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. 19-36

Because of the practice of displaying goods openly to attract consumers, department and specialty stores suffer a high incidence of theft, not only from customers, but from staff - some retailers put employee theft as high as 70 per cent of the total (Sydney Morning Herald 16 August 1990).

Theft prevention strategies range from electronic surveillance mechanisms, closed circuit television (CCTV), access control barriers, use of computers, staff training, increased management attention to the problem, changes in layout of the store, use of security consultants and 'honesty' shoppers to detect thieves, identification of much-stolen items with special stickers and signs, and better internal operations review and audit procedures.

Electronic article surveillance (EAS) systems

Electronic article surveillance (EAS) systems rely on a variety of electronic technologies. Several systems are available, most notably electromagnetic, low frequency and high frequency and all follow the same basic principles of operation.

1. The EAS system creates one or more protective energy field(s) at a store exit.
2. A detector or receiver filters out interference, as it scans the exit area for a particular type of signal-for example, a radio signal from a merchandise tag.
3. The system sounds an alarm if the signal is detected within the field.

An EAS system based on dual high-frequency signals provides the sharpest, state-of-the-art performance for the retailer - broad, precise coverage, low false alarms, flexible installation and an aesthetically pleasing tag.

Electronic article surveillance is now common practice world-wide for clothing

Design of appropriate tags has been a problem for EAS systems. At present, tags must be manually affixed and removed from garments or items. Ultimately, manual merchandise tagging will be eliminated, possibly by some type of 'gun' that will quickly and neatly affix tags to garments and other items. It is also possible that laser scanners will be used to deactivate tags while simultaneously totalling a shopper's purchases (Fancher 1984).

In the absence of EAS, chaining goods will cut down on theft

Case study: EAS in Wherehouse Entertainment home entertainment and informational software chain, California, United States

To combat theft of costly movies and compact disks, Wherehouse Entertainment used computer software, EAS and CCTV and redesigned their stores to facilitate surveillance by staff. As well, they set up a private investigative team to catch professional thieves and established Wherehouse University, a loss prevention training school.

Company policy was to make loss prevention as important as making sales. Listed are some of its loss prevention techniques.

• Installation of over 300 EAS systems in their ground floor video stores. The EAS concept is a pass-around system, where the tag is never deactivated and stays alive. Each store has about five EAS systems-electromagnetic, microwave and voice-activated.
• Formation of a shrinkage (inventory shortage) committee from a cross-section of the Organisation to share intelligence
• Use of CCTV with monitors where customers could see them.
• Installation of customer flow barricades.
• Elevation of counters so clerks operating the terminals could view the store with ease and eventual redesign of the stores so the whole interior is visible from one location.
• Establishment of a retail task force which set up surveillance on thieves' homes, followed them, collected data on their associates and business, set up a sting operation and notified the police. Many people were arrested and the company recovered hundreds of thousands of dollars worth of their own and competitors' products.
• Use of computers to fight shrinkage.
• Establishment of Wherehouse University, a classroom for training staff in loss prevention, with mock-ups of all company systems.
• A system of rewards for individuals who contribute to the company's objectives (Gabbard, Mongang & Leonard 1986).

As well, the 83 top managers make a loss prevention checklist every Friday and visit 97 per cent of the company's stores. Those who perform appropriately are rewarded with an overseas trip.

Integrity or 'honesty' shopping

Acting as ordinary shoppers, trained integrity or honesty shoppers can provide valuable information about sales procedures and customer service and can give retail security managers an abundance of useful information at a reasonable price. The most effective integrity shopping programs are those designed specifically to elicit the information a company needs to make corporate decisions, and where those who receive that information know how to use it properly (Cowden 1987).

Because an honesty shopper can only concentrate on one or two concerns per visit, store managers must first decide what their priorities are - employee theft, adherence to store procedures, customer service, the look of the store, or physical security? Then the integrity shopper should be given a list of details to observe. Finally, the shopper should submit a written narrative-style report or fill out a form.

Using this report, the security manager can identify the store's strengths and weaknesses and implement corrective measures. Information from the reports may be passed on to departments, but the identities of integrity shoppers should be kept secret.

A well-conceived integrity shopping program will act both as a deterrent against employee theft and as positive reinforcement for conscientious work practices.

Using computers to reduce theft

Case Study: using computers and training to reduce employee theft at the Hudson department stores, Detroit, United States

The 1984 merger of Hudson's department stores in Detroit and the Dayton Hudson Corporation in Minneapolis was followed by a major inventory shortage or shrinkage. As well, the number of apprehensions of dishonest employees by the company had dropped from 473 in 1982 to 232 in 1983.

Hudson's responded by forming a task force which identified all possible causes of the shrinkage, and distributed a survey listing these factors to executives, sales managers and employees asking for comment: 43 per cent named inadequate theft-control programs as the primary cause of the rise in shrinkage.

As most retail theft originates at the point of sale, store management used this point to identify dishonest employees. A computerised charge credit analysis of three- and 12-month periods determined the number of transactions for employee charge accounts, and identified individual purchases, the credits and what the rate of return was for employees with accounts. This helped store management isolate rings of employees with high credits. Six or seven prosecutions for dishonesty recovered about $28,000.

Another computer report analysing third-party credits by account or terminal number showed incidents of employees ringing up credits to their own or family members' charge accounts. Another tool was a terminal control report coupled with a terminal detail report. The company also instituted daily and weekly void and no-sale reports to identify repeated unusual transactions by particular clerks.

To catch more internal thieves, the company improved investigations, interviewing, and interrogation training for loss prevention investigators. After training improvements had been implemented and computerised detection methods instituted, the apprehension rate for employees rose to 476 in 1984 and again to 800 in 1985.

Setting up a task to identify all possible factors contributing to shrinkage, surveying employees, analysing the results and using computers enabled Hudson's to save $5 million in a year.

Hudson's management believes its employee theft program worked because the shortage task force tapped into the knowledge of the company's executives, managers and sales staff and responded to their most pressing concerns.

The need for individual analysis

The Selfridges study below shows that global anti-theft measures should not be adopted, but that each department, boutique or shop should be treated as a separate case for the purpose of analysis and design of prevention measures.

Case study: controlling shoplifting in Selfridges Oxford Street department store, London, United Kingdom

Selfridges is one of the largest department stores in the world, with retailing on seven floors and 380 different shops within the one store. Several of these shops have some degree of autonomy, being staffed or stocked by independent concessionaires.

The security department has responsibility for all security arrangements within the store, providing a general cover in the form of electronic surveillance, a team of store detectives and a small number of uniformed security staff. As well as theft by shoppers, their responsibilities cover dishonest staff, terrorist threats, general protection of staff, policing of delivery and storage areas and even lost property.

Selfridges' CCTV system is the largest of its kind in the UK, with about 90 cameras, eight of which monitor external areas and operate as a general theft deterrent. Electronic tagging of merchandise was rejected because of the expense, the large number of entrances and the sheer size of the crowds.

Instead of introducing heavier levels of security to the whole store, Selfridges adopted a more flexible, local approach, monitoring individual departments for losses and introducing appropriate security measures.

One approach was to adopt more traditional methods of sale, for example selling shoes in the old-fashioned manner with personal fittings. Clothing chains were introduced, with chains long enough to allow customers to remove a garment and inspect it, but not try it on without assistance. Special lightweight chains were devised for delicate garments. One chaining system was abandoned as, unless deactivated by staff, it would set off the alarm - upsetting innocent customers and threatening sales - when garments were removed.

Effectiveness: In 1986 Poyner and Webb (1987) evaluated the effectiveness of some of these local security measures using known losses from 1982, 1983 and 1985.

They found that increased management concern for security in the swimwear department - in 1983 a new buyer arrived and introduced a more rigorous regime of stock counting produced a drop in losses from almost 7 per cent to between 1 and 2 per cent. As costumes were displayed on racks with no security devices, the security department concluded that the improvement was mainly due to better staff supervision of stock.

In Reldan, a specialty store, layout was changed and staff numbers increased to improve supervision of stock. The original layout had made it difficult for one sales representative to supervise stock, as well as the fitting rooms and cash point which were some distance away in an adjacent department.

As well, a new buyer introduced better staffing arrangements and supervision of staff in 1983, and losses fell for six months. But when the department was moved and stock increased, losses began to increase again. In December 1985 the department was enlarged and moved into a more central position with staff increased and better supervised. Losses fell again, even though the department was carrying much larger quantities of merchandise.

After a rise in losses coinciding with the opening of Escada, an up-market fashion range normally displayed against one wall of the fashion floor, clothes were chained on free-standing racks and fitting room procedures were tightened. Losses were almost eliminated for three years.

Lighter security chains were eventually introduced to prevent marking of delicate fabrics. Chains do not appear to discourage customers from examining clothes: in fact, with more up-market items, security devices appear to enhance the perceived value of the goods. Experience in the dress department of Selfridges also suggests that chains protect stock even when they are not properly locked.

The Selfridges study demonstrates that conventional security measures such as closed circuit television surveillance and store detectives are not sufficient to prevent shoplifting, but that local security devices, staff supervision, management concern and the physical layout of the store can contribute to a reduction in theft. It also shows how the combination of these security measures can lead to a satisfactory compromise between open merchandising and theft reduction.

Identifying shoplifters' targets

In the following study, shoplifting was reduced by the use of specific signs and stars identifying frequently-shoplifted goods.

Case study: reducing shoplifting by the use of specific signs in Uppsala, Sweden

The object of this study (Carter et al. 1979) was to extend an earlier study (McNees et al. 1976) - showing that signs and stars identifying clothing as frequently taken by shoplifters reduced theft - to other items and see if price and size of the item affected the outcome.

The Uppsala study showed gross sales for the store for 1976 were approximately $21 million, with disappearances costing $411,600 or 1.96 per cent of sales.

Except for an exit guarded by a uniformed guard, shoppers had to pass one of 25 cash registers in the front of the store to get out. At the start of the five-week study the store employed 60 full-time cashiers. During the study 15 part-timers were employed, and by the end of the study the store employed a total of 120 people.

The initial choices for identification as frequently-stolen items were flavoured lip gloss, Elvis Presley records, leather coats and small, adjustable wrenches, which were subsequently dropped from the study and replaced by halogen lights.

Identification marks: Before the study, each of the items was marked with a small red adhesive dot (circa 1 cm in diameter). It was placed on the screw-off cap of the lip gloss and on the price tag of the other items. Cashiers were shown samples of the marked items and instructed to tick a measurement sheet whenever somebody bought one of the marked items. These sheets were returned to the supervisor at the close of the day and a new one issued next morning.

To determine the number of missing items, an observer made inventory checks in the store each morning before opening. All marked merchandise was counted and recorded. Using the cashiers' forms, the number of missing items was calculated. The number of items on the shelf became the starting point for the new day. On 30 per cent of the days a second recorder made independent counts, and on half the days three observers counted the shelf items. An observer also 'bought' coded lip gloss to see if the cashiers were marking the forms.

Warning signs: Following one week of baseline measurement, three signs (17.5 x 27.5 cm) were placed on the cosmetic displays near the entrance to two checkout lanes, and one on a rack near the cash registers. The signs read:

ATTENTION! CONSUMER!
THE ITEMS MARKED WITH A RED CIRCLE
ARE FREQUENTLY STOLEN BY SHOPLIFTERS.

Three red circles approx. 12.5 cm in diameter were taped to the display directly above and behind the target merchandise.

Prior to observation day 15, two signs and four circles were placed on clothing racks containing leather coats, and three signs and three circles near the Elvis Presley records. From observation day 15 to day 21, baseline data were recorded for halogen bulbs, and prior to opening on day 22, three signs and two circles were placed near them. The signs were up for 27 days for lip gloss, 20 for records and leather coats, and 13 for bulbs.

Effectiveness: The number of missing items in all target categories was reduced when the merchandise was specified as being frequently taken by shoplifters, while sales showed moderate to marked increases.

The number of missing lip glosses fell from 18 per cent to 9 per cent; missing Elvis Presley records fell from 9 per cent to 3 per cent; disappearances of leather coats fell from 18 per cent to zero, and the loss of halogen bulbs was reduced from 31 per cent to 10 per cent.

This reduction occurred in conjunction with the intervention and remained below baseline levels for the duration of the study.

Employing outside consultants

Some stores have found it profitable to bring in outside security experts to solve their shrinkage problems.

Case study: employing consultants to reduce shrinkage in Bermans specialty stores, United States

To deal with internal shrinkage, Bermans, a specialty retail chain with stores in 30 states, reduced stock shrinkage in two ways: consultants were hired to raise awareness of staff from selected stores about loss prevention; and investigators were sent in to apprehend staff suspected of stealing. Experienced in investigations, the consultants collected evidence on staff suspected of stealing so they could be prosecuted.

An initial awareness-raising meeting, followed by monthly visits to stores by the consultant, reduced shrinkage by 40 per cent in the first six months at a cost of only 5 per cent of the savings in shrinkage (Gabbard, Montang & Leonard 1986).

As well, undercover investigators were infiltrated into two high-shrinkage stores. This cost $4,500 but uncovered six employees stealing - all but two of the staff in one store. The saving from the highest high-shrinkage store was 33 per cent.

Crime prevention through environmental design (CPTED)

While some CPTED strategies will work in most cases, the key to successful programs is a careful analysis of the specific circumstances and the development of site-specific programs.

Case study: reducing robberies in convenience stores in Tallahassee, Florida, United States with CPTED strategies

Between 1981 and 1985, Jeffery, Hunter and Griswold (1987) tested the influence of internal and external environmental factors on the rate of robberies in 34 convenience stores in Tallahassee. Basing their analysis on the 'opportunity' theory of crime prevention-which maintains that the physical characteristics of a location either present an opportunity to potential criminals or deter them-they analysed individual crime sites and recommended physical changes.

Jeffery and his colleagues found that you cannot judge where crimes will occur just by general location: it does not follow that a particular store will suffer a high rate of robbery simply because it is in a high-crime area. Individual crime site data and analysis are needed to determine how secure a particular store will be.

Their findings were as follows.

Location of cashier: Stores with the cashier located in the centre were robbed less often than were stores with the cashier on the side.

Number of clerks: Stores with more than one clerk were less likely to be robbed than were stores with a solitary clerk.

Visibility within. Stores with unobstructed views within were less likely to be robbed.

Visibility outside. Stores with unobstructed windows and shelving that permit clear viewing of the interior of the store were less likely to be robbed.

Land use. Stores located next to commercial property were less likely to be robbed than were stores near residential property, and these in turn were less likely to be robbed than were stores near vacant lots or wooded areas.

Access to store. Stores with concealed access and escape were more likely to be robbed.

Exterior lighting. Stores with well-lit exteriors were less likely to be robbed.

Evening commercial activity. Stores near areas with evening commercial activity were less likely to be robbed.

Petrol pumps. Stores with gasoline pumps in front were less likely to be robbed than those without.

Cash handling. Stores with good cash handling procedures - and with signs letting shoppers know this - and safes clearly visible were less likely to be robbed. The authors found that '...convenience store robberies are very responsive to both internal and external physical and geographical features which involve the design of the store and the design of the external environment'(Jeffery et al. 1987, p.69). These findings were used to justify an ordinance restricting convenience store operations in Gainesville, Florida, and served as a partial basis for a later state-wide analysis of convenience store robbery (Hunter 1988).

Hunter (1990) re-examined the Tallahassee stores four and a half years later to determine what changes might have occurred. He found a general decline in robberies of 24 per cent from June 1985 through December 1989, with some store robberies decreasing by as much as 86 per cent, while other stores increased by up to 50 per cent.

Both the 1985 and 1989 data demonstrated that crime prevention through environmental design (CPTED) works, but that the influence of environmental factors varies over time.

This means that crime prevention strategies must constantly be upgraded and improved.

In 1988 Hunter and Jeffery undertook a Florida-wide survey of convenience store robberies to test the application of CPTED principles from a local area to a state-wide area. They found that the model had greater applicability at the district level than at the state level (Hunter & Jeffery 1991).

Enhancing accountability and staff awareness

In its survey of business crime prevention initiatives, the UK Home Office (Burrows 1991) found that increased shrinkage losses and recessionary pressures had caused a reassessment of operating cultures, with managerial accountability displacing the entrepreneurial sales orientation of the recent past.

Case study: educating branch staff and developing accountability of individual branch managers at British retailing chain B and Q, United Kingdom

Despite increasing profitability throughout the late eighties, B and Q, an English do-it-yourself retailing chain, found stock losses had risen to an unacceptable level by late 1991.

The company set up a top-level working party representing all the main company functions to formulate a loss-prevention strategy. The working party developed a 95-point action plan and devolved its implementation, and introduced a three-pronged program for educating branch staff and inculcating accountability in branch managers.

Storewatch: An intensive training program for managers and their two deputies was mounted. Over 1100 staff attended sessions which used sophisticated audience-response devices to guarantee participation. Management were set realisable targets for their stores and given training material to help get the message to their staff.

Management 'hit' teams were sent into the worst 25 stores in the chain. Headed by a district manager from another region, these teams of six or seven (audit, security, personnel and training) did intensive week-long investigations, then presented an action plan to the Operations Director.

Audit and security were intensified in the worst regions, reversing a company trend. A regional auditor and security manager were appointed to arrive at a program with the regional management team which targeted stores with the greatest potential payback.

The cost was modest, comprising training material and equipment, hotel accommodation for the hit squad and marginal increases in audit and security staff.

Evaluation: The overall effect of this package of measures in their first full year (February 1990 - February 1991) was to reduce B and Q's stock losses by 25 per cent. The original investment was returned twentyfold in the first year.

Internal operational reviews and audits

Many small businesses could be wasting time and money on security programs because they have mistaken the causes of their shrinkage problem (Moberly 1985). According to Moberly, the following steps are necessary to determine the cause of shrinkage:

1. Identify all operations within the firm that could conceivably affect inventory loss-for example merchandise, shipping, shoplifting, receiving, employee theft, storage, movement of merchandise and accounting.
2. Rank these operations, in descending order, according to proportion of shrinkage assumed to be attributable to each.
3. Moberly suggests taking the first list and turning it upside down to gain a new perspective and perhaps a more accurate picture of inventory flow. Occasionally examination of seemingly insignificant operations reveals practices that require immediate attention.
4. Develop an operational checklist similar to a work simplification/methods improvement process chart (a means of analysing the details of a job operation). This checklist can be used to describe and record each part (who, what, when, where, why, how and how long) of each operation in detailed sequence. Draw upon the experience and knowledge of employees here.
5. Describe the operations to be examined using work simplification/methods improvement by reducing them to some combination of the following five steps.

STEP	SYMBOL	DEFINITION
Operation	0	Doing something to the merchandise
Transportation	T	Moving the merchandise
Delay	D	Any action that causes the movement of merchandise to stop or be delayed
Storage	S	A normal part of the process that requires storing merchandise to prevent unauthorised use
Inspection	I	Any point where an inspection of the merchandise occurs

These steps can be arranged into a flow diagram, like the simplified inventory handling example below:

SYMBOL		OPERATION
s	=	Receiving and storing merchandise at warehouse
D	=	Merchandise lies idle in warehouse/storage area
T	=	Transferring merchandise to selling point/floor
O	=	Taking merchandise from storage cartons and presenting for sale
T	=	Placing merchandise on sales floor
O	=	Pricing merchandise

The flow diagram for this particular operation may appear as S D T 0 T 0. To be effective, an actual diagram would be much more detailed, noting the person, location, time and manner in which each operation was accomplished.

Besides complementing existing procedural review methods, diagrams depicting the flow of merchandise from receipt to point of sale are important tools for:

• scrutinising specific procedures within an operation;
• examining interdepartmental relationships;
• identifying inefficiencies and errors;
• comparing existing merchandise-handling practices with company policy;
• highlighting operational security weaknesses.

For example, the S D T 0 T 0 flow chart illustrated above shows that merchandise was at no stage subjected to inspection for quantity, defects, loss or damage. Further analysis would show who could exploit this vulnerability and how, where and when.

To determine where inspection points should be located, this firm could review previous loss reports and then apply the work simplification concept to the operational flow chart, asking the following questions:

• What type of illegitimate act/s could occur to the merchandise in this operation that would result in a loss?
• How could those illegitimate acts be performed?
• When?
• By whom?
• Where?

Ultimately, however, the effectiveness of the initiative depends largely on these factors: the accurate selection of the operations to be examined; the thoroughness of those examining them; and their ability to recognise points of vulnerability and the ability to secure managerial support for corrective measures.

Preventing fraudulent refunding

Fraudulent refunding involves thieves returning stolen goods to retailers and getting a full refund (Clifford 1988). It can be dealt with in a number of ways:

• enforcing a rule of 'no original receipt, no refund';
• using a phone alert system to other stores to warn them of fraudulent refunders;
• refunding on goods above a certain amount by cheque through the mail from the store's head office;
• requiring management approval on all refunds without a original sales slip;
• keeping a file of all refunds made without a sales slip so they can be reviewed for names, addresses and phone numbers.

To prevent people paying with bad cheques, some retailers code register receipts to show that the item has been paid for by cheque, and refuse to refund the money until the original cheque has been cleared.