Who wants safer cities? Perceptions of public safety and attitudes to migrants among China's urban population

Abstract

Most extant research in the economics of crime literature has focused on explaining variations in crime rates. Public action to prevent crime, however, is often dependent on the level of concern about public safety that is expressed in public perceptions surveys. The economics of crime literature has largely overlooked responses to such surveys as data sources and therefore it has not accounted for the role that public opinion might have in mobilizing public action against crime. We use a unique survey administered in 2003 in 32 Chinese cities to examine the determinants of perceptions of public safety among China's urban population. One of our major findings is that individuals who have a negative perception of rural–urban migrants living in their city have a poor perception of public safety. We also find that the unemployment rate, the masculinity ratio and expenditure on armed police in the city in which the individual resides, whether the individual lives in the coastal region as opposed to the central or western region and average changes in housing prices and average changes in rental prices in the city in which the individual lives are important predictors of perceptions of public safety.

Introduction

The rate of serious crime in China showed a marked increase between the late 1980s and mid-1990s. For example, between 1988 and 1995 homicides increased 71%, assaults 171%, robbery 351%, serious theft 237% and larceny 72% (Guo, 1996, p. 4). The sharp increase in the crime rate has fueled government and public concerns of a crime wave. As one commentator put it, “the government leadership [considers] China [is] currently experiencing a period of criminal ‘high-tide”’ (Ma, 1995, p. 247). As a result of this increase in serious crime, public safety has been a major issue in China since the early 1990s. A 1991 national survey of 15,000 people in China reported that two-thirds of respondents were seriously worried about the level of public safety (Research Institute of Ministry of Public Security, 1991). Concerns about rising crime rates and falling public safety have been linked to China's migration population. It is estimated that 120–150 million peasant workers have relocated to China's cities (Pan, 2002). Migrants are considered to be a main reason for high crime rates in urban China. As Wang and Zuo (1999, p. 278) have put it: “the stereotype of rural migrants is that they are uneducated, ignorant, dirty, and also have high propensities to be criminals”.

Beginning with Becker (1968) and Ehrlich (1973), the economics of crime literature has focused most of its attention on explaining determinants of crime within a rational actor framework. Apart from Pradhan and Ravallion (2003), who examined determinants of public safety in Brazil, there has not been any research in the economics literature on concern about public safety. It is important to begin to address this apparent gap in the literature because public perceptions of safety bear substantially upon both subjective quality of life and on further objective life quality measures, such as government spending on crime prevention and victimization support. Politicians also care about perceptions because how people perceive crime rates is important for political survival for politicians and/or their policies.¹ This study adds to the literature on economic models of the determinants of public safety by drawing on a large survey of urban residents conducted in 2003 to examine the determinants of people's perceptions of public safety in Chinese cities. From a theoretical perspective there are alternatives, but ultimately equivalent ways to envisage perceptions of public safety. One approach is to model the demand for public safety, then derive perceptions of public safety from the demand for public safety function. This would be consistent with the approach in studies such as Clotfelter (1977) and Gyimah-Brempong (1989). An alternative approach, and the modelling approach we employ in this study, is to examine variations in perceptions of public safety as an extension of the ‘supply of offences’ function. This is in the spirit of studies such as Gavaria and Pages (2002) and Pradhan and Ravallion (2003). While the Ehrlich (1973) ‘supply of offences’ function was developed to explain variations in actual crime rates, it can be extended to account well for variance in perceptions of public safety. The reason is that in cities in which the economic model of crime suggests crime rates should be high, one would expect individuals to be more concerned about public safety. Because migrants are blamed for escalating crime rates in Chinese cities we also examine whether urbanite's perception of public safety is influenced by their attitudes to migrants.

The article proceeds as follows. The next section provides an overview of the dataset. Section 3 presents the empirical specification and Section 4 explains the econometric methodology. The results are set out and discussed in Section 5. The final section contains the conclusion and offers some suggestions for future research.