������������������
sults available online . 2 Although the latter is an improvement with respect to information transparency , the process still requires a consumer to choose a restaurant and look it up in advance .
Other areas have chosen to make information available to the consumer at the actual decision point . For example , Toronto requires all restaurants to post their most recent inspection notice at or near the main entrance ( City of Toronto , 2012 ), 3 while the city of Los Angeles requires that inspection scores , in the form of letter grades , be posted near the entry of a restaurant , as well as in an online database ( Simon et al ., 2005 ). In England and Northern Ireland , there is a voluntary program where a score on a five point scale is posted on the window or door of a restaurant . This program is mandatory in Wales ( Food Standards Agency , n . d .). Denmark uses a system of smileys on inspection reports to indicate restaurant compliance with food safety regulations , and these reports must be posted at the restaurant and on the establishment ’ s website , a system that has proven highly popular with consumers ( Ministry of Food , Agriculture and Fisheries , 2011 ).
Previous related research in this area has focused on the economic concept of asymmetric information ( Akerlof , 1970 ), by disclosing inspection results , consumers can incorporate this information in their decision making , and restaurants have an economic incentive to comply with food safety statutes , lest they lose customers ( Chatterji & Toffel , 2010 ; Jin & Leslie , 2003 ; Simon et al ., 2005 ; Weil et al . 2006 ). By giving consumers access to inspection information , transparency in the system would be improved , which could augment consumer trust ; however , the interaction between increased trust and its effect on swings in consumer behavior between hysteria and indifference requires further attention .
Methodology
This article explores transparency
in food inspection systems using an agent-based simulation model ( ABM ). This type of modeling represents complex systems more simply by focusing on the system ’ s individuals and their behaviors ( Railsback & Grimm , 2012 ). Agent-based modeling represents a significant departure from other methods that focus on reducing a system to its component parts , or on aggregating data and looking at averages ; as such , this method is well suited to understanding the interconnected systems present in the risk society .
ABM focuses on “ describing a system from the perspective of its constituent units ” ( Bonabeau , 2002 , p . 7280 ). Agents within the system are given simple rules for decision making at the individual level and depending on these rules , agents can interact with others and react to changes in the environment ( Gilbert , 2004 ). Emergence , a key concept in agent-based modelling , is present when these local interactions give rise to interesting , and unexpected , macro-phenomena due to agents ’ adaptive behavior .
ABMs are useful for the study of situations where real experiments would endanger subjects and therefore be unethical , for example , modelling the spread of disease ( Louie & Carley , 2008 ). Furthermore , the incremental nature of building social com-
2
Saskatchewan Online Restaurant Inspection Information : http :// orii . health . gov . sk . ca / rhaReport . aspx- ? RHA = 6
2
See ( Filion & Powell , 2009 ) for a more in-depth discussion of other jurisdictions and their disclosure systems .
142