The Economics of Suicide in South Africa

Abstract

This study investigates the economics of suicide in South Africa using the Mortality and Causes of Death data from death notification as well as regional economic data for the 2006‐2008 period. Using an inflation rate that varies by month and across province of residence as a proxy for economic performance, the results indicate a negative relationship between inflation and suicide, suggesting that suicides are countercyclical. When controlling for month and province fixed effects, however, the inflation coefficient, albeit remaining negative, is no longer significant, except in the female sample. Suicide is more prevalent among younger individuals, while the greatest proportion of suicide is seen among men. Suicides also exhibit a strong seasonal variation, with peaks in spring and summer, with December having the highest suicide prevalence. The overall results indicate a negative but insignificant relationship between economic performance and suicide in South Africa, with socio‐economic differences and individual characteristics accounting for most of the variation in suicide.

Suicide and poverty in low-income and middle-income countries: a systematic review

“…68 However, gender differences in the relationship between SIB and poverty was not explored due to the limited evidence. Three studies 33,47,49 only explored relationships for both female and male participants. Finally, social factors relating to poverty, such as rurality and access to lethal means, may explain some of the findings.…”

“…When multivariate analyses were performed, results continued to be mixed with two positive associations with per capita real income in Turkey (long-run elasticity of suicide with respect to income -0•41) 44 and with per capita GDP adjusted for inflation in urban (beta=-0•57, SE=0•20) and also in rural China (beta=-0•68, SE=0•06). 66 One 47 null association with per capita GDP in Brazil, and two unclear associations with the inflation rate in South Africa 33 and with GDP per capita in India 40 were found.…”

“…We assessed 29 (78%) studies to be of high [30][31][32][33][34][35][36][37][38][39][40][41][42][43] or acceptable [44][45][46][47][48][49][50][51][52][53][54][55][56][57][58] quality (Table 3). Eight studies [59][60][61][62][63][64][65][66] were of low quality due to problems with risk of bias: performance, attrition and detection bias (interrupted-time series); selection bias, unclear case definitions, detection bias, lack of adjustment for confounding factors (case-control studies); detection bias, lack of adjustment for confounding factors (cross sectional and ecological studies).…”

“…However, seven studies 33,40,44,45,47,60,66 investigated the relationship between suicide and national income. Three studies using bivariate analysis had mixed results: one 33 positive with the inflation rate in South Africa, one 45 negative with Purchasing Power Parity-adjusted GDP per capita across multiple countries, and one 60 unclear association with GDP per capita in Brazil. When multivariate analyses were performed, results continued to be mixed with two positive associations with per capita real income in Turkey (long-run elasticity of suicide with respect to income -0•41) 44 and with per capita GDP adjusted for inflation in urban (beta=-0•57, SE=0•20) and also in rural China (beta=-0•68, SE=0•06).…”

Does unemployment aggravate suicide rates in South Africa? Some empirical evidence

“…Previous studies have focused on provincial or district level suicide data, thus limiting the interpretation of empirical results to specific geographical areas in South Africa. Touching on the cumbersome issue of data availability, Botha (2012) has previously mentioned that the availability of national suicide statistics has been particularly problematic for African economies hence hindering possible research endeavours on the subject matter for these countries. However, recent data released by the World Health Organization (WHO) has made it possible to gather suicide statistics on a national level for periods corresponding to the post-democratic era (i.e.…”