Counting, Measuring and Conceptualising Corruption: what is or can be the use of official statistics?

Transcription

1 Counting, Measuring and Conceptualising Corruption: what is or can be the use of official statistics? Dr. Michael Levi Professor of Criminology Cardiff University Brussels, 2017

2 Corruption data a meaningful construct? 0 What do we want data for, and how serious are we? 0 What do we count, whom do we count and where do we count them? 0 If corruption is a necessary prelude to ongoing crime networks, how often do we count the corruption and where? 0 Counting victimisations, counting consensual illicit acts Activity/input indicators, output indicators and outcome indicators Activities per what? Population? GDP? Contracts? Indicators that provoke discussion (inc official

3 Waving or Drowning in Data? Administrative data and data sharing on actions You only find out who is swimming naked when the tide goes out. (Warren Buffett 2001) A rise in crime rates can be an index of social progress rather than social decay

4 The China Syndrome

5 Non-traditional forms of Serious Crime Containment Community approaches Regulatory Disruption and non Justice system approaches Private sector involvement 1. Community crime prevention 2. Passive citizen participation: - giving information about harms and risks, whistle-blower hotlines 3. Active citizen participation: civic action groups/communist Party 1. Regulatory policies, programmes & agencies (domestic inc. Chinese Central Commission for Discipline and Inspection & foreign, including Council of Europe/EU/FATF/IMF/OECD/World Bank) 2. Faster customs & other regulatory treatment for firms & countries that have approved internal compliance programmes 3. Routine & suspicious activity reports as investigative triggers 4. Tax policy and programmes (HMRC Criminal Taxes Unit) 5. Civil injunctions and other sanctions (RICO, contract vetting, SCPOs) 6. Military interventions 7. Security and secret intelligence services 8. Foreign policy/aid programmes (certification, EU accession) 1. Individual corporate responses 2. Professional and industry associations 3. Special private sector action bodies (CIFAS, IFB, FACT, FFUK) 4. Anti-Identity fraud, anti-corruption and money laundering software 5. Private policing and forensic accounting Levi (2012); see also Levi and Maguire (2004)

6 The Dishonest Politician s Guide to Crime Control 0 Generate a number for prevalence, harm or effectiveness that suits the bureaucratic and/or moral imperative 0 This then becomes a fact by repetition 0 G20 communiqué by Global Financial Integrity But the larger the figure, the less likely it is that any law enforcement of preventative measures will impact significantly on it 0 So why do people almost never match crime costs global, national or local - with control measures in place? 0 Can we resist the media and political demands? 0 If not, what are the consequences?

7 Counting Crimes or Counting Harms? 0 Counting crimes is what we do, ignoring problems of how victims can know the intent of offenders 0 But we sometimes also count organised crime groups nationally plus references to foreign countries or hubs 0 Do we multiply count corruption in every country the drugs or people pass through? 0 How do we evaluate the harmfulness of corruption arising in a third party country and should we include it solely in their crime statistics and/or in ours if one of our citizens or denizens is doing the corrupting? 0 Counting domestic v counting transnational bribery payers, receivers and perhaps launderers of corruption proceeds

8 Harm and Risk Three dimensions of harm and risk: Economic costs (present/future) & impact upon victims and upon third parties who fear being defrauded What the media represent the risks to be drives politics Continuing risks arising from the kind of people who are committing financial crimes Organised criminals with no stake in social respectability to constrain them and growing access to corruptible/ pressurisable staff, especially in recession What they spend their money on (terrorism, political corruption, running for office, business) - though mostly on party-going So do we need to combine victim surveys with offender location studies, & how do we do that for low-clear ups? We need sentinel trend data, not precise numbers

9 M Levi, Cardiff University