On-line version ISSN 2078-5135
Print version ISSN 0256-9574
SAMJ, S. Afr. med. j. vol.98 n.11 Cape Town Nov. 2008
To the Editor: I would like to counter some of the assumptions in Chris Kenyon's article in the July SAMJ.1 Discovering a relationship does not imply causality. The fact that income inequality is linked with crime does not necessarily mean that the wealth gap causes crime. A more reasonable assumption, to my mind, is that it is crime that causes the wealth gap. Crime is an activity of economic sabotage that causes the more wealthy members of society to disinvest from the economy owing to fear of theft and lost worker productivity (from murder and violence), and fears for their own persons and property (with money therefore being put into security and security firms instead of being invested in economically productive activities such as factories). Crime has the effect of reducing investment in property, people and goods - since what is the use of investing money if your property will be vandalised, your workers murdered or hospitalised, and your goods stolen? It can therefore be argued that crime is not the result but the cause of income inequality, effectively separating the rich from the poor and preventing wealth from percolating through society via economic activity. If there were no crime, the income gap would not exist as the restraints on the economy that crime induces would no longer exist, leading to far greater economic growth and far more jobs available with better pay. This logical conclusion implies that it is useless to attempt to reduce crime by poverty alleviation; rather, crime must be effectively combated before an economic environment conducive to poverty alleviation can be brought about.
Kalafong Hospital Pretoria
1. Chris Kenyon. What do the xenophobic attacks reveal about the health of South African society? S Afr Med J 2008; 98(7): 531-532.
Chris Kenyon replies: There is a simple empirical test of Tomek Piorkowski's thesis that crime causes income inequality via decreased economic activity. If this were so, there should be strong correlations between, firstly, high violent-crime levels and low economic activity, and, secondly, between low economic activity and high income inequality. I can find no evidence for the second relationship and only weak evidence for the first. A metanalysis of 34 studies found a weak relationship between poverty and violent crime but a strong relationship between income inequality and violence.1 In addition, the direction of causation was from poverty to crime (opposite to Dr Piorkowski's proposal). Furthermore, it has subsequently been demonstrated that much of the relationship between poverty and violence was itself likely to be due to undercontrolling for income inequality.2 Income inequality, on the other hand, is so tightly linked to levels of violent crime that it explains a tenfold difference in homicide rates related to inequality.2 It is remarkable how some of the richest countries in the world, such as the USA, have very high violent crime rates while some of the poorest, such as Cuba, have very low rates.3 Other rich countries such as Japan have extremely low homicide rates. The striking correlation in all these examples is how accurately the countries' GINI coefficients (level of inequality) predict the levels of violent crime. High income inequality countries (South Africa, USA) have high violent crime levels, and low income inequality countries (Japan, Cuba) have low crime levels. Longitudinal data from a study of 39 countries demonstrate that the relationship between inequality and homicide is very close, and that the direction of causality is again from inequality to homicide.4
South African Community Service doctors earn around R20 000 a month, which puts them (and hence most doctors) in the top 5% of earners in a country which is one of the most unequal in the world. There is compelling evidence that income inequality has a severe effect not only on violent crime but also on a broad range of diseases.2 If it is a doctor's duty to promote the health of the population, then it necessarily follows that we should campaign for more egalitarian social policies and therefore to earn relatively less than we currently do. It is my belief that this conclusion, though perfectly logical, is so unpalatable to our luxurious lifestyles that we would rather suspend our logic than our payments on our BMWs.
1. Hsieh CC, Pugh MD. Poverty, income inequality, and violent crime: A meta-analysis of recent aggregate data studies. Criminal Justice Review 1993; 18: 182-202. [ Links ]
2. Wilkinson R. The Impact of Inequality. New York: The New Press, 2005. [ Links ]
3. Watkins K. Human Development Report. 2007/2008. New York: Palgrave Macmillan, 2007. [ Links ]
4. Fajnzylber P, Lederman D, Loayza N. Inequality and violent crime. J Law Econom 2002; 45(1): 1-40. [ Links ]