The effects unemployment has on crime

Unemployment occurs when people are without jobs and they have actively looked for work within the past four weeks. The unemployment rate is a measure of the prevalence of unemployment and it is calculated as a percentage by dividing the number of unemployed individuals by all individuals currently in the labour force (the International Labour Organization). There never occurs full employment in the country. But there are different methods of combating unemployment. There are some countries that reached a minimum unemployment figure. A good example of this are the post World War II era when there was a huge economic growth occurring. e.g. the United Kingdom in the 1950s and 60s when the average unemployment rate was at about 1.6 % (John Sloman (2004). Economics. Penguin. p. 811), and in Australia in the 1945 government established a policy of full employment, which lasted until the 1970s when the government ran out of money. The latest statistics for UK unemployment is “The unemployment rate stands at 7.7% – down 0.1% over the quarter. 29.19 million people were in work in July to September according to the labour force survey (LFS). The number of people employed was up by 167,000 this quarter and up by nearly 300,000 from last year.” (HRM Guide, 13 October 2010)

As anyone can derive, there is a relationship between crime and unemployment. There have been a lot of studies into this topic as a low level of crime is socially desirable. Therefore, governments of all countries have tried to combat the problem of rising crime levels.

A good example of the latter is post Soviet Russia. Crime rate in the 1980’s increased to very high levels. This was the outcome of the collapse of the Soviet Union, because with it everything collapsed – the law enforcement systems, social security, and minimal standard of living. And, obviously, these are the perfect conditions for an outbreak of crime. Extreme poverty and unpaid wages, which resulted from a suffering economy, have led to property crimes, theft and counterfeiting. By the early 1990s, theft, burglary, and other property crimes accounted for nearly two-thirds of all crime in the country. There was a rapid growth of violent crime, including homicides.( Crime in the Soviet Era Federal Research Division, Library of Congress). In present times, though, Russia has done well in fighting this nightmare of crime. There are many other countries that have fought and are still combating this problem. By the use of the example of Russia, I wanted to show that unemployment is considered a very important cause of crime in any country, together with law enforcement problems and also labour market opportunities.

Economic theories of the effect of crime on unemployment

«… how many resources and how much punishment should be used to enforce different kinds of legislation? Put equivalently, although more strangely, how many offenses should be permitted and how many offenders should go unpunished.” Gary Becker (1968). Becker (1968) is considered to be the first work which formally analyses the issue of crime using an economic model. Becker links the problem of crime to social welfare. It is assumed that criminals are as rational as any other person. And, therefore, if they are involved i n crime activities, it means that they are better off by doing so.

The issue of opportunity cost comes up quite frequently in these discussions. If a potential offender, despite the knowledge that he has about the punishment he will get if he commits a crime, still goes and does it – it means that the reward that he is aiming at justifies the probability of him getting caught and convicted. When looking at crime, the concept of rationality is applied not only to criminals, but to all parties involved – judges, policemen, legislators and potential victims. So, when the institutions that control crime are designed, their main concern is not the crimes that are committed and how much they “weigh”, but the costs that occur to the society. As an example, if it takes half of the population to be turned into judges, policemen, etc in order to decrease the crime rate by 70%, it would probably not happen in the real life, because the cost of doing so is too high. Becker constructed a model, which identifies optimal levels of punishment by minimising the social cost induced by both combating crime and crimes themselves. This model predicts the aggregated supply of offences. The number of offences a criminal would commit, according to Becker, would be negatively related to the probability of apprehension and the severity of punishment. It also includes a certain “u” term, which includes all other variables that can influence the predicted outcome e.g. income from legal and illegal activities, education, risk aversion, etc.

But, this variable “u” is not explicitly studied by Becker. This work was later on continued by Ehrlich (1972, 1973). Ehrlich states that a person is able to spend time on both legal and illegal activities, but the amount of time that one dedicates to any of those activities depends on the amount of utility (return) that he gets from it. By 1973, Ehrlich designed a mathematical model that describes this relationship. Ehrlich bases his model on the “decision making under uncertainty” theory. The assumption used is that a person can switch between legal and illegal activities during their lifetime, depending, as I have mentioned earlier, on the amount of utility they get from them (activities). There is no training and no costs involved in changing between the two.

So, it is obvious, that an increase in opportunities in the legal market, e.g. higher probability of employment and higher wages would increase the expected utility gained from legal activities and so and individual, being rational, would spend more time on legal, rather than illegal deeds. Therefore, keeping a low level of unemployment, in theory will decrease the crime level.

Cohen and Felson (1979) have proposed that in order for a crime to be “done”, it need three factors to be satisfied: motivated offenders, suitable victims and the lack of effective legal punishment system. Leading from this, the increase in crime production is caused by an increase in the first two factors under the third – an ineffective legal system. This argument only supports the prediction made in Ehrlich’s model. We can deduce a positive relationship between the rate of unemployment and the crime rate.

On the other side of the argument, Cantor and Land (1985) have predicted a negative relationship between the rate of unemployment and crime based on the works of Cohen and Felson. They suggested that the higher the unemployment, the less people there are in the second category (suitable victims). If there are more unemployed, those who lost their jobs stay at home and therefore avoid the risk of having their house robbed, for example. Also, a higher rate of unemployment is a sign of a recession, which means that there is less for the offenders to steal, for example. So, higher unemployment may reduce property crimes. Also, using the same logic, we can deduce a reduction in violent crimes again through a reduction in that second factor – suitable victims. Statistically, most violent crimes are committed by strangers, so if you stay at home, you avoid that risk.

Cantor and Land have designed an empirical model which tests the motivation and opportunity effects of unemployment rate on crime. There are several equations under estimation. The dependant variables include the differentiated levels of crimes like homicide, rape, assault, robbery, theft, burglary, etc. Independent variables are either up to date or differentiated unemployment rates. The argument in this paper is that in case of a financial crisis those that become unemployed will receive some unemployment benefits, etc. And therefore they might not turn to criminal activities straight away, but after they are under financial pressure in case unemployment benefits or an alternative source of income expires.

This model has been criticized by several researchers in this area. For example, Greenberg in 2001 has argued that those who become unemployed might not have the sufficient resources to keep themselves, while out of work. This factor might make them engage in illegal activities, which contradicts the outcomes on the Cantor and Land model. He also questioned the mathematical approach of the model. Greenberg (2001) claimed that it is mathematically unacceptable if the differencing procedure is only carried out on the crime rates but not the explanatory variables.

Also, Hale and Sabbagh (1991) have questioned mainly the empirical work of Cantor and Land. The ideas of integration and cointegration were introduced to show that their models are mis-specified. Leading from this, it is argued that any conclusions drawn from their work are probably unreliable.

There are other papers that only try to derive the net effect of unemployment on crime, rather than taking into account motivation and opportunity costs. A good example is a paper written by Fleisher (1963). In it, the author argues that there is a positive correlation between parent unemployment and youth crime. If there is a high level of unemployment in the country, adults become unemployed and it becomes very hard for them to provide for their children, who, therefore, might turn to illegal activities. Fleisher uses time series data to test this relationship because it gives a more thorough view of the relationship of different variable in a long period of time. The prediction of the model is justified using an OLS estimation and shows a positive and significant correlation between unemployment rate and youth crime.

Some researchers preferred to use panel data to investigate this area. For example, a paper written by Raphael and Winter-Ebmer (2001) uses this approach. The argument in this paper is that there is a dependent relationship between crime and unemployment. One might be the cause of the other. E.g. high crime rates in the country will deter investment and so add to an increase in unemployment, which is what happens in countries like Russia. Because Russia is known for white-collar crime, it is a serious obstacle to foreign direct investment. (Forbes.com). Also, people who have had a criminal record might find it very difficult to find a job a so remain unemployed. The results of the OLS estimations of the model are the following: there is a positive relationship between unemployment and property crimes, but, on the other hand, there are insignificant results regarding the relationship between unemployment and more severe crimes like rape and murder. Although, one very interesting point of findings of this paper is the negative relationship between unemployment and murder. The explanation of this can potentially be found in Ehrlich’s three factor model. Unemployment could decrease the potential interaction between a criminal and a victim.