Table 17. Number of Detainees, 2012-2013 Type Capacity Detainees Statistics Border Guard Units 560 N/A Special Holding Facilities 837 17,458 Pre-Removal Centres 4,985 37,345 Police Stations N/A N/A TOTAL 6,867 54,803 *Data provided by the Greek Police, 28 January 2014, 24 June 2014, 1 April 2014 In addition, from 2012-2013 there was an overall increase in the rate of voluntary and forced returns of irregular migrants. Table 18. Returns (forced-voluntary) 2012-2013 Year Forced Returns Voluntary Returns Total 2012 6,798 11,526 20,473 2013 8,780 9,962 16,490 *Data provided by the Greek Police, 24 June 2014 and 1 April 2014 and IOM 2 June 2014 However, in terms of absolute numbers at least, a link between detention and return does not appear to have been established. Assessing the Cost-Effectiveness From an economic perspective, effectiveness does not require 100% enforcement. For every law, there is an “optimal amount of enforcement” which depends on a series of variables: the cost of apprehension, the cost of carrying out the punishment (e.g. imprisonment is more costly compared to fines) and the response of the offender to the enforcement (deterrent effect of conviction). In his seminal work, Gary Becker90, a famous economist, argued 40 years ago that optimal policy equilibrium requires that the costs entailed outbalance the social damages caused by the crime (e.g. loss of society's earnings). Deterrence alone is, thus, not a sufficient reason to counterbalance all other costs; and detention as such weighs heavily within the scale, 90 See G. Becker, “Crime and Punishment: An Economic Approach”, Essays in the Economics of Crime and Punishment, G. Becker and W. L. Landes (eds), 1974, UMI publ., pp. 1-54, available at http://www.nber.org/chapters/c3625.pdf Page 45