Responses to the Problem of Bicycle Theft

Analysis of your local problem should give you a better understanding of the factors contributing to it. Once you have analyzed your local problem and established a baseline for measuring effectiveness, you should consider possible responses to address the problem.

The following response strategies provide a foundation of ideas for addressing your particular problem. These strategies are drawn from a variety of research studies and police reports. Several of these strategies may apply to your community's problem. It is critical that you tailor responses to local circumstances, and that you can justify each response based on reliable data analysis. In most cases, an effective strategy will involve implementing several different responses. Law enforcement responses alone are seldom effective in reducing or solving the problem. Do not limit yourself to considering solely what the police can do: carefully consider whether others in your community share responsibility for the problem and can help police better respond to it. (For more detailed information on shifting and sharing responsibility, see Response Guide No. 3, Shifting and Sharing Responsibility for Public Safety Problems).

What follows is intended to provide an overview of the types of intervention that have been implemented to reduce bicycle theft, the crime reduction mechanisms through which they could operate, and the lessons associated with their implementation. This should help you consider what might be appropriate in your area and help identify some of the issues associated with the implementation of such interventions. In some cases, the responses discussed may not have been subjected to rigorous evaluation, but are included to illustrate the range of tactics possible.

Locks and Locking Practices

1. Educating the public about the use of effective bicycle locks and locking practices. Of paramount importance for any strategy to reduce bicycle theft is increasing the awareness of, and providing the opportunity for, secure locking practice, whereby cyclists lock a bicycle's wheels and frame to appropriate parking furniture. It is important to ensure cyclists know what types of locks are available, which of those are recommended, and why. The security offered by different locks depends on how the lock is used, what it is locked to, and whether bicycle components are secured with quick-release fittings. It is recommended that cyclists use two locks of different types (and secure each to a wheel and the frame), as this will defend the secured bicycle against multiple theft techniques. For example, if a cyclist uses both a D lock and a chain lock, then a thief must apply both "levering" and "cutting" or "striking" to free the secured bicycle. Engaging with local cycle retailers and other relevant stakeholders can alert cyclists to which locks are best, where to get them, and how to use them effectively. In addition, the use of publicity at parking facilities may have a beneficial effect.§

§ See Response Guide No. 5, Crime Prevention Publicity Campaigns.

A number of initiatives have been implemented that include publicity campaigns as part of a range of activities designed to reduce bicycle theft. For example, in response to a large bicycle-theft problem in Leuven, Belgium, a multitactic initiative was implemented in a collaboration between the police, university, and relevant municipal authorities. One element of this initiative involved publicity designed to educate cyclists. In particular, a targeted publicity campaign titled "Lock it or lose it" involved attaching cards to inadequately locked bikes, informing cyclists of how best to secure them. Leuven authorities claimed that, compared with the previous year, there was a 19 percent reduction in recorded bicycle theft during the first six months of the scheme.63

In a more recent study, researchers analyzed the effect of (directly targeted) publicity alone on cyclists' behavior. Between March and July 2006, cyclists' locking practices were observed at five bicycle parking sites across central London.64 The observations focused on what types of locks were used, what parts of the bicycle were secured (one wheel, two wheels, frame, etc.), and what the bicycles were locked to (e.g., cycle stands, other street furniture). To evaluate cyclists' locking practices before the introduction of publicity, a typology of bad, OK, and good locking practice was devised. Cyclists' locking practices were then observed and logged at each site once a week for three weeks in March 2006.

In the first week of April, stickers that promoted secure (good) locking practices were placed in prominent positions on every bike stand at four of the sites. Placing the stickers on the stands themselves meant that cyclists would be likely to see the advice exactly at the point of securing their bicycles. Stickers were not used at the fifth site, so that any changes in locking practices where stickers were not used could be compared with those at places where they were.

The stickers used were brightly colored to attract cyclists' attention, and in addition to a short advisory tag line contained a clear illustration of good locking practice. Depending on your local area, the use of illustrations may be more effective given that the intended audience may not necessarily all be able to read the same language.§ Moreover, a carefully designed image may convey the intended message more powerfully. The sticker used is shown below.

§ For a broader discussion of language barriers and police practice, see Shah, Rahman, and Khashu (2007).

A secure-locking-advice sticker used in London (www.bikeoff.org)

Following the introduction of the stickers, researchers again observed locking practices at the five sites. The follow-up observations indicated that, compared with changes in locking practices at the control site, at those sites where stickers were used, bad locking practices decreased while good locking practices increased. While this intervention's impact on bicycle theft is unknown, the study indicates that publicity can influence cyclists' behavior, and this represents an intermediate outcome of a scheme that is inexpensive and easy to implement.

If analysis of your local problem indicates that the victims are a particular group, such as university students or schoolchildren, then any publicity campaign needs to be tailored to reach, and be applicable to, that group. In South Australia, for example, a police education program was set up in local schools and youth groups warning them of the increased risks associated with leaving bicycles unsecured. To inform adults, however, police used current Neighborhood Watch programs to raise awareness of the same risks.65 Similarly, the University of Minnesota Police Department (UMPD) used a series of online media sources to best reach their intended audience, university students.66

Parking Facilities

2. Reducing flyparking. Flyparking (e.g., locking cycles to trees or street furniture) is common and can contribute to your local crime problem simply because flyparked bicycles are generally less secure than those locked to purpose-built facilities. Reducing flyparking can be achieved by adding additional appropriate facilities (see next section), but in some cases other effort may be required.

At the University of Minnesota, flyparking was identified as an issue, and a "booting" intervention was implemented to tackle the problem. Coordinated by the UMPD, during the first two weeks of May, police and two student security monitors issue warnings to cyclists who flypark bicycles. Thereafter, they fine owners of flyparked bicycles $34. Then they "boot"—lock with a bright orange U-lock—flyparked bicycles, and instruct their owners to contact a student monitor so that they can pay a fine to have the lock removed. The UMPD suggests that bicycle theft has fallen from around 350 incidents per year before intervention to fewer than 150 per year for the two-year period afterward.67

An example of a bicycle flyparked to a tree. John Kleberg

Further benefits of patrolling the campus this way are that student monitors act as capable guardians against crime, and actively engage with potential victims whose current parking practice is increasing their risk of bicycle theft victimization. Moreover, if abandoned bikes can be identified and removed as part of a strategy of this kind, it can reduce unwanted environmental signals that may encourage offending and discourage cycle use.

3. Improving parking furniture. The design and type of bicycle-parking furniture used is important not only from the perspective of how secure it is, but also because it can influence cyclists' locking practices. For example, for cyclists with only one lock, the design of the common Sheffield stand does not encourage them to lock the frame and both wheels (as recommended). An alternative, designed with secure locking practice in mind, is the M stand. The M design removes the opportunity for cyclists to lock the crossbar to the stand, forcing the user to apply a more effective locking practice, such as securing both the wheels and frame.

A recent evaluation examined the impact of bicycle stands such as those shown above on cyclists' locking practices.68 Before intervention, researchers observed locking practices for six months at one site in central London. During this time, the only stands to which cyclists could lock their bicycles were of the traditional Sheffield design. After establishing a profile of locking practice before intervention, six new prototype stands, each designed to promote more-secure locking practices, were installed at the site. Following installation, researchers again observed locking practices. Analysis revealed that locking practices were significantly better for the new stands compared with the Sheffield stands (e.g., cyclists were more likely to lock the frame and both wheels), irrespective of the particular designs, although some designs appeared to encourage better locking practice more than others. Researchers did not measure this intervention's impact on bicycle theft, but the study clearly indicates that changes in the parking environment can influence cyclists' behavior in a way that increases the effort associated with bicycle theft. This represents an intermediate outcome of the scheme.

Sheffield stands make it difficult to lock the frame and both wheels with one lock. John Kleberg

M stand, which encourages secure locking practice (www.bikeoff.org)

The design or type of bicycle-parking furniture will be a particular issue in towns and cities where improvements in cycling facilities are under way, but is unlikely to be the responsibility of the police alone. It is therefore essential to work alongside stakeholders who have the necessary capacity to install appropriate bicycle-parking furniture, if the response to the local problem requires it.

4. Increasing guardianship. While natural surveillance may increase the visibility of bicycle theft, intervention by passersby is not guaranteed. Similarly, CCTV does not guarantee bikes' security, nor will it necessarily act as a suitable deterrent (for more information on CCTV generally, see Response Guide No. 4, Video Surveillance of Public Places).§ Informed, empowered, and motivated guardians such as security guards or other people with an ownership claim to the facility may provide effective guardianship, however.

§ In a project conducted in London, Thorpe (2007) found that on a site covered by three separate CCTV cameras, on average, thieves stole one bicycle per week. Moreover, over a six-month period, police apprehended no thieves using this footage.

For example, bike-rental facilities§§ or bicycle repair shops may be located at parking sites that would benefit from increased guardianship. Such facilities exist in several European cities (e.g., Leiden, Holland), although no published data are available to demonstrate their impact on bicycle theft. In Sint-Niklass, Belgium, a supervised bicycle shed was put up at the train station, and cyclists must subscribe to use the facility. A report suggests that over a one-year period, no one has stolen a cycle from the facility, but there is, of course, an ongoing financial cost to staff this scheme.69

§§ Mayor Richard Daley is reportedly considering implementing a bike-rental system in Chicago similar to the self-service programs found in Paris (www.citymayors.com/news/metronews_americas.html).

Registration and Recovery

5. Using traditional bicycle-registration schemes. Bicycle registration schemes could reduce cycle theft in several ways. Cycle registration would make it easier to identify stolen bikes, and to identify their rightful owners. It may also serve to deter bike thieves by making registered bikes harder to dispose of (e.g., sell).§

§ Car registration has been mandatory in most countries for some time, so a consideration of car registration's effectiveness may be instructive. In reviewing the evidence, Webb (2005) concludes that registration programs' potential effects on crime have been hampered by problems that include database inaccuracies and inadequate enforcement. It is possible that bicycle registration programs could experience similar problems without adequate consideration given to their implementation. Important to this kind of program are coverage and continuity. If records are not maintained or coverage is limited, then such programs are unlikely to have positive effects.

In Appleton, Wisconsin, a registration scheme was implemented as early as 1972. A total of 17,000 bikes were registered, and the police adopted an enforcement strategy that involved constructing a "hot" bike list and monitoring the serial numbers of bicycles parked in racks at junior high schools. They recovered 10 stolen bikes in this way and gained 15 convictions. Unfortunately, it is unclear from the published report over what time period active enforcement took place, how intense it was, and whether there was an impact on bicycle theft.70

In Portsmouth, England, a problem-oriented policing project titled "Operation Mullion" aimed to reduce, among other things, bicycle thefts at a local school.71 In conjunction with the local media and council, a bicycle-marking scheme was implemented at the school and in the surrounding area in the form of road shows. Bicycles were marked using ultraviolet pens or acid etchings, and a 24/7 telephone database was launched to enable cyclists to log details about themselves and their bikes. Though such measures were part of a package of responses, assessment indicated that reported cycle thefts at the school decreased by 39 percent in the year following the marking scheme. In addition, there was anecdotal evidence of a diffusion of benefits, whereby schoolchildren were taking the ultraviolet pens provided home and marking other property.

At Tufts University (Massachusetts), police implemented a sting operation to try to catch offenders involved in what was thought to be an organized bicycle-theft group. The operation resulted in four arrests.72 These arrests were possible only because the police could identify the stolen bicycles' rightful owners.

In Dayton, 5,000 cycles were registered in 1998. Compared with the two previous years, police returned around twice as many recovered bicycles (38%) to their owners. Similarly, in Eugene, Ore., police recovered 14 percent of stolen bikes that had been marked, compared with 5 percent of those stolen unmarked. In Cambridge, of the approximate 1,500 cycles police recover annually, they return about 300 to their owners. To increase the recovery rate, police post pictures of recovered cycles on a police website.73

These reports suggest that registration schemes may particularly help in returning recovered bicycles to their rightful owners. This can be useful for several reasons: it can reduce the number of recovered bikes that police must store (and investigate); it may reduce the cost of crime to the victims, as they will not have to replace recovered bicycles (unless they are damaged); and it can be a good public relations exercise in that the community can see that the police are doing something about the problem.

A potential shortcoming with cycle registration schemes is that they are unlikely to prevent theft from cycles, as only the bicycle's frame is typically marked. Thus, if your local problem is not theft of bicycles, then cycle registration schemes are unlikely to help. In addition, bicycle theft will be prevented only if offenders are aware of the scheme. Offenders will usually be in a hurry to steal a bike, and may consequently fail to notice bike markings that indicate the owner has registered it. The overt marking of registered bicycles is therefore important if the aim of the intervention is to prevent cycle theft in addition to aiding the recovery and return of stolen cycles.

6. Implementing an electronic tagging scheme. A more recent example of a cycle marking and registration scheme that may make such strategies simpler to implement uses Radio Frequency Identification Devices (RFID), which are also widely used in the retail sector for tracing stolen goods and deterring thieves. At Ohio State University, a scheme called Bug-a-Bike™ provides cyclists with the opportunity to have a small RFID tag securely installed in the seat post of their bicycle, or fixed to the frame.74 Striking labels are also fixed to the "bugged" bicycles to warn would-be offenders. Participating cyclists are required to submit their details to a web-based registry system linked to their unique RFID tag. This enables the bike to be registered to the owner, and if stolen, then the police can identify the bicycle using an RFID reader.

Installing an RFID on bicycle frames is an important recent development that allows the bikes to be easily scanned and compared with a "hot list" of stolen bicycles; when RFID tags are installed in the seat post, the seat must be removed before scanning, which is likely to substantially reduce the practicality of the approach.75 To date, Ohio State University's scheme has been successful in the sense that 547 cyclists have registered their cycles, recovered bicycles have been returned to their owners, and students seem to like it.§ The latter is important, as registration schemes' effectiveness will be partly determined by their uptake.§§ 

§ A similar program that was evaluated in Cambridge showed that crime did fall during the intervention period, but as the authors of this study point out, interpretation of the findings is difficult, as police arrested a prolific offender during the evaluation period, and this arrest alone could have been responsible for the reduction observed (Bullock and Tilley 2003).

§§ Sokol (1992) describes a program implemented at Georgetown University (Washington, D.C.) that provided an incentive for cyclists to register their bikes. Here, secure locks were available for loan to students on the condition that they first registered their bicycles. No evaluation of the program's ultimate outcome exists, but the example illustrates a useful way of combining three crime-prevention responses in one (encouraging bicycle registration, providing better locks, and publicizing better locking practices). 

A similar RFID scheme in Southend, England, has taken this approach one step further by implementing a stolen-market-reduction initiative.76 In this case, police provided RFID readers to local bicycle dealers who had agreed to check whether bicycles brought to them for sale or repair had been reported as stolen.77 Although a systematic assessment of this intervention's impact on bicycle theft was unavailable at the time of writing,§ it illustrates the potential for new technology to enhance existing strategies. Potential problems with such interventions are that they depend on a reasonable degree of implementation to be effective, and that they are unlikely to affect the sale of stolen bicycle parts. 

§ The authors of this work claim that, in one area, of the 2,600 bicycles tagged, only 7 have so far been stolen. However, they make no comparison with untagged cycles, so it is difficult to determine whether the intervention has had an impact beyond what one would otherwise have expected.

Offender Detection

7.Setting traps to catch bicycle thieves. In an attempt to detect bicycle thieves in the act, bicycles have also been used as "bait." In schemes implemented in Spokane, Washington; in Gloucester, England; in Wirral, England; and at the University of Toronto, police fit a bicycle with a covert tracking system and leave it (insecurely) locked at a prominent location. If the bicycle is moved, then local police officers are alerted and can track the stolen bicycle with the aim of catching the offender. This type of initiative can also aid in intelligence-gathering. For example, tracking the signal could provide insight regarding the offender's movements after the theft and potential locations of stolen goods and markets. Anecdotal evidence suggests that such interventions may increase arrest rates.78

Early feedback from practitioners involved in the Wirral scheme suggests that those caught stealing the bait bike are often prolific offenders wanted for more-serious offenses such as burglary. If this finding is generalizable, then the use of bait bikes may be a way to target prolific offenders through self-selection.79 To elaborate, it is generally acknowledged that the most committed offenders are quite versatile in their offending, and hence commit a mixture of both minor and serious crimes. The utility of the self-selection approach is that the successful detection of less-serious crimes (in this case, the theft of a bait bike) may require minimal effort (police involved in the Wirral scheme report that they make at least one arrest every day that they use their bait bike), but can bring to police attention offenders who may be involved in more-serious offenses. These may be offenders of whom the police were not previously aware. The approach is thus one way of identifying potentially serious offenders, who select themselves for police attention as a consequence of their own behavior, averting the complications of targeting particular offenders using more ethically dubious methods of (say) offender profiling.

Consultation with government agencies may be necessary before adopting a bait-bike intervention, as Belgium officials, for example, were particularly concerned with supporting what they saw as an incitement to theft.80 Implementing such a scheme also requires thought about the evidential process. For example, would finding a person who had a bait bike be sufficient to convict him or her? It may be necessary to capture an offender on CCTV to show that he or she actually stole the bicycle rather than simply finding it abandoned, in which case the bicycle's location must be in clear view of operational CCTV or reliable witnesses. Finally, such a response has to follow in-depth scanning and analysis of your local problem, as implementing a bicycle trap may be less effective if your problem is theft from bicycles, or is unlikely to be the work of prolific offenders.§

§ See Response Guide No. 6, Sting Operations.