Drivers of violent property crime in South Africa: a system dynamics model focussing on education- and income inequality

Thesis (MEng)--Stellenbosch University, 2021.ENGLISH ABSTRACT: Crime statistics released by the South African Police Service (SAPS) indicate that the total number of annually reported crimes decreased steadily between 2008 and 2019. However, annual reported robberies have been steadily increasing over the same period. This indicates that South Africa’s crime prevention strategies are effectively reducing certain types of crime in the country. Robbery prevention strategies, however, still require further development.This study therefore investigates drivers of robbery, where they form part of violent property crime, by means of a system dynamics methodology to aid the decision-making process for South African robbery prevention strategy development. This study is aligned with the aims of the 2030 National Development Plan (NDP) and conducted in the context of South African education-and income inequality. A scopingliterature review was conducted to gain perspective on mathematical models that had been developed by previous crime-related studies. The results of the review indicate a lack of literature concerned with rigorously validated system dynamics models studying robbery within the South African context.Additionally, a conceptual literature review was conductedto identifydrivers of robberyin South Africa, as well as criminological theory relating to these.The reviewed literature suggests that inequality, one of the sustaining factors of the South African culture of crime and violence, is arguably the core problem of violent crime in South Africa. Among other inequalities, income inequality, as measured by the Gini coefficient, has been shown to be especially high in South Africa. Literature investigating the causes of South Africa’s high income-inequality suggests that unequal education levels among citizens is one of the major causes of income-inequality. Researchers seeking empirical evidence for criminological theories that advocate the relationship between inequality and crime have found some support for their hypotheses, but these relationships are generally weak. However, researchers who sought evidence for this relationship without considering the specific hypotheses of these criminological theories found stronger support, even in the South African context. A dynamic hypothesis, in the form of a system dynamics simulation model was then developed from the arguments found through the conceptual review. The model focuses on relationships between education inequality, income inequality, incidents of robbery and a culture of crime and violence. These relationships are partially based on arguments of general strain theory, social learning theory, differential association theory and routine activity theory. The model was evaluated with validation tests and it was concluded that the model is appropriate for its intended purpose. Several scenarios aimed at reducing South African robbery through increased education levels, were then simulated by the model.Overall, the simulation results suggest that a reduction of robbery incidents through improvementof South African education levelswill require a long time to produce a significanteffect. However, the results do suggest that such interventions also generate long-lasting effects which consequently would require a long time to reverse. Furthermore, the results indicate that combinations of interventions generate stronger effects than the sum of effects produced by interventions applied in isolation.AFRIKAANSE OPSOMMING: Misdaadstatistieke wat deur die Suid-Afrikaanse polisiediens (SAPD) bekend gemaak is, dui aan dat die totale aantal misdade wat jaarliks aangemeld word, tussen 2008 en 2019 geleidelik afgeneem het. Die jaarliks-aangemelde rooftogte het egter gedurende dieselfde tydperk geleidelik toegeneem. Dit dui aan dat Suid-Afrika se misdaadvoorkomingstrategieë sekere soorte misdaad in die land effektief verlaag. Roofvoorkomingstrategieë vereis egter nog verdere ontwikkeling.Hierdie studie ondersoek dus met behulp van ʼn stelseldinamika-metodiek die dryfvere van roof waar hulle deel vorm van gewelddadige eiendomsmisdade. Die doel daarvan is om die besluitnemingsproses vir die ontwikkeling van Suid-Afrikaanse roofvoorkomingstrategieë te verbeter. Hierdie studie is in lyn met die doelstellings van die 2030 Nasionale Ontwikkelingsplan (NOP) en is uitgevoer in die konteks van Suid-Afrikaanse opleiding-en inkomste-ongelykheid.ʼn Verkennendeliteratuuroorsig is gedoen om perspektief te kry op wiskundige modelle wat deur vorige misdaadverwante studies ontwikkel is. Die resultate van die oorsig dui op ʼn tekort aan literatuur wat handel oor streng geverifieerde stelseldinamika-modelle wat roof binne die Suid-Afrikaanse konteks bestudeer. Daarbenewens is ʼn konseptuele literatuuroorsig gedoen om sowel dryfvere van roof in Suid-Afrika as die kriminologiese teorie hieroor te identifiseer. Die hersiene literatuur dui daarop dat ongelykheid, een van die faktore wat die Suid-Afrikaanse kultuur van misdaad en geweld laat voortduur, waarskynlik die kernprobleem van geweldsmisdaad in Suid-Afrika is. Onder ander ongelykhede blyk dit dat die inkomste-ongelykheid, gemeet aan die Gini-koëffisiënt, veral hoog isin Suid-Afrika. Literatuur wat die oorsake van Suid-Afrika se hoë-inkomste-ongelykheid ondersoek, dui daarop dat ongelyke opleidingsvlakke onder burgers een van die belangrikste oorsake van inkomste-ongelykheid is. Navorsers wat empiriese bewyse vir kriminologiese teorieë soek wat die verband tussen ongelykheid en misdaad voorstaan, het ʼn mate van steun vir hul hipoteses gevind, maar hierdie verwantskappe is oor die algemeen swak. Navorsers wat bewyse vir hierdie verwantskap gesoek het, en die spesifieke hipoteses van hierdie kriminologiese teorieë buite rekening gelaat het, het egter sterker steun gevind, selfs in die Suid-Afrikaanse konteks. ʼn Dinamiese hipotese, in die vorm van ʼn simulasiemodel vir stelseldinamika, is toe ontwikkel uit die argumente watdeur die konseptuele oorsig gevind is. Die model fokus op verhoudings tussen opleidingsongelykheid, inkomste-ongelykheid, voorvalle van roof en ʼn kultuur van misdaad en geweld. Hierdie verhoudings is gedeeltelik gebaseer op argumente van algemene spanningsteorie, sosialeleerteorie, differensiële-assosiasie-teorie en roetine-aktiwiteitsteorie. Die model is met stawingstoetse geëvalueer en die gevolgtrekking is gemaak dat die model geskik is vir sy beoogde doel. Verskeie scenario’s wat daarop gemik is om Suid-Afrikaanse roof te verlaag deur die verhoging van opleidingsvlakke is daarna deur die model gesimuleer.Oor die algemeen dui die simulasie-resultate daarop dat dit lank sal neem voor die verbetering van die Suid-Afrikaanse opleidingsvlakke ʼn beduidende effek op die afname in roofvoorvalle sal hê. Die resultate dui egter daarop dat sodanige intervensies ook langdurige effekte oplewer en gevolglik sal dit lank neem om die effekte om te keer. Verder dui die resultate aan dat kombinasies van intervensies kragtiger effekte oplewer as die som van effekte wat geproduseer word deur intervensies wat in isolasie toegepas word.Master