By Dr Oluwole Adeniyi and Professor Andy Newton (Nottingham Trent University), Dr Ferhat Tura and Professor John McAlaney (Bournemouth University)
This blog describes a research project made possible by a Strategic Award from the Hub’s Research Innovation Fund. The Strategic Award funds innovative and interdisciplinary research to understand and tackle gambling harms.
This project investigated the relationship between betting shops and crime in seven cities in England (Birmingham, Bristol, Liverpool, Leeds, Newcastle, Nottingham and Sheffield) and two Police Force Areas (Dorset and Surrey). We conducted spatial and multilevel analysis on data obtained from the Gambling Commission’s register of licensed gambling premises, police-recorded crime data, UK Census, point of interest data from the Ordinance Survey and access to health assets and Hazards (AHAH) from Consumer Data Research Centre across three time points (2015, 2019 and 2022). This project builds on our previous study that found a relationship between betting shops and crime at a national level in England, by exploring whether similar relationships exist at sub-national geographies.
Key Finding #1: There is a relationship between betting shops and crime at the level of cities and Police Force Areas in England. Our cluster analysis (which organises items into groups, or clusters, based on how closely associated they are) provides evidence of co-location in the patterns of betting shops and overall crime across similar neighbourhoods in our seven cities, with strong evidence in Birmingham, Sheffield, Liverpool and Bristol compared to Nottingham and Newcastle.
Furthermore, we used a multilevel model (designed for data where units like neighbourhoods are grouped or clustered within larger units like cities and PFAs) to explore the relationship between betting shops and anti-social behaviour, burglary, bike theft, drugs, public disorder, shoplifting and theft from the person, while controlling for the resident populations and year. We found a significant positive relationship between betting shops and all the crime categories across the seven cities.
In the two PFAs, Dorset showed evidence of co-location of betting shops and overall crime, but Surrey did not. Like the cities, our multi-level models found a positive association between betting shops and all crime types in both PFAs.
It is important to note that our analysis does not show that having betting shops in cities or PFAs causes crime. Rather, our focus is on the co-location of crime with betting shops, and our analysis shows that betting shops are places around which crime occurs.
Key finding #2: The relationship between betting shops and crime is impacted by neighbourhood characteristics, but there are key differences between cities and PFAs. Delving into the data in more detail, we conducted multilevel modelling that controlled for a range of neighbourhood characteristics such as education deprivation, access to healthy assets and hazards, occupation, housing tenure and point of interest data (e.g. bus stops and food and drink places).
This showed that the relationship between betting shops and crime is impacted by neighbourhood characteristics, but there are differences across the cities and PFAs. For instance, the presence of betting shops increases the likelihood of at least one type of crime in Birmingham (ASB and shoplifting), Bristol (all crimes, anti-social behaviour, public disorder and burglary), Liverpool (ASB, public disorder, shoplifting and burglary), Newcastle (shoplifting) and Nottingham (all crime, ASB and shoplifting). On the contrary, there is no relationship between betting shops and crime in Leeds, whereas, in Sheffield, a negative relationship is observed between number of betting shops and bicycle theft. Across the cities and PFAs, the crimes with the greatest association with betting shops were ASB and shoplifting.
We also identified some key correlates of crime. Factors such as higher numbers of private and social renters, food and drink shops as well as bus stops in a neighbourhood further exacerbate the relationship between betting shops and crime. Based on other academic literature, these attributes suggest high residential mobility, low collective efficacy, high population density and high footfall associated with structural and concentrated disadvantages. The concentration of these risk factors of crime together with betting shops will further increase the level of disorganisation and riskiness, which serve as catalysts for crime occurrence.
Potential impacts
This study provides a nuanced picture of the relationship between betting shops and crime across different geographies in England. The results highlight that – irrespective of the similarities in the relationship between betting shops and the different crime types – there are clear differences with cities and PFAs. This emphasises the importance of local policies to tackle the negative impacts of provisioning of gambling activities; and the importance of local councils having sufficient powers to develop tailored approaches that work in their neighbourhoods and for their communities.
About the project team: The project was led by Dr Oluwole Adeniyi, Nottingham Trent University (Nottingham Business School) working with Dr Ferhat Tura, Bournemouth University (Department of Social Sciences and Social Work), Prof Andy Newton, Nottingham Trent University (Department of Criminology and Criminal Justice) and Prof John McAlaney, Bournemouth University (Department of Psychology).