degree rather than direction. This suggests caution in assuming we can speed progress
in infrastructure rollout or efficiency gains based on dramatic new insights from the last
thirteen years. Indeed, given the complexity of institutional reform and the slow pace of
improvements in governance, progress is likely to be incremental. Nonetheless, a new
pragmatism regarding market structures combined with evidence from approaches tried
over the last ten years does provide some guidance. And precisely because many
approaches are yet to be rigorously evaluated and our knowledge base still remains
limited, we should promote risk-taking in and evaluation of projects. This paper
discusses what we have learned about the role of governance in infrastructure, provides
some recent examples of reform efforts and project approaches, and suggests an agenda
for greater engagement primarily at the sector level to improve governance and reduce
the development impact of corruption.


A New Pragmatism in Infrastructure

The last twenty years has seen considerable change in the delivery of infrastructure
services worldwide. The 1980s and 90s saw developing countries, the World Bank and
other donors disappointed with the sustainability of their investments in state-owned
infrastructure --concerned that investment was not having a long-term impact of the
quality and extent of utility and transport services that they desired. Privatization and
competition were proposed responses to this situation, and they have had some successes.
In some sectors, private competitive provision has become the norm. 80 percent of the
world's countries have three or more competing providers of mobile phone services, for
example (World Bank, 2005). Across sectors, there is evidence that private provision can
provide a boost to quality of services and extent of provision (Estache, 2006).

But at the same time it appears that many of the institutional weaknesses underlying poor
provision and low sustainability are not magicked away by allowing private investment,
and in some sectors private provision is likely to have a comparatively minor role. In
developing countries, only 5 percent of private investment in infrastructure has flowed to
water and sanitation over the last ten years, for example. Across infrastructure as a
whole, governments and donors still account for around four fifths of sector investment.                 2

And while private providers have, on average, improved service quality and access,
overall impact has been limited compared to initial hopes, there are numerous examples


2It is worth comparing estimated annual investment needs for energy, transport, water and sanitation
infrastructure in developing countries of $131bn a year to private investment commitments in those sectors
in developing countries estimated at $35 billion in 2005. Private investment figures from the PPI database,
investment needs from Fay and Yepes (2003).


                                                                                                         2

of failed provision, and increasing evidence of public disenchantment with private
providers (Estache, 2006).3 Although telecommunications provision has been
transformed, and at least in some markets the institutions of electricity production has
considerably altered, in other sectors the pace and success of market change has been
limited.

This mixed performance has led to a new pragmatism surrounding infrastructure delivery
in developing countries. This pragmatism involves an acceptance of the likely limits to
private provision under light regulation. In some cases, where competition is viable,
pricing is market-based and the payback time on investment is relatively short, a
liberalized private market has become the norm ­as with mobile telephony. In other
cases, where monopoly provision predominates, where pricing is politically sensitive and
investment payback periods are long, the public sector will frequently remain the
dominant force in the market.

This new pragmatism also involves an acceptance of the vital role for the institutions of
sector governance whether provision is in public or private hands. In a set of industries
where monopoly provision is likely to remain the usual model ­including transport
infrastructure as well as water and sanitation alongside electricity distribution--the role
of government remains vital whether services are provided privately or publicly. If there
is an attempt to attract private investors, in a set of industries where payback periods are
long, the institutional environment covering prices and the broader investment climate
will be a key to attracting finance. If infrastructure remains in the hands of the state, then
the quality of the government institutions managing the infrastructure stock will be vital
to determining outcomes.4 In short, the impact of infrastructure governance remains
regardless of the ownership or management control of assets.5




3Between 1998 and 2002, the percentage of people who disagreed with the statement privatization "has
been beneficial" for their country increased from an average of 45 to 77 percent across Argentina, Bolivia
and Peru (World Bank, 2006b).
4It is worth noting the contrast with mobile telephony in this regard. Because of the comparatively low
sunk costs of investment in mobile provision, private competition has proven itself to be robust even in
conflict and post-conflict environments of minimal governance such as Somalia and Afghanistan. Somalia
had no central government when mobile firms began to compete, and they quickly provided coverage and
pricing which were considerably better than regional averages (World Bank, 2005). Afghanistan only
passed a telecoms law creating a regulator after there were already two competing firms offering services to
over one million customers. Similarly, in countries that get out of the way, the international telephony
business is as competitive as the chocolate industry (Rossotto et. al., 2004). While there remain a number
of regulatory functions in the sector (covering interconnection, for example), the most important role for
governments in this sub-sector is to remove legal impediments on competition.
5The far greater importance of the quality of governance than the quantity of government is confirmed at
the macroeconomic level by growth regressions suggesting that the size of government as a percentage of
GDP is weakly related to growth while the strength of institutions is associated robustly (Kenny and
Williams, 2001, Easterly and Levine, 2002).


                                                                                                          3

Governance Matters for Outcomes in Infrastructure

We have considerable evidence on the importance of governance to investment returns
across sectors. For example, we know that investment in general in Sub-Saharan Africa,
where governance is comparatively weak, has had little impact on growth (Easterly, Pack
and Devarajan, 2003). This also implies a low return to aid flows --nearly two dozen
African countries receive aid worth more than 50 percent of total public expenditures
(Moss, 2005). In addition, we have considerable macro evidence that aid performs better
in the presence of strong institutions (Collier and Dehn, 2001).6

For donors, even where individual infrastructure projects provide strong microeconomic
returns in weak governance environments, their macro impact is reduced because projects
are `cherry-picked' and have little impact on the overall shape or size of a country's
investment program. Indeed, the weak correlation between aid flows and overall
investment suggests that aid may frequently displace local investment in strong projects,
and the local investment resources are instead consumed or invested in white elephants.7

The importance of institutions to infrastructure performance also emerges from an
analysis of the costs of corruption in the sector. Corruption is a symptom of failed
governance, but can also act to further weaken the governance environment. Corrupt acts
not only raise the price of infrastructure but can also reduce the quality of and economic
returns to infrastructure investment.

In Eastern Europe and Central Asia, construction firms report paying an average of 7
percent of government contract values in bribes to win bids or alter terms (Kenny, 2006).
Costs are also raised by the collusion between firms that these bribes can facilitate. Firms
pay additional bribes in meetings with labor and construction inspectors. These activities
will all raise the price of building new infrastructure. The average cost for a upgrading a
two-lane road with a bitumen surface across countries is around $33.2 per square meter.
For those countries with below average costs, the Transparency International Corruption




6Analysis of World Bank projects suggests the same thing (Dollar and Levin, 2005, Wane, 2004). The
extent to which institutional development matters to outcomes in infrastructure is suggested by an OED
(1999) evaluation of infrastructure projects in Africa, where only 18 percent of projects were considered
sustainable in the late 1990s: "infrastructure projects require a level of institutional capacity that simply
does not exist in many Sub-Saharan African countries," the report concluded. It should be noted that recent
scores for sustainability are considerably higher, the 2006 Annual Review of Development Effectiveness
(IEG, 2006) suggests that, across regions, 79 percent of infrastructure projects exiting 2001-2005 are
sustainable, compared to 56 percent in the period 1996-2000. The report also suggests that sustainability of
all projects in Africa has risen from 38.9 in 1996-2000 to 63.8 2001-2005.
7A recent metastudy of 43 papers examining the link between aid and accumulation found that, across the
studies, aid had a small, statistically insignificant positive effect on investment and a small, statistically
significant negative effect on savings (Doucouliagos and Paldam, 2005). Feyzioglu, Swaroop and Zhu
(1998) suggest that concessionary loans to a particular sector do not significantly increase spending in that
sector (see also World Bank, 1998). At the micro level, van de Walle et al (2005) do find that, depending
on their technique, a World Bank project's net contribution to rehabilitated road increments may be as high
as 66 percent (i.e., only one third of the aid displaced local spending).


                                                                                                               4

Perceptions Index averaged 3.6, compared to 2.4 in countries with above average costs ­
suggesting higher costs in countries perceived to be more corrupt (Kenny, 2007).8

Theft of materials and outputs is an additional economic cost of corruption. In Indonesia,
a physical audit of roads built under the oversight of village heads uncovered `missing
materials' worth 24 percent of total expenditures (Olken, 2004). In Bangladesh and
Orissa, in India, leakage due to illegal connections or underbilling accounts for as much
as 30 percent of generated power (Gulati and Rao, 2006). Similarly, Davis (2004)
suggests that unaccounted for water makes up 35 percent of total flows in India. Utilities
are also saddled with the cost of staff in place purely to extract rents. Overstaffing in
public utilities can be significant, as can bribe payments to obtain `ghost' employment
positions (Castalia, 2004).

Consumers pay additional bribes for connections, raising the final cost of infrastructure
services. The average firm in enterprise surveys of Eastern Europe and Central Asia
spends 11 percent of its bribe budget on facilitating connections to utilities, for example
(Kenny, 2006). Davis (2004) used a survey approach to estimate that the average speed
payment or bribe made to get connected to piped water in India works out at $2.64 per
legal customer. Davis also reports on the widespread practice of bribing for falsifying
meter-reading, which reduces the sustainability of service provision.

But the larger impact of corruption extends beyond inflated prices and staffing and the
financial cost of theft. In the case of the Indonesian roads project, for example, each
dollar's worth of stolen materials reduced returns to the project by $3.41, because the
poor construction which resulted significantly reduced road quality and life. Again, in
the Dabhol power project in India, Enron's $20m in payments for what was
euphemistically described as the `education and project development process' led to the
construction of a power plant that is financially unsustainable to run, at a cost of around
$3 billion (Kenny, 2007, see Box 1 for a discussion of the health impacts of corruption in
water).

As the Enron case suggests, failures of governance are often connected to mistargeted
spending. In Indonesia, at the start of the 1990s, road maintenance accounted for about
47 percent of the central government grants to districts for roads. By the end of the
decade this had dropped to 15 percent. Expenditures should be more than 17 times their
current level to maintain the existing road stock in good condition (Kenny, 2006). There
is anecdotal evidence connecting this fall in maintenance expenditure to the greater
opportunities for rent-seeking presented by new construction.

Finally, bribes to regulators can lead to the provision of low-quality services or over-
generous contract terms. Construction-related permitting appears to be a regulatory area
particularly prone to corruption, and this can carry a high human cost (Kenny, 2007). In
1999, more than half of all buildings in Turkey failed to comply with construction



8A regression analysis suggests that lower corruption is significantly associated with lower costs in this
sample, a result weakly robust to the inclusion of GDP ­see Kenny (2007) for more details.


                                                                                                          5

regulations. One result of this evasion was a considerable number of avoidable fatalities
in the 1999 earthquake in which 11,000 people died.


Box 1: Estimating The Health Impacts of Corruption in Water
A survey of evidence of corruption in Water provision in South Asia in 2001-2 suggested that contractors
frequently paid bribes to win contracts worth between one and 6 percent of contract values, followed by
kickbacks during construction of between 6-11 percent of the contract value, as well as forming
`sanctioned' cartels that raised prices by fifteen to 20 percent. In addition, the survey reported that
kickbacks went in part to cover low quality work, in which materials worth between 3 and 5 percent of the
contract value were not supplied (Davis, 2003). Assuming an economic impact of each dollar of such
missing materials of between $3-4 in terms of shorter life and limited capacity, this suggests an economic
cost equal to 9-20% of already inflated contract prices. These two forms of corruption together may have
raised the price of a sustainable water connection by 25 to 45 percent.

What is the economic and social cost of this corruption? An analysis of household survey data for 43
developing countries suggested that increased access to water had a significant impact on child mortality,
for example. Each additional percentage point of household access to water was associated with reduced
under-five mortality by one death for every two thousand children born (Liepziger et. al., 2003). Best
practice prices for a household water connection are around $400 (Fay and Yepes, 2003). If this price is
raised by 45 percent to $580 because of corruption, fewer households can be connected.

Given a cost of $400 per household connection, each $1m investment in low income country piped water
projects might save nineteen children a year who would otherwise die from water-related disease.9 At a
cost of $580 per connection, $1m in water investments might save only 13 children each year. Over twenty
years, additional costs imposed by corruption would mean that an additional 113 children die before the age
of five for each $1m invested in water projects. One recent estimate suggested that low income countries
could invest $29 billion in water projects over the ten-year period 2000-2010 (Fay and Yepes, 2003). If all
of this investment took place in a low-corruption environment, this might lower the global toll of child
deaths by over 540,000 each year, compared to 375,000 children if all of the investment took place in a
higher-corruption environment.

This is a very partial accounting of the impact of corruption in the sector. Limited household access to
water has impacts beyond increased childhood mortality ­it increases illness and death amongst older
children and adults as well. This has follow-on effects in terms of educational outcomes and income
generation. Corruption and weak governance also lowers overall sector investment ­not least by deterring
private finance of rollout. And limited access is only one impact of corruption, which also increases the
cost and reduces the quality of water provision. Nonetheless, even as a partial and approximate indication
of the impact of weak governance in the sector, the access and child mortality calculation suggests that
corruption in water is a matter of life and death.



Failed governance can lead to the construction of the wrong infrastructure, poor
construction and quality of provision, insufficient maintenance and high levels of thefts
and losses, then. This can dramatically reduce economic returns not just to individual
projects but to the entire infrastructure stock, and lead to lower levels and less efficient
provision of infrastructure provision ­a result that has been empirically illustrated in a
number of studies (Estache and Kouassi, 2002, Kaufmann et. al. 2005, Castalia, 2004, Bo
and Rossi, 2006). Box 2 discusses the example of corruption in Jakarta's water supply
system.

9Based on an average household size of five people and a crude birth rate of 30 per 1000 people (the
average for low income countries). The exact estimates are 18.75 and 12.93 deaths averted, respectively.


                                                                                                          6

Box 2: Ten Years of Governance Weaknesses in Jakarta's Water Supply
In the late 1980s, only 14 percent of Jakarta's population lived in households connected to the municipal
water system. Fully one third relied on water from street vendors. They paid between three and fifty times
as much for water as did people with a household connection. Water vendors got their supplies from public
taps. There were very few of these taps ­1,200 serving at least 2.5 million people. This increased the labor
costs of the street vendors, who usually transported water using jerricans piled on handcarts. Nonetheless,
the vendors were earning hourly wages two to three times the average for men with primary or lower
education. The public taps were controlled by operators, who (in turn) charged three to six times their per
liter costs for water.

Why were prices so high ­so much higher than any reasonable calculation of labor costs and wholesale
prices suggested that they should be? And why were there so few public taps? It wasn't that the cost of
construction was out of reach ­five to ten thousand additional taps could have been constructed for less
than the price of one of the city's underutilized water treatment plants.

It appears that the street vendors were running a cartel, supported by bribes to local officials and water
utility staff. After subtracting the costs of labor, equipment and payments to public tap operators, nearly
one half of the price paid to vendors for water by households was accounted for by cartel rents and bribes.
In turn, after accounting for official payments to the utility for water and operating costs, about 60 percent
of vendor payments to public tap operators were available for excess profit and bribe payments. And it
appeared that the number of public taps had been optimized to maximize the rents available for water
vendors and utility staff.10

Governance in Jakarta's water sector faced additional challenges in 1997 when two European water
companies, one partnering with a son of President Suharto, another with a close associate of the President,
were awarded concessions for the city's water supply without a competitive bidding process. In this case,
contracts allowed for limited financial oversight and, additionally, had to be renegotiated in the aftermath
of the Asian financial crisis. While 90 percent of the population of Jakarta now has 24 hour access to
water, non-revenue water remains 50 percent of the total supply and a number of performance targets
(along with sufficient tariff increases) have been missed (Castalia, 2005).

Jakarta's experience suggests the complexity and impact of failures in governance. Under public provision,
low income households in Jakarta were paying as much as 7 percent of their meager resources on water in
the dry season ­and around two thirds of that was due to the costs of collusion, bribery and skewed
investment decisions related to maximizing rents. And the case of Jakarta, alongside a number of other
challenging experiences with water concessions linked to non-competitive award around the World, has
dampened both firm and government interest in private sector investment in the sector. Alongside limited
and inefficient public investment, this has dramatically curtailed the rollout of water services to the
unconnected ­leaving many more people reliant on low quality and expensive water from vendors or
frequently unreliable borehole water.



As a result, the macroeconomic impact of an infrastructure project can be significantly
reduced by governance failures even if the project itself is carefully chosen, well
designed and corruption-free.11 Not least, other investment decisions may remain
mistaken and O&M budgets in many developing countries are too low to sustain existing


10 A simple solution to this problem adopted by the Jakarta government was to allow any householder with
a tap to sell water. As a result, end water prices to unconnected households dropped between 30 and 60
percent. Source: Lovei and Whittington, 1991.
11For example, Infrastructure at the Crossroads (World Bank, 2006b) noted that World Bank staff, "in the
context of the broader sector dialogue, spend a substantial amount of time steering government officials
away from poorly designed projects." This is an important effort, but may only have a macro impact if the
effort is to stop such investments occurring at all rather than stop them occurring under Bank projects.


                                                                                                              7

stocks of infrastructure ­as reflected in poor indicators of quality. Taking a project
approach to corruption would miss these impacts, which are frequently at the core of the
overall development impact of failed governance (Kenny, 2006). In turn, this suggests
that `ring fencing' a project against corruption only impacts a small part of the total
expenditure portfolio of a particular sector, and it may impact the part of the portfolio that
needs it least.12

These lessons suggest that for the majority of developing countries a broad focus on
governance is the most effective method to improve the macroeconomic returns to
infrastructure projects, and similarly the most effective method to reduce the
development impact of corruption in the sector. And while private involvement is likely
to change the nature of governance challenges, it does not remove those challenges. As a
result, the governance agenda is a high priority across sectors and investment projects
regardless of market structure or the strength of the case at the micro-economic level for
an individual investment.


Improving Governance in Infrastructure Sectors

The record of donor attempts to strengthen institutional structures in developing countries
is mixed. For example, a number of sobering World Bank evaluation department reports
conclude (inter alia) that "[w]ithin just a few years, the Bank has developed and
mobilized a variety of tools... that bring the quality of public sector institutions into the
spotlight. So far there is little evidence that governance is improving..."13 and that "[t]he
Bank does not apply the same rigorous business practices to its capacity building work
that it applies in other areas. Its tools... are not effectively used... most activities lack
standard quality assurance processes..."14 These general lessons regarding governance
are as likely to apply in infrastructure as elsewhere.

This mixed record is a result of the complexity of the political economy of reform, the
need to fit institutional solutions to varied country and sector circumstances and a limited
knowledge base regarding what works. Furthermore, there are limits to the progress that
we can expect to catalyze in this area. Institutions change slowly, and their quality is
strongly correlated with GDP per capita.




12Furthermore, a general lesson from the evidence surrounding the success of anticorruption efforts is the
following: "because corruption is itself a symptom of fundamental governance failure, the higher the
incidence of corruption, the less an anticorruption strategy should include tactics that are narrowly targeted
at corrupt behavior and the more it should focus on the broad underlying features of the governance
environment" (Shah and Shacter, 2004).
13OED (2004) which also suggests regarding programs that promote empowerment "both the intended and
actual poverty impact of this type of intervention remain to be demonstrated."
14OED (2005a). See also OED (2005b) on investment climate work: "World Bank Group strategies for
improving the IC [investment climate] have suffered from a lack of knowledge about what types of
institutional arrangements will work... The feasibility of reform depends on the political economy of the
reform process..."


                                                                                                             8

Nonetheless, some countries are doing much better job at the same GDP per capita in
providing infrastructure outputs ­better policies and institutions can make a difference.
Countries at the same level of income per capita have notably different access rates to
infrastructure as well as transmission and distribution losses, for example (see Figures 1,
2 and 3). Given this, and bearing in mind the caveat that our knowledge base of
rigorously evaluated governance interventions is extremely limited, it is worth examining
what the experience has been in governance reforms to improve general development
outcomes and sector infrastructure performance.

Macro-level responses such as fiscal reform for transparency and development focus in
budgets, civil service reforms including competitive selection and merit-based pay,
reform of general procurement rules and auditing standards, legal reform and increased
freedom of information should all have an impact on corruption in infrastructure sectors
as well as elsewhere (Levy, 2007). Indeed, much of the sector agenda for improved
governance builds upon the same principles and approaches as these macro interventions,
but tailored to the particular features of infrastructure provision. These include efforts to
increase competitive pressures, reduce unnecessary regulation and better monitor
necessary regulation, improve planning and budgeting processes, increase civil society
participation, reduce the discretionary power of individual bureaucrats and improve
financial and physical auditing. These reforms can be replicated at the local level, an
increasingly important element of governance reform since the spread of decentralization
(see Box 3).


Box 3: Local Level Governance Reforms in Campo Elias
A World Bank Institute-supported program in the municipality of Campo Elias in Venezuela provided
support to citizen participation in governance that significantly improved survey scores of satisfaction with
government services and reduced perceptions of corruption and inefficiency. The Campo Elias Action Plan
involved (i) management reforms including standardization of administrative procedures and integrating
administrative data into a computerized system (ii) participatory budgeting processes and (iii) publication
of government documents and transactions histories alongside the creation of a tripartite auditing
committee made up of business, citizens and government officials (World Bank, 2000).



Sector Structure

We have seen that the overall record of privatization and competition has not been as
positive as initially hoped by those who promoted them. Having said that, it does appear
that private and especially competitive provision is broadly associated with greater
efficiency and improved outcomes (Estache 2006). The cities in Kolkata and Mumbai
have long been served by private electricity utilities, and commercial losses in those cities
are between 12-15 percent as compared to 30-35 percent with state-owned electricity
utilities in India, for example (Gulati and Rao, 2006). Estache and Kouassi (2002) find
that private provision is linked to greater efficiency in African Water utilities. Private
competition also reduced the level of bribes paid to utilities for services in a study of 21
transition economies (Clarke and Xu, 2004).




                                                                                                            9

Conversely, evidence from corruption surveys in transition countries suggests that state-
owned firms are at least as if not more likely to bribe for contracts and licenses.
Government-managed construction operations are particularly prone to materials theft, as
the case of the Indonesian roads project discussed earlier makes clear. Evidence from
Tanzania also suggests that that one of the most egregious violators of labor standards on
construction sites was the National Housing Corporation (Kenny, 2007). A model in this
area is Chile, which has extended private competitive provision in public works
infrastructure considerably further than most developing countries ­and at the same time
is expanding the model of separating the policy, regulatory and client functions of
government (Box 4).


Box 4: Chile's Reform of Public Works Governance
Chile's Ministry of Public Works relies heavily on private provision, with 65 percent of investment
resources for public works provided by the private sector. An independent regulatory body, the
Superintendent of Public Works, will oversee transparent concession contracts oriented towards outputs
and performance rather than inputs or investment mandates. A new set of benchmarked service standards
will drive contract design and oversight, leading to `conservation' concessions for road networks, for
example. Model contracts will be designed to maximize competitiveness in the bidding process as well as
focus on outcomes. The regulator will oversee both implementation and any contract renegotiation.



Having said that, the level of competition and the role for private participation will vary
considerably by country and sector. In construction and telecommunications, there
appears to be little role remaining for monopoly providers, public or private. But in other
sectors plausible market structure is likely to vary considerably depending (not least) on
the size of the market. Around half of the developing world has embarked on power
market reform, for example, but private risk investment with unbundled competition in
both wholesale and retail markets is unlikely to be plausible in the majority of them
(Besant-Jones, 2006). Although regional cooperation in areas such as power pools may
extend the role for competition to some smaller countries (See Box 5), for most smaller
low-income markets, management contracts may be the most plausible role for private
participation.


Box 5: Extending Romania's Competitive Energy Market
Romania's initial market liberalization in energy involved bilateral contract negotiation between suppliers
and distributors. In some cases, supply contracts were signed by managers of state-owned generation
companies at prices below market value as part of a deal that was alleged to be corrupt. Under further
sector reform, all power contracts are now competitively auctioned by the market operator, OPCOM.
Romania hopes to develop OPCOM into a regional market operator for the Energy Community of South
East Europe, providing services and selling shares in the company to neighboring countries.



The introduction of private provision, especially absent competition, requires improved
governance both in the initial privatization but also in the regulation of providers. The
process of involving the private sector has itself has been a significant source of rents.
The widespread problems with Independent Power Producers are perhaps some of the
best known examples. For example, only 40 percent of generated power from 42 IPPs in


                                                                                                         10

the Philippines is used at a cost 33 percent higher than state-owned production (Gulati
and Rao, 2006). Fifteen IPPs were authorized in Malaysia without competitive tender
providing electricity at prices up to 50 percent higher than state-owned production under
contracts that required purchase, and owners included numerous close friends of Dr
Mahatir (Smith, 2003). This suggests the complexity of private provision and the
extensive possibilities for governance failure.

Transparent (largely) price-based international competitive bidding involving a strategic
investor appears to have been a more sustainable model for introducing the private sector
in electricity (Besant-Jones 2006). With privatizations of existing assets, beyond
international competitive bidding, there is also an important role for reduced discretion
through, for example, the use of standard charter documents and increased transparency
of valuation estimates and the methods and results of the process (Kaufmann and
Siegelbaum, 1997). Governments can also bring in trusted international transactions
advisors under the sponsorship of international institutions such as the International
Finance Corporation.

Concession contracts are a common form of privatization in infrastructure, and a form
that has run into significant implementation problems. In Latin America, approximately
90 percent of water and transport privatizations and over half of energy privatizations
over the 1990s were in the form of concessions. These concessions did improve
efficiency, but did little to reduce final prices or extend services (in part because of
increasing indirect taxation). Furthermore, concessions were frequently renegotiated (in
three quarters of cases in water), on average after just over two years of signature.
Renegotiation was far less likely in countries with a pre-existing regulator established
under law, where concessionaires were selected based on price for concession award
rather than lowest tariff, and where tariff regulation was based on rate of return rather
than a price cap (Estache et. al. 2003). This confirms other findings on the importance of
sequencing in privatization (creating regulatory institutions first). It also suggests the
importance of empowering the regulatory body with access to quality accounts on which
to make rate of return estimates. In turn, this suggests the need for corporate governance
reforms within state-owned enterprises as well as concessionaires. These are subjects
taken up in the following sections.


Regulation

The governance of regulation can have a significant impact on outcomes. Governments
should set standards to which infrastructure firms and operators can be held accountable
and which citizens can understand and monitor. The plausibility of such rules are
important ­setting service standards that cannot be met or enforced merely weakens
credibility and sector efficiency. Consistency of regulation is also vital ­with significant
sunk costs and long payback periods, infrastructure providers need comfort that the
business environment will remain stable over the long term (Parker, 2002).




                                                                                          11

The roles and objectives of institutions responsible for regulation need to be clearly set
out in primary legislation and contracts. Predictability can be increased by publishing ex
ante the parameters that will be used in making decisions ­such as price adjustment
formulas and timetables of events. Stakeholders should also be able to participate in the
discussion of draft regulatory decisions.15 Contracts, licenses, regulatory decisions,
consultation papers, legislation, tariff review procedures and methodologies, timetables
and the results of consultations can all be public documents, yet a PPIAF survey of
regulators in 2005 found that less than 30 percent of regulators currently publish
contracts and licenses (Bertolini, 2006).16

In larger countries with comparatively strong institutions and in certain sectors (most
notably telecommunications), independent regulation has proven a successful model.
Andres et. al. (2007) find that regulators in Latin America that were established under
law, well funded by regulatory levy, and with a fixed-term regulatory commission
screened by legislators, were considerably better at aligning cost of capital and rate of
return. Gasmi et. al. (2006) find that independent regulation (with tenure) financed from
operator contributions with clear authority (decisions that could only be overruled by the
courts) as well as transparency (open meetings open and textual explanations of
decisions) and power over licensing interconnection, pricing and quality of service
produce better outcomes in telecoms in terms of network rollout.

Where the fully independent model is adopted, regulators should be able to recruit their
own staff based on needs and pay market rates (not necessarily civil-service rates) to
attract and retain expertise. They should have access to their own source of funds and
decide how they are used (subject to rules regarding transparency, accountability, audits,
and other controls on the abuse of power). Regulatory officials should only be able to be
dismissed before the end of their term on certain grounds, such as incompetence,
malfeasance, and conflicts of interest. The agency ought to be able to enforce decisions
through fines, penalties and, ultimately, preventing a firm from operating.

Even with these features, regulators in developing countries will have comparatively
limited capacity, suggesting the importance of simplicity in regulatory design. The US
Federal Communications Commission budget of $304 million, for example, is probably
larger than total telecoms sector revenues in around 89 of 139 low and middle income
countries for which we have data.17 A focus on a few precise and simple regulatory
instruments over theoretically more efficient but complex rules is particularly vital in
countries with limited regulatory capacity (Besant-Jones, 2006).

One important step towards regulatory simplicity is to remove overlap between agencies.
Mongolia, a country of just two million people, has over 20 separate governmental

15The regulatory process should require interested parties to be consulted before the government comes to
a decision and should require the government to explain its decision. Given the importance of price control
decisions, it may make sense to allow the regulated company to appeal on grounds of the substance of the
regulator's reasoning, not just on matters of law.
1671 percent of East Asian regulators disclosed procedures and decisions, while 42 percent disclosed
  licenses and contracts, in a 2004 survey (Muzzini, 2006).
17Estimated on the basis that telecoms revenues are approximately two percent of 2005 market GDP.


                                                                                                       12

entities with price-control powers over infrastructure, and regulatory powers overlap both
between national agencies and between national and local authorities. New Zealand, with
an economy 40 times as big as Mongolia's, has one institution regulating prices in
infrastructure. In the Philippines, an independent water regulator had authority to set
standards and apply rules regarding financial renumeration in the sector.                            But the
contracts governing private provision of water in Manila also set out standards regarding
financial arrangements and standards. This confusion led to public disputes, complex
appeals processes and, in the end, political intervention in tariff-setting (Ehrhardt et. al.,
2007).

Regarding the removal of unnecessary or unenforceable regulation, pricing regulation can
be limited to cases where there is evidence that a monopoly or dominant provider of a
service without inter-modal competition might plausibly abuse that power to raise prices.
Quality regulation can be limited to cases where there is a strong public interest
overcoming that of greater flexibility to provide low quality, low cost services to low-
income consumers and where the regulation is plausible to enforce.

In construction, for example, it is not clear that high regulatory burdens are improving
outcomes -there is no correlation between the number of procedures involved in getting
permission to build a warehouse and the number of worker accidents across countries.18
This may be related to weak enforcement or inadequate rules ­in either case this may be
grounds to reconsider regulatory design. Easing the process of regulatory compliance
will also have a role in reducing governance failure, through approaches such as
consolidated clearances and time limits to license issuance (if a license is not issued or
denied within a certain period, it is automatically granted).

Regarding new institutional structures, given that institutional change is a slow process, it
is also an important principle to work as much as possible within the existing framework
of organizations rather than create new structures that require significant adaptation. The
creation of new regulatory agencies which are themselves institutionally weak has led to
government interference in decisionmaking regardless of de jure independence ­one
notable case was the reversal of a 100 percent tariff increase in Bogota's water rates
proposed by the regulator and over-ruled by the president (Ehrhardt et al 2007).

One example of an institutionally simple approach is to combine a monitoring agent
within the appropriate ministry with provider contracts that define service standards and
tariff setting (or operator remuneration). This removes the need for an independent
regulatory agency and has worked with some success in countries including Senegal,
Cote d'Ivoire and Vanuatu (Ehrhardt et al 2007). Related approaches involve the
creation of an advisory regulatory agency with its own budget which is empowered to
provide public and independent advice regarding issues such as tariff structures and
quality, but where the legal authority to direct change remains lodged with the concerned
Ministry.



18See Djankov et al. (2000) for a broader discussion of the limited impact of regulation of entry on quality
of service provision, but its stronger link with the extent of corruption, across a range of sectors.


                                                                                                          13

At the same time, it is worth noting that the survey of Latin American infrastructure
concessions implemented in the 1990s cited above found that the incidence of early
contract renegotiation was 40 percent in cases where the regulatory framework was
embedded in the concession contract compared to 17 percent where the regulatory
framework was embedded in law (Guasch, 2004). It appears that there are no universally
superior alternatives in this area, and that country circumstances will determine different
optimal responses.

Information disclosure can be a tool to encourage self-regulation. Mandatory disclosure
requirements in the US regarding toxic releases by firms has been linked to greater
emissions reduction efforts, for example (Konar and Cohen, 1997), and a similar
approach has worked in developing countries. Related to this, broader measures to
improve private sector corporate governance can reduce the supply of bribes ­including
strengthening disclosure rules of local financial markets and requirements for
independent directors and audits. In particular, governments might demand minimum
corporate governance standards for companies involved in large public infrastructure
transactions (Castalia, 2004).

Furthermore, there is a role for holding company officials criminally liable for regulatory
lapses as a tool to encourage self-regulation. The UK allows for conviction of senior
company officials in cases where gross negligence leads to death. Cases include fatal
accidents arising out of unsafe systems of work or the provision of unsafe goods or
services (Kenny, 2007).

Pricing regulation can be a particularly contentious issue, and one which has frequently
derailed private participation. Governance reform can improve the likelihood of
satisfactory outcomes in this area as well. In a number of West African water
concessions, prices paid to operators are set by the regulator at cost-recovery levels, but
prices paid by consumers are set by the government, which subsidizes providers for the
gap between consumer prices and cost-recovery tariffs. In a transparent system, where
both cost recovery and consumer tariffs, as well as the subsidy, are advertised, the costs
and benefits of low tariffs will be clear to consumers, operators, and taxpayers alike.
Community participation can further ease the political economy of pricing reform which
is frequently a key component of the governance agenda (see Box 6).


Box 6: Community Participation in Sustainable Pricing
The local water company in Banjar in South Kalimantan, Indonesia faced water losses of 40 percent,
provided water for only 12 hours a day and suffered considerable revenue shortfalls. As part of a
restructuring plan to improve the quality and sustainability of provision, the water company proposed price
increases which were initially rejected by the local government. The company took the plan to its
customers, and entered into a contract with the community promising better customer service in return for
higher tariffs. With community support, the local council passed the tariff increase, and water quality and
supply (along with customer service) have increased markedly in the period since then (World Bank, 2005).




                                                                                                          14

SOE Management

Attempts to improve the efficiency of State-Owned Enterprises through governance
reforms are not new ­frustration with an earlier round of such efforts in the 1970s and
1980s was one factor behind the push for private participation. Many approaches have
not been carefully evaluated, and what we do know suggests mixed impacts (Irwin and
Yamamoto, 2004, Gomez-Ibanez, 2006). Nonetheless, where full privatization is not an
option, state-owned enterprises may still benefit from a number of internal and external
governance measures, even if hopes for dramatic improvement should be tempered.

A first step is to extend alternatives to centralized monopoly provision where possible.
Extending competition through deregulation or contracting out of non-core operations is
one possibility. For smaller firms, (for example, local water operatives), customer-
governed enterprises such as cooperatives may be an attractive option (Gomez-Ibanez,
2006).

Corporatization is an additional tool. This is designed to strengthen SOE autonomy,
through a separate statutory authority, with a distinct legal identity, separate accounts and
its own board of directors (preferably including members from the private sector and
stakeholder groups as well as government) (Gomez-Ibanez, 2006). Full corporatization
requires the enterprise to be incorporated under the same laws that govern private
corporations.19 Management is given considerable autonomy in areas such as HR,
financial management, pricing (under regulation) and service provision. Measures of
autonomy in these areas have been associated with considerably better performance in
terms of coverage and quality in an econometric analysis of water utilities from Africa,
Asia and the Middle East (Braadbaart et. al., 2007).

The World Bank has been involved in a number of such SOE reform efforts in
infrastructure (See Box 7). The case of Chile's performance in public works points out
that there is also a role to extend the independent state-owned agency model beyond
infrastructure to construction. Tanzania's TANROADS agency has also moved some
distance towards this model ­supported by a hypothecated roads fund, emphasizing
maintenance over investment and utilizing performance agreements with private
contractors. A somewhat similar effort to establish a Road Sector Development Team
overseeing output-based contracts in Nigeria is under development.

The experience of developed countries suggests that corporatization alone is a necessary
but not sufficient condition to improve performance. Focusing only on the internal


19SOEs might be owned by a body independent of sectoral ministries (on the model of Temasek,
Singapore's national holding company overseen by the Ministry of Finance, or the Indonesian Ministry of
State Owned Enterprises). Uganda's Parastatal Monitoring Unit has the responsibility of publishing reports
based on SOE submission of audited accounts and a certification of the safekeeping of the enterprise's
assets that summarize financial and operational performance, agreed actions to improve that performance,
the value of direct and indirect government subsidies and the value and terms of debt stock (Irwin and
Yamamoto, 2004). At the same time, such approaches may create another level of decision-making, and
their ability to prevent government intervention in management on short-term political grounds has proven
limited (performance contracts have often suffered the same fate) (Gomez-Ibanez, 2006)


                                                                                                       15

governance dimension is unlikely to improve SOE performance in a sustainable manner
(Vagliasindi, 2007). In particular, SOE management does not face any credible threats
for non-performance from the external environment in which it operates ­there is no
plausible threat of takeover or bankruptcy as an ex ante performance and monitoring
instrument, especially in cases where the SOE is a monopoly.20 As a result, there is a role
for additional techniques to increase transparency and efficiency.


Box 7: Three Cases of SOE Reform
A reform project involving the Phnom Penh Water Supply Authority included putting in a management
team with performance-based incentives, corporatization, water meters installation and automated billing
and accounting, customer surveys and a new tariff structure based on cost-recovery models and
accompanied by a public information campaign. Similarly, the Bangladesh Rural Electrification and
Renewable Energy Project supported rural cooperatives with management operating under a performance
contract, with market-based salary structures (World Bank, 2006b).

In the Dominican Republic, a reform program has supported a considerable improvement in energy bill
collection rates. In 2004, only 45 percent of power generated in the country was paid for by consumers.
Theft and non-payment of bills by government, the private sector and households alike was rampant.
Under the reform program, new legislation criminalized electricity theft, and provided for easier access to
customer premises. Performance-based management contracts were introduced for distribution companies,
metering was extended along with meter inspection, bill payment processes were simplified and illegal
connections regularized. Energy was rationed on circuits with large losses to provide and incentive for
regularization. Auditing and publication of accounts was combined with social surveys and the use of a
`social contract' between providers and stakeholders promising better quality in return for better payments,
all complimented by an active communications program. By January 2007, the percentage of power
generated that was paid for by consumers had risen to 56 percent



In some countries, an SOE audit committee is chaired by an independent non-executive
director and externally audited accounts meeting International Financial Reporting
Standards (IFRS) are published. Additionally, SOEs that borrow from the private sector
without the benefit of a government guarantee will face the scrutiny of lenders and credit-
rating agencies. Listing a minority of shares will create another level of monitoring by
shareholders.

One of the key challenges in addressing public sector governance is that SOEs are
frequently given broad developmental objectives, such as ensuring employment, or
creating necessary infrastructure for economic development. If there are multiple goals
and especially if they cannot be clearly quantified, then performance cannot be accurately
measured ­and SOE managers cannot be rewarded on the basis of performance. Where
possible, it may be simpler to set a clear target of financial performance, utilizing
regulatory oversight and transparent subsidy tools to achieve non-financial objectives.21
Governments might consider quantifying costs and benefits of meeting equity objectives


20Continued state-directed credit, equity injection and finance deficits have created perverse incentives for
managers of SOEs and perpetuated soft-budget constraints (Kornai et al. 2003).
21In this model, a regulatory body oversees (or publicly advises on) pricing (and, if necessary, quality) of
provision. The concerned sectoral ministry uses transparent subsidy mechanisms to support social goals
related to pricing and rollout (World Bank 2006).


                                                                                                           16

and entering them explicitly in the accounting system. In South Africa, SOEs and the
executive authority agree on a "Corporate Objective Statement" which expresses in terms
of outputs the financial and non-financial performance that the SOE is expected to meet.
This document is public.

Regarding petty corruption, performance improvement can be considerable. In Andhra
Pradesh, transmission and distribution losses were reduced from 38 to 26 percent 1999-
2003 in large part through metering, theft control and the regularization of 2.25 million
unauthorized connections (Gulati and Rao, 2006). As part of a process of enterprise
reform, a number of human resources tools can be used to help improve performance and
reduce corruption amongst public officials including training (especially on codes of
conduct), whistleblower protection, complaints mechanisms, improved disciplinary
procedures, transferring staff in posts considered `at risk,' random checks on performance
by investigators posing as customers and reducing anonymity through name tags and
other information disclosure (Sohail and Cavill, 2007).

On the demand side, citizen's charters can clarify the standards that consumers can
expect in terms of quality, timeliness, cost and coverage of services. These can be based
on regulatory standards where appropriate but should be actively disseminated to the
public alongside information on complaint and redress mechanisms. An additional tool is
the use of freedom of information acts. These have been passed with little impact in a
number of countries,22 but when well designed, used to provide specific information on
services and supported with active civil society right to information campaigns, they have
had some notable successes in improving infrastructure outcomes (See Box 8).                   23



The Concerned Citizens of Abra for Good Government in the Philippines was founded to
provide a monitoring role utilizing information provided by governments and its first case
in 1987 involved following up on an announcement by the Department of Public Works
and Highways that 20 projects had been completed in Abra. The group was able to
document that some projects had not begun and others had been completed with
substandard materials. As a result of its investigation, eleven officials were found guilty
of misconduct and dismissed (Sundet, 2004).

Citizen report cards are a related approach. The report card process surveys households
about their experience and satisfaction interacting with public officials and receiving
government services. They can be combined with a social audit where official reports of
works and expenditures are reviewed and compared to the results of customer surveys.
The World Bank-supported Indonesia Urban Sector Development Reform Project also



22More than 50 countries have freedom of information laws, and 15­20 more are considering them. But the
mere existence of a freedom of information law is clearly inadequate. A recent survey of the effectiveness
of freedom of information laws in Armenia, Bulgaria, Macedonia, Peru and South Africa found that, on
average, only 35 percent of requests for information were fulfilled (sources:
http://www1.worldbank.org/prem/PREMNotes/premnote93.pdf and
http://www1.worldbank.org/publicsector/legal/freedom.htm)
23Good design incorporates a presumption in favor of information release, clear and simple procedures for
requests, responses and timings, an appeal mechanism and penalties for non-complying officials.


                                                                                                        17

uses NGO oversight including routine checks on specifications in each construction
project location, as well as citizen score cards and satisfaction surveys.


Box 8: Providing Information on Services and the Role of Citizen Involvement
In Brazil, the Sistema Nacional de Informacoes Sobre Saneamento provides national information on water,
sanitation and solid waste service provision. The SNS collects data on 77 operational, financial and quality
indicators from 26 regonal and 260 municipal service providers. The information is published on the
Internet as well as in a year book and CD.

A clear case of a gap between the de jure and de facto state of information disclosure is provided by India.
A 2006 survey of water and sanitation utilities found that only one third of the information they were
legally required to be publish ­covering information on licenses, permits budgets and subsidies-- was
actually available. Nonetheless, where information is disclosed, Indian experience suggests that this can
help to improve services. Parivartan, a citizen's group, uses the Delhi Right to Information Act to demand
information on the status of infrastructure construction budgets, construction projects and repair work. In
one case, Parivartan supported residents living alongside an unmaintained road to file a series of
information requests which revealed that official records suggested the road had been repaired multiple
times, that the work had officially been carried out by the engineering department of the Corporation of
Delhi, and that materials had been issued to staff for the repair work. The series of requests pinpointed
materials theft within local government as the cause of road deterioration in this case.24

The World Bank-supported Orissa State Roads Project will use implementation of the state's Right to
Information Act as the underpinning of governance and anticorruption elements built into project design.
A project disclosure policy based on RTI requirements will be supported by new document and information
management systems alongside a complaints management system and corruption perceptions surveys. In
addition, the project reports business process re-engineering to simplify and standardize department
processes, procurement reform including e-procurement, the creation of a Chief Vigilance Officer and
third-party monitoring.



Budgeting and Investment Planning

Evidence of the quality of budget management in 25 heavily indebted poor countries
(HIPCs) suggests there is considerable need for progress, with the average country
meeting just over five out of 13 performance benchmarks in the areas of
comprehensiveness, credibility, execution and external scrutiny.25 Budget transparency
has been an important target of efforts such as the Extractive Industries Transparency
Initiative (See Box 9). But similar weaknesses regarding transparency and efficiency are
also apparent at the sector level, where planning and project selection procedures are
often very weak.

Planning and technical and economic capacity to evaluate projects should both help to
provide better criteria for project selection (avoiding the introduction of white elephant
investments at the last minute on an `emergency' basis). Improved budgeting will also
reduce cases where funding is inadequate to cover ongoing contract payments ­such
situations place government officials in the position of deciding which contracts get paid
first. For example, Paraguay is developing a five year itemized sector budget, a road

24Sources: Satyanand, (2007) and www.parivartan.com, accessed 04/25/07.
25Calculated from p. 36 of Levy (2007).


                                                                                                          18

sector strategy and a road social contract in consultation with stakeholders that will guide
expenditures and provide an agreed framework for measuring successful delivery. In
Chile, the Ministry of Public Works is moving towards an in integrated planning process
which will use a supply-chain approach to ensure investments are made based on multi-
modal analysis using improved economic and engineering assessment.


Box 9: EITI
The Extractive Industries Transparency Initiative is a global program aimed at improving transparency and
accountability in resource rich countries. The heart of the program is publication and verification of
payments made to governments by extractive industries. Over the 2003-2006 period, 25 countries
committed to implement the EITI, and the Bank was providing assistance in more than half of these
countries in 2006. Multi-stakeholder committees to oversee EITI implementation had been set up in 12
countries and auditors had been appointed in 8 countries. The World Bank administers the EITI Multi-
Donor Trust Fund, which is used to provide technical and financial assistance to countries implementing
the program (World Bank, 2006).



Demand estimation will also improve the economic return to projects (World Bank,
2006b). This involves not only improved economic analysis of projects, but also better
cost forecasting. Predicted outcomes should be compared with predicted and actual
outcomes of a relevant reference class of past projects to determine their likely accuracy,
and forecasts should be published, publicly debated, and subject to independent review
(Kenny, 2007).

There is a further need to improve expenditure tracking through levels of government. In
some cases, infrastructure services involve multiple levels of government simultaneously,
and this can create particularly complex governance challenges. Rural roads is one such
case (see Box 10). Tanzania's Pro Poor Expenditure Tracking Survey in 2001 covered
four sectors including water and rural roads. The Tanzania PETS found significant
leakage. One district had recorded only 10 percent of receipts from central government
for road funds over 1999-2000, for example ­this encouraged the Treasury to advertise
all transfers from central to local government in the media (Sundet, 2004).

The response to expenditure tracking analysis can be improved if civil society is actively
engaged. For example, in South Africa the Public Service Accountability Monitor based
at Rhodes University provides a database of information on budget allocations and
resources available to government departments as well as plans for their use, progress
reports and recommendations made by oversight bodies (Sohail and Cavill, 2007).




                                                                                                       19

Box 10: The Governance Challenges of Rural Roads in Africa
Rural roads are often the legal responsibility of local governments, but they are often planned and built by
national road authorities. In addition, thousands of kilometers of roads have been constructed by
agricultural projects, food for work programs or NGOs, for example ­usually without the input of those
who might bear the responsibility of maintenance. The majority of African countries lack a legal
framework governing the rights and responsibilities of communities regarding local roads and encouraging
local maintenance of roads has proven a particular challenge. Budgets for such maintenance are often as
little as a tenth of the amounts required. These problems are exacerbated by the low capacity and limited
pay of local roads officials.

A first step in a reform agenda would be to ensure planning for rural roads that involves communities and
local governments. In addition, a legal framework governing ownership, a budgetary framework ensuring
adequate maintenance and a management framework that provides capacity is required. One approach tried
in Madagascar in the late 1990s involved the creation of road users associations to carry out financing and
management of tertiary and access roads. These associations operate under a management contract signed
by the local government, which also provides about one third of association funds. Additional funding is
provided by local taxes and tolls (Calvo, 1998).



Improving Governance in Project Design and Execution

Identifying which government official is responsible for ensuring delivery of a good
quality outcome is an important first step in ensuring accountability. That official should
be empowered with the necessary (technical and financial analysis) resources to monitor
implementation. The official should be able to act under the guidance of comprehensive
and transparent rules governing both oversight and payment procedures, and face
consequences for failure to act according to this guidance (Kenny, 2007). Sufficient
simplicity to allow for oversight is also important. The World Bank's Second Sulawesi
Urban Development Project in Indonesia (41 participating local governments, 150
implementing agencies, carried out in the midst of a rapid and ambitious decentralization
program) suggests the increased vulnerability of complex projects to corruption ­more
than 100 firms and individuals have been debarred as a result of corruption in the project
(World Bank, 2006b).

In order to help monitoring groups to determine if citizens are getting what they paid for,
contracts and amendments can be published. Many countries provide access to contracts
if a specific case is made under a Freedom of Information law, and the government of the
Australian State of Victoria publishes all contracts (including contract revisions) for
contracts worth in excess of AUS$10m (around USD $7.7m) as a matter of course
(Kenny, 2006).26 DfID's Construction Sector Transparency Initiative, which the Bank is
supporting, is loosely modeled on the Extractive Industries Transparency Initiative and is
designed to bring a greater level of openness to government contracting in construction.
It will foster publication and review of key project details including budgeted and actual
payments, a project description and any project evaluation.



26 Maximum disclosure of contract details surrounding consultants is perhaps particularly important,
because this is the area where price-based selection for well-defined deliverables is least plausible (which
in turn provides considerable discretion to selection committees).


                                                                                                           20

Within projects, community scorecard approaches can be used to monitor progress. In
contrast to citizen report cards, these approaches involve a discussion at the community
level under the guidance of a facilitator who collects information on project progress and
implementation to provide immediate feedback to those overseeing the project. Further
enhancing the role for demand-side approaches in governance, community-driven
development involves giving communities or locally elected bodies direct control over
the decision making, management and use of development funds (See Box 11).


Box 11: Community Driven Development
In Indonesia, a 1997 survey of 48 villages found that less than 3 percent of village development requests
proposed through the government's development planning system received funding. The Kecamatan
Development Program (KDP), begun in 1998, emphasized participatory appraisals for project selection
with financing provided by a combination of village and local funding and direct central government
support. These features help to ensure that local priorities were the key to setting project prioritization. In
the KDP, project budgets, financing and procurement decisions are discussed publicly and displayed on
village information boards, each village has an independent committee to oversee contracts and
implementation, and journalists and NGOs are invited in to act as watchdogs. In addition, there is an
anonymous complaints mechanism which channels concerns to project authorities. KDP projects that met
high local demand with close local oversight and involvement produced savings of between 25 to 56
percent over conventional infrastructure projects and carried economic rates of return ranging from 33 to 83
percent (Wong and Guggenheim 2005).27

The World Bank's Uttar Pradesh Swajal project is another example of using community-driven approaches
as part of an effort of reform to ensure cost recovery for maintenance and partial cost-recovery for capital
costs in water and sanitation. Under the project, resources were transferred to rural communities which
procured materials, services and works themselves. NGOs assisted with community mobilization and
private firms provided technical design specifications, inspection and monitoring. 92 percent of water
supply schemes and 100 percent of latrines constructed under the project remain fully functional (World
Bank, 2006b).



Output-Based Aid provides an attractive vehicle to reduce the potential development
impact of failures of governance. OBA is a strategy for using explicit performance-based
subsidies to support the delivery of basic services. It involves delegating service delivery
to a contractor, under contracts that tie disbursements to the services or outputs actually
delivered (Kenny and Mumssen, 2007). The outputs defined are `technology neutral'
(delivery of electricity services of a given quality to a given community) rather than
technology specified. As such, the OBA process is less susceptible to corrupt agents
designing project specifications in such a way as to favor a particular bidder. The process
is also likely to limit cases where unnecessary or unsustainable projects are undertaken
because they provide greater opportunities for corruption. The outputs of an OBA project
(water or power connections, a pothole-free road) are also easily monitored, and payment
is withheld until outputs are achieved. As a result, even if corruption does occur in OBA
projects, the risk that it significantly reduces the development impact of the financing is
limited. An OBA project supported by the World Bank in Dakar, Senegal, provides on-


27Similarly, community-based construction of schools in Zambia and Mauritania cut costs by one half to
two thirds over national competitive bidding approaches ­although some considerable part of this saving
was due to lower architectural standards (Theunynck, 2006).


                                                                                                             21

site sanitation facilities to poor households. All payments to local builders constructing
the sites (as well as to non-government agencies implementing the scheme) are withheld
until after site construction, with some financing held back until after a period of
monitoring usage for several years, to ensure sustainability in construction and demand.

Within the procurement process itself, there is scope for interventions covering e-
procurement, financial management information systems, transparency and third-party
(NGO or external professional) oversight rules that are applicable across sectors.
Companies that have previously broken laws, that do not have an internal system of
controls in place, or that have been selected in non-competitive bids might be subject to
additional oversight by an independent inspector acting as an integrity monitor, with
access to internal documents and deliberations on a confidential basis. The World Bank's
Paraguay Road Maintenance project combined a number of features to improve
governance and reduce corruption at both the sector and project level (See Box 12).


Box 12: The Paraguay Road Maintenance Project
The Paraguay Road Maintenance Project used two workshops held by the Ministry of Public Works to
discuss sector governance weaknesses ­these provided input into sector reform proposals and project
design. Results elements adopted included a framework of monitorable output indicators to be tracked, but
also baseline and follow-up surveys to provide evidence of socio-economic project impact. Transparency
measures included an active program of dissemination prior to and during the project, as well as electronic
publication of procurements, contract award and implementation progress, complaints procedures and other
project and sector-related documents and policies. Participatory processes were used in the design of the
local (unpaved) roads component of the project, which also involved local cost-sharing. Local universities
will also be involved in monitoring physical quality of outputs for output-based project components.

Combined with World Bank led fiduciary assessments, the consultation process also prompted the
development of a number of mitigation measures regarding project fiduciary arrangements, including new
financial management systems and improved audits, and an enhanced supervision process involving the
monitoring of `red flags' such as price variance in bids and procurement delays, laid out in an annex to the
project documents.28



Implications for Infrastructure Investment

Governance plays a key role in determining the development impact of investment
operations in infrastructure. Even projects that are chosen on the grounds of strong
demand and that are well built and corruption-free, will have considerably reduced
macroeconomic impact if sector governance is weak. Furthermore, as global
anticorruption efforts involving the UN, the OECD and multilateral development banks
ramp up, there is a growing fear that countries which are perceived to be corrupt will lose
access to the international skills, knowledge and finance of firms unwilling to face the
risk of global sanctions (Box 13). Given this, sectoral analysis of governance and sector-
level responses in terms of institutional capacity building should have an increased
importance.

28Available here:
http://web.worldbank.org/external/projects/main?pagePK=64283627&piPK=73230&theSitePK=40941&m
enuPK=228424&Projectid=P082026


                                                                                                          22

Box 13: Potential Negative Spillovers of a Global Anticorruption Drive
International anti-corruption initiatives such as the OECD Anti-Bribery Convention as well as institutional
initiatives such as World Bank investigation and blacklisting of firms that bribe both raise the potential
global cost of corrupt activities. In countries and sectors where the perceived norm is to bribe, international
firms have the choice to exit or face an increased risk of home country prosecution and global blacklisting.
The Confederation of International Contractors Associations and the World Economic Forum report
anecdotal evidence that companies which have adopted strict anti-corruption procedures have, as a result,
lost business in some markets. Additional anecdotal evidence from World Bank staff suggests an
increasing reluctance of international contractors to bid on World Bank projects in countries perceived to
be corrupt. There is evidence that the 1977 Foreign Corrupt Practices Act had this impact on US firms,
with US foreign direct investment diverted from countries perceived to be corrupt towards those perceived
to be less so (Hines, 1995). In countries where corruption is perceived to be widespread this may leave
construction or concession contracts to be bid on by a few local firms who face lower potential costs of
corruption and which may lack the capacity to carry out the work. This suggests the vital importance of
combining global approaches to anticorruption with strong actions at the national level, or counties that are
perceived to be corrupt may face the consequence of lower competition, reduced quality and (potentially)
an increase in corruption at the national level.29



Regarding the focus of intervention, the introduction of private competition, where it is
feasible, should remain a cornerstone of reform in infrastructure, and there is still some
considerable scope for additional progress in telecommunications, power production and
infrastructure construction for public clients. Improved state-owned utility operation as
well as greater private sector participation require further development of both robust
privatization processes and the institutions of regulation. The importance of regulatory
approaches that can work in environments of limited institutional capacity suggests the
need to move beyond regulatory independence towards the design of low-discretion, low-
capacity regulatory approaches suited to developing countries.

The continued presence and importance of state-owned operations in infrastructure
suggests the need for a renewed search for methods to improve their performance.
Again, this would include an enhanced emphasis on transparency, outside monitoring and
audit and stakeholder engagement through tools including scorecards and NGO
monitoring. The need for support begins at the level of the sector, helping ministries to
improve planning and budgeting as well as extending transparency and participatory
involvement.

In addition, governments and donors have fiduciary obligations to ensure that their funds
are used effectively for the intended purposes. Corruption joins capacity limitations,
limited knowledge and capture by special interests as a hurdle to overcome in meeting
this obligation. As a result, there is a role for increased awareness of the corruption issue,
including the development of generic lists of `red flags' for types of infrastructure project
that could inform project teams as they monitor project implementation.

At the same time, it is worth emphasizing that our knowledge of corruption at the sector
level is inadequate to suggest, beyond a very broad range, which infrastructure projects in
which countries are likely to be most at risk, and therefore justify increased monitoring

29A suboptimal response on the part of donors would be to further ring-fence projects to encourage
international bidders.


                                                                                                            23

(Box 14). Under these circumstances, the only justifiable procedure regarding increased
focus due to concerns with corruption would be to demand greater oversight on all
projects (at significant cost) or a very few where we have particularly strong reasons for
concern beyond survey and perceptions indicators. Once more, this suggests that
considerably increased focus by donors or governments on individual procurement
processes should be a `last line' strategy in cases where broader approaches at the project
or sector prove impossible and there are strong reasons to believe that corruption is a
particularly severe problem.

Box 14: Measuring Progress in Governance and Anticorruption: Do We Have `Actionable
Indicators' in Infrastructure?
Measuring the status and progress of infrastructure governance will be important to evaluate the success of
approaches designed to improve it. At the same time, our knowledge of the exact extent of corruption or
quality of governance in infrastructure by sector, country or activity is limited. What evidence we have
suggests that general perceptions measures are poor proxies for governance weaknesses or levels of
corruption in infrastructure. For example, taking construction firms in the 1999-2000 BEEPS dataset
which reported on the level of corruption in Eastern European government contracting, only around 14
percent of the variation in answers is accounted for by country of residence (Kenny, 2006). It is clear that
we cannot say with any certainty which of these countries is more or less corrupt than average when it
comes to government construction contracting, and this on a measure of bribe payments rather than the
impact of corruption on development outcomes.

The available survey evidence covering infrastructure in particular paints a complex picture of within-
country variation in the levels of corruption by activity and area. And that available survey evidence is
patchy --not least, it is hard to survey monopoly providers on an anonymous basis as the extent of
corruption in their industry, for example. Furthermore, we have almost no evidence regarding the
prevalence of different types of corruption ­payments for orchestrated collusion versus payments for
changes in contract terms after award, as it might be--despite their significantly different potential to derail
sector performance and project outcomes.

This limited granularity of sectoral corruption indicators in particular suggests that it would be a mistake to
use the available indicators to decide on infrastructure lending levels or policies towards particular
countries. At the same time, we have seen that corruption is a symptom of failed governance, and it is of
concern because it can reduce the development impact of infrastructure investments. Given this, the need
for accurate measures of corruption per. se. is reduced.

To measure progress in the effort to improve governance requires measures of the nature of governance
institutions (the status of the regulator, the nature of corporatization, transparency regarding concessions
and contracts, and so on), and measures of outcomes to determine if governance changes are having a
development impact (infrastructure rollout, prices, `technical' losses).30 One route through which
improved governance will improve outcomes is through a reduction in the impact of corruption, but a
measure of corruption itself is not required to judge success or failure of governance initiatives. If
`actionable indicators' regarding governance and anticorruption are required to set policy or determine
progress, these can be selected from the more objectively verifiable of input and output measures related to
institutional progress and outcomes. The role for survey and perceptions indicators of corruption itself
should probably be limited to use in analytical work designed to accommodate for the level of noise such
indicators contain.


30Input measures might include indicators of transparency (contract publication), the presence of a
planning document including major sector investment decisions and the extent to which it is followed,
estimates of the adequacy of maintenance funding based on international benchmarks, SOE externally
audited accounts, indicators of sector structure and competition, and so on. Output indicators might include
percentage of roads in good condition, traffic volumes, non-technical losses, percentage of production paid
for, extent of access to infrastructure and so on.


                                                                                                            24

This is not to say that there is nothing that can or should be done to sensitize project
teams to red flag warnings. The Paraguay Roads Project discussed above included a
project-specific list of red flags ­many of which are also indicators of potential problems
with project delivery linked to failures in capacity or other causes not necessarily the
result of corruption. Development of such an indicator list can be a useful tool not only
to monitor for potential cases of corruption but also, more broadly, to provide an
overview of project health and the potential need for greater supervisory support for
projects carried out in an environment for weak governance.

Furthermore, although it is difficult to differentiate levels of corruption at the sector level
across countries, different types of projects and components carry distinctly different
levels of risk. Large, complex, one-of-a-kind components where there are few potential
bidders and where change orders are likely will be considerably more susceptible to
corruption than simple procurements based on standard or commodified products. This
suggests the potential for the development of a risk-based menu of anti-corruption
options based around project or component type. Where components are large and
complex, for example, the return to third party monitoring, independent physical and
financial audit, integrity pact mechanisms and/or increased procurement transparency is
likely to be considerably higher in terms of improved outcomes.

The above analysis of experience suggests that a range of approaches have already been
tried in projects to address governance and anti-corruption at the sector and project level.
There is a considerable role for greater application and testing of these existing
techniques. This will require a further shift in donor and ministry incentives from
lending or budget volume to lending or budget outcomes, but also a greater acceptance of
project risk. The danger of a failed project appears greater in institution-building than in
infrastructure construction. This reflects in part a perhaps blinkered approach to project
outcome analysis which focuses too narrowly on asset rates of return rather than the
project's impact on country-wide returns to investment. But the appearance of risk also
reflects the fact that institution-building is complex and frequently depends on a set of
political and economic circumstances that can rapidly change.

In the particular case of work in more challenging institutional environments, this also
suggests the need for movement away from a `risk-management system' to a `risk-
weighting method.' The higher potential for poor or slow execution or diversion of funds
should be weighted against a potentially greater payoff in terms of development impact,
and (where payoffs appear considerable) greater resources should be provided for support
of institutional development and project management rather than abandoning the project
in favor of a simpler operation.

An additional incentive to sector-level governance reform in infrastructure would be a
greater number of success stories alongside analyses of failed attempts. Many of the
approaches suggested above have not been subject to rigorous analysis ­and very few to
repeated rigorous analysis allowing for a solid understanding of success factors. We lack
sufficient data both on governance inputs (details of regulatory or SOE structure) as well
as outputs (transmission and distribution losses, road quality, prices) in order to make



                                                                                             25

definitive statements on the types of reform that carry the highest payoffs in particular
environments.


Conclusion

A new pragmatism surrounding the tarnished bullet of private participation should not
foster commitment to a new silver bullet of institutional reform. Not least, evidence that
corruption remains a significant issue in the wealthiest countries in the world suggests the
complexity of progress ­and that zero tolerance is an implausible stand.31 Experience of
the process of governance and institutional reform to date is that it is a complex and
lengthy process prone to false starts and setbacks.

Improved governance involves transactions costs, and so there is an optimum level of
governance which avoids high-cost limited impact interventions. Anti-corruption
measures, much like other regulation designed to minimize market or government failure,
can carry higher costs than the economic benefit of the reduced corruption with which
they are associated. Ben Olken's (2004) examination of anti-corruption interventions in
the Indonesia KDP project suggested that sending out invitations to village meetings to
discuss projects may have fallen into that category, for example. Similarly, international
competitive bidding may reduce opportunities for local collusion, but it can also greatly
extend procurement processes and add significantly to costs.32 Again, even if we set the
goal of reducing corruption to zero, it is not clear that we have the tools to accomplish
such a goal. For example, the Olken study found that an estimated 24 percent of
expenditures were "lost" in materials theft, suggesting that community-driven
development is not a panacea even if it can be a powerful tool to improve outcomes.

Having said that, governance reform, particularly at the sector level, can carry high
payoffs. And ignoring governance weaknesses can significantly dampen economic
returns to infrastructure projects. Even with (especially because of) imperfect knowledge
of what works in the sector, greater effort and evaluation regarding infrastructure
governance should be a priority.




31The recent corruption scandal involving Randy Cunningham, a Congressman from San Diego in the US,
is a case in point: he received $525,000 in return for $6m in government contracts (Newsweek 3/20/06
"The CIA: Questions About a Contract"). This is a payoff as a percentage of the contract value of a little
under 10 percent ­around the same as the level of bribe payments reported for government contracts in
Eastern Europe and Central Asia. That the price demanded by Cunningham was so low suggests an active
market for corruption even in the US.
32In the case of school construction, bundled ICB approaches have been abandoned due to high cost in
favor of delegation to contract management agencies, NCB and community-based approaches by the 1990s
(Theunynck, 2006).


                                                                                                         26

Bibliography

Andres, L., J. Guasch and S. Straub (2007) Does Regulation and Institutional Design
Matter for Infrastructure sector Performance? Mimeo, World Bank.

Bertolini, L. (2006) how to Improve Regulatory Transparency PPIAF Gridlines Note No.
11.

Besant-Jones, J. (2006) Reforming Power Markets in Developing Countries: What Have
We Learned? World Bank Energy and Mining Sector Board Discussion Paper 19.

Bo, E. and M. Rossi (2006) Corruption and Inefficiency: Theory and Evidence from
Electric Utilities, mimeo, UC Berkeley.

Braadbaart, O., N. Eybergen and J. Hoffer (2007) Managerial Autonomy: Does it Matter
for the Performance of Water Utilities? Public Administration and Development 27 111-
121.

Calvo, C. M. (1998) Options for Managing and Financing Rural Transport Infrastructure,
World Bank Technical Paper 411.

Casartelli, G. and E. Wolfstetter (2007) World Bank Policy on Selection and
Employment of Consultants: Study of Its Effectiveness, mimeo, World Bank.

Castalia (2004) Integrity in the Provision of Infrastructure: The Way Forward in Control
of Corruption and Accountability, mimeo, World Bank.

Castalia (2005) Sector Note on Water Supply and Sanitation for the Infrastructure in East
Asia and the Pacific Flagship. Mimeo, World Bank.

Clarke, G. and L. Xu (2004) Privatization, Competition and Corruption: how
Characteristics of Bribe Takers and Payers Affect Bribes to Utilities, Journal of Public
Economics 88.

Collier, P. and J. Dehn (2001) Aid, Shocks and Growth World Bank Policy Research
Working Paper 2688

Davis, J. (2003) Corruption in Public Service Delivery: Experience from South Asia's
Water and Sanitation Sector, World Development 32, 1.

Djankov, S., R. La Porta F. Lopez-de-Silanes and A. Shleifer (2000) The Regulation of
Entry, NBER Working Paper 7892

Dollar, D. and V. Levin (2005) Sowing and Reaping: Institutional Quality and Project
Outcomes in Developing Countries World Bank Policy Research Working Paper 3524.

Doucouliagos, H. and M. Paldam (2005b) Aid Effectiveness on Accumulation: A Meta
Study, University of Aarhus Department of Economics Working Paper 2005-12.

Easterly, W. and R. Levine (2002) Tropics, Germs and Crops: How Endowments
Influence Economic Development, NBER Working Paper 9106.

Easterly, W., R. Levine and D. Roodman (2003) New Data, New Doubts: A Comment on
Burnside and Dollar's "Aid, Policies and Growth" NBER Working Paper 9846.




                                                                                        27

Ehrhardt, D., E. Groom, J. Halpern and S. O'Connor (2007) Economic Regulation of
Urban Water and Sanitation Services: Some Practical Lessons World Bank Water Sector
Board Discussion Paper No. 9.

Estache, A. and E. Kouassi (2002) Sector Organization, Governance and the Inefficiency
of African Water Utilities world Bank Policy Research Working Paper 2890

Estache, A., J. Guasch and L. Trujillo (2003) Price Caps, Efficiency Payoffs and
Infrastructure Contract Renegotiation in Latin America World Bank Policy Research
working Paper 3129.

Fay, M. and T. Yepes (2003) Investing in Infrastructure: What is Needed from 2000 to
2010 World Bank Policy Research Working Paper 3102

Gasmi, F., P. Noumba and L. Virto (2006) Political Accountability and Regulatory
Performance in Infrastructure Industries: An Empirical Análisis, World Bank Policy
Research Working Paper 4101.

Gomez-Ibanez, J. (2006) Alternatives to Privatization Revisited: The Options for
Infrastructure, mimeo, World Bank.

Guasch, J. L. (2004) Granting and Renegotiating Infrastructure Concessions: Doing it
Right Washington DC: World Bank.

Hines, J. (1995) Forbidden Payment: Foreign Bribery and American Business After 1997,
NBER Working Paper 5266

IEG (World Bank Independent Evaluation Group) (2006) Annual Review of Development
Effectiveness Washington DC: World Bank.

Irwin, T. and C. Yamamoto (2004) Some Options ofr Improving the Governance of
State-Owned Electricity Utilities, Energy and Mining Sectoer Board Discussion Paper
No. 11.

Kaufmann, D. and P. Siegelbaum (1997) Privatization and Corruption in the Transition,
mimeo, World Bank.

Kaufmann, D., F. Leautier and M. Mastruzzi Governance and the City: An Empirical
Exploration into Global Determinants of Urban Perfromance World Bank Policy
Research Working Paper 3712.

Kenny, C. (2006) Measuring and Reducing the Impact of Corruption in Infrastructure,
World Bank Policy Research Working Paper 4099.

Kenny, C. (2007) Corruption in Construction, mimeo, world Bank.

Kenny, C. and Y. Mumssen (2007) Output Based Aid as a Tool to Reduce the Impact of
Corruption on Development Outcomes in Infrastructure, mimeo, World Bank.

Kenny, Charles and David Williams (2001) What Do Economists Know About Economic
Growth ­Or Why Don't They Know Very Much? World Development Vol. 29 No.1.

Konar, S. and M. Cohen (1997) Information as Regulation: The Effect of Community
Right to Know Laws and Toxic Emissions, Journal of Environmental Economics and
Management 32.



                                                                                    28

Leipziger, D. M. Fay, Q. Wodon and T. Yepes (2003) Achieving the Millennium
Development Goals: The Role of Infrastructure World Bank Policy Research Working
Paper 3163.

Levy, B. (2007) Governance Reform: Bridging Monitoring and Action World Bank:
Washington DC.

Lovei, L. and D. Whittington (1991) Rent Seeking in Water Supply World Bank
Infrastructure and Urban Development Department Discussion Paper 1991.

Moss, T. (2005) Ten Myths of the International Finance Facility, Center for Global
Development Working Paper 60.

Muzzini, E. (2006) Consumer Participation in Infrastructure Regulation: Evidence from
East Asia and Pacific Region. World Bank Working Paper No. 66.

OED (World Bank Operations Evaluation Department) (1999) Sub-Saharan Africa:
Lessons from Four Sectors Washington DC: World Bank.

OED (World Bank Operations Evaluation Department) (2004) Annual Review of
Development Effectiveness Washington DC: World Bank.

OED (World Bank Operations Evaluation Department) (2005a) World Bank Support for
Capacity Building in Africa Washington DC: World Bank.

OED (World Bank Operations Evaluation Department) (2005b) Improving Investment
Climates: An Evaluation of World Bank Group Assistance Washington DC: World Bank.

Olken, B. (2004) Monitoring Corruption: Evidence from a Field Experiment in Indonesia,
NBER Working Paper 11753.

Parker, D. (2002) Economic Regulation: A Review of Issues Annals of Public and
Cooperative Economics 73, 4.

Rossotto, C. et al (2004) Competition in International Voice Communications, World
Bank Working Paper No.

Satyanand, P. (2007) Requiring Indian Utilities to Report: Harnessing Disclosure
Legislation ot Improve Water and Sanitation Service, mimeo, World Bank.

Shah, A. and M. Shackter (2004) Combating Corruption: Look Before You Leap,
Finance and Development 41, 4.

Smith, T. (2003) Privatising Electric Power in Malaysia and Thailaind: Politics and
Infrastructure Development Policy Public Administration and Development 23.

Sohail, M. and S. Cavill (2007) Partnering to Combat Corruption in Infrastructure
Services: A Tool-kit, mimeo, Loughborough University

Sundet, G (2004) Public Expenditure and Service Delivery Monitoring in Tanzania:
Some International Best Practices and a Discussion of Present and Planned Tanzanian
Initiatives, HakiElimu Working Paper No. 7.

Van de Walle, D. and D. Cratty (2005) Do Donors Get What they Pay For? Micor
Evidence on the Fungibility of Development Project Aid World Bank Policy Research
Working Paper 3542.



                                                                                    29

Wong, S. and S. Guggenheim (2005) Community-Driven Development:
Decentralization's Accountability Challenge in World Bank (ed) East Asia
Decentralizes: Making Local Government Work World Bank: Washington DC.

World Bank (1998)Assessing Aid: What Works, What Doesn't and Why New York:
Oxford University Press.

World Bank (2000) Reducing Corruption: Lessons from Venezuela PREM Notes No. 39.

World Bank (2005) Connecting East Asia, A New Framework for Infrastructure,
Washington DC: World Bank.

World Bank (2005) Financing Information and Communication Infrastructure Needs in
the Developing World, world Bank Working Paper No. 65

World Bank (2006) Held by the Visible Hand: The Challenge of State-Owned Enterprise
Corporate Governance for Emerging Markets Washington DC: World Bank.

World Bank (2006) Implementation of the Management Response to the Extractive
Industries Review Washington DC, World Bank.

World Bank (2006b) Infrastructure at the Crossroads: Lessons from 20 Years of World
Bank Experience, Washington DC: World Bank.




                                                                                  30

Figure 1: Access to Water

   of                                                                 100
                                                 )
     %(ec                                                              90
                                                                       80
                                                  ccessa               70
         oursretaw                                      htiw           60
                                                                       50
                                                                       40
                  de                                                   30

                    ovrp                                    noitalupop 20
                                                                       10
                        mI                                              0
                                                                        100    1000              10000    100000

                                                                                GNI Per Capita (PPP)



Figure 2: Electricity Household Access


                          d                     100
                           ieifrtcelE                                 80

                                                                      60


                                     holdse                           40

                                                                      20
                                           Hous
                                               %                      0
                                                                       100            1000                10000

                                                                                GNI per Capita PPP




                                                                                                               31

Figure 3: Electricity T&D Losses


   oni                                   60

      butirtsiD                )         50
                                put
                                   utO 40
               nda                 %( 30
                  onissi           ses   20

                        m             Los10
                         nsa              0
                            Tr
                                          100    1,000             10,000    100,000

                                                   GNI per Capita (PPP)




                                                                                        32