Crime prevention series

Preventing retail crime in department and specialty stores

Published in:
Preventing retail crime
Susan Geason and Paul R Wilson
Canberra : Australian Institute of Criminology, 1992
ISBN 0 642 17047 9 ; ISSN 1031-5330
(Crime prevention series) ; pp. 37-46

The major crime problems affecting supermarkets are shoplifting and the theft of shopping trolleys. In the United States, robberies of supermarkets are increasing, and demonstrations of consumers against certain products are causing problems for management. Occasionally, supermarket chains are targets of extortionists who tamper with goods or food - for example, the Tylenol case in the US and a recent Australian case where criminals contaminated toothpaste and mouthwash with arsenic.

A 1990 survey by the American Food Marketing Institute (FMI) showed an average of 62 apprehensions per store for shoplifting from the 41 supermarket companies who responded (FMI March 199 1). The average value of each item stolen by the 250,000 shoplifters was nearly US$18, and the companies recovered more than US$4.5 million worth of merchandise.

Both in the United States and Australia, shoplifters' favourite targets are cigarettes (60 per cent), health and beauty items (17 per cent) and meat products (14.3 per cent). In some states in the US, alcohol is sold in supermarkets and is popular with thieves.

In the five-year period 1985-1989, shoplifting arrests in the US increased 30 per cent, but fewer than 35 per cent of the people apprehended for shoplifting were arrested (FMI February 1991). The arrest rate has not changed over the past 10 years.

EAS in supermarkets

Electronic article surveillance (EAS) systems rely on a variety of electronic technologies. Several systems are available, most notably electromagnetic, low frequency and high frequency.

Supermarket chains began to install EAS systems in the US in the early 1980s (Ohlhausen 1987).

Typically, EAS users tag only a small percentage - 2-3 per cent - of supermarket items, such as the most stolen items mentioned above. The tags can be disguised as bar codes or otherwise hidden.

Some supermarkets install sensors at each checkout aisle. After groceries are rung up, the customer walks between two arched steel bars, and any tagged items hidden on the person will cause an alarm to sound. The cashier can then ask the customer to step back and walk through again, and suggest that the person has forgotten something he or she intended to pay for.

Other supermarket operators put EAS sensors at the store's exit. This is obviously cheaper than placing them in each aisle, and does not require each checkout to be rewired. Most importantly, cashiers do not have to be retrained and are not expected to handle shoplifter problems - a security officer at the store's exit makes sure shoppers walk through the scanner and responds to alarms.

Closed circuit television (CCTV)

Closed circuit television is in wide use as a management tool and a deterrent to crime. One of the most recent innovations in the technology is systems which trigger only when activated by alarm sensors or other signals.

Case study: use of CCTV at Tesco Stores Ltd, United Kingdom

One of England's largest retailers, Tesco, is vulnerable to crime such as attacks on staff in the course of robbery, shoplifting and internal theft. Disillusioned with the piecemeal security arrangements for their supermarkets, particularly their larger superstores, and concerned about the cost of store detectives, Tesco has begun installing TISS - the Totally Integrated Security System.

Over recent years a number of new and existing Tesco stores have been fitted with this internally-developed security package, the TISS program. TISS incorporates changes in store design and procedures, but its central component is centralised CCTV monitoring of all vulnerable areas inside and outside the store.

Notices to customers are backed up by closed circuit television in many supermarkets

Three areas are covered by TISS, and CCTV coverage and each is relayed both to a security control room and the manager's office:

1. Central cash handling areas have been modified into a suite of rooms comprising a cash room and a safe roomseparated by a sterile, airlock area. This internal 'fortress' incorporates internal alarms, CCTV cameras and monitors and hardened doors. Safes are only accessible when management and cash carriers unlock in tandem.
2. To protect against cash snatches at the cash register, as well as staff theft and collusion, check-out lanes are all separately monitored by CCTV and cash lifts from the registers are carried out by the use of an air conveyancing system to the 'fortress'.
3. Vulnerable external areas like staff entrances, rear doors and waste disposal areas are kept under surveillance.

In most cases, monitoring is not continuous, but is triggered by defined usage or activated by the operator.

The capital cost of this system is high - about £150 for an average superstore - and running costs in the form of security staff are about £15-20,000 a year.

Evaluation: The first installation of TISS at an existing store made an immediate impact. Unknown losses dropped from some £12,000 a week to £5000, recouping the capital expenditure outlay in six months. In addition, cash losses from the cash registers dropped from about £500 a week to £S20, and violent incidents in the store almost disappeared (Burrows 1991).

Performance at eight stores running complete systems indicates that unknown losses can be reduced by between 40 and 60 per cent at high loss stores, while violent incidents virtually disappear. And when professional thieves are arrested, taped evidence increases the likelihood of guilty pleas in court.

Loss of shopping trolleys

Shopping carts have a way of disappearing, especially where customers walk to the store instead of driving.

In Australia shopping trolleys cost about $200 each and most supermarkets have about $25,000 worth of them at any given time. About 5 per cent of all supermarket shopping trolleys disappear (personal communication, Australian Supermarket Institute 1991).

Shopping cart corrals: Supermarkets have dealt with trolley disappearances in a number of ways, for example many stores have installed barriers to keep carts at the customer loading zone in front of the store, though this does not stop all trolley theft.

Some markets allow customers to take carts to their cars and send employees out to scour the neighbourhood for carts people have taken home; others offer rewards for finding lost carts.

Cart corrals allow people to use carts however they want, but encourage them to return them on their own. This is how they work. A customer walks over to the cart storage area and puts a coin or a token in a device that has locked a cart to the cart in front of it. The first cart is then freed and the customer can use it in the store and the parking lot. The customer's coin is returned when he or she pushes the cart back into the cart corral.

Rewards for returned trolleys: Friction has arisen in Australia where some local councils have impounded trolleys found on streets and released them on payment of a $30 bounty or auctioned them off. The supermarkets feel this is too expensive (although they admit passing on the cost to the consumers).

A fairer system, the supermarkets believe (personal communication, Australian Supermarket Institute 1991) would be a smaller payment to council employees who alert a supermarket to the location of trolleys.

Loss through the back door

Not all items stolen from supermarkets exit by the front door: some losses occur through the back receiving area. Vendors might deliver stale goods or fewer goods than the store is paying for, and receiving clerks can collude with them (Ohlhausen 1987).

Educating receiving clerks is a common method of preventing this sort of loss. Some chains conduct a formal certification program for receivers, and others have security supervisors checking receiving as part of their regular round.

Securing cigarettes

As well as being one of the items most frequently shoplifted from supermarkets, cigarettes are of special concern to supermarket management in Australia because it is against the law to sell them to people under 18 years of age. To make sure they are within the law, Australian supermarkets now keep cigarettes behind the checkout counter where only the cashier can reach them. This also prevents theft.

Shopping bag searches

Most Australian supermarkets reserve the right to search customers' shopping bags. Searching bags is a contentious issue, with implications for individual rights and privacy, and pressure from consumer organisations has required supermarkets to agree to a code of practice and to post signs informing customers of their rights and obligations. The NSW RTA members' sign reads:

DEAR SHOPPER,
IT IS A CONDITION OF ENTRY THAT YOU WILL PRESENT ANY BAGS,
CARTONS, PARCELS AND CONTAINERS FOR INSPECTION BY OUR
STAFF BEFORE LEAVING OUR STORE.
THANK YOU

One of the admitted weaknesses of the existing bag search system is that it is impossible to tell - especially since the introduction of bar coding - whether specific items originated in the supermarket doing the search or another shop. Negotiations are under way between customers' rights organisations and supermarkets about the provision of lockers for people to store pre-purchased goods.

Robbery

Robbery is not a serious problem for Australian supermarkets yet. In the United States, where supermarkets open for long hours, sometimes around the clock, robbery is a growing problem.

In the US Food Marketing lnstitute's 1991 Security and Loss Prevention Issues Survey, responding supermarket companies reported one robbery per 10 stores in 1990. Most robbers were armed with a gun or knife and most made threats. The average loss per robbery was US$3789. A guard was on duty during about 6 per cent of the incidents. Thieves favoured Saturdays and Tuesdays.

Case study: preventing retail theft in a grocery store in Uppsala, Sweden by identifying and inventorying high-risk items

In 1985, retail theft cost an estimated $500 million in Sweden, five times the 1977 rate. Carter, Holmstrom, Simpanen and Melin (1988) carried out a project to measure the role of employees in preventing retail theft in a grocery store.

The study was conducted in an independently owned and operated grocery store in Uppsala, Sweden, population 130,000. The store grossed $7.2 million in 1985 and lost $216,000 or 3 per cent. There were 60 employees, of whom 45 were part-timers.

Existing security measures included a sign notifying customers that the store was patrolled by a plain-clothes guard and that shoplifters would be prosecuted. Such a guard did visit the store irregularly, but was not present during the intervention. Four convex ceiling mirrors had also been installed.

First the researchers identified frequently stolen products from three groups of items and twice weekly graphed losses for the separate groups.

The store used an optical price-scanning system. Most items in the store were enclosed in factory packaging containing European Article Number (EAN - A unique 13-digit code readable by scanning equipment). The cash registers were connected to a central computer that maintained data on sales, number of transactions, stock, prices and other relevant information.

The intervention included three components: an oral presentation about the study was made to employees, a list of target items was distributed to employees and posted in the lunchroom, and data on losses for each of three groups of target items - sweets, personal hygiene products, and jewellery - were graphed bi-weekly on a diagram in the lunchroom following each inventory. The target items were selected on the basis of industry statistics showing them to be frequently stolen.

After the products were identified and losses graphed, thefts from the three groups dropped from 8 per day to 2 a day.

The number of missing items for each of the three target categories was reduced in conjunction with the introduction of the intervention and remained below baseline levels for the duration of the study.

For sweets, theft rates averaged 4.7 per day before intervention and 1.2 after; for personal hygiene products, theft rates averaged 1.6 per day before and 0.8 per day after; for jewellery, the theft rate during baseline was 2 per day, whereas only I item was stolen during the three-week intervention.

Although the intervention was directed solely at employees, it is unclear whether employee or customer theft was affected. The possibilities are:

• employees were made aware that taking products was noticed and considered to be stealing;
• specific identification of target items could have increased the threat of detection;
• customer theft could have been reduced by increased employee attention to target items; and
• although only a minority of employees questioned indicated that they had changed their behaviour, subtle but effective changes may have occurred in the behaviour of several employees.

Taken with other findings, this study shows that identifying high-risk items can be more effective than identifying a suspected individual or group in preventing retail theft.