Seeing Is Believing: Analyzing Rising Theft Through ARCSS Data

Analyzing Rising Theft Through ARCSS Data

There are few publicly available sources of information on retail loss. For many years, the National Retail Security Survey (NRSS) has been the primary source and it has been invaluable in helping the public understand the challenges retailers face in maintaining safety and security in their stores. In recent years, the NRSS has confirmed what customers and employees have observed firsthand. Brazen thefts, dramatic displays of violence, and an unprecedented number of store closures have become frequent subjects of news articles. In response, several organizations and media outlets have attempted to downplay the severity of retail security challenges, primarily criticizing the quality and availability of retailer data.

For the past three years, the Loss Prevention Research Council (LPRC) has compiled a database of shrink, incidents, and security measures as part of its American Retail Crime, Shrink, and Security (ARCSS) initiative.

One limitation of the NRSS is its reporting of shrinkage and theft in broad terms. While we can note, for example, that shrink has increased from 1.4 percent in 2021 to 1.6 percent in 2022, we lack the ability to delve deeper into the data to examine differences between sectors or regions. Although we can state that 70 percent of retailers reported increased theft from repeat offenders, we lack precise numbers on recorded thefts or repeat offender involvement. While we do not doubt the alarming trends reported by retailers, unfortunately, there are skeptics who do.

- Sponsors -


How can retailers put their existing data into action and stop the controversy around the shrink and theft numbers? For the past three years, the Loss Prevention Research Council (LPRC) has compiled a database of shrink, incidents, and security measures as part of its American Retail Crime, Shrink, and Security (ARCSS) initiative. Our goal is to provide the industry and the public with more detailed information on retail security. We aim to answer specific questions such as “How much has shrink increased for drug stores in the past five years?” or “How many more violent incidents occurred in stores in Los Angeles compared to New York?” This data will allow the LPRC to provide concrete numbers and address skeptics.

To date, fourteen retailers representing twenty-three brands have contributed some form of data to the ARCSS initiative. Of these, thirteen have provided shrink data for their stores, nine have contributed incident counts at the store level, and two have contributed incident-level data. Most of the data in the ARCSS database covers the years 2021 to 2023, with some retailers contributing data dating back to 2009. This has enabled us to identify trends and compare recent years to historical records.

Is Theft Decreasing?

In January 2024, the LPRC utilized the ARCSS database to address the question “Has retail theft decreased post-pandemic?” A report released by the Council on Criminal Justice (CJJ) in November 2023 suggested that contrary to retailers’ claims, shoplifting had decreased from 2019 to 2023. The CJJ utilized publicly accessible data from the police departments of 24 major cities. However, there are several limitations to using police data.

The primary limitation is that not all crimes are reported to the police. In recent years, retailers have limited the number of thefts they report to the police for various reasons. One reason is nuisance abatement laws—penalizing property owners for criminal activity on their premises. Such laws reduce calls for service, making crime appear lower than it is. Another reason is the lack of response from law enforcement, as many cities struggle to hire enough police officers. With limited resources, more serious crimes take precedence over property crimes.

Despite the limitations of police data, major news outlets such as the New York Times, NPR, and CNN cited the CJJ report as evidence that claims of increased shoplifting are overblown. In response, several retail members urged the LPRC to conduct an independent analysis using data from the ARCSS database. In February 2024, the results of this analysis were presented to over one hundred retail executives at the LPRC’s IGNITE meeting in Gainesville, Florida.


Our analysis included data from ten retailers representing nineteen brands, comprising 953 stores in twenty-nine major cities. We split our analysis into two parts: the first examining year-over-year changes from 2022 to 2023 and the second examining longer trends from 2019 to 2022. This two-part design was necessary because some retailers only contributed data covering specific time spans.

At a brand level, we found that, on average, shrink increased by 24 percent from 2022 to 2023, rising from 2 to 2.4 percent. Brands also reported 1,153 additional thefts in stores within the studied cities, a 9 percent year-over-year increase. When analyzing changes from 2019 to 2022, we found that average shrink increased by 19 percent, and reported thefts grew by 39 percent. We observed a decrease in violent incidents year-over-year and from 2019-2022. However, these results are inconclusive due to limited data from only a few retailers on these types of incidents.

Our analysis has its limitations, primarily the limited dataset. We cannot make statements representing the broader industry as the NRSS can, given that far more retailers contribute to the NRSS than to the ARCSS database. It should be noted that both the NRSS and ARCSS share the initial hesitancy from retailers to get involved and share data.

Retailer Collaboration

Decades ago, during the early days of the NRSS, retailers didn’t see the benefit of participating in a survey they received through a fax machine. Those who did contribute spent considerable time and effort gaining legal and executive approval within their organization. Today, however, the NRSS has proven invaluable for benchmarking and even lobbying Congress.

The LPRC aims to build on the success of the NRSS, providing the industry with a more regular and detailed analysis of the state of retail loss prevention. This can only happen with the support and contributions of retailers. If you’d like to learn more about how you can contribute or have any questions about the project, please contact LPRC’s Director of Research Cory Lowe ( In addition to supporting loss prevention research, you can have a voice in the narrative surrounding retail crime in the United States.

James Martin

James Martin, MS, specializes in geospatial analysis and data science. He has applied statistical and machine learning approaches to advanced research in the fields of ecology, public health, and criminology. James received a BS in Public Health from the University of South Florida in 2017 and an MS in Geography from the University of Florida in 2019. Currently, James works as a research scientist at the LPRC, where he collaborates with retailers to evaluate the effectiveness of technologies and policies in reducing theft, fraud, and violence. His recent focus involves compiling retailer data and establishing data pipelines between organizations to facilitate larger-scale analyses. Contact James at

Stay Updated

Get critical information for loss prevention professionals, security and retail management delivered right to your inbox.