Following 15 years of high policy attention to curbing corruption in developing countries, this thesis concerns the effects of polices induced by the international donor community on curbing corruption in sub-Saharan Africa. I approach this question by assessing, in three stand-alone empirical chapters, the effects increased political competition, economic liberalisation, and the use of judicial punishment for corruption-related crimes have had on curbing corruption in sub-Saharan Africa. In the first empirical chapter, I assess the effect on corruption from increased political competition following the third wave of democratisation. While popular theories propose that political competition helps curb corruption by inducing political accountability, I find instead, in the sub-Saharan African context that in times of tense political competition the incumbent ensures his victory by buying the loyalty of the elite through distributing state resources for private means. This prebendal politics is, in turn, associated with higher levels of corruption. In the second empirical chapter, which concerns economic liberalisation and its effect on corruption, I ask what happens to corruption as the formal institutions governing the market change. Using insights from a case study on Rwanda, I find that corruption transforms rather than disappears in the advent of economic liberalisation. The third empirical chapter concerns the use of punishment for corruption-related crimes. By using politically contextualised information on prosecutions, I find that such anti-corruption interventions risk being used for political ends instead of curbing high level impunity. The overarching conclusion from this research endeavour is that corruption in the subSaharan African context has a political function which makes the reforms prescribed by the donor community difficult, or illogical, to fully comply with. The political functionality of corruption must therefore constitute the analytical cornerstone when developing anti-corruption policies in order to set realistic expectations and avoid unintended consequences.

This work measures whether MPs are held individually accountable for their actions through a novel analysis of the 1997 and 2010 UK general elections. Previous research suggests that MPs’ behaviour has little effect on their careers; however, developments in the media’s aggressive reporting style, the rise of personality politics and decline in traditional voting patterns indicate that this is an opportune time to examine the effect of political controversies (including scandals) on MPs’ careers. This analysis focuses on three crucial stages that form a chain of accountability: (1) exposure: the media publicises the controversy and a perception is formed; (2) internal sanction: an MP retires before an election; (3) electoral sanction: voters punish MPs at the polls. Data on MP-specific controversies between the 1992 and 1997 and the 2005 and 2010 elections was sourced from The Times, The Guardian, The Daily Telegraph and their respective Sunday editions. This work also contains an original analysis of the 2009–2010 MP expenses scandal that utilises British Election Study panel survey data to examine how information on MP malfeasance affects voters’ perceptions of MPs. The findings indicate that political controversy is linked to whether an MP retires, with those MPs from the governing party driving the result in both the 1997 and 2010 elections. Overall, voters do not hold MPs responsible for their actions at the polls. Analysis of the expenses scandal supports these general findings: constituent perceptions of their MPs’ expenses behaviour respond to public information, but do not translate into election results. Internal sanction is shown to be the most powerful form of political accountability in the chain. While identifying any individual MP accountability is novel, the overall results are in line with traditional analyses of the strength of party politics, and indicate the importance of electoral system design for accountability.

This thesis studies the markets, organizations, and people involved in the American drug trade in three sites: El Paso, Phoenix, and Chicago. The multisited analysis explores the relationships that exist within each site and provides an overview of how the drug trade differs in various contexts. This study illustrates that, contrary to the rhetoric of politicians and law enforcement actors over the past decade, there is not a vast, singular criminal conspiracy that organizes the drug trade with the intent of taking over cities in the U.S. Instead, there exists a flexible network of actors that operates largely in the margins of society. These actors participate in the network for various lengths of time and supply the ongoing demand for illicit drugs and/or the protection required for such activities in various U.S. markets. These participants include street and prison gangs, transportation organizations, and other groups and individuals, who often work as subcontractors, for employers they may not know, accepting risky tasks for what they may view as a lucrative opportunity. Moreover, this study shows that settings are important in determining how actors in the drug trade behave. This study demonstrates that when the state can effectively respond to violent acts, the criminal entrepreneurs seek to minimize publicly visible acts of violence, thus shifting acts of violence from the street to places out of public view or to prisons or jails. Ultimately, context must be taken into account if policy which seeks to reduce the harms of the drug trade in terms of unwanted usage and violence are to be effectively operationalized.

Vandalism is a problem for many communities across Scotland negatively affecting the lives of people who live in them. Whilst there has been recent research into the broad phenomena of anti-social behaviour, there has been little recent research into the specific phenomena of vandalism. In particular, little is understood about why vandalism often persistently re-occurs year in year out in particular locations. Wider research into crime and place suggests that, opportunities to commit crime, levels of relative deprivation, and the capacity or belief that a community can work together (collective efficacy) may be relevant factors. Other theories suggest some areas may act as crime attractors or be more criminogenic than others. There is a strong need for a better understanding of why certain places experience persistent vandalism and others do not. This thesis seeks to redress that gap by suggesting that to understand the nature of vandalism it is best understood as a crime of place rather than property. By drawing on theoretical and methodological approaches from both criminology and geography the thesis explores whether some areas experience high and low concentrations of vandalism year on year; if patterns change over time and whether areas with differing levels of vandalism share characteristics. Exploring issues related to crime and place presents specific methodological challenges. In criminology there has been much debate about whether it is best to consider crime and place processes at the micro or macro level. This thesis contributes to this debate by contending that it is necessary to employ a multi-method approach which integrates both micro and macro levels of investigation to properly understand crime and place. The results presented here are based on secondary analysis of six years of recorded crime data on vandalism supplied by Lothian and Borders police covering the period 1 April 2004 to 31 March 2010 for a case study area within Edinburgh with a broad mix of socio-demographic contexts. The thesis investigates the value of taking an Exploratory Spatial Data Analysis approach combining GIS based Crime Mapping and LISA (Local Indicators of Spatial Autocorrelation) analysis with Group Trajectory Analysis. This is complimented by data acquired from holding focus groups with Police Officers responsible for neighbourhood policing who used shaded maps to aid discussion of characteristics of areas with high and low vandalism. Findings suggest there are distinct High, Low and Drifting areas of vandalism with particular characteristics influenced by crime attractors, routine activities, relative deprivation and collective efficacy. By using an innovative multi-method ESDA quantitative and qualitative approach, important insights into the nature of vandalism as a place crime are gained; using a multi-spatial and temporal approach was found to be crucial. Findings are somewhat confined as they relate to a single case study area and a small number of focus groups were undertaken only with police Officers and not other community actors which may limit generalisabily. These concerns are discussed along with recommendations for future policy on vandalism and theoretical and methodological approaches for researching crime and place.

There has recently been a marked increase in the number of reported cases of corruption in international sport. As such, a growing number of sponsors have taken remedial action in order to protect the reputation of their brand. This study has been designed to analyse how sponsors respond to the threat of corruption in sport and to identify the contextual factors that influence such a response. Maennig (2005) identified two forms of corruption – ‘management corruption’ and ‘competition corruption’. It is the latter, defined by Gorse & Chadwick (2010) as “any illegal, immoral or unethical activity that attempts to deliberately distort the outcome of a sporting contest (or an element within the contest) for the personal material gain of one or more parties involved in the activity” that provides a focus and context for this study and includes such behaviour as doping, tanking, match fixing and spot fixing. In order to identify how sponsors respond both to the threat of being associated with a property affected by this type of behaviour, also referred to as sporting transgression in this study, and to analyse instances of actual corruption, a four-stage research methodology has been employed. A database of cases of sporting transgression was created to provide contextual background and further rationale for the focus of this study; a series of preliminary interviews were conducted with professionals in and around the sponsorship industry to highlight the potential implications of corruption for sponsors; and a number of case studies were developed, recognising the key stakeholders in sponsorship management. A series of in-depth semi-structured interviews were then conducted with multiple stakeholders in the sport industry - sponsors, legal professionals with expertise in sponsorship and commercial managers in governing bodies of sport. Interviews were recorded, transcribed and then analysed using grounded theory coding techniques, allowing for key themes to emerge and responses to corruption in sport discussed. Analysis of the interview data indicates that sponsors adopt a ‘wait-and-see’ approach when dealing with the potential impact of corruption, relying on a number of factors to decide upon remedial courses of action. These factors have been conceptualised and a Sponsor Response to Sporting Transgression, or SRST, Model is proposed.

Globally, the palm oil industry has been linked to practices that fit the most conventional definitions and perceptions of crime as well as the types of social and environmental harm that do not fit strictly legalistic definitions and understandings of crime. This thesis examines both the perceptions and realities of harm in the context of palm oil production in Colombia’s Pacific coast region, attending to the perspectives of corporate executives, public officials, industry representatives, small growers of oil palm, local palm oil critics, and NGOs with a critical stance towards agroindustrial palm oil production. The theoretical and analytical approach put forward to this end redirects the harm debate from a central concern with the academic contestation of harm within criminology, toward a focus on the on-the-ground contestedness of harm. The central research question that underpins the study is: “How are perceptions, practices, and realities of harm linked to palm oil production in the Colombian Pacific coast region contested, and what are the implications of this for debates on harm within green criminology?” Via a rich field-based account of the constructions, practices, and the lived and perceived realities of harm related to palm oil production, and the interrogation of the mechanisms and relations of power that thereby invest practices and discourses of harm, the study contributes empirically and theoretically to the green criminological analysis of the extractive industries, encouraging green criminology to engage with the notion of harm in more complex and nuanced ways. This approach enhances criminological understanding of the power dynamics that draw and keep in place the boundaries between legal harm, tolerated illegal harm, and non-tolerated illegal harm, and the hegemonic notions and practices of legality that thus operate to reproduce the status quo in ways that generate harm to human beings and the natural environment.

My original contribution to knowledge is the assessment of two ballistics analysis systems using the same bullets and cartridge cases to assess interoperability potential. The results are discussed in the context of policing issues representing a multidisciplinary approach to combating gun crime. Microscopic comparison of bullets and cartridge cases allows inferences to be made that objects bear marks from the same weapon. Parts of this process have been automated. Digital images of objects are stored in a database and correlations undertaken to find potential matches. An expert will decide on the most probable match based upon a range of potential candidates. All evidence should be utilised to the fullest extent, including data from ballistics systems. The success rate of the most widely used system has been quoted at between 50% and 95% suggesting that links to other crimes remain undiscovered. There are different ballistics systems available but research has only been conducted on one. There is no interoperability between systems. Data cannot be shared between different systems. An essential pre-requisite to any work on interoperability, is an understanding of the different systems and the data produced. The research aims were to design a methodology to enable the assessment of systems and to produce ammunition that can be used repeatedly as required. The aim was to conduct an experiment with two of the currently available systems. The results show variance between systems and their accuracy needs improvement. An error rate has been defined and applied to each system. The results suggest that complete interoperability of systems will only be possible with the full cooperation of the manufacturers. A limited form of interoperability focussing on data sharing may be possible. The results have implications for experts using the systems and suggest that a matching standard should be developed to make forensic ballistics analysis an objective discipline.

The vast corpus of works on corruption in Africa focuses almost exclusively on 'grand corruption' and political elites (so-called 'Big Men'), and hardly on 'everyday corruption' and ordinary actors. When everyday corruption appears in the literature, it is frequently explained away as petty and/or normal - something expected and accepted. In this study, I take issue with this predominant narrative, couched in an equally dominant but narrow Weberian notion of corruption. Grounding corruption in the micro-politics of urban public transport in Lagos, Nigeria's commercial capital and Africa's largest city, I argue that ordinary actors detest the corruption that they encounter daily. At the same time, their power(lessness) in the face of its banality compels them to constantly devise tactics to find a way around it or to make it productive for their ends. Structured into six chapters, the study begins by probing the popular imagination, discourse, and spatiality of corruption. It then shows how corruption is embedded in routine socio-economic relations, how it conditions ordinary lives and social livelihoods, and how everyday actors encounter it, exploit it, resist it, or become its victims each day. The study required eight months of ethnographic fieldwork grounded on the routine experiences and lifeworlds of road transport workers in Lagos, Nigeria. My direct experience of the 'surrounds' of these urban actors, the 'junctions' that constitute the spatial hinge of violent extortion and complicity, and routine participation in the omnipresent 'danfos' (commercial minibus-taxis) enabled access to a sense of how this complex system works.

This thesis examines the way in which corruption discourses are embedded in processes of state formation. It builds on the theoretical premise of social constructionism, namely that ‘corruption’ is not an entity that exists in reality but that it is an agreed-upon classification of certain types of behaviours. These processes of social and political construction are foundational for corruption discourse, conceptualised as a political practice through which the legitimacy of power and authority, of either persons, behaviours or institutions, can be challenged. As a socially and politically constructed entity, corruption discourse is shaped by political processes and in turn also shapes political processes. The comparison of corruption discourses in Britain in the 19th century and in Germany in the late 19th to mid-20th centuries endeavours to demonstrate the different ways in which they were shaped, as well as in turn shaped, contextual state formation processes. The two countries represent two different pathways through which high levels of corruption control were achieved, one democratic, the other authoritarian. While anti-corruption measures in Britain were introduced alongside democratisation processes in the 19th century, various German states implemented measures top-down in their 18th century efforts to modernise state administration. This study looks at the times when corruption discourses became a matter of public interest, and traces their role vis-a-vis subsequent institutional developments. In Britain this starting point is located in the early 19th century, in Germany in the Kaiserreich of the 1870s. Three case studies each exemplify and illustrate the different sequences in which corruption discourse unfolded. In Britain, these are the 1809 Duke of York case, exemplifying a ‘discovery phase’, in which corruption discourse first showed signs of becoming weaponised for political discourse; the 1830 to 32 Electoral Reform discourse exemplifying a ‘contestation phase’ in which corruption allegations were strategically used to undermine the legitimacy of Parliament and the system through which it was elected; and the 1889 Corrupt Practices Act discourse, exemplifying a ‘consolidation phase’ in which anticorruption measures became normalised rather than being subject to contest. In Germany, the 1896/97 Tausch Affair represents a different kind of discovery phase, one that is restrained and corrupted by authoritarian intervention; the Erzberger-Helfferich case of 1919 represents a different kind of contestation phase, one that is characterised by the hyper-mobilisation of corruption discourse that contributed to the eventual failure of the Weimar Republic; and the Spiegel Affair of 1962, in the context of the Spiegel’s role in post-war Germany more broadly, represents a successful consolidation phase in the Bundesrepublik, in which authoritarian intervention failed to corrupt corruption discourse. The cases thus highlight different ways in which corruption discourse was shaped by, and in turn shaped, state formation processes. They showcase a range of different institutional and political framework conditions as well as a variety of institutional outcomes, of reform, consolidation and destruction. The thesis argues that corruption discourse was thus a central driver of state formation processes, and that concepts of corruption were integral to the idea of the modern state.

Induced abortion is an understudied topic in part because of a lack of reliable data due to underreporting of abortion in surveys. The debate about whether teen pregnancies push women into socioeconomic disadvantage would benefit from longer follow-up periods and reliable data. This thesis studies which socio-demographic characteristics are associated with the likelihood of abortion, provides more precise estimates than previous studies on the socioeconomic risk factors of repeat abortion, and examines socioeconomic outcomes of women with different teen pregnancy histories. I use register data over the reproductive life span of Finnish women born in 1955–59, 1965–69 and 1975–79 (N=274,908). There is no underreporting of abortion in these data. The thesis consists of four sub-studies. The first examined the socioeconomic gradient in the risk of first abortion using event-history analysis. Low education was associated with higher risk of abortion and the gap increased over time. Selection into education and varying access to family planning services were the likely mechanisms. The second study found an educational gradient in the likelihood of repeat abortion, which has become more common over time, indicating a need of an intervention. The third sub-study examined the association between the timing of abortion and union dissolution using multi-process modelling. There were correlated unobserved characteristics associated with both unstable relationships and a higher likelihood of an abortion. Finally, I compared socioeconomic outcomes of women who had a birth, an abortion, or no pregnancy in adolescence. Results using the Karlson-Holm-Breen mediation method showed teen abortion did not mediate the relationship between parental and own socioeconomic position but teen birth did through accumulation of disadvantage. Overall the results show there is a group of women who do not benefit from contraceptive services as much as others. These findings are of importance, as reliable information on abortion is not typically available.