In the world of loss prevention, sometimes it is important to think like criminals. By understanding their thoughts and decision-making, interventions and strategies can be designed to disrupt and alter those thought patterns. This forms the basis for rational choice theory, which focuses on the rational calculation of benefits and costs associated with certain actions. However, what is and is not rational is often boiled down into two concepts: choice-structuring properties and bounded rationality.
Choice-structuring properties relate to the “internal checklist” that is required to do certain actions. In this way, everyone is capable of committing crime, but we all have different thresholds and conditions necessary before we do so. Think about stealing money from a table in your office. Some people may steal the money if there’s only $20 there and no one is watching. Others may require it to be after work to avoid suspicion. Regardless, anyone can be a criminal under certain conditions, and research can attempt to discover what those conditions are. Once those conditions are known, practitioners can then alter their presence to dissuade and discourage those behaviors.
Bounded rationality suggests that we can only make decisions based on the knowledge we have, not the total amount of information that exists. In other words, our decisions are “bounded” by the information we have at that point in time. No offender has all the information about security measures, but we can shape what someone knows to be part of that decision-making.
To better understand some of those characteristics, let’s explore two frameworks used to understand how offenders view their targets and what forms part of those “choice-structuring properties.”
What Is CRA(A)VED?
Originally conceptualized as the characteristics of “hot products” by Clarke, the CRAVED framework describes the characteristics of targets that would be most enticing for offenders. CRAVED is an acronym that suggests certain choice-structuring properties that would encourage certain types of criminal acts (usually theft): concealable, removable, available, valuable, enjoyable, and disposable. However, Pires and Clarke expanded “Available” to instead incorporate “Abundance” and “Accessible,” hence the sometimes altering of the spelling to CRA(A)VED to include both of these components.
Prior Research
CRA(A)VED has largely been explored within wildlife crime, such as through parrot poaching, illegal fishing, illegal crab and lobster fishing, and livestock theft. Beyond wildlife crime, some research has investigated the application of CRA(A)VED to sexual homicides and, most importantly to this discussion, theft of various consumer goods.
Though not as heavily investigated by researchers within a wildlife crime context, some studies have explored theft in retail spaces and the impact of CRA(A)VED characteristics on certain products.
For shoplifting, Smith investigated the theft rate of four different categories
of products within one retailer: over-the-counter drugs, personal care items, beauty and cosmetics, and general merchandise. Based on an analysis of shrinkage due to “likely theft” as designated by the retailer, Smith found that not only were there “hot products” (products that were targeted over others) within each of the categories, but that the top three theft categories were nail cosmetics, film and memory cards, and cell phone accessories. Additional analyses also showed that the two strongest predictors for theft were if the product was “concealable” and “disposable,” though products that were viewed to be “valuable,” “enjoyable,” and “available” were also important in predicting theft. However, the CRA(A)VED characteristics only explained roughly 30 percent of why the theft occurred, leading to some follow-up studies to try to investigate this further.
Smith and Clarke tried to include additional measures associated with “disposability” that they believed would help improve the predictive power of CRA(A)VED on shoplifting. Here, rather than use an indirect measure of using the profit margin as an indicator of demand, Smith and Clarke broke it down further into three other ways to measure “disposability”: looking at differences in brand name products, looking at products that needed regular replenishment (like razor blades), and products that had specific roles in illicit drug use.
However, they found that the earlier analyses, those that did not account for these additional measures, were less predictive of theft than the earlier analyses by Smith. This suggests two important findings. First, CRA(A)VED is not the silver bullet in completely predicting crime. It is a useful tool that can predict theft better than randomly guessing, but it is not perfect. Second, CRA(A)VED is only one calculation of many that offenders may do when choosing their targets. What is also likely, though not captured in CRA(A)VED,
is their perception of current security tools, such as EAS and security wrap alarms, which is something offenders will actively consider in their choices.
Why Shift to CAPTURED?
There are a few concerns when only looking at the CRA(A)VED model to understand “rational” target selection. First, CRA(A)VED only focuses on the characteristics of the product rather than its context within the store. In other words, CRA(A)VED only applies to the product’s shape, size, and value rather than also considering where it’s located within the store and what security measures are utilized with it.
Second, one of the major limitations when using CRA(A)VED in trafficking scenarios (including the movement of products within a supply chain) is that different stages of the movement of products could be rated differently by different actors. For example, if we look at an ORC operation involving “boosters” (individuals who steal products from stores to be sold) and “fencers” (the sellers of that stolen property), they may have different preferences of viable targets. “Boosters” may want to focus on concealable and disposable targets to remove them from the stores and get rid of them quickly. “Fencers,” on the other hand, may wish to have products that are more valuable, enjoyable, and available so that they can sell them at a higher profit. As such, a different approach may be necessary to understand supply chain and organized retail crime-specific issues, such as the CAPTURED model initially developed for wildlife trafficking.
As mentioned previously, offenders, under a rational choice theory perspective, will calculate the benefits and costs of certain actions. This is part of the “choice-structuring properties” described earlier, where certain conditions are necessary for a crime to be committed. However, CRA(A)VED is generally limited to focusing on the initial theft of the product rather than its movement through other criminal networks. What is valued by the initial thief may not be valued by the seller of those stolen goods. CAPTURED then expands on those ideas to understand the priorities of different criminal groups (thieves versus traffickers, for example).
The CAPTURED Framework
Similar to the CRA(A)VED model, CAPTURED is an acronym that describes the characteristics of products that are enticing to offenders.
There are some additions and alterations from the original CRA(A)VED model. First, CAPTURED tries to shift away from only focusing on the monetary worth of products from the term “valuable.” Rather, it broadly accepts that some items have priorities beyond financial elements. This is also incorporated with its inclusion of “usability” and “enjoyability” as characteristics. For instance, drugs may be valued not just for the price, but also for the effects they have on the person consuming them. Similarly, for perishable items, especially grocery items like meat and fruits, traffickers will need to consider the transportation needs of certain products over others, lending the addition of the “transferability” of that product.
Second, it highlights important elements in the movement of those products between user groups, traffickers, and storefronts (both legal and illegal, as might be the case for laundering products). For instance, looking at thefts of razor blades from stores, the “booster” (individual stealing the product from the store) may prioritize the concealability of the product while end users may instead emphasize the usability of those products. Similarly, benefit-denial mechanisms may impact those end users rather than the initial “boosters,” becoming a key concern for the reseller rather than the booster.
Third, CRA(A)VED does not necessarily account for situations in which products are reintroduced back into either illicit or legitimate markets. Because CR(A)VED generally focuses on the initial theft, other situations (such as double brokerage in the supply chain or reselling through online marketplaces) related to the initial theft are ignored. Rather, CAPTURED views each stage of the full crime as interconnected, though discrete, events.
Unlike CRA(A)VED, which primarily focuses on the actor-focused qualities of the product (in essence, the emphasis on the offender), CAPTURED also accounts for stage-focused considerations of the product (the separation of the entire crime into discrete components). In other words, CRA(A)VED is limited to only focusing on the characteristics that offenders believe are advantageous for theft; CAPTURED, in contrast, also accounts for the characteristics that are favored by those that engage in trafficking, or those that sell to the end users. This is explored a bit more in the next section.
The Theoretical Application to ORC and Cargo Theft
As alluded to in the previous section, CAPTURED is advantageous for discussing a broader, market-driven view of theft rather than solely focusing on just the characteristics of the target itself. In this way, CAPTURED attempts to showcase how narrowly focusing on just the theft of the product misses the criminal ecosystem and market dynamics in which that theft is situated, which has been an issue when discussing ORC.
To better understand this, it is important to know some of the dynamics associated with organized crime, and it is important to get a working vocabulary in the discussion of some key terms. Within an organized crime structure, there are groupings of people that have specific roles, such as those that either produce a specific product (such as those within drug rings), those that remove products from existing markets (such as “boosters” in retail crime, cargo theft within a supply chain, or poachers in wildlife crime), and those that sell those products into either illegitimate or legitimate markets (such as the “fencers” in retail crime).
Similarly, there is an element of an enterprise when describing ORC, but, unlike in a typical business, ORC can vary in its “looseness,” similar to other trafficking rings. Here, different groups may only know a few members of other groups, but, as a whole, those intermixing groups orchestrate an economically driven stealing, defrauding, and selling process. The challenge, however, is that simply removing one member in any of the groups does not dismantle the entire business; rather, they can be easily replaced with other members, and lack of knowledge of all members in the “organization” can be advantageous since it limits visibility for law enforcement and retailers as to who all the key actors in that ORC ring are.
CAPTURED, however, can highlight how each of these different “actors” prefer different characteristics of the product. We can see some examples of what this can look like depending on the specific ORC case (adapted from Moreto & Lemieux, 2015).
Conclusion
Overall, CAPTURED presents a useful framework for understanding organized retail crime by expanding upon the earlier principles of CRAVED and looking beyond the initial theft of a product. However, a few key limitations exist within the CAPTURED framework that should be discussed here. First, tests for CAPTURED have been limited to content analyses of news articles and interviews. Unlike studies of CRA(A)VED, limited studies have assessed the predictive power of each CAPTURED component on the different crime types embedded within organized crime (such as the initial theft, trafficking, illegal resale of products, etc.) In other words, it is not exactly clear which components of CAPTURED are important for each segment of the organized crime event (the theft to the trafficking to the reselling of the product) other than through this theoretical lens.
Second, studies of CAPTURED have primarily been limited to understanding wildlife trafficking (and partially to labor trafficking; see Moreto et al., 2020) rather than other forms of organized crime (such as drug trafficking and ORC). Though there are likely similarities between the two, there are some questions as to how certain components of CAPTURED translate into retail crime. For instance, poaching of certain reptiles could lead to either the removal of the live animal or the trafficking of certain by-products (meat, skin, etc.) In this sense, the latter’s value for the end user could require the target (the snake) to be processed.
However, in a retail setting, this is not as easily defined. Perhaps a processed product is one devoid of its security measures, but this leaves many products already on shelves to be considered “processed.” A processed product could instead be considered one without its original package, which could be one strategy offenders use when defeating EAS systems since the source tagging would be left on the shelf. Further discussions with practitioners and academics alike could better define what these terms mean and how to develop strategies that would assist in crime prevention.
EDITOR’S NOTE: Find more information on the sources used for this article at LPresearch.org.
Christina Burton, PhD, is a research scientist at the Loss Prevention Research Council. She received her PhD in criminal justice at the University of Central Florida in 2023 with her dissertation “Understanding the Interactions Between Conservation Agencies and Rangers: A Mixed-Methods Approach.” She primarily focuses on survey design, situational crime prevention, intelligence-led policing and crime analysis, and environmental criminology. She previously received a master’s degree in criminal justice, a graduate certificate in crime analysis from UCF, and bachelor’s degrees in anthropology and biology from the University of Florida. She has presented research at multiple conferences, including the American Society of Criminology, the Academy of Criminal Justice Sciences, the Southern Criminal Justice Association, and Ranger Rendezvous. She has published several peer-reviewed journal articles. ATHENA International has also recognized her with their Emerging Women Leadership Fellowship award for such scholarship. She also enjoys teaching and has been awarded the Outstanding Doctoral Student Educator Award within the Department of Criminal Justice at UCF.