The United Nations Convention against Corruption
National
Anti-Corruption
Strategies
A Practical Guide for
Development and
Implementation
UNITED NATIONS
New York, 2015
UNITED NATIONS OFFICE ON DRUGS AND CRIME
Vienna
The United Nations
Convention against Corruption
National Anti-Corruption
Strategies:
A Practical Guide for Development
and Implementation
© United Nations, September 2015. All rights reserved, worldwide.
The designations employed and the presentation of material in this publication do not
imply the expression of any opinion whatsoever on the part of the Secretariat of the
United Nations concerning the legal status of any country, territory, city or area, or of
its authorities, or concerning the delimitation of its frontiers or boundaries.
This publication has not been formally edited.
Publishing production: English, Publishing and Library Section, United Nations Ofce
at Vienna.
iii
Acknowledgements
This Guide is a product of the Corruption and Economic Crime Branch of the United
Nations Ofce on Drugs and Crime (UNODC) and has been developed in line with the
thematic programme entitled “Action against corruption, economic fraud and identity-
related crimes (2012-2015)” and pursuant to resolution 5/4 of the Conference of the
States Parties to the United Nations Convention against Corruption.
UNODC wishes to thank its consultants Richard Messick and Matthew Stephenson for
their substantive contributions to the drafting of this Guide.
UNODC also acknowledges with profound gratitude those who have contributed their
expertise and experience at various stages of the development of this Guide and the
experts who participated in the international expert group meeting held in Vienna from
6 to 8 May 2015: Roberta Solis Ribeiro, Chief, International Affairs Advisory, Ofce
of the Comptroller-General, Brazil; Levan Kakava, Head, Investigation Service, Ministry
of Finance, Georgia; Levan Shanava, Specialist, Investigation Service, Ministry of
Finance, Georgia; Daviti Simonia, Investigator of Especially Important Cases, Anti-
Corruption Agency, Georgia; Orsolya I-Valde, Legal Adviser, National Protective
Service, Hungary; Kaushik Goburdhun, Chief Legal Adviser, Independent Commission
Against Corruption, Mauritius; Theunis Keulder, Chair, Namibia Institute for Democracy;
Lise Stensrud, Policy Director, Norad, Norway; José das Neves, Deputy Commissioner
for Education, Campaigns and Research, Anti-Corruption Commission, Timor-Leste;
Zelia Trindade, Deputy Prosecutor-General, Timor-Leste; Phil Mason, Senior Anti-
Corruption Adviser, Financial Accountability and Anti-Corruption Team, Department for
International Development, United Kingdom of Great Britain and Northern Ireland;
Jennifer Widner, Professor of Politics and International Affairs, Princeton University,
United States of America; James Wasserstrom, Strategy Adviser, Special Inspector-
General for Afghanistan Reconstruction, Washington, D.C., United States; Gillian Dell,
Transparency International; Anga Timilsina, Programme Manager, Global Anti- corruption
Initiative, United Nations Development Programme; and Jonathan Agar, Associate Legal
Ofcer, Ofce of Legal Affairs of the Secretariat.
UNODC wishes to acknowledge the contributions of UNODC staff members Ronan
O’Laoire and Constantine Palicarsky, who were responsible for the development of the
Guide, and also acknowledges the expertise and comments provided by the following
staff members: Dimitri Vlassis, Candice Welsch, Oliver Stolpe, Robert Timothy Steele,
Jason Reichelt, Samuel De Jaegere, Zorana Markovic, Virginia de Abajo-Marqués,
Annika Wythes, Troels Vester, Claudia Sayago, Shervin Majlessi, Jennifer Sarvary
Bradford and Akharakit Keeratithanachaiyos.
UNODC wishes to express its gratitude to the Government of Australia for its generosity
in providing funding for the development of this Guide.
v
Contents
Introduction ............................................................... 1
I. Drafting process ........................................................ 5
A. Assign responsibility for drafting the strategy to a small,
semi-autonomous group .............................................. 5
B. Ensure the continued support and involvement of senior political leaders ..... 6
C. Consult regularly with all government agencies that will be affected by
the strategy ......................................................... 7
D. Solicit the views of the political opposition whenever possible ............... 8
E. Engage all sectors of society in the drafting process ....................... 8
F. Emphasize communication, transparency and outreach throughout
the drafting process .................................................. 10
G. Allocate sufficient time and resources to drafting the strategy ............... 10
H. Take advantage of other countries’ experience and expertise ................ 11
II. Preliminary diagnosis and situation analysis ................................. 13
A. Conduct a preliminary diagnosis of corruption challenges .................. 13
1. Self-assessments, peer reviews and other assessments ................ 15
2. Cross-country comparisons of corruption or governance ................ 16
3. Country-specific surveys of corruption perceptions .................... 17
4. Surveys of actual experience with corruption. . . . . . . . . . . . . . . . . . . . . . . . . . 18
5. Internet platforms and social media ................................. 19
6. Information collected by government agencies ........................ 20
7. Comparisons of different data sources ............................... 21
8. Vulnerability assessments ......................................... 22
9. Proxy measurements ............................................. 23
B. Assess obstacles to effective reform .................................... 23
1. Evaluate resource constraints ...................................... 23
2. Address potential opposition and support ............................ 24
III. Formulating anti-corruption measures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
A. Tailor the strategy to the diagnosis, taking account of constraints ............ 27
B. Be ambitious but realistic ............................................. 27
C. Identify concrete, specific measures to be taken .......................... 28
vivi
D. Describe the objective of each reform element ............................ 29
E. Consider the costs, benefits, burdens, opposition and support
for each element. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
F. Pay attention to prioritization and sequencing ............................ 30
G. Specify implementation times for different reforms ........................ 31
IV. Ensuring effective implementation ........................................ 33
A. Put a single, high-level entity in charge of coordination and implementation ... 34
B. Provide the coordination and implementation body with sufficient authority .... 35
C. Foster cooperation between the coordination body and the implementing
agencies ........................................................... 36
D. Harness the power of reputation ....................................... 36
E. Have each agency agree to an implementation, monitoring and
evaluation schedule .................................................. 37
F. Do not underestimate the challenges of coordinating implementation ........ 38
V. Monitoring, evaluating and reporting ....................................... 39
A. Monitoring and evaluating implementation ............................... 40
1. Disaggregate policy reforms into discrete steps ....................... 41
2. Select one or more indicators of progress ............................ 41
3. Choose a baseline for each indicator ................................ 43
4. Establish realistic targets for each implementation indicator ............ 43
5. Watch for indicator manipulation ................................... 44
6. Be cautious when using agency self-evaluations ...................... 44
7. Utilize evaluations to adjust implementation targets and strategy goals ... 44
8. Allocate sufficient time and adequate resources for evaluation ........... 45
B. Monitoring and evaluating impact ....................................... 45
1. Do not use year-to-year changes in corruption index scores to measure
strategic impact ................................................. 46
2. Select impact indicators that can be compared over time ............... 47
3. Isolate the impact of the policies implemented pursuant to the strategy ... 47
4. Be sensitive to the cost and time required ............................ 49
5. Involve civil society organizations, scholars, research organizations
and citizens ..................................................... 50
6. Provide methods for ongoing revision to the strategy ................... 50
C. Public reporting of results of monitoring and evaluation .................... 50
1. Internal reporting as part of coordination mechanisms, ensuring the
accountability of the implementation process ......................... 50
2. Reporting to higher authorities in the executive or parliament ........... 50
3. Public reporting ensures transparency in the implementation of
the national strategy .............................................. 50
Annex. Kuala Lumpur statement on anti-corruption strategies ................... 53
1
Introduction
The United Nations Convention against Corruption is a comprehensive international
instrument intended to combat the scourge of corruption around the world. As observed
in the preamble to the Convention, corruption not only threatens the stability and security
of societies, the institutions and values of democracy, ethical values and justice, sustain-
able development and the rule of law, but is also a transnational phenomenon that affects
all societies and economies, making international cooperation to prevent and control it
essential. With ratication by 176 States parties (as at 24 July 2015), the Convention
has established opposition to corruption as a global norm and made the elimination of
corruption a global aspiration.
States parties to the Convention must undertake effective measures to prevent corruption
(chapter II, articles 7 to 14), criminalize corrupt acts and ensure effective law enforce-
ment (chapter III, articles 15 to 42), cooperate with other States parties in enforcing
anti-corruption laws (chapter IV, articles 43 to 50) and assist one another in the return
of assets obtained through corruption (chapter V, articles 51 to 59). Moreover, in addition
to calling for effective action in each of these specic areas, article 5 imposes the more
general requirements that each State party: (a) develop and implement or maintain effec-
tive, coordinated anti-corruption policies; (b) establish and promote effective practices
aimed at the prevention of corruption; and (c) periodically evaluate relevant legal instru-
ments and administrative measures with a view to determining their adequacy to prevent
and ght corruption (see box 1). Furthermore, under article 6, each State party is required
to ensure the existence of a body or bodies, as appropriate, that prevent corruption by
implementing the policies referred to in article 5 and, where appropriate, overseeing and
coordinating the implementation of those policies. Thus, one of the most important
obligations of States parties under the Convention, and to which they are to be held
accountable under the Mechanism for the Review of Implementation of the Convention
established under article 63,
1
is ensuring that their anti-corruption policies are effective,
coordinated and regularly assessed.
1
See Conference of the States Parties to the United Nations Convention against Corruption resolutions 3/1 and 4/1
(contained in documents CAC/COSP/2009/15 and CAC/COSP/2011/14).
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
2
Box 1. Article 5 of the United Nations Convention against Corruption (Preventive
anti-corruption policies and practices)
1. Each State Party shall, in accordance with the fundamental principles of its legal
system, develop and implement or maintain effective, coordinated anti-corruption
policies that promote the participation of society and reflect the principles of the rule
of law, proper management of public affairs and public property, integrity, transparency
and accountability.
2. Each State Party shall endeavour to establish and promote effective practices
aimed at the prevention of corruption.
3. Each State Party shall endeavour to periodically evaluate relevant legal instruments
and administrative measures with a view to determining their adequacy to prevent and
fight corruption.
4. States Parties shall, as appropriate and in accordance with the fundamental prin-
ciples of their legal system, collaborate with each other and with relevant international
and regional organizations in promoting and developing the measures referred to in
this article. That collaboration may include participation in international programmes
and projects aimed at the prevention of corruption.
For many countries, achieving the objective of article 5 of the Convention against
Corruption may entail the drafting, publication and implementation of a national anti-
corruption strategy; in other words a blueprint for a realistic, comprehensive and inte-
grated plan for reducing corruption in that country. A formal, written strategy is by no
means required for compliance with articles 5 and 6; States parties can maintain effective,
coordinated anti-corruption policies without promulgating a strategy document. But, as
referred to in the Kuala Lumpur statement on anti-corruption strategies (see annex),
publishing a national anti-corruption strategy can be an effective way for States parties
to ensure they meet their obligations under article 5. In the statement, which the
Conference of the States Parties took note of in its resolution 5/4, entitled “Follow-up
to the Marrakech declaration on the prevention of corruption”, it is also recognized that
anti- corruption strategies could provide a comprehensive policy framework for actions
to be taken by States in combating and preventing corruption and could be a useful tool
for mobilizing and coordinating the efforts and resources of Governments and other
stakeholders, for policy development and implementation and for ensuring the monitoring
of policy implementation. Indeed, to date, UNODC has identied over 70 countries that
have issued either a single national anti-corruption strategy or a set of documents that
together constitute a comprehensive, coordinated anti-corruption framework of the kind
foreseen in the Kuala Lumpur statement.
Existing national anti-corruption strategy documents vary widely in length, detail, scope
and emphasis. This is both expected and desirable. There is no “one-size-ts-all” approach
to producing an effective strategy, particularly given that different countries have very
different legal, cultural and political traditions and face very different challenges, oppor-
tunities and constraints. But there has also been considerable variation, and often con-
siderable disappointment, in how effective different countries’ strategy documents have
been in achieving the goals outlined in the Kuala Lumpur statement. For example, in a
2014 report to the Council of the European Union and the European Parliament, the
European Commission found that although in some cases, work on national anti-
corruption strategies was a catalyst for genuine progress, in some others, impressive
strategies had little or no impact on the situation on the ground.
2
Other assessments of
2
European Commission, “Report from the Commission to the Council and the European Parliament: EU anti-corruption
report”, document COM(2014) 38.
3
Introduction
the experience with national strategies have likewise found considerable variation in their
quality and impact.
3
Moreover, even though there is no single correct approach to devel-
oping a national anti-corruption strategy, there are common challenges and difculties
as well as lessons to be learned from past successes and failures.
Responding to a request by the Conference of the States Parties to the Convention in
its resolution 5/4 to identify and disseminate good practices among States parties regard-
ing the development of national anti-corruption strategies, and consistent with the respon-
sibility of the United Nations Ofce on Drugs and Crime (UNODC) to provide technical
assistance to States parties to help them meet their obligations under the Convention,
4
this Guide offers recommendations for countries considering drafting or revising a
national anti-corruption strategy. Although the focus of the present Guide is on recom-
mendations for the development of formal national anti-corruption strategy documents,
many of the recommendations may also be relevant to countries that plan to meet their
obligations under article 5 without adopting a formal strategy.
The analysis is broken down into the following ve key aspects of an effective national
anti-corruption strategy document:
1. The drafting process for the strategy should be overseen by a body that has
sufcient autonomy, expertise and political backing, and should involve substantive
input from key stakeholders from both inside and outside the Government;
2. The strategy should contain a preliminary evaluation and diagnosis of the main
corruption challenges that the country faces, including the obstacles to the imple-
mentation of an effective anti-corruption policy. The preliminary diagnosis should
also identify gaps or limitations in current knowledge or understanding of those
issues;
3. Based on the preliminary evaluation and diagnosis, the strategy should contain
an anti-corruption policy that lays out ambitious but realistic objectives, identies
top priorities in both the near term and longer term and establishes the appropriate
sequencing of reforms;
4. The strategy should include an implementation plan in which responsibility for
overseeing its execution is assigned to a coordination unit and mechanisms to ensure
the various agencies carrying out different aspects cooperate with one another are
provided for;
5. The strategy should contain a plan for monitoring and evaluating the plan’s
implementation and impact to ensure that the elements of the policy plan are properly
executed, that they are having the desired impact and that they can be revised as
necessary.
Two preliminary observations are in order before proceeding. First, a national anti-
corruption strategy is not merely a technocratic document whose implementation depends
on whether national leaders are sufciently determined to do so. One purpose of devel-
oping and promulgating a national anti-corruption strategy is to help generate and
3
See United Kingdom of Great Britain and Northern Ireland, Department for International Development, Why Cor-
ruption Matters: Understanding Causes, Effects and How to Address Them. Evidence Paper on Corruption (London,
January 2015); Southeast Europe Leadership for Development and Integrity, Anti-Corruption Reloaded: Assessment of
Southeast Europe, rev. ed. (Soa, Center for the Study of Democracy, 2014); United Nations Development Programme,
Anti-Corruption Strategies: Understanding What Works, What Doesn’t and Why? Lessons Learned from the Asia-Pacic
Region (Bangkok, 2014); Maíra Martini, “Examples of national anti-corruption strategies”, Anti-Corruption Helpdesk
Answer (Berlin, Transparency International, 2013); Karen Hussmann, ed. Anti-Corruption Policy Making in Practice:
What can be Learned for Implementing Article 5 of UNCAC? Report of Six Country Case Studies: Georgia, Indonesia,
Nicaragua, Pakistan, Tanzania and Zambia (Bergen, Norway, U4 Anti-Corruption Resource Centre, 2007).
4
See CAC/COSP/2009/15, sect. I.A, resolution 3/4, entitled “Technical assistance to implement the United Nations
Convention against Corruption”.
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
4
maintain the leadership, citizen demand and broad support necessary to act effectively
against corruption. A strategy document that simply identies a series of sound, sensible
policies will not advance these objectives. The document, as well as the processes for
drafting it and for ensuring post-promulgation implementation and revision, should create
a sense of ownership and commitment among those both inside and outside the
Government, and should establish targets and benchmarks that enable both domestic and
foreign audiences to hold the Government accountable to the goals it has set for itself.
Second, an overarching theme of this Guide is that strategy drafters must strike a balance
between ambition and realism. A strategy that is too cautious and limited—one that
proposes only easy-to-achieve policy changes—may not address a country’s most press-
ing corruption concerns and will thus not generate the excitement, focus and pressure
needed to make signicant progress in preventing and combating corruption. On the
other hand, an overly ambitious strategy, with grand objectives that are not feasible given
the country’s resources and capacity, will be ignored or dismissed as impractical, and
might further contribute to a sense of frustration and resignation among the citizenry.
The balance between ambition and realism is not an easy one to strike, and the right
approach will vary by country, but those responsible for developing the national anti-
corruption strategy must be mindful of it at every stage of the process.
5
Drafting process
The rst step in formulating an effective national anti-corruption strategy is designing
the process for drafting the document. Decisions regarding process are crucial, as they
affect both the content of the document and the likelihood of its successful implemen-
tation. The particular institutions and individuals who will take responsibility for drafting
the strategy will vary from country to country, and political considerations, resource
constraints and other factors will dictate at least certain aspects of the process. None-
theless, there are some general principles that may be worth following, or at least
following as closely as possible, when setting up a process for drafting a national anti-
corruption strategy. The eight principles listed below are worth emphasizing in
particular.
A. Assign responsibility for drafting the strategy
to a small, semi-autonomous group
Designing the drafting process is critical and great care should be taken at this crucial
rst stage. A relatively small committee or organization should have primary responsi-
bility for drafting the strategy document and should have a reasonable degree of auton-
omy in developing the draft. A clear directive from a senior political leader, preferably
the Head of the Government, setting out the drafting unit’s mandate and responsibilities,
can be useful for clarifying responsibilities and avoiding “turf battles” among different
agencies.
5
Ideally, the drafting body should be chaired by an individual with sufcient stature,
legitimacy and political inuence to act as an effective “champion” for the drafting body,
and ultimately for the strategy itself. The chair should serve both as an effective liaison
with other senior political leaders and as the public face of the national anti-corruption
strategy-drafting process. He or she need not necessarily be involved in the details or
technical aspects of the drafting process, but should have an overall leadership and
managerial role.
5
In the United Kingdom, the Prime Minister issued a directive putting the Home Ofce and the Cabinet Ofce in
charge of drafting the national strategy. See John Bray, “UK national anti-corruption plan: coming soon”, Integrity Matters,
No. 13 (London, Control Risks, 2014).
I.
6
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
A balance must be struck when selecting the members of the drafting body because—as
will be emphasized below—getting broad-based participation and support from a range
of stakeholders is essential to producing an appropriate and effective strategy. At the
same time, however, a national anti-corruption strategy is not likely to be effective if it
is a long “wish list” or a haphazard collection of disconnected, unorganized ideas and
suggestions. It is almost always advisable to identify a small number of primary drafters
responsible for the language of the document. Such individuals must be carefully selected
because they need to be able to consult broadly with a range of stakeholders and experts,
but not have an attachment to any preconceived agenda. At the same time, the primary
drafters also need to construct a strategy that has a coherent vision. They should have
sufcient stature to be respected and sufcient technical expertise to be able to produce
a rigorous, high-quality document. They should also be open to input from different
sources and should represent a diverse group of people so that they are trusted by a
wide range of stakeholders. These qualities are not always found in the same person, so
it may be important to assign the primary drafting responsibilities to a diverse
committee of individuals with different but complementary skill sets, who can work
together effectively.
One approach that a number of countries have followed is to establish a high-level
drafting committee supported by a small group of technical experts responsible for
actually writing the strategy. In Cameroon, a six-person drafting team reported to an
eight-member national commission representing relevant ministries, the legislature, the
supreme court, civil society and the business community. In Pakistan, a small project
team drawn from the anti-corruption bureau and key agencies worked under a steering
committee chaired by a high-ranking government ofcial and composed of senior gures
from across State institutions, civil society, the media and the private sector.
6
B. Ensure the continued support and involvement
of senior political leaders
It is vital that the national anti-corruption strategy has high-level political support, and
that such support is maintained throughout the process. Ideally, senior political leaders,
preferably the Head of State, should openly and publicly endorse the strategy-drafting
process at the outset. During the drafting process, support from senior leaders can be
demonstrated in a variety of ways, including speeches on the importance of having a
comprehensive national anti-corruption strategy and public statements urging citizens to
provide input to the committee. In addition to public endorsements at the beginning and
end of the process, it is also helpful if senior political leaders are kept informed and
involved throughout the development of the strategy, perhaps sometimes attending meet-
ings of the drafting committee.
Once the drafting process is complete, a formal, legal endorsement of the nal document
strategy may be required. The form of this endorsement will depend on the nation’s
legal and political traditions. Some countries may require the strategy to be submitted
to parliament for approval by way of a resolution or even legislation; in others, an
executive decree may sufce. Even where formal endorsement is not necessary, it can
be a useful way of cementing support for the strategy. For many countries, a legislative
vote of approval will be the best means of doing so.
6
Unless otherwise noted, information about a State’s strategy is taken from the text of its national anti-corruption
strategy.
7
Chapter I. Drafting process
Admittedly, senior political leaders may sometimes have different views about the causes
of corruption or the priorities for ghting it, and there are legitimate concerns that if
such leaders are involved in the drafting process, they might use their inuence in ways
that many anti-corruption advocates (perhaps including some members of the drafting
body) might view as weakening the strategy. But a national anti-corruption strategy is,
by its nature, a “top-down” mechanism that relies on the efforts of a range of government
ministries, agencies and institutions for its success. Without high-level political support
and without the sense throughout every level of government that the senior leadership
backs the strategy and its drafters, a national anti-corruption strategy is unlikely to be
effective, no matter what it says on paper. If the senior leadership will not support a
robust anti-corruption strategy, a better course may to be to focus on the development
of a broad-based consensus on problems and solutions, and defer drafting a strategy
until some future date.
C. Consult regularly with all government agencies
that will be affected by the strategy
The national anti-corruption strategy-drafting process should ensure the appropriate
participation of representatives of any government agency responsible for carrying out
any part of the anti-corruption strategy. This will include executive branch agencies such
as the ministries of justice and interior, the specialized anti-corruption agency, the
ombudsman, the police, nancial intelligence units, agencies responsible for public
procurement and the civil service commission or other entities in charge of recruiting,
promoting and disciplining public employees. Where the strategy will affect or require
action by agencies outside the direct control of the executive—the judiciary, the legis-
lature, independent entities such as the national audit agency, independent regulatory
commissions and regional or local governments—they too must be part of the process.
In Ghana, for example, strategy drafters sought input from parliamentarians and the
judiciary, while in Estonia consultations were held with a number of entities independent
of the Government, including the public prosecutor and the competition law agency. In
Peru, special workshops were organized to hear the views of members of the judiciary
and the supreme audit agency, and input from the ombudsman (defensor del pueblo)
was sought throughout the drafting process.
The inclusion of various agencies in the strategy-drafting process is important for numer-
ous reasons. Most obviously, representatives of different agencies have different kinds of
expertise that may improve the quality of the strategy, for example by identifying problems
and challenges that the lead drafters may have missed (or misunderstood) and by sug-
gesting creative solutions. It is also vital that at least some participants in the drafting
process can provide accurate nancial and budgetary information, as these considerations
will be critical to effective implementation. But perhaps even more importantly, different
agencies will typically have responsibilities when it comes to implementing the strategy,
and their acceptance and active support is critical if the strategy is to succeed. Participation
in the drafting process will give them more of a sense of “ownership” of the strategy.
There is always a risk that representatives of government agencies will be overly resistant
to reforms (a common, though not always justied, complaint about bureaucrats every-
where). The views of government agencies should be respected but not considered dis-
positive. On balance, it is better to involve representatives of all relevant government
departments in the process early, and to have a chance to address their concerns, rather
than to present them with a strategy that they were not involved in developing and thus
may resist. Partly for that reason, the consultations should extend not only to the political
8
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
leadership of the various implementing agencies, but also to the technical staff or career
civil servants who will play a key role in implementing the strategy’s recommendations.
Another advantage of involving multiple agencies in the strategy-drafting process is that
it can improve these agencies’ ability to coordinate and cooperate on anti-corruption
issues; their interaction in the drafting process may facilitate cooperation on implemen-
tation, monitoring and evaluation—all of which are critical to the strategy’s success.
D. Solicit the views of the political opposition
whenever possible
Perhaps more controversially, it may be advisable during the strategy-drafting process
to involve not only members of the current Government, but also members of the political
opposition (the parliamentary opposition, or possibly leaders of regional or local gov-
ernments from opposition parties). Their involvement can be useful for many reasons.
First, input from the opposition is a gesture of good faith by the Government—an
endorsement of the idea that ghting corruption is not a partisan issue—and it may
reduce the degree to which the national anti-corruption strategy is viewed as a political
tool of the current administration. Second, opposition parties are often well-positioned
to raise and represent the views of constituencies that might otherwise be marginalized
in the drafting process; the opposition can also raise concerns if the strategy-drafting
process is ignoring issues or concerns relating to certain communities. Third, Govern-
ments change—yesterday’s opposition can become tomorrow’s ruling party—and the
long-term sustainability of a national anti-corruption strategy often depends on whether
it continues to enjoy support following a change in Government.
The inclusion of opposition representatives in the drafting process is, however, not with-
out risk. The presence of opposition representatives in drafting discussions may chill the
free exchange of ideas, and in some cases opposition parties may exploit their role to
advance their political interests or even to sabotage the process to deny the Government
a politically popular achievement. Even when those fears are overblown, the Government
may nonetheless be reluctant to include the opposition in the drafting process. Where
direct participation by opposition representatives is not feasible or desirable, it would
still be appropriate for the strategy’s drafters to consult with leading opposition gures
as part of a more general consultation with a range of experts.
E. Engage all sectors of society in
the drafting process
The strategy drafters should nd appropriate ways to engage broadly with those outside
of the Government: civil society organizations, the business community, the media,
academics, the general public and other stakeholders. The reasons are straightforward:
rst, such stakeholders may have valuable information and useful recommendations for
crafting a more effective strategy that is better tailored to the country’s particular needs
and circumstances; second, inclusion of a broad range of voices in the drafting process
may help build a common vision and increase the legitimacy of the strategy, and hence
political support for it, in the wider society. Those who feel that their voices were heard
in the creation of a policy are more likely to be allies in pushing the strategy forward
and ensuring its effective implementation.
7
7
Elinor Ostrom, “Beyond markets and States: polycentric governance of complex economic systems”, American
Economic Review, vol. 100, No. 3 (2010), pp. 641-672.
9
Chapter I. Drafting process
Broad-based consultations are not without their risks, and although steps can be taken
to minimize them and ensure a productive dialogue (see box 2), strategy drafters need
to bear these risks in mind. Too much emphasis on achieving consensus or trying to
reect the input of all stakeholders can result in a watered-down strategy, one full of
generalities and aspirations but short on concrete choices among competing priorities
and approaches. It is also possible that including outside stakeholders in the drafting
process might backre: if participating stakeholders feel that their views were ignored,
or that the consultations were a sham, they may feel alienated by the strategy-drafting
process, and thus be less likely to become strong allies in its implementation. The strat-
egy drafters must do their best to ensure that the various stakeholders feel that their
views are taken seriously and that they have an impact on the drafting process. At the
same time, drafters must be clear that, ultimately, the draft will integrate a range of
views and perspectives into a coherent document, meaning that the nal national anti-
corruption strategy will not always do or say exactly what every stakeholder would like.
Box 2. Ensuring a productive dialogue with stakeholders
To improve policy implementation, the Government of Australia issued a guidance note
to ministries that included a chapter on stakeholder consultation. It recommends that
before engaging stakeholders, the following questions be addressed:
• Is the purpose and benefit of stakeholder engagement during the implemen-
tation clear?
• Have the right stakeholders been identified?
• Has the cost-effectiveness of different communication channels been
considered?
• Is a wide consultation process using social media beneficial?
• Has sufficient consideration been given to how stakeholder interactions will be
managed during the implementation phase?
• Is there clear accountability for stakeholder engagement, including managing
expectations?
• When is the best time for stakeholders to be engaged?
• How will the information obtained through stakeholder engagement be acted on?
• Is there a communication strategy for consulting with stakeholders?
• If government advertising is proposed, have relevant guidelines been consid-
ered and applied?
• Have staff been provided with guidance on identifying potential conflicts of inter-
est? In particular, are there satisfactory arrangements to manage perceptions
or instances of conflict of interest that may arise from consulting with
stakeholders?
Source: Australia, Department of the Prime Minister and Cabinet and Australian National Audit Office,
Better Practice Guide: Successful Implementation of Policy Initiatives (Canberra, 2014), p. 36.
These considerations have important implications for the timing of consultations with
stakeholders as well. A key question is whether to wait until a rough draft of the strategy
document is ready before engaging with stakeholders, or whether consultations should
begin earlier in the process. Both alternatives carry risks. If drafters wait until they have
a text, they might be accused of using the consultations not as a genuine effort to solicit
outside input, but simply as a way to win support and legitimacy for what they wanted
to do anyway. On the other hand, open-ended consultations can produce long, unfocused
lists of suggestions. Not only can this be inefcient, but it also risks alienating those
whose suggestions are not ultimately included in the nal draft. One way to navigate
10
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
between these pitfalls might be to structure consultations in stages, rst soliciting general
input about the country’s corruption challenges, then producing a preliminary draft or
outline and then seeking more focused input. Another, complementary approach might
be to structure early consultations around recent ndings concerning the country’s cor-
ruption problems. For instance, if there is a current survey reporting citizens’ experiences
with bribery in different government agencies, the strategy drafters could seek input from
stakeholders on their perspective on these ndings, and what might be done about them.
F. Emphasize communication, transparency and
outreach throughout the drafting process
In addition to (and perhaps in conjunction with) the consultations with stakeholders, the
drafters of the national anti-corruption strategy should endeavour to be transparent about
the process to the extent feasible (though of course sometimes condentiality might be
required, for example in the case of consultations with other government departments
on sensitive internal matters). Besides maintaining transparency, the Government should
have a plan to publicize the work of the national anti-corruption strategy-drafting body,
so that the general public and interested stakeholders know that their input is welcome
and to explain why ghting corruption, and having a national anti-corruption strategy,
are considered priorities. An effective media and communications strategy serves many
functions: encouraging more, and more diverse, input into the process; increasing the
legitimacy of the strategy-drafting process; sending the signal that the Government is
committed to ghting corruption; and sustaining interest and attention to the strategy-
drafting process, particularly given that the process is likely to take time.
G. Allocate sufficient time and resources
to drafting the strategy
The strategy-drafting process needs adequate time and resources. It is simply not realistic
to suppose that within the space of a few weeks the drafting committee will be able to
consult widely with a broad range of stakeholders inside and outside of Government,
assemble and review the relevant data and evidence, gather new data where there are
critical gaps in the knowledge base required to produce a sound national anti-corruption
strategy and develop a set of well-reasoned, coordinated policy proposals. Providing
adequate time and resources is therefore critical.
How long it will take to draft the strategy will depend on several factors: does the
strategy follow on from an earlier one? How much background research has already
been done? How easy will it be to obtain information from different agencies about their
activities? Another important consideration will be the time required to consult with
different stakeholders. How difcult will it be to hold meetings across the country? How
long will it take to schedule them?
Even with a good deal of background information at hand, drafters in Cameroon found
it a major challenge to prepare a rst draft in the three months allotted for that task.
8
Drafters in Pakistan were able to produce a draft in a little over 10 months, but much
of the research they needed had already been done, and signicant support from the
donor community helped them accelerate the process. A 2014 survey by the United
8
Cameroon, Commission Nationale Anti-Corruption, Stratégie nationale de lutte contre la corruption 2010-2015
(Yaoundé, 2010), p. 13.
11
Chapter I. Drafting process
Nations Development Programme (UNDP) of strategies in the Asia-Pacic region found
that drafting a strategy usually takes between one and two years.
9
The point about the time and resources required for drafting a good strategy is worth
emphasizing because the development of a national anti-corruption strategy is often
prompted by a scandal, putting leaders under pressure to act quickly and decisively. In
other cases, a newly elected Government may want to put an anti-corruption strategy in
place quickly to demonstrate its commitment to a campaign pledge to ght corruption.
Proponents of a national anti-corruption strategy may need to take advantage of these
windows of political opportunity. In some cases, it might be necessary to accelerate the
process of drafting a strategy for other reasons, such as completing the process within
the current budget cycle. In contrast, when an election is pending, it may be unrealistic
to expect signicant progress on drafting a national anti-corruption strategy until after
the election is concluded. Whatever the circumstances, all too often, countries allocate
too little time and too few resources to producing a high-quality strategy document, and
as a result the strategy has little impact in the long run.
H. Take advantage of other countries’
experience and expertise
Strategy drafters should make productive but judicious use of experts from other nations
with experience of drafting, advising on or implementing national anti-corruption strat-
egies. Many States parties to the Convention, particularly developing countries and coun-
tries in transition, may have signicant technical capacity limitations, but all countries
are likely to benet from the expertise and differing views that outside experts can bring
to the drafting process. Such knowledge exchanges are encouraged under article 60 of
the Convention, in which States parties are called upon to consider affording one another
the widest measure of technical assistance in their respective plans and programmes to
combat corruption, including the mutual exchange of relevant experience and specialized
knowledge. Foreign consultants can add signicant value, both through the additional
technical capacity they can bring to bear and their ability to draw from the experiences
of other jurisdictions.
At the same time, however, it is vital that each country retains full ownership of its
national anti-corruption strategy, and that the strategy is appropriately tailored to the
specic needs and circumstances of that country. The drafting of a national anti- corruption
strategy should not be “outsourced” to foreign experts. This is not only because foreign
experts may not always fully understand the unique challenges and circumstances of a
particular country, but also because a national anti-corruption strategy is likely to be
effective only if there is a strong sense of domestic ownership and commitment. Too
often, countries hire foreign consultants to draft their national anti-corruption strategy
documents and there is only minimal input from the domestic ofcials who are entrusted
with implementing the strategy and the citizens who the strategy is supposed to benet.
In these cases, when the foreign experts depart, the strategy is unlikely to be implemented
effectively, and in many cases may languish in obscurity.
A related, and important, point concerns the role of donor support for the national
anti-corruption strategy-drafting process. If taken seriously and done right, producing a
national anti-corruption strategy will, as explained above, be resource- and time- intensive.
For less afuent countries, therefore, international donor support can be essential.
9
Anti-Corruption Strategies: Understanding What Works, What Doesn’t and Why?, p. 14.
12
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
Further more, international donors, and the increasing attention that the donor community
pays to issues of corruption and good governance, may—along with the pressure arising
from the Government’s obligation to comply with the provisions of the Convention against
Corruption—provide countries with the impetus to move forward with a serious, substan-
tive national anti-corruption strategy. At the same time, however, donors may not always
have a clear understanding of a particular country’s circumstances, and may sometimes
push in directions that are not appropriate for that country. Moreover, a national anti-
corruption strategy that is produced mainly to please international donors is unlikely to
have the sort of genuine domestic support and legitimacy that are prerequisites for long-
term, lasting results. Thus, it is important to consider carefully the appropriate relationship
between international donors and recipient countries, so that donor support can be used
most productively in the development of an effective anti-corruption strategy.
13
Preliminary diagnosis and situation analysis
An effective national anti-corruption strategy must be based on an accurate assessment
of the problems and challenges that the particular country faces in combating corruption.
This preliminary diagnosis, sometimes termed a “situation analysis”, consists of two
main elements: an assessment of the nature, extent and impact of the country’s corruption
problems and an assessment of the obstacles that may hinder the implementation of
effective anti-corruption reform. These two elements are described in more detail below.
A. Conduct a preliminary diagnosis of
corruption challenges
Although corruption is a universal phenomenon, countries face very different types of
corruption problems and challenges. It is impossible to combat a problem that is not
understood. Therefore, as stressed in the Technical Guide to the United Nations Con-
vention against Corruption, the rst step in drafting an effective national anti-corruption
strategy is the collection and analysis of information about the extent and nature of the
nation’s corruption problem.
10
How severe is it? Who is affected the most by it? What
kinds of harm is it causing? What forms does it take? What sectors of the economy are
most affected? In what government agencies is it most prevalent? The more information
the strategy drafters have about the answers to these questions, the greater their ability
to set priorities, allocate resources and align the strategy’s objectives with the problems
that confront the country.
Although a preliminary diagnosis of the country’s corruption problem is a critical input
into a strategy, few strategies devote sufcient attention to this essential rst step. A
content analysis done for this Guide of 53 national strategies found that only 13 con-
tained a detailed discussion of the state of corruption in the country.
11
The remaining
40 contained little diagnosis, merely referencing a few statistics or noting that certain
sectors were more corrupt than others; some simply stated that corruption was a major
10
UNODC and United Nations Interregional Crime and Research Institute (Vienna, 2009), p. 6.
11
Examples of good preliminary diagnoses include the 2009 strategy for Armenia, the strategy for the former Yugoslav
Republic of Macedonia for the period 2011-2015 and two strategies for Estonia, one for the period 2008-2012 and the
other for the period 2013-2020.
II.
14
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
14
problem. The lack of an adequate preliminary analysis in these strategies is consistent
with the results of a 2007 review of national anti-corruption strategies by the U4 Anti-
Corruption Resource Centre,
12
and a 2014 review by UNDP of anti-corruption strategies
in the Asia-Pacic region.
13
Both of these reviews also found that the great majority of
anti- corruption strategies contained little, if any, analysis of the corruption problems the
country faced.
One reason policymakers may be tempted to skip an initial diagnosis and situation
analysis is that they think corruption problems are obvious.
14
Obvious corruption prob-
lems, however, may not be the most severe or most damaging, but simply those that are
most visible or salient. Focusing on such problems is often a misleading basis for design-
ing public policy;
15
the most harmful problems may be hardly noticeable, or even invis-
ible. Even when an obvious problem merits priority attention, an initial diagnosis is still
crucial, because it can help policymakers understand the reasons for the problem and to
craft a more effective response. More importantly, a key part of an initial assessment is
estimating how severe different corruption problems are. This leads to the creation of a
baseline against which the impact of the strategy can later be evaluated, another critical
element in a national strategy discussed in chapter V, section A, subsection 3, below.
Moreover, the preliminary diagnosis of a country’s corruption challenges is not simply
an assessment of the scope or extent of corruption in the country. It must also include
at least a preliminary attempt to evaluate the country’s current experience in combating
corruption. Are there important gaps in the laws? Are the laws enforced effectively? Do
the agencies responsible for investigating and prosecuting corruption have sufcient
resources and technical skills? Which agencies have responsibility for combating
corruption or addressing associated issues? Are their roles and responsibilities clear, or
are there gaps or redundancies? What other aspects of the country’s laws and institutions
may be contributing to the corruption problem? These and other questions should be
taken into account when evaluating the challenges a country faces.
Assessing the nature and extent of a country’s corruption problem, however, is challeng-
ing. Many of the available techniques are time- and resource-intensive, perhaps prohib-
itively so for some countries. As a rst step in the process of developing a national
anti-corruption strategy, the drafters will need to formulate a realistic plan for gathering
information from a range of sources. All information sources have their strengths and
weaknesses, and the drafters need to keep this in mind and use a diverse range of mate-
rials to develop at least a preliminary assessment of the main corruption problems. The
eight sources of information listed below may prove useful in conducting this preliminary
assessment, although, as noted, each has its own limitations and drawbacks.
12
See Hussmann, ed., Anti-Corruption Policy Making in Practice: What can be Learned for Implementing Article 5 of
UNCAC?.
13
Anti-Corruption Strategies: Understanding What Works, What Doesn’t and Why?.
14
That is the reason Georgian ofcials gave for issuing an anti-corruption plan in 2000 without doing an evaluation
rst. See Jessica Schultz and Archil Abashidze, “Anti-corruption policy making in practice: Georgia—a country case
study”, in Anti-Corruption Policy Making in Practice: What can be Learned for Implementing Article 5 of UNCAC?,
Hussmann, ed., p. 69.
15
Daniel Kahneman, “Maps of bounded rationality: psychology for behavioral economics”, American Economic Review,
vol. 93, No. 5, pp. 1449-1475.
1515
Chapter II. Preliminary diagnosis and situation analysis
1. Self-assessments, peer reviews and other assessments
The corruption assessments and reviews done by international organizations and non-
governmental organizations can be a valuable and easily accessible source of information.
All parties to the Convention against Corruption have completed or are in the process
of completing the rst cycle of the Implementation Review Mechanism, during which
compliance with chapter III (Criminalization and law enforcement) and chapter IV (Inter-
national cooperation) was reviewed.
Each implementation review begins with the State party under review submitting a
self-assessment in which it identies the steps taken to comply with the Convention,
including legal provisions, national policies and proof of implementation of these through
case summaries and statistics, as well as an outline of how the institutions forming part
of the national anti-corruption framework operate, at both the domestic and international
levels. Two States parties then act as peer reviewers and, working on the basis of con-
sensus, make observations on gaps that have been identied and recommendations on
how these can be addressed in order to strengthen the national anti-corruption framework.
They also identify good practices. The State under review may also identify technical
assistance needs. In many cases, one of the needs for technical assistance identied is
the need to develop a national anti-corruption strategy or action plan in order to imple-
ment the recommendations and to strengthen anti-corruption reform (examples from the
Convention review process appear in box 3). During the sixth session of the Implemen-
tation Review Group, a large number of countries reported on the national implementa-
tion efforts that had been triggered by the country reviews. Most frequently, these actions
related to the development of national strategies in line with the Convention; legislative
reforms, including in respect of transnational bribery, money-laundering and illicit enrich-
ment; strengthening investigative, prosecutorial and judicial capacity; enhancing inter-
agency coordination; and establishing measures to protect witnesses, experts and reporting
persons.
Box 3. Linkages between the findings of country reviews and the development of
national anti-corruption strategies
• Tunisia is in the process of adopting a national plan of action that will be a
cornerstone of the national anti-corruption strategy that is also being devel-
oped, and incorporates the recommendations arising from the review of imple-
mentation of the Convention.
• The Philippines has used the momentum generated by the implementation
review process to establish a presidential inter-agency committee in order to
formulate and develop plans, policies and response strategies to enhance com-
pliance with the Convention.
• In Kenya, the State President has repeatedly affirmed his commitment to imple-
menting the recommendations emanating from the review. This creates the
political conditions to strengthen the existing national anti- corruption
strategy.
In addition, the Convention on Combating Bribery of Foreign Public Ofcials in Inter-
national Business Transactions of the Organization for Economic Cooperation and Devel-
opment (OECD), the Inter-American Convention Against Corruption, the African Union
Convention on Preventing and Combating Corruption, and the Criminal Law Convention
on Corruption and the Civil Law Convention on Corruption of the Council of Europe
also require parties to undergo a review process to assess compliance. These reviews,
which are also compulsory and include an element of peer review, often highlight areas
16
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
16
where the examined State is not in full compliance with its obligations under the relevant
convention and include recommendations for further actions to address such areas.
In addition to these formal peer review reports, UNDP, the World Bank, the regional
development banks and a number of non-governmental entities—domestic and interna-
tional civil society organizations, universities and research centres—conduct country
reviews that focus on or at least include assessments of corruption, integrity and gover-
nance. Prominent examples include Transparency International’s National Integrity Sys-
tem assessments, Global Integrity’s reports and the U4 Anti-Corruption Resource Centre’s
country evaluations. Other relevant information may be gathered through assessments
and evaluations on combating money-laundering and countering the nancing of terror-
ism, such as the mutual evaluations carried out by the Financial Action Task Force, as
well as the informal national risk assessments carried out by the World Bank.
All of these reports and reviews are particularly helpful because they not only provide
information about the extent and nature of corruption, but they often contain essential
information about gaps or weaknesses in a country’s anti-corruption laws, policies and
institutions. Many States have used the process of evaluation and assessment as a starting
point for reform. Of course, the fact that these reviews are qualitative rather than quan-
titative also introduces an element of subjectivity. Nonetheless, when such reviews exist,
they should be carefully considered by the strategy drafters when conducting the pre-
liminary diagnosis of the country’s main challenges in ghting corruption.
2. Cross-country comparisons of corruption or governance
To get a general sense of the extent of a country’s corruption problem, strategy drafters
often consult international indexes that purport to compare the level of corruption or the
quality of governance among different countries in a region or across the world. Strategy
drafters (and others) may be particularly interested in how their country ranks in relation
to other countries in the same region or with a comparable level of economic develop-
ment. These international indexes are typically based on the opinions or perceptions of
country experts, though some incorporate other data as well.
Two of the most commonly used indexes are Transparency International’s Corruption
Perceptions Index and the World Bank’s Worldwide Governance Indicators.
16
The U4
Anti-Corruption Resource Centre, the World Bank, the Vera Institute of Justice and
Transparency International’s Gateway Project have catalogued a wide range of additional
measures. The U4 Anti-Corruption Resource Centre has compiled a guide to 14 of the
most well-known comparative corruption indicators with a summary of what each
purports to measure.
17
The World Bank has identied 39 cross-national corruption or
integrity indicators
18
and the Vera Institute has compiled 53 different sources that compare
governance performance across nations.
19
The Gateway Project’s search engine examines
over 500 different data sources not only for cross-country indicators but for country-
specic studies as well.
20
16
Both the Corruption Perceptions Index and the Worldwide Governance Indicators are composite measures. That is,
they take data from other sources, standardize them and aggregate them to create a single number summarizing the
underlying sources. They are both, in a sense, a “poll of polls”. See Transparency International, “Corruption Perceptions
Index 2014: full source description” (Berlin, 2014); Laura Langbein and Stephen Knack, “The worldwide governance
indicators: six, one, or none?”, Journal of Development Studies, vol. 46, No. 2, pp. 350-370.
17
Soa Wickberg, “How-to guide for corruption assessment tools”, Expert Answer No. 365 (Bergen, Norway, U4
Anti-Corruption Resource Centre, 2013).
18
World Bank, AGI Data Portal, Links to Governance Indicators. Available at www.agidata.org.
19
Vera Institute of Justice, Rule of Law Indicator Instruments: A Literature Review. A Report to the Steering Committee
of the United Nations Rule of Law Indicators Project (New York, 2008).
20
Transparency International, Gateway: Corruption Assessment Toolbox. Available at http://gateway.transparency.org.
1717
Chapter II. Preliminary diagnosis and situation analysis
While a country’s ranking on these various indicators can provide a rst estimate of
how severe the corruption problem is, care is required when interpreting these measures.
The index scores are derived from different kinds of data, each with strengths and
weaknesses that the strategy drafters must bear in mind if they decide to use these index
scores when developing the initial corruption assessment. For example, the index scores
based on subjective evaluations by experts can be distorted because experts read one
another’s writings and talk among themselves, and their perceptions may reect a
“conventional wisdom” rather than their independent judgement.
21
Moreover, while infor-
mation on the perceived overall level of corruption in the country can be useful for
mobilizing the public behind the strategy—for example, by demonstrating that corruption
is viewed as a serious problem—this sort of aggregated national-level data are of little
value in formulating the content of the strategy. To develop an appropriately tailored
strategy to combat corruption, it is not enough to know how corrupt a country is per-
ceived to be relative to other countries. What is required is precise, detailed information
on the nature and source of corruption, and the leading cross-national indicators do not
provide such information.
3. Country-specific surveys of corruption perceptions
Strategy drafters may also be able to draw on (or commission) in-depth surveys of the
nation’s citizens or businesses. These country-specic surveys are usually more detailed
than cross-country surveys, with questions about the extent and nature of corruption, its
seriousness and whether it has lessened or worsened over time, as well as questions
about specic types of corruption or about corruption in particular sectors or government
agencies. In addition to the general public, country-specic surveys may target investors,
business executives, members of the news media or other experts in particular areas.
Their views can be especially useful in areas such as bid-rigging and the purchase of
defence equipment in which ordinary citizens are unlikely to have the specialized knowl-
edge required on which to base their perceptions.
22
The principal problem with perception surveys is that there can be a signicant gap
between perceptions of corruption and the actual extent of it.
23
The reasons for the
divergence are several and may include the reluctance of respondents to truthfully answer
questions about corruption,
24
the inuence of respondents’ characteristics (e.g., education,
income and ethnic afliation),
25
media coverage
26
and different views among respondents
21
See Stephen Knack, “Measuring corruption: a critique of indicators in Eastern Europe and Central Asia”, Journal of
Public Policy, vol. 27, No. 3 (2007), pp. 255- 291; Daniel Treisman, “What have we learned about the causes of
corruption from ten years of cross-national empirical research?”, Annual Review of Political Science, vol. 10 (2007),
pp. 211-244.
22
These expert perception surveys are distinct from and serve a different function to the expert consultations, which
are discussed in the preceding chapter and should be part of the strategy-drafting process. In some cases, the two may
be linked.
23
See Benjamin A. Olken, “Corruption perceptions vs. corruption reality”, Journal of Public Economics, vol. 93, Nos.7
and 8 (2009), pp. 950-964; Bonvin Blaise, “Corruption: between perception and victimization: policy implications for
national authorities and the development community”, in Challenging Assumptions on Corruption and Democratization:
Key Recommendations and Guiding Principles (Bern, Swiss Agency for Development and Cooperation, 2008); Knack,
“Measuring corruption”; Treisman, “What have we learned about the causes of corruption from ten years of cross-national
empirical research?”; Michael Johnston, “Measuring the new corruption rankings: implications for analysis and reform”,
in Arnold J. Heidenheimer and Michael Johnston, eds., Political Corruption: Concepts and Contexts, 3rd ed. (New
Brunswick, New Jersey, Transaction Publishers, 2002), part XIII, chap. 44, pp. 865-884.
24
See Jan Sonnenschein and Julie Ray, “Government corruption viewed as pervasive worldwide: majorities in 108 out
of 120 countries see widespread problem”, Gallup, 18 October 2013; Aart Kraay and Peter Murrell, “Misunderestimating
corruption”, World Bank Policy Research Working Paper, No. 6488 (Washington, D.C., World Bank, 2013).
25
See Diylan Donchev and Gergely Ujhelyi, “What do corruption indexes measure?”, Economics and Politics, vol. 26,
No. 2 (2014), pp. 309-331; Olken, “Corruption perceptions vs. corruption reality”.
26
Richard Rose and William Mishler, “Explaining the gap between the experience and perception of corruption”, Studies
in Public Policy, No. 432 (Glasgow, Centre for the Study of Public Policy, 2007).
18
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
18
about what constitutes corruption in the rst place.
27
Perceptions of corruption can be
important to understand in their own right and, at least in some cases, they may reect
an underlying reality, particularly when perception data are corroborated by other sources.
Nonetheless, perception-based data should always be treated with caution and should
never be the exclusive source for diagnosing a country’s corruption challenges.
4. Surveys of actual experience with corruption
Those responsible for developing the national anti-corruption strategy can also use or
commission surveys that ask about respondents’ actual experiences with corruption, in
addition to, or instead of, asking them about their perceptions of corruption. Such expe-
rience surveys may, for example, ask citizens or businesses
28
about how often they have
paid a bribe in the past year, and can also include questions on the agency to which the
bribe was paid, how much was paid and the purpose of the payment. Some of the most
detailed methodologies in this area were developed and tested by UNODC in its cor-
ruption research in Afghanistan, Iraq and the western Balkans. The UNODC household
and business sector surveys
29
focus on the experience of those surveyed as victims of
corruption, allowing for both a detailed identication and analysis of specic corruption
scenarios and for pinpointing the most vulnerable sectors, agencies and groups. This
kind of survey also provides data that may serve as a baseline for future monitoring and
evaluation purposes. Experience surveys are an attractive source of information about
corruption because they can offer a more objective, quantitative measure of at least some
forms of corruption.
Because of the nature of this kind of survey, it is mostly suitable for detecting the typical
cases of petty, administrative corruption. For obvious reasons, the majority of the
population may have no direct experience with “grand corruption” involving political
and business elites, trading in inuence or procurement fraud. To address the inherent
weaknesses of household surveys, experience-based surveys have also been carried out
with the business sector, focusing on business practices, dealings with governmental
authorities and participation in public procurement. In addition to the UNODC surveys,
examples of such business surveys include the international crime against businesses
survey by the United Nations Interregional Crime and Justice Research Institute, the
UNODC crime and corruption business surveys, the Global Competitiveness Survey by
the World Economic Forum, the World Business Environment Survey by the World Bank
and the World Competitiveness Yearbook of the International Institute for Management
Development.
There are other limitations that should be considered in relation to experience-based sur-
veys. There is evidence that in some countries, respondents may be unwilling to admit
that they have personally engaged in corrupt transactions, even when they are the victims
in those exchanges.
30
In addition, differences in respondents’ willingness to participate in
a survey or to answer questions honestly may introduce biases into the results; this issue
is too often neglected, but it can be a serious problem, especially when non-
response rates are very high and very likely not to be random. In countries where large
27
Sandra Sequeira, “Advances in measuring corruption in the eld” in New Advances in Experimental Research on
Corruption, Danila Serra and Leonard Wantchekon, eds. (Emerald Group Publishing, Bingley, United Kingdom, 2012),
pp. 145-176.
28
The World Bank regularly surveys rms in its client countries, asking detailed questions about whether they had to
pay a bribe to obtain power, water or phone service or a construction or import licence and other matters associated with
starting or operating a business. See World Bank, “World Bank’s enterprise survey: understanding the questionnaire”
(Washington, D.C., January 2011).
29
See conference room paper entitled “Quantitative approaches to assess and describe corruption and the role of
UNODC in supporting countries in performing such assessments” (CAC/COSP/2009/CRP.2).
30
Kraay and Murrell, “Misunderestimating corruption”.
1919
Chapter II. Preliminary diagnosis and situation analysis
segments of the population may not have a telephone or may be hard to reach in person,
biases of this kind have been detected.
31
Collecting original survey data can be expensive
and time-consuming and may require specialized skills, which may pose challenges, par-
ticularly in poorer countries where the required infrastructure is lacking. Nonetheless, when
used properly, survey data can be a valuable source of information on corruption.
The costs of these surveys on corruption can be reduced in several ways. A preliminary
assessment using other sources can be conducted, and the survey could be limited to
those areas where it would be most productive. In some cases, questions about corruption
can be added to a survey already planned by the Government, an international organi-
zation or a private entity. This will be far less expensive than commissioning an entirely
new survey.
A number of countries and organizations have now chosen to combine experience and
perceptions questions into a single survey to help offset the weaknesses of each. The
most extensive cross-national experience and perception survey to date is Transparency
International’s 2013 Global Corruption Barometer, during which 114,000 citizens in
107countries were asked not only about their perception of corruption in their country,
but also if anyone in their household had had to pay a bribe to obtain a public service
within the previous two years. UNODC has also conducted a number of mixed perception
and experience surveys and can provide model questionnaires. The Government of
Zambia made good use of a combined perception and experience survey in drafting its
2007 national anti-corruption strategy: 1,500 households, 1,000 public servants and man-
agers of 500 businesses were asked about both their direct experience with corruption
and their perceptions. The Government then used the responses to focus prevention efforts
on departments where the results showed corruption to be especially pronounced.
32
5. Internet platforms and social media
Although the traditional method for gathering information about citizens’ personal expe-
rience with corruption has been through surveys, a recent innovation for gathering such
information, spearheaded by civil society organizations, involves the use of websites or
social media platforms where citizens can anonymously report a demand for a bribe by
a government employee.
33
Many sites provide space for citizens to add details: the amount
of the bribe requested, for example, and the agency employing the person who demanded
the bribe. Site operators periodically report how many bribes have been requested, from
which departments and what the average amount requested was, in addition to other
relevant information.
34
The technique is still quite new, the reliability and validity of the
data reported remain open to question
35
and many of the concerns noted above about
more traditional surveys (such as response bias) may apply even more strongly to these
new platforms. Nonetheless, when combined with other information, crowdsourced data
on corruption can be useful and may emerge as an important source of quantitative data
that strategy drafters and implementers can use for measuring corruption or monitoring
31
See Ben Leo and others, “Do mobile phone surveys work in poor countries?”, Working Paper No. 398 (Washington,
D.C., Center for Global Development, 2015).
32
Laurent C. W. Kaela, “Zambia: national governance baseline survey—report” (Lusaka, University of Zambia, Depart-
ment of Political and Administrative Studies, 2004); Karen Hussman and Davies Chikalanga, “Anti-corruption policy
making in practice: Zambia—a country case study”, in Anti-Corruption Policy Making in Practice: What can be Learned
for Implementing Article 5 of UNCAC?, Hussmann, ed., p. 217.
33
See Dieter Zinnbauer, “Crowdsourced corruption reporting: what petried forests, street music, bath towels, and the
taxman can tell us about the prospects for its future”, Policy and Internet, vol. 7, No. 1 (March 2015), pp. 1-24.
34
One of the earliest and most celebrated of these efforts is described in Karthik Ramanna and Rachna Tahilyani, “I
paid a bribe (dot) com”, revised ed., Harvard Business School Case Collection, Case No. 112-078 (Boston, United States,
Harvard Business School, 2014).
35
Derek Ruths and Jürgen Pfeffer, “Social media for large studies of behavior”, Science, vol. 346, No. 6213 (2014),
pp. 1063 and 1064.
20
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
20
changes in corruption levels. Indeed, Governments may wish to consider closer collabo-
ration with organizations gathering information on corruption through social media, or
perhaps even launch their own social media platform as part of the process of gathering
data for the development of their national anti-corruption strategy.
6. Information collected by government agencies
Many government agencies collect information in the course of their operations that can
be used when making a preliminary diagnosis of the corruption problem (see box 4 for
an example). Data on corruption complaints, corruption court cases and allegations of
procurement fraud, for instance, provide direct evidence about the incidence and location
of corruption. Other data offer evidence of the effectiveness of anti-corruption controls
or point to areas vulnerable to corruption. The percentage of public servants failing to
le their personal nancial disclosures on time is one indicator of how effective integrity
controls are. The number of requests an agency makes to deviate from the standard pro-
cedures for awarding a government contract, the number and amount of cost overruns on
such contracts, the percentage by which an agency exceeds its budget and a failure to
obtain a clean audit letter can all be indicators of corruption. Public expenditure tracking
surveys and audit reports can reveal “leakages” (i.e. missing funds) that suggest the pos-
sibility of corruption.
36
Comparisons of the ratio of spending on particular items, such as
infrastructure, to the quantity provided may also provide evidence of corruption.
37
As with any data source, the accuracy of information generated by an agency’s operations
must be veried, as agencies may falsify data for many reasons, such as pressure to meet
performance targets, increase budgets or hide performance problems.
38
But where opera-
tional data are available and reliable, they are a useful and inexpensive source of infor-
mation. Furthermore, because they are gathered on a regular basis, they can be a good
source of continuing information on the strategy’s impact. If an agency is not regularly
publishing data on its operations, the lack of data may itself be an indicator of corruption:
the less data an agency produces, the easier it is to hide its activities from scrutiny. If
strategy drafters nd that an agency is publishing little or no information on its operations,
the strategy should contain a call for that agency to regularly produce sufcient data for
it to be properly monitored and have its leadership held accountable.
Box 4. Using agency data: a lesson from Uganda
The Inspectorate of Government of Uganda has made particularly good use of admin-
istrative data to track corruption trends and identify areas where additional reforms are
required. In perception surveys, Ugandans consistently rank the police, the civil service
and the judiciary as the three institutions most prone to corruption. But the two cate-
gories of public servants against which Ugandans file the most corruption complaints
are local government administrators and public officials. The difference may be because
citizens do not know how and where to file a complaint against the police. As the
Inspectorate noted in its fourth annual report, ensuring that citizens have the requisite
information about registering complaints against the police is an important issue to
36
See Ritva Reinikka and Jakob Svensson, “Local capture: evidence from a central government transfer program in
Uganda”, The Quarterly Journal of Economics, vol. 119, No. 2 (2004), pp. 679-705; Asli Gurkan, Kai Kaiser and Doris
Voorbraak, “Implementing public expenditure tracking surveys for results: lessons from a decade of global experience”,
PREM Notes, No. 145 (Washington, D.C., World Bank, 2009).
37
See Miriam A. Golden and Lucio Picci, “Proposal for a new measure of corruption, illustrated with Italian data”,
Economics and Politics, vol. 17, No. 1 (2005), pp. 37-75.
38
Zoe Radnor, “Hitting the target and missing the point? Developing an understanding of organizational gaming”, in
Performance Information in the Public Sector: How it is Used, Wouter Van Dooren and Steven Van de Walle, eds. (London,
Palgrave Macmillan, 2011), pp. 98-109.
2121
Chapter II. Preliminary diagnosis and situation analysis
address. The Inspectorate’s fourth annual report also illustrates how administrative data
can be used to measure areas that surveys cannot. The effectiveness of an agency’s
anti-corruption prevention programme, for example, cannot be observed directly; there
is no way to measure how much of a reduction in corruption in an agency is attributable
to a prevention programme. One indicator closely related to how well the agency pre-
vents corruption is the efficacy of its financial controls. The Inspectorate thus includes
data from the Auditor General’s office reporting the percentage, if any, by which each
agency exceeded its annual budget. The Inspectorate also gathers data on whether an
agency received a clean audit certificate and what percentage of public contracts ran
over time and budget.
Source: Uganda, Inspectorate of Government, Tracking Corruption Trends in Uganda: Using the Data
Tracking Mechanism, 4th annual report (Kampala, 2014).
Of course, the data generated by government agencies—even if accurate—must be inter-
preted carefully, as much of it will be open to more than one interpretation. Government
agencies may fail an audit not because of corruption but because their staff lack the
proper training to comply with audit procedures. Public contracts may run over budget
because of inadequate preparation, unforeseen events or simple mismanagement. Simi-
larly, law enforcement and corruption complaint data can be misleading when considered
in isolation: a large number of corruption prosecutions may mean more corruption, or
it may mean better detection or more aggressive prosecutors; a lack of citizen complaints
about corruption may mean corruption is low, or that citizens are unaware of corruption,
unaware they have the right to complain, or too intimidated do so. In some cases, the
correct interpretation will be obvious from the surrounding environment; in others, cor-
rectly interpreting the quantitative operational data will require consulting additional
information sources.
7. Comparisons of different data sources
An emerging technique for measuring corruption is to compare different sources of quan-
titative data that would be expected to give similar results in the absence of corruption
and to use the difference between them as evidence of corruption. For example, the
difference between the recorded value of exports from country A to country B and the
recorded value of imports to country B from country A can be evidence of corruption in
customs agencies.
39
Actual data can also be compared with the predictions of an economic
model, a method used to identify collusion in public procurement.
40
These techniques,
which are becoming more widespread, can provide useful measures of corruption,
41
and
in some cases the data required are readily available. Their principal disadvantage is that
they measure only narrow forms of corruption. In addition, these techniques require accu-
rate data from the different sources and the technical skills to conduct the comparisons—
prerequisites that can be a challenge for resource-constrained States.
39
See Raymond Fisman and Shang-Jin Wei, “Tax rates and tax evasion: evidence from ‘missing imports’ in China”,
Journal of Political Economy, vol. 112, No. 2 (2004), pp. 471-496; Raymond Fisman and Shang-Jin Wei, “The smuggling
of art, and the art of smuggling: uncovering the illicit trade in cultural property and antiques”, American Economic Journal:
Applied Economics, vol. 1, No. 3 (2009), pp. 82-96; Sandra Sequeira and Simeon Djankov, “Corruption and rm behavior:
evidence from African ports”, Journal of International Economics, vol. 94, No. 2 (2014), pp. 277-294; Sandra Sequeira
and Simeon Djankov, “Trade costs and corruption in ports”, Port Technology International, No. 53 (2013).
40
Josef Ålander, “Screening for cartels in procurement procedures: lessons learned”, Competition Policy International
(2014).
41
For a survey of these “forensic economics” techniques, see Eric Zitzewitz, “Forensic economics”, Journal of Economic
Literature, vol. 50, No. 3 (2012), pp. 731-769.
22
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
22
8. Vulnerability assessments
Strategy drafters may consider conducting or commissioning vulnerability assessments,
also called risk analyses, to identify areas where a government ministry, agency or
department may be at particular risk for corruption. Such an analysis typically begins
by categorizing the organization’s functions into different activities and evaluating three
issues within each activity: the general control environment, the inherent risk of corrup-
tion and the adequacy of existing safeguards.
42
Although a simple corruption vulnerability
analysis can sometimes be completed in a few days, a more comprehensive assessment
may take several months, particularly if agency staff members are uncooperative. And,
as is true with respect to peer reviews and reports of non-governmental organizations,
these assessments rely substantially on subjective, and potentially contestable, judgement
calls by the assessors. It may be possible to assuage potential controversy over the results
of such analyses by providing all those with an interest in the analysis with an oppor-
tunity to participate in the process and review the results, but of course there is the risk
that the ndings will be watered down to achieve consensus or avoid controversy.
9. Proxy measurements
While most of the tools and methods described above aspire to measure corruption
occurrence directly, it is also possible to assess the scope of the integrity-related problems
in a country by measuring “proxies”, or studying facts that may, through interpretation,
give an idea of corruption-related problems. Examples of such proxy indicators may
include the size of the grey economy in a country, time needed to complete a certain
administrative procedure or, instead, measuring the availability in the country of a strong
anti-corruption institutional infrastructure, assuming that if this infrastructure is in place,
corruption is less likely to occur.
43
All information sources have their advantages and disadvantages, and they need to be
combined in appropriate ways. Strategy drafters should begin by reviewing what infor-
mation is readily available and then consider what additional information might be most
useful and can be obtained at a reasonable cost within a reasonable time. If the country’s
Convention against Corruption implementation review has revealed gaps in its anti-
corruption laws, or if Transparency International’s Global Corruption Barometer or a
World Bank enterprise survey has shown high levels of corruption in particular govern-
ment departments, the strategy drafters can follow up with targeted studies to identify
the causes and suggest potential solutions. Where information is simply not available to
the drafters, they should acknowledge that fact and tailor the policy proposals accord-
ingly. They might also recommend additional information-gathering as part of the strat-
egy itself, an approach adopted by the drafters of the national strategies for Cabo Verde,
Greece, India and the United Republic of Tanzania.
The principal dilemma here is that developing the knowledge required for good policy-
making can be expensive, time-consuming and intellectually challenging. Yet policy-
makers are often under pressure to act; in many countries the corruption problem is an
urgent, pressing issue; the public is demanding a solution and policymakers want to
address the nation’s ills quickly. Spending too much time analysing the problem before
42
Robert Klitgaard, Controlling Corruption (Berkeley, California, University of California Press, 1988), pp. 84 and 85.
See also Council of Europe, “Project Against Corruption in Albania (PACA): technical paper—corruption risk assessment
methodology guide”, document CMU-PACA-02/2011; Tony Scott, Alistair Carstairs and Doug Roots, “Corruption pre-
vention: the Hong Kong approach”, Asian Journal of Public Administration, vol. 10, No. 1 (1988), pp. 110-119; Mauritius,
Independent Commission Against Corruption, Public Sector Anti-Corruption Framework Manual (Port Louis, 2009).
43
See UNDP and Global Integrity, A Users’ Guide to Measuring Corruption (Oslo, UNDP Oslo Governance Centre,
2008).
2323
Chapter II. Preliminary diagnosis and situation analysis
producing a strategy document risks squandering a window of political opportunity.
Indeed, sometimes those who benet from the status quo will use the excuse of needing
more research to dissipate political pressure to take action. But at the same time, the
effectiveness of a national anti-corruption strategy depends on the depth and breadth of
the knowledge base on which it rests. Without a thorough assessment of the prevalence
of corruption and the ways in which it manifests itself, a strategy is likely to have little,
if any, impact on corruption. If time or resource constraints prevent the construction of
a robust knowledge base, then the ambitions of the strategy must be scaled back.
Matching the means available with the desired goals and objectives is the essence of
strategy—and a constant challenge.
B. Assess obstacles to effective reform
The second step when conducting a preliminary diagnosis is to identify the most import-
ant obstacles to adopting and implementing the necessary reforms. For a national
anti-corruption strategy, the two obstacles listed below are worthy of special attention.
1. Evaluate resource constraints
There is substantial variation between countries when it comes to the resources that they
are able to devote to anti-corruption efforts, and a national anti-corruption strategy should
explicitly take that into account and contain a realistic assessment of the resources (of
different kinds) available for such efforts. Although some anti-corruption measures, such
as changes in the formal law, may not require signicant investments of time or material
resources, their implementation may (see box 5 for an example). Other initiatives, such
as dramatically stepped-up enforcement of anti-corruption laws, may be quite resource-
intensive. Further, some anti-corruption measures require not only resources but special-
ized skills in forensic accounting, languages, information technology and other technical
areas. Even seemingly straightforward measures such as amending a law or regulation
may prove difcult in practice, as legislative or regulatory change can consume time,
attention and space on a limited agenda.
Box 5. Implementing a right-to-information regime
Implementing a full-fledged access-to-information law takes time and resources.
Thanks to the many model statutes available, drafting and enacting a new law can be
done quickly. Ensuring that agencies can respond to citizen requests, however, often
requires reforms to their record-keeping standards and procedures. A World Bank study
suggests that an access-to-information law should be phased in gradually, beginning
with a duty to publish key information about different agencies and their rules, and then
slowly expanding the duty to include responding to individual requests. Such an approach
not only gives agencies the time to modernize record-keeping standards and proce-
dures, but also spreads the costs out over several years.
Source: Toby Mendel, “Legislation on freedom of information: trends and standards”, PREM Notes,
No. 93 (Washington, D.C., World Bank, 2004).
Resource needs and constraints can be integrated into the discussion of the specic
elements of the anti-corruption policy as well as into the implementation and monitoring
plan. As discussed in chapter III, section E, below, the strategy should ideally contain
realistic estimates of the resources required for each element of those plans. The pre-
liminary diagnosis should also provide a general discussion of available resources for
24
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
24
anti-corruption efforts and how to link nancing to the country’s budget cycle. Explicit
discussion of resource constraints can be a helpful disciplining device, reminding strategy
drafters of the need to be realistic and to make sometimes difcult trade-offs. A focus
on resource limitations in the preliminary diagnosis may also reveal where more resources
are required to combat corruption. If that is the case, exactly what is needed (more
funding or specialized skills and institutional capacity), and why it is needed, should
form part of the strategy.
2. Address potential opposition and support
Paying attention to resource constraints is important, but the principal obstacle to effec-
tive implementation of anti-corruption reforms is usually not technical or resource-based,
but political. Individuals and organizations proting from corruption will resist changes
that leave them worse off, and they will therefore try to thwart the implementation of
an anti-corruption strategy. Although, as explained in box 6, the essence of a strategy
is anticipating opposition to its execution, few national anti-corruption strategies acknowl-
edge the potential for opposition, let alone contain possible ways of countering it. Of
the 53 strategies reviewed for this Guide, only a handful mentioned the possibility of
opposition, and of those that did, most said it would be overcome through political will.
None, however, dened what was meant by “political will”, how to recognize if it was
present, or what to do if it was not.
44
Dismissing the risks of opposition to anti-corruption
reforms by invoking the idea of “political will”, rather than analysing how it can actually
be overcome, is a common failing in much writing on anti-corruption policy,
45
and it is
thus not surprising that most national anti-corruption strategies do not take potential
opposition into account.
Box 6. Strategy defined
Strategy remains the best word we have for expressing attempts to think about actions
in advance, in light of our goals and our capacities.
A strategy is much more than a plan. A plan supposes a sequence of events that allows
one to move with confidence from one state of affairs to another. Strategy is required
when others might frustrate one’s plans because they have different and possibly oppos-
ing interests and concerns.
Source: Lawrence Freedman, Strategy: A History (New York, Oxford University Press, 2013), pp. x-xi.
There are, however, ways that opposition to combating corruption can be overcome.
Indeed, the development and execution of a national strategy can itself be one. The
creation of a solid, widely shared understanding of what the nation’s corruption problems
are, ensuring widespread consultations on how to address them, and, as will be discussed
in chapter V below, issuing regular updates on the progress made in ameliorating them
are all ways to build the political will required to overcome opposition to the strategy.
44
As one commentator explains, political will is “the slipperiest concept in the policy lexicon. It is the sine qua non
of policy success which is never dened except by its absence.” (Linn Hammergren, Political Will, Constituency Building,
and Public Support in Rule of Law Programs (Washington, D.C., United States Agency for International Development,
1998), p. 12.)
45
See Matthew Andrews, “Creating space for effective political engagement in development”, in Governance Reform
Under Real World Conditions: Citizens, Stakeholders, and Voice, Sina Odugbemi and Thomas Jacobson, eds. (Washing-
ton,D.C., World Bank, 2008).
2525
Chapter II. Preliminary diagnosis and situation analysis
A critical step to checking opposition is identifying what groups are most likely to
support the strategy and what groups are most likely to resist it.
46
Once supporters and
opponents are identied, the strategy can be designed in a way that builds support for
it and overcomes opposition. Phasing in controversial changes slowly while implementing
popular ones quickly is one technique. Understanding the reasons for opposition is
another important step. Civil servants might oppose the introduction of a stringent new
ethics code for fear they would unknowingly run afoul of its provisions. If this is the
case, strategy drafters can reduce opposition by coupling the introduction of the new
code with the creation of a system allowing them to seek advice or approval in advance
for conduct that might be subject to question.
47
Finally, as explained in box 7, facilitated
deliberations among representatives of different groups can help achieve consensus on
a reform programme.
Box 7. Overcoming opposition through deliberation
A deliberative process is one way to identify not only who will resist reform and why,
but also to overcome that resistance. The World Bank has employed this technique to
create support for anti-corruption action plans in a number of countries. Meetings of
citizens, businesses, elected leaders and civil society groups were held, during which
evidence of the harm corruption was doing to the nation was presented, along with
alternatives for addressing it. A facilitator moderated a discussion that was aimed at
developing a consensus around an effective reform agenda.
Sources: Jane Mansbridge, “Deliberative and non-deliberative negotiations”, HKS Faculty Research
Working Paper Series, No. RWP09-010 (Boston, United States, Harvard Kennedy School, 2009); Carolien
Klein Haarhuis, “Promoting anti-corruption reforms: evaluating the implementation of a World Bank
anti-corruption program in seven African countries (1999-2001)”, ICS dissertation series, No. 114
(Utrecht, Utrecht University, 2005).
Public attitudes towards corruption are also important to understand, and might be some-
thing that can be surveyed (in addition to public perceptions of or experiences with
corruption). It would be useful to know, for example, whether citizens are outraged by
corruption or grudgingly tolerant of it, or whether there are certain practices that they do
not consider corrupt. Another area that could be explored is whether there are differences
in how different groups in society—the business community, women, the poor, public
servants and elected ofcials, for example—feel about these issues. It is useful to consider
too how the politics of anti-corruption might evolve over time, in recognition of the fact
that the politics around this issue are dynamic rather than static, and that certain policies
may bring about dedication to the successful implementation of the strategy or staunch
opposition to it.
46
See Michael Johnston and Sahr Kpundeh, “Building a clean machine: anti-corruption coalitions and sustainable
reform” (Washington, D.C., World Bank, 2002).
47
Richard E. Messick, “Policy considerations when drafting conict of interest legislation”, in Corruption and Conict
of Interest: A Comparative Law Approach, Jean-Bernard Auby, Emmanuel Breen and Thomas Perroud, eds. (Cheltenham,
United Kingdom, Edward Elgar Publishing, 2014), part II, chap. 8, pp. 113-123.
27
Formulating anti-corruption measures
After conducting a preliminary diagnosis of the problems and challenges, the next step
in drafting a national anti-corruption strategy is to formulate the actual policy goals and
the practical ways of achieving them. This could take the form of a workplan or a
broader statement setting out the way ahead. The steps foreseen may contain a combi-
nation of new legislation, organizational changes and other reforms that the country
should pursue to combat corruption. The list of measures will vary from country to
country, because nations differ both in their corruption problems and in their capacity
to address them. No matter how the actual substantive part of the strategy is drafted,
however, the following guidelines will help in formulating it.
A. Tailor the strategy to the diagnosis,
taking account of constraints
The policy should follow from, and be tailored to, the diagnosis of a country’s major
corruption problems and challenges, taking into account the relevant constraints. This
may seem obvious, but few strategy documents reviewed for this Guide even attempted
to link the diagnosis of problems and challenges to the specic elements of the proposed
anti-corruption strategy. The strategy document should explicitly connect each proposed
measure to the preceding diagnosis of the problems and challenges. On a closely related
point, the strategy drafters should consider what measures would require new legislation
or signicant new budgetary allocations, and what measures can be implemented under
existing legal authority or through redeploying existing resources.
B. Be ambitious but realistic
The point of being ambitious but realistic, which is stressed repeatedly throughout the
Guide, is especially important when identifying the substantive goals and proposals which
may be formulated in a separate action plan. The balance between ambition and realism
is a difcult one to strike. The goal of a strategy is not merely to announce a blueprint
for policy reform, but to create pressure for change, and the strategy document should
therefore set high standards and ambitious goals that go beyond minor changes to the
III.
28
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
status quo. At the same time, if the strategy document contains proposals and objectives
that are unrealistic in view of the constraints that a country faces, the strategy may be
brushed aside as an impractical “wish list”. When such a national anti-corruption strategy
later fails to deliver, it can fuel disillusionment and cynicism among citizens.
Strategy drafters should keep in mind that it is unlikely that the measures included in
the national anti-corruption strategy will be implemented exactly as set forth. The national
anti-corruption strategy is part of a process—one that is about building and sustaining
political momentum to address corruption, as well as charting an initial course of action.
That means that it will be continually revised and updated as circumstances change and
new information comes to light. The fact that a national anti-corruption strategy is a
political document as much as it is a technocratic document must be borne in mind
when striking the appropriate balance between ambition and realism in the action plan.
C. Identify concrete, specific measures to be taken
The strategy should outline specic, concrete steps that will help control corruption within
the country. As noted above, it is unlikely that every item will be implemented exactly
as laid out in the strategy. Moreover, actual implementation of specic policies may
require a level of detail that would be out of place in a national strategy document, so
it may be necessary to adopt a further action plan to complement the strategy. But it is
still important that the strategy contains a set of concrete recommendations, rather than
simply stating objectives in broad, abstract terms without connecting them to any specic
reforms or policy changes. Because converting even concrete recommendations into spe-
cic action might require a level of detail that would be inappropriate for a national
anti-corruption strategy document, the strategy might set out the recommended policy
reforms and then call upon the implementing agencies (which, again, should ideally have
had a role in the drafting process) to develop more specic implementation plans.
As for the areas the strategy should cover, or the types of reforms it should be aimed
at advocating, this Guide does not offer specic substantive recommendations. The need
to tailor the proposals to each country’s specic circumstances, and to make difcult
trade-offs, means that a general Guide like this one cannot specify what particular mea-
sures a given country’s strategy should contain. At the same time, the Convention against
Corruption supplies a useful template, or point of departure, for national anti-corruption
strategy drafters to consider when developing the content of the action plan. The drafters
can consider each of the main categories of anti-corruption activities that States parties
are obliged to undertake under the Convention (see box 8), and further consider what
sorts of actions, if any, might be appropriate in the particular circumstances of the country
in question to full those obligations.
Box 8. United Nations Convention against Corruption: a framework for a national
anti-corruption strategy
The United Nations Convention against Corruption calls for a range of actions against
corruption that drafters may find useful in formulating their own action plan. The Con-
vention emphasizes the need for action in the following areas:
(a) Strengthening anti-corruption law and law enforcement, including in such areas
as:
(i) Laws against embezzlement or other misappropriation of property, in both
the public and private sectors (articles 17 and 22);
29
Chapter III. Formulating anti-corruption measures
(ii) Anti-bribery laws, both domestic and foreign (articles 15 and 16);
(iii) Laws against improper trading in influence or other abuse of the functions
of public office (articles 18 and 19);
(iv) Laws targeting the proceeds of corruption, including laws to counter
money- laundering and measures to facilitate the identification, freezing,
confiscation and return of illegally obtained assets (articles 14, 23, 24, 31
and 52 to 57);
(v) Adequate protection for whistle-blowers and witnesses (articles 13, 32 and
33);
(vi) Legal tools or reforms that may facilitate the effective prosecution of cor-
rupt activities (to the extent that they are consistent with domestic legal
traditions and basic principles of fairness) including liability for illicit
enrichment, obstruction of justice, participation and attempt; recognition
that knowledge or intent may be inferred from factual circumstances; lia-
bility of legal (as opposed to natural) persons; and appropriately extended
statutes of limitations (articles 20, 25 and 26 to 29);
(vii) Meting out appropriate sanctions for corrupt behaviour (including both civil
and criminal remedies) and doing so fairly (articles 30 and 34 to 35);
(b) Implementing more effective measures to prevent corruption measures while
reducing the incentives and/or the opportunities for corruption. Such reforms might
include:
(i) Measures to prevent corruption in the public sector, including through
promotion of integrity in public administration, strengthening public sector
ethics, addressing corruption risks in procurement and in the budgeting
process and promoting transparency and civic participation (articles 7 to
10 and 13);
(ii) Measures to prevent corruption in the private sector, including transpar-
ency and reporting measures, accounting controls and auditing (article 12);
(c) Establishing education programmes, publicity campaigns and awareness-
raising initiatives intended to change attitudes towards corruption, shift cultural norms
and encourage reporting (article 13);
(d) Bolstering institutional capacity, providing adequate training and ensuring the
effectiveness, autonomy and integrity of the government bodies that implement anti-
corruption policy (articles 6, 11, 36, 58 and 60);
(e) Improving both domestic and international coordination between law enforce-
ment authorities and other agencies responsible for combating corruption, as well as
between the public and private sectors (articles 37 to 39 and 43 to 50);
(f) Enhancing the collection and analysis of data and other information on cor-
ruption, in order to address uncertainties and knowledge gaps (article 61).
D. Describe the objective of each reform element
In addition to laying out reform proposals, the strategy should contain an explanation
of what each reform is intended to accomplish, and why and how it will be accomplished.
Again, these objectives should be grounded in the diagnosis of a country’s current sit-
uation with respect to corruption challenges, and should be ambitious but realistic in
terms of both number and reach. Articulating the objectives connected to each proposal
(or a combination of proposals) is helpful not only when designing the strategy, but also
in fostering accountability for implementation and generating feedback for ongoing revi-
sion, two topics discussed later in this Guide. It is often best to frame these objectives
in terms of specic intermediate outputs (for example, increases in the number of cases
prosecuted or citizen satisfaction), rather than deterrence of corruption or government
legitimacy or similar general goals.
30
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
E. Consider the costs, benefits, burdens,
opposition and support for each element
The considerations governing resource costs and political constraints discussed in chapter
II, section B, on diagnosing the problem, should be applied to each of the specic
reforms and initiatives outlined in the policy. Thus, where possible, the strategy (or the
more specic implementation plans produced by the implementing agencies) should
provide at least a rough estimate of the resource requirements for each proposal, as well
as a more general assessment of feasibility and cost-effectiveness.
48
National anti-
corruption strategy drafters and implementing agencies should also consider and discuss
how the burdens and benets of implementing the proposal would be distributed (with
respect to both the institutions responsible for implementation and the parties affected),
and assess the likely political support and opposition for the proposals and what steps
might be taken to increase the former and decrease the latter. In most cases, those
political considerations would not appear in the formal published strategy, or would be
referenced only obliquely and euphemistically. But they should still be considered and
discussed explicitly when formulating the strategy, and elements of political consider-
ations might be helpfully incorporated in the document itself, for example through
discussions of the need to create “political space” for reform or the need to build
supportive coalitions.
F. Pay attention to prioritization and sequencing
Corruption is a complicated, multifaceted problem requiring a comprehensive, multi-
pronged solution. At the same time, it is simply impossible to do everything, and not
every possible reform is likely to be equally effective or equally feasible. One of the
most important things drafters of a strategy document can do is to identify the areas of
highest priority and the sequence in which a country should attempt to implement various
reforms or tackle various aspects of the corruption problem. As a recent review of
national anti-corruption strategies in South-Eastern Europe concluded, the failure to
prioritize rendered these strategy documents simply an exhaustive list of all possible
aspects of an anti-corruption plan, rather than a realistic road map for bringing corruption
under control.
49
In setting priorities, it is important to consider three factors with respect to each reform
that might be included in the national anti-corruption strategy: (a) the signicance of
the problem targeted (the amount of harm it does to the society and the citizenry);
(b) the likely effectiveness of the proposed policy reforms in reducing the problem; and
(c) the expected cost of the proposed reform (including not only budgetary or economic
cost, but also the amount of “political capital” or technical capacity or attention that
must be expended to implement the reform effectively). The evaluation of the rst factor,
the magnitude of the targeted problem, should follow from the strategy’s preliminary
diagnosis. The consideration of the second and third factors is equally important, how-
ever, because it is not always the case that the most resources should be devoted to the
biggest problems. The relevant question is not the magnitude of the harm to be targeted,
but how much any given reform can reduce the harm to society.
48
For how to conduct cost-benet analyses of governance and anti-corruption policies, see Jesper Johnsøn, “Cost-
effectiveness and cost-benet analysis of governance and anti-corruption activities”, U4 Issue No. 10 (Bergen, Norway,
U4 Anti-Corruption Resource Centre, 2014).
49
Southeast Europe Leadership for Development and Integrity, Anti-Corruption Reloaded: Assessment of Southeast
Europe, p. 40.
31
Chapter III. Formulating anti-corruption measures
Moreover, considering the impact of individual reforms in isolation is not sufcient. The
discussion of prioritization and sequencing should include two other factors as well.
First, close attention should be paid to how different elements of the overall reform
strategy t together, and what legal, institutional or social prerequisites must be in place
before particular elements of the strategy are likely to be effective. An aggressive law
enforcement strategy may require legal reforms that give investigators and prosecutors
new tools to combat corruption along with extensive training on how to use them. When
setting priorities and outlining a sequence of reforms, these sorts of links and interde-
pendencies should be taken into account in a national anti-corruption strategy. Second,
how different interventions might affect political support for anti-corruption efforts, and
the rest of the strategy, should be considered. As discussed above, it may sometimes
make sense to start with initiatives that can yield tangible and politically popular results
in the relatively short term as a way to build condence and political support (see box9
for an example). In some situations, it may also sometimes be prudent to delay initiatives
that would provoke erce political opposition in the short term, in order to build support
and momentum for reform over time. On the other hand, in some circumstances it may
be important to tackle certain challenging problems rst, while there is a political window
of opportunity (created, perhaps, by recent corruption scandals) that might close if too
much time passes. Again, there is no right answer; the important thing is that the drafters
of the national anti-corruption strategy pay explicit attention to these issues.
Box 9. Benefits of an early attack on corruption in education
Reports by the United Nations Educational, Scientific and Cultural Organization
(UNESCO), Transparency International and UNDP have revealed that corruption infects
every level of education in insidious ways, from the sale of grades to the purchase of
degrees to the demand for sexual favours in return for a pass mark. An early attack
on corruption in the education sector can therefore be advantageous both substantively
and politically. Substantively, having a well-educated citizenry is critical for economic
growth, political stability and social well-being; corruption thus undermines national
development. Politically, for many citizens, providing their children with a high-quality
education is a high priority, and programmes to stamp out corruption in education are
likely to draw enthusiastic support from a wide swath of society. Nor is an attack on
corruption in education likely to generate the kind of opposition that it might in other
sectors, because education corruption often involves small groups in different schools
and different regions who do not have connections to, or the support of, powerful
interests.
Sources: UNESCO, Education for All: 2000-2015—Achievements and Challenges (Paris, 2015); Transpar-
ency International, Global Corruption Report: Education (Abingdon, United Kingdom, Routledge, 2013);
UNDP, Fighting Corruption in the Education Sector: Methods, Tools and Good Practices (New York, 2011).
G. Specify implementation times
for different reforms
Different reforms take different amounts of time to implement and different lengths of
time to have an effect. As explained in box 5 above, a right-to-information law can be
enacted quickly, but implementation often takes several years, and its effect on corruption
may take several more. The strategy should contain an explicit acknowledgement of the
different time periods required for different measures to take effect and an explanation
of why, and should emphasize the need for patience with respect to those that will take
years to achieve. A strategy promising citizens that a fully-edged system allowing access
to information will be quickly established can create disillusionment or cynicism when
32
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
it does not immediately materialize. Identifying realistic timeframes for implementing
reforms is one area where the experience of other nations can be particularly valuable.
To help manage citizens’ expectations, it is important to provide periodic reports on
progress, and setbacks, in achieving the objectives, a point stressed in chapter V below.
Given the variation in the time needed to implement various anti-corruption measures,
proposed policy reforms and objectives should be categorized by timeframe within the
strategy document, with reform objectives sorted into those that are short term (for
example, those that can be implemented within the next 6 to 18 months), medium term
(within two to ve years) and long term. Short-term goals should, where possible, be
connected to the current budget cycle. For those countries with a national planning
process, which often spans three to ve years, it may make sense to link the national
anti-corruption strategy to more general national development plans, particularly for the
medium-term goals. Longer-term goals must be understood as more aspirational. Cir-
cumstances and Governments change, the situation evolves and it is difcult to envision,
let alone plan and budget for, anti-corruption efforts that need to be implemented over
a decade (or longer). Nonetheless, setting longer-term goals can still perform a useful
function in the planning process and can help keep up the pressure on Governments to
follow through on more ambitious objectives. It can also be helpful for the strategy to
contain a discussion of the links between the short- or medium-term goals, which can
be described in more concrete terms and implemented within the next budget cycle or
national planning period, and the longer-term objectives of the strategy.
33
Ensuring effective implementation
The successful implementation of a national anti-corruption strategy requires concerted
action by a range of ministries, departments and agencies and will often require coor-
dination with entities that operate independently of the Government: prosecution services,
supreme audit agencies and regulatory authorities. Implementation may also require
cooperation from the judicial and legislative branches of Government and from regional
and local governments as well. Coordinating the implementation of a national anti-
corruption strategy is a major challenge. One particular challenge is that some individuals
or agencies charged with implementation may benet from the corrupt status quo and
may therefore look for ways to undermine implementation. Even putting that problem
aside, the changes put forward in a strategy document often put additional burdens on
government agencies, making it more difcult for them to carry out their missions and
thus creating resistance among agency staff, clients and others.
50
The need for coordi-
nation and cooperation across agencies may pose additional challenges for organizations
that are not accustomed to working with one another and that may have different cultures,
objectives and interests.
51
Simply formulating a strategy that calls on government agencies to adopt new rules and
policies to ght corruption is therefore not enough. A successful national anti-corruption
strategy must not only lay out a comprehensive set of substantive policy reforms, but
also provide the means for ensuring coordinated implementation—at a minimum for the
activities of executive branch agencies, and in many instances for coordinating the work
of independent agencies, separate branches of Government and subnational government
units. A number of strategies have been adopted in the past on the assumption that high-
level support from the Head of Government would sufce to ensure effective and coor-
dinated implementation of the strategy. However, although high-level support is necessary,
it is rarely sufcient over the longer term. Heads of Government and cabinet ministers
50
For example, the primary mission of a public works ministry is planning, procuring and maintaining roads, bridges
and other infrastructure. Instituting new procurement processes, introducing more audits or requiring other changes in
the name of controlling corruption will slow the construction of new works and the maintenance of existing ones and
will also drive up costs (at least in the short term). Even if the net benet of such changes will be positive, agency
personnel, construction companies, engineering and consulting rms, trucking companies and power consumers, to name
but a few with an interest in infrastructure, are unlikely to see it that way, and they are likely to undermine, if not resist,
measures that slow down infrastructure development or appear to increase its cost.
51
Geert Bouckaert, B. Guy Peters and Koen Verhoest, The Coordination of Public Sector Organizations: Shifting
Patterns of Public Management (Basingstoke, United Kingdom, Palgrave Macmillan, 2010), pp. 25-32.
IV.
34
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
generally cannot give the implementation of the strategy—and the inter-agency coordi-
nation such implementation often requires—the frequent and continued personal attention
that is needed.
Unfortunately, the strategies reviewed for this Guide contained little in the way of dis-
cussion or analysis of the coordination issue, a nding that is consistent with an analysis
by the U4 Anti-Corruption Resource Centre of an earlier generation of strategies.
52
How
best to coordinate the implementation of an anti-corruption strategy will depend upon a
variety of country-specic factors, but experience with implementing earlier strategies,
as well as studies of public sector management, offer six lessons strategy drafters should
observe. These are listed below.
A. Put a single, high-level entity in charge of
coordination and implementation
A particular body or agency should be put in charge of coordinating implementation
and monitoring compliance with the national anti-corruption strategy’s policies and
objectives. If a country has a specialized anti-corruption agency, that agency may seem
like a natural candidate for this role. However, as a 2014 review by UNDP of the expe-
rience with anti-corruption strategies in the Asia-Pacic region found,
53
anti-corruption
agencies may not be well suited to this task. They may have neither sufcient resources
nor adequate power to compel other departments and agencies to follow their direction.
This conclusion was also reached by the U4 Anti-Corruption Resource Centre in its case
studies on the implementation of an early generation of strategies.
54
This is not surprising;
most anti-corruption agencies were created relatively recently,
55
and may be outside
governmental hierarchies or not have cabinet-level status, thus putting them at a disad-
vantage when dealing with government ministries or other powerful, well-established
agencies with direct access to the Head of Government.
One alternative to putting the anti-corruption agency in charge of coordination is entrust-
ing this responsibility to a centrally staffed unit or designated ministry that reports to
the president, prime minister or other senior gure in Government. This is a common
method for coordinating any policy that involves several agencies, and it is usually
accompanied by giving the coordinating unit the authority to pass orders issued from
the senior ofcial in charge to the different agencies and to monitor their compliance.
56
No organizational set-up will be right for every Government, but in the Kuala Lumpur
statement it is recommended that, whatever entity is the designated coordinator, it should
be housed within a high-level government agency.
A second alternative is to create a commission chaired by a senior ofcial and comprising
representatives from key ministries to coordinate implementation. For example, the 2012
strategy for Croatia established the Monitoring Committee for Implementation of
Corruption-Suppression Measures, headed by the Minister of Justice. And in Bulgaria,
as explained in box 10, a separate technical secretariat was created to support a
52
See Hussmann, ed., Anti-Corruption Policy Making in Practice: What can be Learned for Implementing Article 5 of
UNCAC?.
53
Anti-Corruption Strategies: Understanding What Works, What Doesn’t and Why?.
54
See Jessica Schultz, “Anti-corruption policy making in practice: Pakistan—a country case study”, and Hussmann and
Chikalanga, “Anti-corruption policy making in practice: Zambia—a country case study”, in Anti-Corruption Policy Making
in Practice: What can be Learned for Implementing Article 5 of UNCAC?, Hussmann, ed.
55
Samuel De Jaegere, “Principles for anti-corruption agencies: a game changer”, Jindal Journal of Public Policy, vol.1,
No. 1 (2012), pp. 79-120.
56
Bouckaert, Peters and Verhoest, The Coordination of Public Sector Organizations, pp. 38 and 39.
35
Chapter IV. Ensuring effective implementation
high-level coordination unit. Furthermore, even when the national anti-corruption agency
is not well-suited to lead strategy implementation, it may be very well placed to serve
as a consultant or adviser to the main implementation unit. Again, there is no single
correct approach, but States may nd a two-level structure along these lines helps strike
the right balance between the political and technical aspects of coordinating effective
implementation of the national anti-corruption strategy.
B. Provide the coordination and implementation
body with sufficient authority
More important than the coordination unit’s structure are its mandate and authority. In
some countries, the unit responsible for coordinating the implementation of the national
anti-corruption strategy has lacked the authority to compel other agencies to align their
policies and actions with the national strategy, thereby undercutting implementation. In
other countries, in contrast, the coordinating entity enjoys considerably more authority,
with the result that other agencies are more likely to work together to implement the
strategy (see box 10). The coordinating entity should therefore have a clear mandate,
57
in the form of a law, regulation or decree setting out its responsibilities and powers, and
the Head of Government needs to make it clear he or she fully supports that mandate
and expects all other government departments and agencies to cooperate with it. Coor-
dinating units should not overreach, however, by trying to redene an implementing
agency’s mission or to micromanage it. Doing so would be seen as an invasion of the
implementing agency’s responsibilities, making the unit a competitor rather than a neutral
judge of performance. Coordination units have neither the skills nor the staff to micro-
manage other government departments, and attempts to do so, or to compel departments
to assume new responsibilities, will only create friction.
58
Box 10. Coordination units: lessons from Bulgaria and the United Republic of
Tanzania
In the United Republic of Tanzania, the Good Governance Coordination Unit, a small
office under the President, coordinated the first national anti-corruption strategy. In
Bulgaria, a small office served as the secretariat to the commission responsible for
coordinating actions against corruption, headed by the Minister of Justice, with senior
officials from key agencies as members. The Unit in the United Republic of Tanzania
had no power to force agencies to align their policies and actions with the national
strategy; it could only urge them to do so. The secretariat of the coordination commis-
sion in Bulgaria, by contrast, had the authority to convene meetings with each agency
to assess its compliance with the national strategy’s objectives; the commission rein-
forced the secretariat’s authority through weekly meetings where it reviewed the sec-
retariat’s work and issued directives to agencies where compliance was lagging. In the
United Republic of Tanzania, agencies went their own way, ignoring the Good Governance
Coordination Unit’s requests. In Bulgaria, the coordination secretariat succeeded in get-
ting ministries and agencies to work together to implement the strategy.
Sources: Hussman, ed. Anti-Corruption Policy Making in Practice: What can be Learned for Implementing
Article 5 of UNCAC?; Council of Europe, Group of States against Corruption, “First evaluation round
compliance report on Bulgaria” document Greco RC-I (2004) 3E, pp. 3-4.
57
Organisation for Economic Co-operation and Development, OECD Integrity Review of Italy: Reinforcing Public
Sector Integrity, Restoring Trust For Sustainable Growth (Paris, 2013), p. 49; See also Hussmann, ed., Anti-Corruption
Policy Making in Practice: What can be Learned for Implementing Article 5 of UNCAC?.
58
See James Q. Wilson, Bureaucracy: What Government Agencies Do and Why They Do It (New York, Basic Books,
1989), pp. 270-274.
36
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
Moreover, because a chief source of power in government bureaucracy is control over
budgets and personnel, consideration might be given to providing the coordinating unit
with the authority to provide input into budget development and execution. This would
increase its inuence and help ensure that the agencies charged with implementing var-
ious elements of the national anti-corruption strategy have sufcient resources to do so.
C. Foster cooperation between the coordination
body and the implementing agencies
The coordinating unit’s formal authority gives it “hard power” over implementing agen-
cies. But as all government managers soon learn, formal legal authority often does not
sufce to control the actions of the staff of those agencies.
59
Hence, the coordinating
unit must also cultivate the respect of the implementing agencies to further voluntary
compliance (a form of “soft power”). For example, it is important that the coordination
unit has the capacity to intelligently evaluate agency reports; the more skill it displays
in evaluating agency reports, the more respect it will engender from agency professionals,
strengthening its authority and lessening the chance of agencies ignoring its directions.
It is also important to remember the need for the coordinating unit to maintain the
support and respect of not only the senior ofcials at the implementing agencies but
also the mid-level agency staff who are critical to the success of the implementation of
the anti-corruption strategy.
There are a variety of other ways to foster inter-agency cooperation, such as including
each agency’s contribution to inter-agency coordination as part of their overall perfor-
mance assessments, seconding employees of one agency to another and holding joint
training sessions.
60
More generally, coordinating units can promote compliance with the
national anti-corruption strategy by appealing to shared values and by engaging in an
ongoing dialogue about the implementation of the strategy, which is an approach fol-
lowed in Brazil (see box 11). These softer approaches may be especially important with
respect to agencies and institutions over which the coordinating unit lacks any direct
supervisory power, such as independent agencies, the legislature and the judiciary,
61
and
autonomous subnational government entities.
D. Harness the power of reputation
Somewhere between “hard” compulsion and “soft” persuasion, the coordinating unit can
attempt to harness the power of reputation—using public praise or criticism to motivate
agencies to act effectively. This is another technique that may be particularly appropriate
59
A classic study from the United States on the challenge of controlling “street-level” bureaucrats is Michael Lipsky,
Street-Level Bureaucracy: The Dilemmas of the Individual in Public Services (New York, Russell Sage Foundation, 2010).
The inability of internal audit controls to curb agency behaviour is recounted in Michael Power, The Audit Society: Rituals
of Verication (Oxford, Oxford University Press, 1997).
60
A variety of different methods for fostering collaboration across agencies are listed in Patricia M. Noonan, Mary E.
Morningstar and Amy Gaumer Erickson, “Improving interagency collaboration: effective strategies used by high-perform-
ing local districts and communities”, Career Development and Transition for Exceptional Individuals, vol. 31, No.3(2008),
pp. 132-143. The authorities in the province of Alberta, Canada, found that collaboration was enhanced when senior
ofcials’ performance, and thus their pay, was dependent in part on how well they facilitated inter-agency coordination
(Australia, State Services Authority of the government of Victoria, Victorian Approaches to Joined Up Government: An
Overview (Melbourne, 2007), p. 34).
61
As stated in UNODC, The United Nations Convention against Corruption: Implementation Guide and Evaluative
Framework for Article 11 (Vienna, 2015), p. 3, the primary responsibility for strengthening integrity among members of
the judiciary rests with the judiciary itself.
37
Chapter IV. Ensuring effective implementation
with respect to more independent entities. For example, regular reports on the progress
the judiciary is making in resolving corruption cases or on the prosecution service’s
success rate in prosecuting corruption cases might encourage these organizations to
improve their performance. Cultivating broad-based support for the national anti-
corruption strategy and its implementation among the public, the media and civil society
organizations will also help with the coordinating unit’s mission. The coordinating unit
can use its position, and the national anti-corruption strategy document, to encourage
more media coverage and public scrutiny of implementation, which may give other
agencies more of an incentive to comply with their obligations under the strategy.
E. Have each agency agree to an implementation,
monitoring and evaluation schedule
As noted in chapter I, the inclusion of senior representatives from each agency that will
be responsible for implementing elements of the national anti-corruption strategy in the
drafting process increases the chances that those agencies will support and work to
implement the strategy. Besides including those agencies in the deliberations on drafting,
it is often advisable for the strategy drafters, as part of the process of preparing the
policy, to reach explicit agreements with each agency on what measures that agency will
take to implement the strategy, and how and when each agency’s implementation per-
formance will be evaluated.
62
This helps ensure that those responsible for overseeing the
coordinated implementation of the strategy will receive ongoing feedback on the strat-
egy’s progress, including any delays or bottlenecks.
Box 11. Coordinating anti-corruption policy in Brazil
The National Strategy for Combating Corruption and Money-Laundering (Estratégia
Nacional de Combate à Corrupção e à Lavagem de Dinheiro) is the coordinating body for
policies on combating money-laundering and corruption in Brazil. Over 70 federal
government organizations are members, including the prosecution service, different
offices within the police, the judiciary and the legislature. Independent control and
oversight bodies (the Comptroller General, the Federal Audit Court, the Brazilian Secu-
rities Commission, the Council for Financial Activities Control, the National Pension
Plan, the Central Bank, the Brazilian Intelligence Agency and the Brazilian Federation
of Banks) are also members. The Ministry of Justice serves as the secretariat, and civil
society organizations and a variety of state and local agencies also participate.
Member agencies meet once a year in plenary to examine the work of the past year
and agree a plan for the coming year. Working groups assigned to carry out particular
measures meet throughout the year, and an office that monitors implementation meets
every two months.
The Strategy’s achievements include developing and implementing a national training
programme on combating corruption and money-laundering; standardizing requests for
obtaining information on suspects’ bank accounts; creating a group to improve the work
of the Brazilian judicial system; setting up Wiccla, a “wiki” encyclopaedia with informa-
tion about patterns of criminal activity and relevant criminal laws; and developing
anti-corruption legislation.
62
Jody Zall Kusek and Ray C. Rist, Ten Steps to a Results-Based Monitoring and Evaluation System: A Handbook for
Development Practitioners (Washington, D.C., World Bank, 2004), pp. 66-67.
38
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
F. Do not underestimate the challenges of
coordinating implementation
As emphasized throughout this Guide, the substantive objectives laid out in the national
anti-corruption strategy must be drafted with an eye towards what can be realistically
achieved. Where coordination is likely to prove excessively difcult—for example where
regional governments are in the hands of different parties or the judiciary or independent
agencies have resisted meaningful participation in the drafting of the strategy—the strat-
egy’s objectives and elements should be adjusted accordingly. The strategy might be
scaled back in the light of the obstacles to implementing more ambitious measures, or
might incorporate efforts to build support and implementation capacity for more ambi-
tious action at a later date. It remains critical that strategy drafters resist the temptation
to create a far-reaching strategy that is simply not possible to implement in the current
environment. While the failure to effectively implement the strategy’s recommendations
can always be blamed on a lack of political will, that can all too easily be used to excuse
faulty analysis during the crafting of a strategy.
39
Monitoring, evaluating and reporting
Regular monitoring, evaluating and reporting are essential if a national anti-corruption
strategy is to have its desired effect. Monitoring provides data on the strategy’s progress
over time in achieving its objectives; evaluation uses these data to determine which
elements are working, which ones are not, and what changes need to be made. Reporting
the results of monitoring and evaluation helps maintain support for the strategy, not only
by keeping stakeholders and citizens abreast of progress, but also by candidly revealing
setbacks and explaining what is being done to address them. Periodic reporting helps
citizens hold Government accountable for strategy results
63
and provides opportunities
for citizen input into strategy execution.
64
The importance of monitoring, evaluation and reporting for the successful realization of
the strategy’s goals are underlined in the Kuala Lumpur statement on anti-corruption
strategies (see box 12). However, as analyses of anti-corruption strategies by the U4Anti-
Corruption Resource Centre
65
and UNDP
66
have found, in the past, few national anti-
corruption strategy documents have paid sufcient attention to monitoring and evaluation.
This does appear to be changing: several strategies reviewed during the preparation of
this Guide do list the policies to be monitored and evaluated for effectiveness, and the
measures that will be used to do this. Nonetheless, most strategies continue not to address
this essential element in sufcient detail, or else leave vague how the monitoring and
evaluation will be conducted.
63
Steven Van de Walle and Floor Cornelissen, “Performance reporting: what it is and where it comes from” in The
Oxford Handbook of Public Accountability, Mark Bovens, Robert E. Goodin and Thomas Schillemans, eds. (Oxford,
Oxford University Press, 2014), chap. 7, pp. 441-455.
64
Alfred Tat-Kei Ho, “Reporting public performance information: the promise and challenges of citizen involvement”,
in Performance Information in the Public Sector: How it is Used, Van Dooren and Van de Walle, eds., pp. 198-201.
65
Monitoring and evaluation have been almost entirely ignored and there has been a lack of information for and
communication with implementing agencies, political actors, the media and the public about the progress of strategies
(Hussmann ed., Anti-Corruption Policy Making in Practice: What can be Learned for Implementing Article 5 of UNCAC?,
p. viii).
66
Anti-Corruption Strategies: Understanding What Works, What Doesn’t and Why?, p. 4. According to this review, the
most prevalent weakness of anti-corruption strategies is lack of adequate attention to monitoring and evaluation.
V.
40
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
Box 12. Monitoring and evaluation: the Kuala Lumpur recommendations
Monitoring and evaluation are so critical to the success of a national anti-corruption
strategy that the Kuala Lumpur statement on anti-corruption strategies contains six
separate recommendations addressing their inclusion and use in national strategies:
(a) Make monitoring, evaluation and reporting an integral part of the strategy by
incorporating them into it at the design phase;
(b) Develop clear measures for monitoring progress and establish baselines for
comparing changes over time;
(c) Ensure that reliable data for each measure are available;
(d) Report regularly on the monitoring and evaluation data;
(e) Ensure that clear lines of responsibility exist for monitoring, evaluating and
reporting;
(f) Be sure the agency or agency assigned to monitor, evaluate and report on the
strategy has sufficient resources to meet those responsibilities.
There are a variety of ways to monitor policy interventions, and many techniques for
evaluating them.
67
For a national anti-corruption strategy, the best approach will usually
combine two types of monitoring and evaluation: one that assesses implementation and
a second that reviews impact. To illustrate the distinction between these approaches, we
can consider the example of a national anti-corruption strategy under which delays in
prosecuting corruption cases must be reduced, procedures to obtain a business licence
must be streamlined and agencies must increase access to information about their activ-
ities. Implementation monitoring and evaluation would focus on whether the measures
required to achieve these three objectives are being put into effect. Are delay-reduction
measures being introduced? Have rules streamlining the licensing procedure been written
and promulgated? Impact monitoring and evaluation, in contrast, asks what effect these
policy changes are having. Are corruption cases being resolved faster? Are business
licences being issued with less red tape? Is more information about Government in the
hands of citizens? More generally, and more importantly, impact monitoring and evalua-
tion will assess whether these measures have contributed to a reduction in corruption.
Although implementation monitoring and evaluation and impact monitoring and evalu-
ation are frequently done in parallel, they are analytically distinct and are described
separately below.
A. Monitoring and evaluating implementation
Each element of an anti-corruption policy should be monitored to ensure that it is being
implemented in accordance with the timetable and dictates of the strategy. While the
rst step is to establish a coordinating authority with sufcient mandate, autonomy and
inuence to oversee implementation, that alone is not sufcient. The coordinating agency,
as well as the agencies responsible for implementing different elements of the strategy,
the political leadership and the general public, all need ways to measure whether the
67
For useful reviews, see Jesper Johnsøn and Tina Søreide, “Methods for learning what works and why in anti-
corruption: an introduction to evaluation methods for practitioners”, U4 Issue No. 8 (Bergen, Norway, U4 Anti-Corruption
Resource Centre, 2013); Elliot Stern and others, Broadening the Range of Designs and Methods for Impact Evaluations:
Report of a Study Commissioned by the Department for International Development, Working Paper, No. 38 (London,
Department for International Development, 2012); UNDP, Handbook on Planning, Monitoring and Evaluating for Results
(New York, 2009); Peter H. Rossi, Mark W. Lipsey and Howard E. Freeman, Evaluation: A Systematic Approach, 7thed.
(Thousand Oaks, California, Sage Publications, 2004).
41
Chapter V. Monitoring, evaluating and reporting
various components of the strategy are being implemented. The particular techniques for
appropriate implementation monitoring will depend on the circumstances—the type of
policy reform or initiative in question, the country’s capacity and other factors. None-
theless, experience with implementing national anti-corruption strategies, as well as with
other public policies, shows the value of following the eight guidelines listed below.
1. Disaggregate policy reforms into discrete steps
It will often be useful to break down the implementation of each programme into a
series of steps and evaluate progress step by step. Even in the seemingly simple case
of whether a new law has been enacted, breaking the enactment process into a series of
steps (for example, drafting by a ministry, submission to the cabinet for approval, trans-
mission to the legislature, referral to a committee, approval by that committee and a
vote by the full legislature) can be helpful. Implementation monitoring would then consist
of regular updates on which steps have been accomplished, how long it took to clear
each step in the process, and how long a measure has been pending at a particular stage.
Evaluation would consist of determining whether there has been an inordinate delay at
any step and, if so, identifying the reason for that delay. This approach can help speed
up implementation because those responsible for each step in the process will know that
any tardiness in fullling their duty will be revealed. An example of how to implement
a more complex element, an anti-corruption education programme, is shown in box 13.
2. Select one or more indicators of progress
The second step is to select an indicator or indicators that measure progress in putting
each separate action or element into effect. A good indicator should be “SMART”
(specic, measurable, achievable, relevant and time-bound).
68
How these sensible guide-
lines are translated into specic indicators will depend on a variety of factors, many of
them specic to the agency and policy programme in question. For the examples above
(enactment of a law and establishing an anti-corruption education programme), several
implementation indicators are straightforward, and can take the form of a simple
“completed” or “not completed” assessment for each step (or “milestone”) on the path
to implementation. In other cases, the indicator might be quantitative. If the strategy
calls for training all civil servants on ethics within three years, one indicator could be
the number trained each quarter or each year, and a second could be the total number
trained to date. If the strategy provides for strengthening the enforcement of the anti-
corruption laws, an indicator might be the number of investigations opened in a given
period, for example six months or a year, with progress being a steady increase in that
number over time.
A problem with many indicators is the difculty of attributing changes in the indicator
to the programme or policy being monitored. Consider, for example, the indicator sug-
gested above, on the number of investigations opened. The number of investigations
opened in a given period will be dependent on more than the investigating agency’s effort.
The underlying (and unobserved) rates of corrupt or suspicious activity will also affect
the number of investigations opened, as will changes in the agency’s budget and other
factors outside of the implementing agency’s control. The ideal indicator is one where
any change can be attributed solely to the actions of the agency being monitored. In
practice, however, it will often be difcult to nd such indicators. In some cases statistical
techniques can be used to isolate the effect of an agency’s action on the indicator;
68
Handbook on Planning, Monitoring and Evaluating for Results, see p. 58.
42
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
Box 13. Model structure for monitoring an anti-corruption education programme
PARTICIPATION OF SOCIETY, AWARENESS-RAISING AND EDUCATION
PRIORITY 1 ANTI-CORRUPTION EDUCATION
Description Efforts will be made to introduce anti-corruption education
at all stages of the Georgian educational system, including at the
primary, high school and university levels. This will require coop-
eration between government departments, particularly the Minis-
try of Education, and academic institutions. Support can also be
provided by the United Nations Office on Drugs and Crime.
Outcome 1 Integrity education integrated into Georgian primary schools
Outcome
indicators
• Percentage of schools teaching integrity issues
• Percentage of children reaching the required standard as
part of the course
Benchmark
Benchmark
completion
Comple-
tion date
Implementa-
tion bodies Cost
1.1 Survey
conducted of
existing anti-
corruption and
integrity teach-
ing in Georgian
primary schools
Survey has been
completed, with the
participation of over
50 per cent of
Georgian primary
schools from all
regions of Georgia
February
2014
Ministry of
Education
5 000
1.2 Workshop
held to evaluate
present gaps in
integrity
education
Workshop concluded
with agreement on key
gaps in integrity
education and key
features of possible
curriculum
March
2014
Ministry of
Education,
primary school
teachers, civil
society
5 000
1.3 Draft
integrity curricu-
lum developed
for primary
school students
Draft curriculum
completed and
amended following
consultation with
workshop participants
July
2014
Ministry of
Education
15 000
1.4 Curriculum
piloted in a
limited number
of schools
Curriculum is taught
successfully in a
minimum of 30
primary schools across
Georgia
September
2014
Ministry of
Education,
selected
primary
schools
5 000
1.5 Curriculum
revised in line
with feedback
from piloting
institutions
Updated curriculum
completed incorporat-
ing comments from all
schools participating in
pilot programme
January
2015
Ministry of
Education,
selected
primary
schools
10 000
Source: UNODC training materials for national workshops held in Egypt and Georgia in 2014.
in others, reliance on different contextual factors will be necessary to try to interpret the
meaning of changes in the indicator value.
The frequency with which data on an indicator is collected will also depend upon a host
of agency-specic factors. Arrest rates for corruption may be available as often as
43
Chapter V. Monitoring, evaluating and reporting
monthly, whereas data on implementing prevention measures may take longer to compile
and report. What is important is that the data are reported often enough and in a timely
enough manner to be useful.
3. Choose a baseline for each indicator
A baseline is information about the value of the indicator at the beginning, or just prior
to, the monitoring period. It is the starting point against which changes during the period
of strategy execution are measured. For the passage of a new law or the adoption of a
new programme, the baseline will be zero. Baselines for quantitative indicators are
sometimes less straightforward. The baseline for the indicator of the total number of
civil servants completing ethics training might be zero, but if some civil servants had
been trained before the strategy was initiated, it would not. If 3,000 had already been
trained, then the baseline would be 3,000. The baseline for the number of investigations
opened would be the number of investigations opened in the period before the strategy
commences. This could be easy or difcult to determine, depending on the state of
record-keeping at the relevant agency or agencies. In some cases, a baseline study may
need to be commissioned to determine conditions at the outset.
69
When a baseline cannot
be determined for a particular indicator, the indicator must be replaced by one for which
a baseline can be ascertained.
4. Establish realistic targets for each implementation indicator
Once an indicator has been selected and a baseline determined, an implementation target
should be set. For the passage of a law, it would be the time required for enactment. For
an indicator on the number of investigations opened, the target might be a certain percent-
age increase each year. Targets should be ambitious but realistic, which requires attention
to past performance and to the availability of resources. If the legislative process is par-
ticularly slow and cumbersome, and it normally takes at least 18 months to enact new
legislation, setting a six-month target for passing a new law is not realistic; an ambitious
one might be 15 months. If the strategy calls for expanding the use of existing e-procure-
ment methods to all government departments and in the past it has taken two years and
signicant resources to convert one department from a paper system to an e-procurement
system, a target of converting three departments within the rst 18 months of the strategy
is unrealistic. Likewise, if the government budget is under pressure owing to a downturn
in the economy or some other reason, it may be unrealistic to expect that the funding
needed for the rapid conversion of a large number of departments will be forthcoming.
The evaluation process should include a comparison of current indicator levels against
the targets and, more generally, consideration of how the indicator values change over
time, in order to measure the pace of progress.
70
As noted earlier, it is important that
the coordinating body and the implementing agencies agree on realistic targets for the
implementation indicators, and that implementing agencies are held accountable for meet-
ing those targets. This is an area where monitoring by civil society can be particularly
valuable. In the Republic of Moldova, the East Europe Foundation monitors agency
compliance with the strategy, issuing regular updates showing which agencies are ful-
lling their obligations and which ones are lagging behind.
71
69
International Federation of the Red Cross and Red Crescent Societies, “Baseline basics” (Geneva, 2013), p. 2.
70
Kusek and Rist, Ten Steps to a Results-Based Monitoring and Evaluation System, pp. 129-136.
71
Mircea Manoli, “Evaluation report on implementation of the national anti-corruption strategy for the years 2011-2015
(2012-2013 period)” (Chişnşu, East Europe Foundation Moldova, 2014).
44
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
5. Watch for indicator manipulation
When selecting indicators, and when evaluating progress using those indicators, the
coordinating unit must be mindful of a common problem: when agencies or individuals
know they will be evaluated according to certain metrics, they have incentives to behave
in ways that improve performance on those metrics but that might not actually represent
meaningful, substantive improvement on the dimension of performance that the metric
is supposed to measure.
72
If a law enforcement unit is evaluated according to how many
successful convictions it secures, it might spend considerable resources on low-level,
easily proved violations to improve its performance in terms of number of convictions
(or conviction rate), even if those cases are not important. Were all aspects of imple-
mentation easy to dene and measure, “indicator gaming”, as it is sometimes called,
would not be a problem. In most cases, however, the indicators that are selected are
imperfect. The unit responsible for monitoring and evaluation must be aware of the
problem of implementing agencies altering their behaviour in ways that improve their
score on the indicators without improving the genuine implementation of the strategy.
6. Be cautious when using agency self-evaluations
An agency’s own report on its progress in implementing a strategy component can be
a useful indicator. It is inexpensive, based on direct observation, and can reduce the
potential for friction or antagonism between the coordinating agency and the implement-
ing agencies. Indeed, resource and technical capacity constraints mean that, in many
cases, there will be no alternative but to assign each implementing agency primary
responsibility for monitoring its own implementation performance. Experience shows,
however, that agencies are often overly optimistic when assessing how well they are
doing in implementing a programme, and many strategies have fallen short because of
an excessive reliance on the implementing agencies monitoring their own progress. The
agency responsible for coordinating implementation of the strategy should therefore
ensure, to the extent possible, that there is some independent evaluation of progress
toward implementing the strategy. In some cases, it may be advisable to request that an
agency conduct a preliminary self-evaluation and then compare the results to an inde-
pendent evaluation or audit (perhaps by the coordinating unit, another government depart-
ment or an external monitor such as a civil society group or outside consultant). Among
other advantages, doing so might help the implementing agencies understand how they
can better assess their own performance.
7. Utilize evaluations to adjust implementation targets and
strategy goals
Monitoring data alone will show only whether the different elements of the strategy are
being implemented on time. It is the analysis of the data (the “evaluation”) that identies
the reasons why implementation of a particular element is behind schedule or over budget
and that will allow for suggestions on ways to address delays and cost overruns. Eval-
uations serve other purposes as well. Where implementation is proceeding smoothly, or
an agency has devised an innovative solution to a problem, the evaluation can lead to
lessons that will be applicable in other contexts. What is critical is that the evaluations
are used to make adjustments to the implementation process and even to strategy design.
It is simply not possible for the strategy drafters to get all the details of implementation
and design correct at the outset. This is one reason why it is critical that, as explained
72
Radnor, “Hitting the target and missing the point?”.
45
Chapter V. Monitoring, evaluating and reporting
below, policymakers and the public should be provided with clear, easy-to-read reports
on the evaluations. They cannot act on the evaluations, or push for action on them, if
they are not aware of them or do not understand them.
8. Allocate sufficient time and adequate resources
for evaluation
Although it has been suggested that, as a rule of thumb, a typical implementation eval-
uation takes approximately three months and requires basic policy analysis skills,
73
at
best this is a very rough estimate. The time, cost and technical skill required to conduct
an evaluation will vary depending upon what is being evaluated and what the monitoring
reveals. An indicator for enhancing corruption prevention might be the number of cor-
ruption vulnerability assessments conducted every six months. If the monitoring data
reveal a sharp decline in assessments during the previous reporting period, the evaluation
might require no more than a phone call to the agency responsible for vulnerability
assessments to determine the reason for the drop. If, on the other hand, an evaluation
is undertaken to determine why fewer corruption complaints were led in the current
year than the previous one, more in-depth research will be needed. That research could
include an examination of the overall trend in corruption in the country, changes in the
incentives to report and other issues that might have affected reporting levels.
B. Monitoring and evaluating impact
In section A, above, the importance of monitoring and evaluation to ensure that the
different components of a national anti-corruption strategy were actually being imple-
mented was emphasized. That type of monitoring and evaluation, although essential, tells
only whether the policies adopted are being put into place. It does not reveal whether
those policies are reducing corruption. Monitoring and evaluating policies to determine
their effect on the level of corruption are also vital, perhaps more so. After all, because,
as discussed earlier, knowledge about the best way to combat corruption is limited, and
because the particular circumstances of each country are so different, even a well-
designed national anti-corruption strategy may contain signicant errors, omissions and
limitations that must be corrected over time. Thus, an essential element of an effective
national anti-corruption strategy is monitoring and evaluating its impact (as well as the
impact of the individual policies that comprise it) on corruption, and providing mecha-
nisms for revising and updating those policies, and the strategy as a whole, as new
evidence comes to light.
A number of the techniques for monitoring and evaluating the impact of the strategy on
corruption will parallel the techniques examined in chapter II for conducting an initial
assessment of a country’s corruption problem. Many of the information sources used to
make that initial assessment (country evaluation reports, surveys, government data, etc.)
can be used again to help determine whether the overall corruption situation in the
country has improved, worsened or stayed the same. The strengths and weaknesses of
those various data sources were discussed in chapter II and will not be repeated here.
However, some special issues arise when using these data sources not just to provide a
“snapshot” of the corruption situation, but to evaluate progress over time and the impact
of individual policy interventions. Such considerations entail six additional
73
Gloria M. Rubio, “Key choices for a menu of evaluations”, in Building Better Policies: The Nuts and Bolts of
Monitoring and Evaluation Systems, Gladys Lopez-Acevedo, Philip Krause and Keith MacKay, eds. (Washington, D.C.,
World Bank, 2012), pp. 124 and 125.
46
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
recommendations concerning best practices for monitoring and evaluating the impact of
the national anti-corruption strategy and its component policies on corruption. Those
recommendations are listed below.
1. Do not use year-to-year changes in corruption index scores
to measure strategic impact
Countries sometimes try to measure their success or failure in combating corruption by
tracking whether the country’s score on the Transparency International Corruption Per-
ceptions Index or other international indexes has improved or worsened. However, year-
to-year changes in these index scores are generally not meaningful or useful measures
(see box 14). Therefore, even though these indexes may be useful for some purposes,
they should not be employed to evaluate the impact of the national anti-corruption strat-
egy on corruption levels. Indeed, xating on these index scores can be distracting, con-
fusing and counterproductive. For that reason, the national anti-corruption strategy should
not only eschew their use as an evaluative tool, but should explain why it does. There
remains a great deal of misunderstanding about these indicators,
74
and one purpose of a
strategy document is educational: to explain what the nation’s corruption problems are,
how they arose and what can be done to eliminate them. That explanation should include
a discussion of why some of the widely publicized measures of corruption are not useful
for assessing the nation’s progress in combating corruption. This would help to ensure
changes in the scores from year to year do not mislead citizens into thinking the strategy
is failing or succeeding when it is not.
Box 14. Why yearly changes in corruption perceptions are not a reliable measure
Many of the major international corruption indexes, such as the Transparency Interna-
tional Corruption Perceptions Index, are not comparable across years. Year-to-year
changes in the scores can be, and often are, caused by the addition or subtraction of
data sources, or by changes in the underlying data sources that themselves are not
comparable across years.
The statistical margins of error associated with these indexes are often quite large, and
even though the better indexes (like the Corruption Perceptions Index and the World
Governance Indicators) report these margins of error, they are often ignored when mak-
ing year-to-year comparisons. In reality, only a tiny number of year-to-year changes
are statistically significant, and there is so much inherent “noise” in the data that even
substantively meaningful changes might not show up as statistically significant.
Perceptions of corruption are not only often inaccurate, but they are also slow to change
over time, and genuine changes may take years to be captured in perceptions.
Changes in perceptions can also be affected by changes in things other than the actual
level of corruption, including the anti-corruption strategy itself, an effect that could, in
theory, go in both directions. The adoption of a strategy may convince evaluators that
the country is serious about tackling corruption, causing its score to improve even if
the strategy has not made a real impact. Alternatively, adoption of the strategy could
underline the country’s corruption problems, worsening its perception score even if the
strategy is in fact having a positive effect.
74
See, for example, Claire Provost, “Is Transparency International’s measure of corruption still valid?”, The Guardian
(London), 3 December 2013.
47
Chapter V. Monitoring, evaluating and reporting
Even if these perception indexes could be used to detect meaningful changes (or a lack
thereof) to a country’s level of corruption, it would be very difficult to know what that
implies for the impact of the strategy, because the critical question is: “How would the
country be doing without the strategy?” To answer that question, it is not enough to
know how the country’s perceived corruption differed before and after the adoption of
the strategy; what is needed is information about what would have happened (what the
change would have been over the same time period) had the strategy not been adopted.
And such information is generally impossible to find out from the available data.
2. Select impact indicators that can be compared over time
The problems of relying on international corruption perception indexes are part of a
larger measurement issue: to be useful for impact monitoring and evaluation, the indi-
cators selected should be comparable over time. For many indicators, intertemporal com-
parisons are fairly straightforward: for example, the number of corruption complaints
received, the percentage of people saying they were solicited for a bribe or the amount
of money found missing in a public expenditure tracking survey can all be compared
year to year. Indeed, although measuring corruption and assessing the impact of policy
interventions may seem like a daunting task, it is sometimes both possible and cost-
effective to measure changes in corruption (or likely corruption) relatively directly,
through investigations, audits and spot checks. Even for relatively poor countries, audits
and spot checks can be a cost-effective technique for assessing corruption and related
outcomes of interest. Financial audits can reveal the amount of money that disappears,
without adequate explanation, from government programmes. Spot checks can reveal
whether public employees such as teachers
75
and health-care workers
76
are actually show-
ing up to their jobs.
But even when appropriate indicators that can be compared over time have been iden-
tied, it is still important to be careful not to misinterpret the signicance of year-to-year
changes in these indicators. For example, as noted earlier, an increase in corruption
complaints might indicate that corruption is getting worse, or it might indicate that
citizens are more aware of or more willing to report incidents of corruption. This is
perhaps not such a concern with regard to audits and spot checks, but it should still be
taken into account. A decrease in the amount of “missing money” detected in audits or
expenditure tracking surveys might mean that corruption has decreased, but it might also
mean that corrupt actors have become better at fooling the auditors or have shifted to
other, less easily detected modes of corruption. As stressed in the discussion on con-
ducting a preliminary analysis, data on corruption are often open to more than one
interpretation, and it is important to look at the context to determine the correct
interpretation.
3. Isolate the impact of the policies implemented pursuant to
the strategy
Even if valid and reliable measures or proxies for corruption can be identied, assessing
the impact of the various policies adopted to implement the national anti-corruption
strategy will still be challenging, as it is often difcult to disentangle the effect of the
75
Esther Duo, Hanna Rema and Stephen P. Ryan, “Incentives work: getting teachers to come to school”, American
Economic Review, vol. 102, No. 4 (2012), pp. 1241-1278.
76
Abhijit V. Banerjee, Rachel Glennerster and Esther Duo, “Putting a band-aid on a corpse: incentives for nurses in
the Indian public health care system”, Journal of the European Economics Association, vol. 6, Nos. 2 and 3 (2010),
pp.487-500.
48
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
policy from all of the other factors that may have had an effect as well. Again, before-
and-after comparisons, while sometimes helpful, can be misleading, because what is
important when evaluating a policy is the comparison of what happened with the policy
in place with what would have happened had it not been there.
One way to deal with this issue, rst used to test the effectiveness of new drugs and
increasingly employed to evaluate public policies,
77
is the use of randomized controlled
trials. In a randomized controlled trial, a new policy is applied to certain randomly
selected units (such as particular villages or ofces), but not to others (which then serve
as controls). The randomization allows any difference in average outcomes between the
two groups to be attributed to the policy intervention rather than confounding factors.
Randomized controlled trials are considered the most rigorous method for evaluating a
policy’s impact and have been used to evaluate, for example, the effectiveness of different
methods for reducing corruption in road construction
78
and different techniques to ensure
that teachers come to work.
79
These trials can and should be used much more often in
the context of anti-corruption policy. They may be particularly appropriate when a new
policy is to be introduced in stages, for example with a pilot project. If the implemen-
tation of the pilot project is randomized and the outcomes are assessed, a wealth of
useful, reliable information can be provided on the policy’s impact.
Randomized controlled trials do, however, have a number of limitations. For many coun-
tries, the most important of these limitations is cost: a well-designed randomized con-
trolled trial takes time, resources and expertise. Additionally, while they are well suited
to addressing the impact of specic, narrowly dened interventions, they are less well
suited to assessing complex interactions among multiple policy interventions and cer-
tainly could not be used to assess the impact of the anti-corruption strategy taken as a
whole.
80
There are other limitations as well. For example, sometimes the short-term
impact of a pilot programme differs from the impact observed when a policy is
implemented more broadly. Despite these limitations, randomized controlled trials are
increasingly being used for policy evaluation, and national anti-corruption strategy
drafters should consider, in the context of their impact evaluation plans, opportunities
for expanding their use.
When randomized controlled trials are not feasible or appropriate, other impact evaluation
techniques will be required to separate out the impact of the national anti-corruption
strategy, and the individual anti-corruption reforms that comprise that strategy, from
confounding factors. Sometimes, the best alternative will be to try to take as many other
confounding factors into account and explicitly control for them in a statistical model.
81
When this is not possible either, those overseeing the strategy need to explore other
options, including gathering qualitative or survey data (assessing what those close to the
policy reform think, based on their subjective impressions of its impact). None of these
alternatives is ideal, but what is critical is that the impact evaluation process at least
makes some attempt to disentangle the effects of the policy reforms associated with the
strategy from confounding factors, rather than relying on simple before-and-after
comparisons, which can often be misleading.
77
Rossi, Lipsey and Freeman, Evaluation: A Systematic Approach, pp. 238-252.
78
See Benjamin A. Olken, “Monitoring corruption: evidence from a eld experiment in Indonesia”, Journal of Political
Economy, vol. 115, No. 2 (2007), pp. 200-249.
79
See Duo, Rema and Ryan, “Incentives work: getting teachers to come to school”.
80
See Chris Elbers and Jan Willem Gunning, “Evaluation of development programs: randomized controlled trials or
regressions?”, World Bank Economic Review, vol. 28, No. 3 (2014), pp. 432-445.
81
See Johnsøn and Søreide, “Methods for learning what works and why in anti-corruption”.
49
Chapter V. Monitoring, evaluating and reporting
4. Be sensitive to the cost and time required
The cost of an indicator or evaluation technique in terms of money, time and attention
is always a relevant factor. As is true with the setting of substantive goals and policy
proposals, techniques for evaluating a strategy’s impact must be realistic, taking into
account the constraints under which the country must operate. A public expenditure
tracking survey, which compares the funds the central Government allocates to sub-
national units for various programmes with the amount that each recipient actually
receives, can provide an accurate measure of how much money is lost to corruption or
other “leakage” along the way. Implementing such a survey can, however, cost anywhere
from 75,000 to 400,000 United States dollars, depending on the size of the country and
the number of services covered. It can also take a year to conduct.
82
For many countries,
this investment is well worth it given the rich data such a survey may provide, to say
nothing of its potential deterrent effect. But in other countries, public expenditure tracking
surveys are too expensive and take too long to undertake on a regular basis.
For some policy interventions, it will simply be impossible to measure their impact on
corruption directly, or even to nd reliable, affordable proxies.
83
In these cases, the best
that can be done is to base the policies on the best available theory and evidence, and
to work to ensure appropriate implementation. There are also second-best ways of judg-
ing impact. Assessing what those close to the policy reform think is one. Companies
that bid for public contracts can be asked whether they think reforms to the public
procurement law have reduced corruption. Another alternative is to probe deeper into
policy implementation.
84
How much effort is being devoted to it? How carefully is it
being put into effect? How efciently? A third option is to commission an open-ended
research programme looking at all aspects of the programme: the theory of change behind
it, stakeholder feedback, programme management and other real-time, operational,
action-research that can ll in gaps in the evidence and theory behind the policy.
85
What
is critical is that some form of evaluation is conducted on each element of the strategy.
Otherwise, policymakers and citizens will be asked to take it on faith that it is working
as planned.
5. Involve civil society organizations, scholars, research
organizations and citizens
There are several reasons for broadening participation in monitoring and evaluation
exercises. Civil society groups, academic researchers and citizens may be able to provide
rich, helpful qualitative information on how the national anti-corruption strategy is work-
ing, or at least on how it is perceived. In addition, keeping civil society groups and
citizens involved in the evaluation process can help maintain their support for the strategy.
Regular interaction between the coordinating entity, the government agencies responsible
for implementation, civil society groups, researchers and the general public can also
generate new ideas and approaches, leading to the continual improvement of the strategy
over time.
82
Gurkan, Kaiser and Voorbraak, “Implementing public expenditure tracking surveys for results”, p. 7.
83
The challenges include abundant possible causes, insufcient theory, weak data, competing estimation techniques
and the possibility of deep heterogeneity (Robert Klitgaard, Addressing Corruption Together (Paris, Organisation for
Economic Co-operation and Development, 2015), p. 47).
84
Theodore Poister, Measuring Performance in Public and Nonprot Organizations (San Francisco, California, Josey-
Bass, 2003).
85
Stern and others, Broadening the Range of Designs and Methods for Impact Evaluations, p. 4.
50
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
6. Provide methods for ongoing revision to the strategy
A major purpose of monitoring and evaluating the impact of a strategy is to allow for
its ongoing revision as new information comes to light. The national anti-corruption
strategy, although issued at a particular point in time and intended to be comprehensive,
must be a living document. A strategy document can provide a helpful set of guidelines
and a preliminary plan. However, given that corruption is such a complicated issue, to
be dealt with over a long time period in a complex, uid environment, the strategy must
incorporate within it a mechanism for constant feedback, re-evaluation and potential revi-
sion. The strategy should, when possible, explicitly discuss and describe the process for
feedback and revision over time, so that the document remains relevant both as a technical
guide to sound policy and as a means for building political support and fostering political
accountability for the effective implementation of anti-corruption policy.
C. Public reporting of results of
monitoring and evaluation
Reporting on the implementation of the anti-corruption policy and on the ndings con-
tained in the monitoring and evaluation reports is an important element of the overall
accountability mechanism that should be built into every strategy. Reporting improves
coordination, enables organizations to learn from past mistakes, ensures transparency of
the process, stimulates public involvement and may generate support for anti-corruption
efforts, sometimes even increasing pressure in favour of anti-corruption reforms.
Reporting may have multiple dimensions, all feeding into the expected results outlined
above. The three types of reporting set out below are used to work towards achieving
slightly different policy goals.
1. Internal reporting as part of coordination mechanisms,
ensuring the accountability of the implementation process
Internal reporting is an indispensable element of the process of managing the implemen-
tation of anti-corruption policy. It enables the leadership of an organization to have a
clear picture of whether and how specic activities within the framework of the overall
anti-corruption strategy are being implemented.
2. Reporting to higher authorities in the executive or
parliament
As a part of the overall coordination and implementation mechanisms, State bodies share
information on how they implement the policy and provide information to the central
institution in charge of policy coordination and implementation. Such reporting ensures
the internal accountability of the system to the higher-level bodies of the executive
(council of ministers, prime minister, president) or to the parliament. Sometimes,
specialized anti-corruption structures may also be the recipients of this type of report
(particularly in the case of specialized policy coordination commissions).
3. Public reporting ensures transparency in the
implementation of the national strategy
The importance of reporting to the public, although critical, is often overlooked. Sharing
the results with the public may have an educative effect about the strategy and the
51
Chapter V. Monitoring, evaluating and reporting
challenges in implementing it. This will both keep up public pressure for full implemen-
tation of the national anti-corruption strategy and help temper expectations when dif-
culties arise. Regular public reporting on progress towards implementing the strategy
will also increase citizen support for the strategy when it faces political roadblocks. The
reporting process should include ways the public can provide input into strategy execu-
tion, as citizens will often have different perspectives or insights from those managing
the strategy.
86
Ultimately, reporting to the public is also required in order to ensure the
transparency of the process and to strengthen public support for anti-corruption reforms.
Such reporting could be a part of an overall communication strategy, aiming at strength-
ening public support, informing the public and reducing opposition to sometimes difcult
reforms. In this context, development of a public communication plan is of crucial
importance; it should take into account and utilize all possible information mediums,
including social media, without prejudice to traditional and tested approaches such as
town hall meetings and public hearings.
Monitoring and evaluation reports should therefore be aggressively publicized, shared
with media outlets and civil society organizations and posted on the Internet. More
generally, a communication plan may be part of the overall anti-corruption strategy.
87
To
facilitate effective citizen understanding of and engagement with the implementation,
monitoring and evaluation process, public reports should be presented in a clear, simple
and easily understandable format, keeping technical language to a minimum, avoiding
large “data dumps” and using charts, graphs and other visual aids when appropriate.
88
While the reports will usually originate with the implementing agencies, the coordination
entity should be ultimately accountable for their timely publication and distribution.
Therefore, when budgeting for this entity, provision should be made for a public affairs
ofce or similar unit to ensure the capacity for effective public reporting.
86
Tat-Kei Ho, “Reporting public performance information”, pp. 198-201.
87
For how to craft a media plan, see Elaine Byrne, Anne-Katrin Arnold and Fumiko Nagano, Building Public Support
for Anti-Corruption Efforts: Why Anti-Corruption Agencies Need to Communicate and How (Washington, D.C., World
Bank, 2010).
88
Kusek and Rist, Ten Steps to a Results-Based Monitoring and Evaluation System, pp. 129-137.
53
Annex. Kuala Lumpur statement on
anti-corruption strategies
a
On 21 and 22 October 2013, high-level representatives of anti-corruption authorities and
national planning authorities from South, East and South-East Asia and anti-corruption
experts from around the world gathered in Kuala Lumpur at the invitation of the United
Nations Ofce on Drugs and Crime (UNODC) and the United Nations Development
Programme (UNDP), in partnership with the Government of Malaysia, to discuss a set
of guidelines for anti-corruption strategies that could instruct the process of developing,
designing and implementing sustainable anti-corruption strategies.
The participants included several ofcials of anti-corruption authorities, other anti-
corruption practitioners and representatives from national planning authorities and exec-
utive, legislative and judiciary branches. Representatives of UNDP, UNODC the World
Bank, the Anti-Corruption Initiative for Asia and the Pacic of the Asian Development
Bank and the Organization for Economic Cooperation and Development, Transparency
International and the U4 Anti-Corruption Resource Centre also took part in the
proceedings.
The participants reviewed and discussed country experiences from around the world,
with a particular focus on South, East and South-East Asia, relating to the process of
developing anti-corruption strategies, the design and content of anti-corruption strategies
and the monitoring and evaluation of anti-corruption strategies.
Recalling the United Nations Convention against Corruption
b
which, in its articles 5 and
6, obliges States Parties, in accordance with the fundamental principles of their legal
systems, to develop and implement or maintain effective, coordinated anti-corruption
policies that promote the participation of society and reect the principles of the rule
of law, proper management of public affairs and public property, integrity, transparency
and accountability,
Recognizing that anti-corruption strategies can provide a comprehensive policy frame-
work for actions to be taken by States in combating and preventing corruption, a useful
tool for mobilizing and coordinating efforts and resources by Governments and other
a
Originally issued in conference room paper CAC/COSP/2013/CRP.12. The Conference of the States Parties to the
United Nations Convention against Corruption took note of the statement in its resolution 5/4.
b
United Nations, Treaty Series, vol. 2349, No. 42146.
54
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
stakeholders, for policy development and implementation, and for ensuring monitoring
of policy implementation,
Considering that a number of countries in South, East and South-East Asia have imple-
mented anti-corruption strategies or are in the process of drafting new strategies and
revising old ones,
Recognizing that there are a large variety of models of and approaches to the develop-
ment of anti-corruption policies and strategies and common lessons to be learned from
experiences in developing these strategies in the region,
The participants:
1. Recommend the following:
Anti-corruption strategy development process
• POLITICAL, SOCIAL, ECONOMIC AND CULTURAL CONTEXT: States
must take into account their particular political, social, economic and cultural
context when designing anti-corruption strategies.
• POLITICAL WILL: Committed political leadership, ideally from the highest
levels of the State, and broader political support to steer the overall process and
mobilize necessary resources, are necessary conditions of an effective anti-
corruption strategy development process.
• STAKEHOLDER INVOLVEMENT (INCLUSIVE PROCESS) AND OWNER-
SHIP: Broad engagement of stakeholders builds ownership and helps to ensure
acceptability and effectiveness of strategies adopted. State institutions (executive,
legislative and judiciary) at the national and subnational levels, civil society
organizations, private sector, media, professional societies, trade and industry
associations and labour unions, academic institutions, youth and cultural organ-
izations, can serve as important allies and partners in the development of anti-
corruption strategies and can reduce the vulnerability of the reform efforts to
changes in political leadership.
• CLEAR AND TRANSPARENT PROCESS: The process of developing strategies
needs to be clear and transparent from the outset.
• COMMON VISION: A consensus should be built around a common vision and
intended objectives of strategies.
• STRENGTHENING COORDINATION: Anti-corruption strategies should focus
on enhancing inter- and intra-agency coordination during the development
process as well as the implementation and monitoring phases.
• SOUND KNOWLEDGE BASE: Development, implementation and monitoring
of strategies should be informed by sound diagnostics, needs and evidence of risk
and vulnerability areas and gaps in anti-corruption policies and institutions.
• SUSTAINABILITY AND INSTITUTIONALIZATION OF THE PROCESS:
Development of strategies should be institutionalized to ensure continued rele-
vance and timely modication of the anti-corruption strategies.
• ALLOCATING AND MOBILIZING RESOURCES: Necessary resources should
be mobilized at the time of development of strategies to ensure effective
implementation and monitoring of strategies.
• PUBLIC COMMUNICATION AND ENGAGEMENT: Anti-corruption and
national planning authorities shall communicate and engage with the public
regularly in order to ensure public condence and channel feedback for the
effective implementation of anti-corruption strategies.
55
Annex. Kuala Lumpur statement on anti-corruption strategies
Anti-corruption strategy design and content
• RATIONALE CORE OBJECTIVES AND REALISTIC GOALS: Core objectives
and goals, and rationale for interventions should be dened based on national
priorities, and identied gaps and needs.
• NATIONAL DEVELOPMENT STRATEGY/PRIORITIES AND BROADER
CONTEXT: Anti-corruption strategies should be incorporated within broader
national development initiatives currently in focus and should take into account
international/regional obligations.
• INTEGRATION WITH OTHER RELEVANT NATIONAL PROGRAMMES/
REFORM AGENDAS: Anti-corruption strategies should take into account and
establish links with other relevant national strategies (e.g., judicial sector, public
administration reform, open government, etc.) and should seek to form synergies
with other agencies.
• COMPREHENSIVE AND COORDINATED APPROACH: Anti-corruption strat-
egies should be organized under an overarching/holistic approach while taking
into account sector-specic needs.
• CLEAR AND UNDERSTANDABLE DOCUMENT: Strategies have to be clear,
concise and easily understood.
• STRUCTURE AND DESIGN: While there is no simple formula for the proper
design, content or implementation of anti-corruption strategies, the United Nations
Convention against Corruption can be used as a framework for anti-corruption
strategies, taking into account relevant data, particular needs and national
capacities.
• PRIORITIZATION AND SEQUENCING: Strategies need to be realistic on what
is achievable in the short, medium and long term, set clear priorities and sequence
actions based on priorities. Strategies could be designed with the aim of enhanc-
ing the credibility of leadership and ensuring quick tangible results to strengthen
the national commitment to reform.
• IMPLEMENTATION MECHANISM: It is imperative that strategies provide for
an implementation mechanism in the form of an action plan with clearly iden-
tied responsibilities and timelines for implementation with focus on results. The
agency designated to coordinate implementation of strategies should be within
high-level government agencies.
• SUBNATIONAL IMPLEMENTATION: Where applicable, particular attention
should be paid to strategies’ implementation at the subnational and local levels.
• INSTITUTIONAL AND FINANCIAL SUSTAINABILITY (NEEDS and
CAPACITIES): Strategies should provide for their institutional and nancial
sustainability and should take into account capacity for implementation.
Anti-corruption strategy monitoring and evaluation
• INTEGRAL PART OF STRATEGY DESIGN: Monitoring and evaluation mech-
anisms are an integral part of national anti-corruption strategies. Elements of
evaluation and data-collection systems should be built into strategies from the
design phase.
• INDICATORS WITH CLEAR BASELINES AND TARGETS: Measurable indi-
cators, with established baselines and tracking mechanisms, are needed to deter-
mine whether targets are being achieved.
• NEED FOR DATA-GENERATION TOOLS: Effective monitoring and evaluation
require reliable data that are generated based on multiple sources.
56
National Anti-Corruption Strategies: A Practical Guide for Development and Implementation
• REGULAR REPORTING: Regular monitoring and reporting allow authorities to
gauge progress in implementation and achieving results in curbing corruption.
• EVALUATION VS. PROGRAMME MANAGEMENT: It is important to distin-
guish between programme management monitoring (activities/outputs) as opposed
to evaluation (outcomes/impact) and between implementation responsibilities as
opposed to monitoring and oversight responsibilities.
• RESPONSIBLE AUTHORITY: National body/bodies should be entrusted with
the responsibility for monitoring, implementation and regular reporting and be
provided with sustainable institutional and nancial support. An independent
evaluation should ensure accurate monitoring and reporting at regular
intervals.
2. Encourage anti-corruption and national planning authorities to promote these
recommendations within their respective agencies, countries and regional/international
networks in a time-bound manner.
3. Call upon anti-corruption and national planning authorities to promote these recom-
mendations in order to assist members of the executive and the legislature and the
judiciary, and the public in general, to better understand and support anti-corruption
strategies in their development, design, implementation and monitoring.
4. Welcome the commitment of the host country to raise these recommendations at the
fth session of the Conference of the States Parties to the United Nations Convention
against Corruption and encourage other participating countries to support this
initiative.
5. Express appreciation and gratitude to the Government of Malaysia for hosting the
Asia Regional Meeting on Anti-Corruption Strategies organized by UNDP and UNODC
to reect on and promote these recommendations for anti-corruption strategies.
Vienna International Centre, P.O. Box 500, 1400 Vienna, Austria
Tel.: (+43-1) 26060-0, Fax: (+43-1) 26060-5866, www.unodc.org
V.15-04910
