Illicit Activity: The Economics of Crime, Drugs and Tax Fraud

The Economics of Crime, Drugs and Tax Fraud

Ziggy Macdonald, David Pyle, Ziggy Macdonald, David Pyle, Ziggy Macdonald, David J Pyle

• Publisher: Taylor & Francis
• Year: 2018
• Print ISBN: 9781138737723
• eBook ISBN: 9781351733403
• Edition: 1
• Topic: Social Sciences
• Subtopic: Sociology
• Index: Social Sciences

This title was first published in 2000: A collection of research papers on the theme of illicit activity, all written by either members or associate members of the Public Sector Economics Research Centre in the Department of Economics at the University of Leicester. The work reported covers three areas of activity: crime (especially property-related crime); consumption of illegal substances (drugs); and income tax evasion.

1. Introduction

ZIGGY MACDONALD AND DAVID PYLE

Overview

This book contains a collection of research papers on the theme of illicit activity, all written by either members or associate members of the Public Sector Economics Research Centre in the Department of Economics at the University of Leicester. The Oxford English Dictionary defines 'Illicit' as anything which is 'unlawful; not allowed'. The work that is reported in this volume covers three broad areas of such activity, namely crime (especially property related crime), consumption of illegal substances (drugs) and income tax evasion.

Crime and drug taking, especially, are regarded as major social issues in many advanced societies. For example, the UK Prime Minister (Tony Blair) in announcing the Government's ten-year strategy on drugs in 1998 said that, life in Britain 'could be so much better if we could break once and for all the vicious cycle of drugs and crime which wrecks lives and threatens communities' (Home Office, 1998, p. 1). The Strategy Document went on to say, 'drugs are a very serious problem in the UK...(they) are now more widely available than ever before and children are increasingly exposed to them. Drugs are a threat to health, a threat on the streets and a serious threat to communities because of drug-related crime' (ibid, p. 2). In the US, President Clinton announced, in 1998, a strategy to cut the use of illegal drugs by 50 per cent by 2007. This followed a report from the National Institute of Health showing that in 1996 over 40 per cent of High School Seniors had used drugs in the last year. In the UK, the British Crime Survey 1998 showed that 31 per cent of 16-19 year olds had used drugs in the last year.

Concern about crime is widespread throughout the Western World. The 1996 International Crime Victimisation Survey (Mayhew and van Dijk, 1997) shows that in eleven OECD countries, on average 22 per cent of respondents felt unsafe when out alone after dark. In England and Wales, the proportion was as high as one-third. However, the good news is that fear of crime has fallen in Britain during the 1990s, as the crime rate has come down (British Crime Survey, 1998). The same survey shows that in eight of the 11 OECD countries included in the sample, the crime victimisation rate was at least 24 per cent in 1995, with The Netherlands and England heading the league table with victimisation rates of 31 per cent. Despite this, in several countries (US, Canada and England and Wales, for example) rates of recorded crime have been falling since the early 1990s. Nevertheless, rates of recorded crime are still high by historical standards and the reasons for the decline in recent years are not entirely clear (but see Chapters Two through Six below).

For obvious reasons, the extent of income tax evasion cannot be known with any degree of certainty. However, various ingenious devices have been used by economists in order to try to measure it (see Pyle, 1989 for a survey) and it is common to see estimates of 'hidden' economic activity amounting to between ten and 15 per cent of Gross Domestic Product in a whole range of countries. This might be thought to have a number of economic consequences, the most obvious of which is the loss of tax revenue for the Government. In fact, loss of tax revenue is not obvious, provided tax evaders buy goods and services from others, which will generate income for the providers and tax receipts for the Government. However, tax evasion will certainly distort macroeconomic indicators such as unemployment statistics and measures of national income and output. Such distortions may cause governments to make totally inappropriate macroeconomic interventions. Also, tax evasion may distort resource allocation, if the opportunities to evade are greater in certain occupations than others. This may lead some individuals to choose an occupation that is in their own financial interest, but not in the best interests of society (for example, university professors may prefer to become self-employed odd job people, if by doing so they can avoid paying a great deal of income tax).

In response to evidence of increasingly widespread engagement in illicit activity by members of the general public, governments have been increasing their expenditure on policing, prisons, courts, and law enforcement generally. This represents a social waste, for the resources used in this way could be used for more productive purposes. For example, in the UK, expenditure on the criminal justice system increased in real terms from £8.6 billion in 1988-9 to £11.3 billion in 1997-8, i.e. by about 30 per cent over nine years (Social Trends, 1999).

What have economists got to say about such phenomena? Most people's view (if they have any view at all) is that economists examine either issues such as unemployment, inflation, and interest rates or the structure and performance of firms and industries and the role of markets. In other words, economists focus upon what can be described as narrowly economic and financial issues. Nothing could be further from the truth. Over the last 30 years or so, an increasing number of economists have been applying economic methods to analyse all kinds of social phenomena, including activities such as crime, drug taking, marriage and divorce.

There are two aspects of illicit activity to which economists can contribute. First, in dealing with the problems which illicit activity creates, government agencies use a great deal of resources and it is an economic issue about how much resource should be put into dealing with these problems and how best to allocate that resource amongst competing areas and policies. For example, should we spend more on police (prevention and detection) or prisons (punishment), say, as a way of reducing theft? Should we spend more on education and rehabilitation programmes for drug users than on catching and punishing drug dealers? If we had additional resources should we use them to reduce tax evasion or allocate them to burglary prevention programmes instead? This kind of analysis is meat and drink to an economist. Second, the decision to engage in illicit activities can be viewed as an economic one about how individuals allocate their time to competing activities, some of which are risky i.e. you might get caught and punished. Economists are used to analysing decisions of this type e.g. labour supply decisions, choice in risky situations, etc. What distinguishes the economic approach to criminal participation is the assumption that individuals engaged in such activity are rational decision makers who attempt to optimise some welfare function. In this volume you will see a number of papers that explore the implications of this approach for modelling participation in illicit activity.

The economic approach to participation in illicit activities was first advanced by the American economist Gary Becker. In fact, Becker applied the economic approach to a much wider range of 'social' issues, such as crime, marriage and divorce, discrimination, drug addiction and religion. Somewhat modestly, Becker attributes his interest in applying economics to 'social' problems to advice given to him by the Chicago economists Milton Friedman and Theodore Schultz when he was a young graduate student (see Tommasi and Ieurelli, 1995).

This collection of papers has a unifying theme, which is the approach that the contributors have adopted to the study of their subject. The approach adopted at PSERC to analyse illicit activity is primarily empirical, although it is quite broad. It encompasses aggregate time-series studies, micro (cross-sectional and panel data) analysis, and mathematical simulation. Members of PSERC have been at the forefront of the application of economic methods to the study of illicit activity in the UK and this volume reflects the breadth of the work that is being undertaken by them. Members of PSERC have recently published articles outlining this research in a number of academic journals. Whilst the papers contained in this volume are based upon that published work, they have been specially written for this book.

In the rest of the Introduction we outline the contents of each chapter. We have grouped this discussion under three headings, which are crime, drugs and tax evasion.

Crime

In recent years research into crime at PSERC has focussed upon an economic explanation of criminal activity. In particular much analysis has been undertaken of the relationship between the incidence of crime, especially property crime, and the state of the nation's economy. In simple terms this work confronts the question whether there is any evidence that when the economy is in recession property crime increases and that when the economy enters a boom property-related crime diminishes. This work has concentrated upon the experience in England and Wales using aggregate, annual time-series data. Chapters Two to Four report some of that work. In Chapter Five Denise Osborn, a Visiting Fellow at PSERC examines the relationship between crime and the economy using quarterly aggregate data. In Chapter Six, Amor Diez-Ticio, another visitor to PSERC, applies the underlying technique to data for the US. Chapter Seven, written by another visiting scholar at PSERC (Ingolf Dittmann) presents a theoretical analysis of the reasons why States tend to use both fines and imprisonment as alternative forms of punishment for the same crime.

In Chapter Two, Derek Deadman and David Pyle begin by presenting a theoretical model of criminal participation, which treats involvement in crime as a labour supply decision. This is the essence of the economic approach to crime, which treats potential criminals as rational individuals who respond to incentives. They conclude that the amount of time someone will spend engaged in criminal activity will then depend upon 'wage rates' in both legitimate and criminal activity, the probability of being caught and convicted of committing a crime, the size of the punishment if caught and the probability of being unemployed in legitimate activity. The bulk of Deadman and Pyle's chapter is devoted to applying the economic approach outlined earlier in their chapter to data on residential burglaries in England and Wales between 1950 and 1995. They use aggregate national level data (i.e. data for the whole of England and Wales). This model contains a number of deterrence, economic and 'taste' variables, which are the probability of being convicted, the conditional probability of being imprisoned, the number of police officers, the expected length of imprisonment, the number of males aged 15 to 24 years, the unemployment rate, and per capita real personal consumption. Deadman and Pyle provide a discussion of some of the issues, which confront this type of analysis e.g. measurement error in the crime variable (their research uses recorded crime data, which are notoriously error prone. This is issue is addressed more fully in Chapter Three) and possible simultaneity between the dependent and explanatory variables. They also discuss a major problem in modelling time series aggregate data, which are the time-series properties of their data. This is an important issue for regression analysis, because both crime and economic data in post-war England are dominated by time trends and meaningful regression results can only be found when the data are freed of these trends. This has proved a controversial issue, which is explored in several chapters in this volume. Turning to their results, Deadman and Pyle find that the majority of the explanatory variables, and especially the deterrence and economic variables, are statistically significant and have their expected signs. Deterrence seems to work, especially in the short term, in lowering crime and economic variables have both short and long term effects upon the rate of residential burglary (this is explored more fully in Chapter Four).

In Chapter Three, Derek Deadman, Stephen Pudney and David Pyle address the issue of measurement error in recorded crime. Recorded crime data are well known to be error prone (data from the British Crime Survey suggests that perhaps only one third of all crime is reported to and recorded by the police). Unfortunately, there is insufficient time-series victimisation data available to enable the estimation of a model of the type reported in Chapter Two. Deadman et al. investigate whether the use of recorded crime data can lead to erroneous conclusions being drawn about the influence of both economic and deterrence variables upon the 'true' crime rate. Their theoretical analysis indicates that the effect of measurement error in an economic model could be quite complex, particularly where one of the explanatory variables (the conviction rate) is measured by the ratio of convictions to recorded crime. As recorded crime is an underestimate of the true amount of crime, the estimated conviction rate will overstate its true value. This will tend to impart a negative bias to the effect of the conviction rate upon crime. However, the real situation is even more complex than that, because of the presence in the time-series model of lagged values of variables. As a result, it becomes virtually impossible to assess the likely impact of measurement error. Deadman et al. address this issue by incorporating extraneous information on the reporting of crime to the police, which comes from the British Crime Survey (BCS) and the General Household Survey (GHS). These are two household surveys, which include questions about respondents' experience of crime and whether incidents were reported to the police. Deadman et al. focus upon residential burglary and use Monte Carlo simulation to examine whether the parameter estimates in a time series economic model of residential burglary are affected by the presence of measurement error. They find that whilst this does lead to biases in parameter estimates, these are not particularly serious and would not lead a 'naïve' investigator to draw incorrect conclusions.

The relationship between the incidence of crime, especially property crime, and the state of the economy is something which has received a great deal of attention from economists and criminologists in the UK in recent years. Derek Deadman and David Pyle have played a leading role in that discussion and in Chapter Four they both review that debate and offer some fresh results on the short- and long-term nature of the relationship. In 'theoiy' the direction of the link between crime and economic activity is not immediately obvious, because of the conflicting effects of 'opportunity' and 'motivation'. This uncertainty may explain the often unclear conclusions that emerged from early (1970s) empirical work that examined the link between crime and unemployment. In the 1990s, British economists began to examine the link between crime and other economic variables, especially real consumption expenditure and Gross Domestic Product (GDP). Deadman and Pyle pioneered the adoption of modern time series econometric methodology in this area. This enabled them to build dynamic models of crime which separate out the short term and long-term effects upon crime of the different variables. The results of this kind of work show that there is a much stronger link between recorded crime and economic indicators than had been thought in the early 1980s. In particular there is evidence that when the economy enters a recession (manifested by lower real consumption spending and higher unemployment) then the incidence of property crime increases. In the longer term, crime increases with increases in real consumption and unemployment. In Chapter Four, Deadman and Pyle report estimates of such a model for residential burglary in England and Wales between 1950 and 1995 and find exactly these effects.

In Chapter Five, Denise Osborn offers some further analysis of the time-series relationship between recorded crime and economic factors in England and Wales. She uses quarterly data between 1975 and 1993 to model separately the series for burglary, theft and handling of stolen goods, criminal damage and personal crime (crimes of violence and sex offences). She also models the series for all property crime, i.e. the first three series listed above. The choice of quarterly data restricts the number of explanatory variables and, unlike Deadman and Pyle, Osborn does not include deterrence and demographic variables in her model. In modelling the long-term relationship between crime and economic variables, she concludes that there is stronger evidence of a relationship between crime and consumption than there is between crime and GDP and that unemployment is not cointegrated with crime in the long run. Even the addition of unemployment to the crime-consumption relationship weakens the evidence for the existence of cointegration between crime and economic variables. Therefore, she dismisses any role for unemployment in explaining long-run changes in crime in England and Wales. In fact, she is unable to find any economic explanation for the rise in personal crime in England over the 20-year period of her data. However, Osborn does find that changes in unemployment appear to have a role in explaining short-term changes in personal crime. She also concludes that there is little role for changes in consumption and unemployment having any role in determining short-term movements in property crime, although there is some evidence that unemployment has an asymmetric (i.e. ratchet) effect. A rise in unemployment leads to an increase in property crime that does not disappear when unemployment falls again.

In Chapter Six Amor Diez-Ticio applies the modelling approach suggested by Deadman and Pyle in Chapter Four to time-series data for the US. This is an interesting comparison, for there is now a great deal of evidence from both recorded crime statistics and victimisation data that crime in the US has been falling for considerably longer than it has in the UK. For example, the burglary rate in the US has been on a downward trend for almost 20 years and is now one half of what it was in 1980. Indeed, with the exception of homicide (murder) statistics, crime rates in the US (whether recorded or 'real') are often lower than in England and Wales (see Langan and Farrington, 1998). Diez-Ticio analyses annual time-series data for the period 1950 to 1996 for three separate crime groups: burglary, robbery and auto-theft. Explanatory variables are the clear up rate for the crime itself, the unemployment rate, real personal consumption per capita, the proportion of the population aged between 15 and 24 years and a measure of income inequality (the share of income received by the poorest 20 per cent of the population relative to the richest 20 per cent). She finds evidence that economic variables have a role in explaining the behaviour of US crime rates, although that role may be different across crimes and between the short-term and the long-term. For example, both unemployment and real consumption have long-term influences upon robbery rates, but only unemployment has a short-term effect. For burglary, unemployment has a long-run effect and consumption has a short-term effect. Only unemployment affects auto-theft rates, both in the short-term and the long-term. Perhaps somewhat controversially, she finds that the effect of unemployment on crime rates in the long-term is negative, which is contraiy to the findings of Deadman and Pyle for English and Welsh data. For robbery, she also finds that the inequality variable has both a short and long-term effect, although it is not significant for other crimes.

The focus of Chapter Seven, by Ingolf Dittmann, is upon punishment and in particular whether an economic model of punishment can explain why sometimes the same crime is punished by either a fine or a sentence of imprisonment. Becker's seminal article (Becker, 1968), and the subsequent literature on the economics of punishment, suggests that the use of fines is more efficient than imprisonment because fines are cheaper in social cost terms. For example, they do not lead to the loss of output which imprisoning someone implies, the capital costs of prison building and the recurrent costs of employing prison warders. Only if the size of the fine is limited by the level of an individual's wealth should imprisonment be used. Dittmann explores a number of recent attempts to reconcile the prescription of the economic model that punishment should take the form of a fine with the observed use by the courts of both fines and imprisonment (and a host of other forms of punishment) for the same type of offence. He shows that these approaches can explain the use of fines and imprisonment as alternative methods of punishing the same offence. However, only his model can explain the existence mandatory sentences of imprisonment, i.e. situations in which judges have no discretion but to impose a gaol term no matter how rich the offender. He posits a ...

Table of contents

• Cover
• Half Title
• Dedication
• Title
• Copyright
• Contents
• List of Figures
• List of Tables
• List of Contributors
• 1 Introduction
• 2 An Economic Model of Criminal Activity
• 3 Measurement Error in Economic Models of Crime
• 4 Crime, Deterrence and Economic Factors
• 5 An Investigation into Quarterly Crime and its Relationship to the Economy
• 6 The Relationship Between Economic Conditions and Property Crime: Evidence for the United States
• 7 Imprisonment versus Fines: A Theoretical Perspective
• 8 The Victims of Property Crime
• 9 The Under-Reporting of Property Crime
• 10 The Social and Economic Determinants of Illicit Drug Use
• 11 The Labour Market Consequences of Illicit Drug Use
• 12 Income Tax Evasion: A Theoretical Analysis
• 13 Income Tax Evasion: An Experimental Approach