--- title: "Gypsy Cabs in Tonsberg" type: "pdf" year: "2004" canonical: "/projects/1262" --- # GYPSY CABS IN TØNSBERG A CASE FOR PROBLEM ORIENTED POLICING VESTFOLD POLICE DISTRICT, NORWAY, 2004 THE PROBLEM: A problem of gypsy cabs operating during the weekend evenings had developed in Tønsberg, a small Norwegian town of 36,000 inhabitants. The passengers and drivers of these unlicensed cabs sometimes got into fights, and there were also reports of drivers committing sexual assaults and rapes against female passengers. ANALYSIS: Police crackdowns failed to stop the gypsy cabs from operating and, as result of media pressure, the police decided to mount a problem-oriented project to deal with the problem. RESPONSE: Working with partners in the community they (1) blocked off pick up areas used by gypsy cabs, (2) made buses and taxis more accessible, (3) introduced stiffer penalties for those caught driving gypsy cabs, (4) informed suspected gypsy cab drivers about these consequences, and (5) informed the public about the risks involved and about legal alternatives. ASSESSMENT: $\quad$ These measures substantially reduced the problem. According to the local taxi drivers and the police, fewer gypsy cabs are now operating and, even if there has been some displacement of the gypsy cab operations, they are much less blatant than they formerly were. Finally, there was no evidence of a predicted deterioration in public order. # Table of Contents - [GYPSY CABS IN TØNSBERG A CASE FOR PROBLEM ORIENTED POLICING](#gypsy-cabs-in-tønsberg-a-case-for-problem-oriented-policing) - [SCANNING](#scanning) - [Analysis](#analysis) - [Response](#response) - [Assessment](#assessment) - [Judge's Commentary](#judges-commentary) - [Introduction](#introduction) - [ANALYSIS](#analysis) - [RESPONSE](#response) - [ASSESSMENT](#assessment) - [CONCLUSION](#conclusion) - [APPENDIX A](#appendix-a) ## SCANNING Four methods of data collection were utilized: 1. Direct observation. 2. Record checks. 3. Interviews with passengers and other people affiliated with gypsy cab drivers. 4. A meeting with the Tønsberg Taxi Association. In summary, gypsy cab drivers, many with criminal records, had more or less come to dominate the pick up locations that provided the best supply of passengers in the downtown area during the weekend evenings. ## Analysis Four causal factors associated with the gypsy cab operations were identified: 1. An acute shortage of transport in the weekend evenings/early mornings. 2. A simple means of earning money for the divers. 3. The inconvenient location of pick up areas for buses and licensed taxis. 4. A lack of awareness among customers of the consequences. ## Response The proposed measures were: 1. Blocking off pick up areas used by gypsy cabs. 2. Making buses and taxis more accessible. 3. Introducing stiffer penalties for those caught driving gypsy cabs. 4. Informing suspected gypsy cab drivers about these consequences. 5. Informing the public about the risks involved and about legal alternatives. Discussions with those most closely involved in the problem were initiated in October 2003 and all measures had been implemented by the end of December. ## Assessment The assessment consisted of four elements. 1. A follow-up meeting with the Tønsberg Taxi Association. 2. Repeat observations. 3. Follow-up interviews with people affiliated with gypsy cab drivers. 4. A check of recorded acts of public disorder and assaults. In summary, it was found that fewer gypsy cabs were operating less intensively and less obviously by comparison with the situation prior to the introduction of the measures. There was no evidence of a predicted deterioration in public order. In conclusion, the measures implemented contributed to making illegal means of transport less accessible and legal means more accessible, with no unintended negative effects. ## Judge's Commentary This project undertaken in Norway exemplifies several of the important characteristics of a successful problemoriented policing project. First, it was focussed on a highly specific problem that of gypsy (unlicensed) cabs operating during weekend evenings in the downtown area of a small city. The drivers of these cabs sometimes got into fights with passengers and committed sexual assaults against female passengers. The licensed taxi drivers complained that the gipsy cabs were undercutting them because they paid no taxes on their earnings. Second, the project undertook a thorough analysis of the causes of the problem, which suggested some novel ways for the police to intervene, including making it harder for gypsy cabs to pick up passengers and making it easier for passengers to use the buses and licensed taxis. Third, the project undertook a multi-faceted assessment of the results achieved, which found that the interventions had succeeded in reducing the problem. ## Introduction Society creates opportunities for crime by shaping our different daily routines. One example is the increase of assaults in the wake of weekend entertainment activities. Gypsy cab operations constitute another, recent phenomenon associated with the provision of weekend entertainment. An additional cost of this phenomenon can be the victimization of passengers. This forms the background for the Gypsy Cab POP project carried out in Tønsberg, a small Norwegian coastal town of 36,000 inhabitants. For its size the town has a vibrant night-life, particularly in the summer. Downtown Tønsberg There are 30 licensed premises restauarants, pubs and other pleaces of public entertainment - in the downtown area of Tønsberg, which is divided into two parts by highway 308. Since a large number of people pass back and forth along the highway in a more or less inebriated condition, traffic is diverted from this part of the city on weekend nights by means of a pair of barriers (see Figure 1). ![img-0.jpeg](https://popdatasets.blob.core.windows.net/popdatasetmdimgs/04-35%28F%29/img-0.jpeg) Figure 1. Downtown area of Tønsberg. The barriers are controlled by the police, who close off this section of the highway on Friday and Saturday evenings and open it up again on Saturday and Sunday mornings. Some of the area's parking lots are not available for parking at these times, but taxi cabs and buses are permitted to use this temporarily closed section of road. The gypsy cab problem Driving an unlicensed cab is illegal in Norway. Drivers who get paid for transporting passengers without permission are acting in breach of Paragraph 4 of the Occupational Transport Act. Attempting to do so is also an offense. It is not, however, an offense to make use of such transport. The gypsy cab problem in Tønsberg was confined to weekends. According to the police, the drivers were for the most part non-Norwegians from temporary residential facilities ("camps") provided for immigrant asylum-seekers. The police suspected that their activities were well organized, with a downtown kiosk functioning as a dispatch center. The definition of the problem was vague and mostly based on impressions and rumors. The police carried out crackdowns in an attempt to close down these operations and drafted in additional personnel. Four crackdowns cost the equivalent of US\$ 62,000 . These appeared to have no effect on the problem. This failure, in combination with media pressure, motivated the police to try an alternative strategy, and it was decided to solve the problem using the POP-approach. A team was organized under the direction of Police Constable Knut-Erik Søvik, with Professor Johannes Knutsson from the National Police Academy serving in an advisory capacity. The project began in the spring of 2003 with a seminar for all senior officers of the Vestfold Police District, and later on one for those of middle rank. In the course of this process, Knutsson conducted on-site visits on a total of four occasions. The scanning phase was carried out in August, analysis in September, the response phase from October to December 2003 and assessment in April and May of 2004. The equivalent of US\$ 15,000 was made available to administer the project. SCANNING In order to lay the ground for the subsequent stages of the process, the following questions were formulated: - Where were the gypsy cab operations carried out from? - How? - At what times? - By whom? - How many cars were involved? - What kind of cars? - How much money was involved? - How many crimes were committed? Four methods of data collection were employed. 1. Systematic direct observation was used to establish the number of cars used, from where, how and at what times. 2. Record checks were conducted to collect information about drivers suspected of operating gypsy cabs and their cars, as well as about crimes committed. 3. Information about gypsy cab activities were gathered by means of interviews with passengers and other persons affiliated with gypsy cab drivers. 4. A meeting with the Tønsberg Taxi Association was arranged to provide insights into the legal taxi cab drivers’ opinions of gypsy cab driving. Observations Nighttime observations were carried out on three weekends. Earlier experience had indicated the central importance of three specific locations. The observations made from concealed positions revealed that two parking lots to the northwest of the center, as well as a section of a street at the other end of the area were being used as pick up points. The gypsy cabs did not show up until after 0.30 am and their activities continued until 4.00 am, with the most intense period of activity taking place between 3.00 and 3.30. Some drivers left their cars, walked towards the center of the downtown area, and returned with customers before leaving, while others waited in their cars. Many returned on several occasions to pick up new passengers. In some of the cars, another person accompanied the driver in the front seat. The location of the suspected dispatch center was observed particularly intensely, but there were no signs that persons in the kiosk made phone calls subsequent to which cars came to pick up passengers. The observers registered all suspected cars using the following criteria: - The car was parked and the driver was waiting in the car. - The car was parked and the driver had left to get passengers. - Passengers had been recruited by a person without a car. - Passengers came from the kiosk or similar place. - The car had been observed with passengers in the car. In all, 52 separate cars were noted on 92 occasions and it seemed that the criteria were too inclusive. In order to err on the side of caution, the following more restrictive criteria were then applied for a car to be classified as a gypsy cab: it had to have been observed on at least three or more occasions on the same night, or on at least two weekends, in accordance with the criteria described above. On the basis of this definition, the number of active gypsy cabs in late August was determined to be at least 14; a figure that can be compared with the 50 or so legal cabs usually in operation. Interviews Information relating to the names of passengers was obtained from the 42 offenses against the Occupational Transport Act reported between June 2002 and September 2003. The project manager conducted phone interviews with 47 (67 percent) of the 70 persons identified in this way. The passengers had an average age of 25 years and 60 percent were males. The main findings of the interviews are presented in Table 1. Table 1. Results from interviewed passengers $(\mathrm{N}=47)$ Themes | Riding alone | $20 \%$ | | :-- | :-- | | Availability as reason for ride | $93 \%$ | | Amount paid | US\$ 15-46 | | Bad experience | $21 \%$ | Most of the passengers had ridden together with others, but one-fifth had traveled alone. For the vast majority ( 93 percent), availability constituted the decisive reason for traveling by gypsy cab because gypsy cabs were so much easier to get hold of. Most had paid US\$ 15-23, with a maximum of US\$ 46 having been paid for a single journey. One-fifth of the passengers reported a bad experience. For some, the price negotiated had increased during the journey, some hade been threatened following arguments with the driver, and some were worried about reckless driving. On one weekend night people affiliated with gypsy cab drivers were interviewed by a woman officer in civilian clothes. (It was decided not to interview active drivers because of potential legal problems).These people assisted with finding passengers, they brought drinks and refreshments to waiting drivers and sometimes rode with the drivers in the front of the cab. Contacts with these respondents were established in the downtown area. The answers can be summarized as follows: an active driver could earn more than US\$ 150 per night. Some had regular jobs and others lived on social welfare. The best time for operating was between one and five o'clock. At most a driver could complete eight tours per night. The drivers often hung around the kiosk, picking up customers. Besides being motivated by economic gain, a few wanted to meet people and to relax. For some it was partly a means of making contact with women and they sometimes took advantage of women who had drunk too much and lost control. Records check The records check was limited to the same period as above and was focused on obtaining information about: 1. Individuals engaged in illegal taxi driving and their cars. 2. Cars recorded in the observation study and their owners. 3. Serious crimes committed by gypsy cab drivers. Drivers and cars The country of origin of the 36 identified gypsy cab drivers is presented in the Table 2. Most drivers came from the Middle East, particularly from Iraq. Table 2. Country of origin of persons sanctioned for illegal transport provision in Tønsberg during the period June 2002 to September 2003. N=36. | Country | Proportion (\%) | | :-- | :--: | | Iraq | 44.4 | | Norway | 13.9 | | Lithuania | 13.9 | | (former) Yugoslavia | 8.3 | | Iran | 5.6 | | Somalia | 2.8 | | Turkey | 2.8 | | Sudan | 2.8 | | Unknown | 5.6 | | Total | 100.0 | Of the 36 persons identified, 16 (44 percent) had a prior criminal record. Most common were traffic offenses ( 36 percent), acts of violence (19 percent) and crimes against property ( 17 percent). Four drivers lacked a driver's license. The majority was living in communities near Tønsberg and only one lived in an immigrant camp. With the exception of six cars with Lithuanian plates, the cars were registered in Norway. They were on average 16 years old. ${ }^{1}$ Crimes committed during gypsy cab journeys Record checks were conducted of all complaints relating to assault, robbery and rape reported during the period referred to above. One case of aggravated assault had taken place in the center of the city following an argument over prices. Two rapes had been committed. In one case the perpetrator was a man accompanying the driver and in the other the driver himself. In both cases, the perpetrator was non-Norwegian. During the period in question a total of six rapes were reported in Tønsberg involving strangers as perpetrators. Two gypsy cab drivers, both with prior criminal records, were reported for having robbed their passengers. Meeting with the Tønsberg Taxi Cab Association About 40 taxi cab drivers took part in the meeting. The association had complained to the police on several occasions and to begin with the atmosphere was rather tense. The taxi cab drivers complained that they had to work harder to produce the same net income because gypsy cab drivers did not pay any taxes. Gypsy cab drivers had also more or less come to dominate the best pick up locations, with a better supply of passengers. This created the impression that few legal cabs were available. The theme of acts of violence committed by gypsy cab drivers gave rise to heated feelings and a number of very serious crimes were alleged to have been committed that had not been reported to the police. ## ANALYSIS Four causal factors associated with the gypsy cab operations were identified. These were: 1. A temporary shortage in the supply of means of transport. 2. A simple means of earning money. 3. The location of pick up areas for legal means of transport. 4. A lack of awareness among customers of the consequences. Shortage of supply Gypsy cab activities are concentrated in a few hours during the night at weekends and are linked to the way the entertainment sector is organized. Discos, bars, pubs and restaurants stop serving alcohol at 3.00 and close at 3.30 am . Within a short period of time, a huge crowd leaves for home. The first factor is thus a temporary and acute shortage of supply in the means of transport. A simple way of earning money This situation creates readily available opportunities to make easy money. The cars employed are for the most part old and cheap and the expenses involved relate primarily to gas. Gross earnings may amount to approximately US\$ 150 for a couple of hours work and around US\$ 1,200 a month for those who drive every night on the weekends. During the summer and the Christmas holiday season, opportunities for profit are greater. Since gypsy cab drivers do not pay tax on their earnings, the net income, considering the number of hours actually worked, is fairly high. According to the information obtained from customers, the prices varied from US\$ 15 to 23 per trip; prices that hardly differed from those charged by legal taxis. Thus gypsy cab drivers do not compete on price, which leads us on to factor three. Location of pick up areas For traffic safety reasons, the stand used by the night buses had been located at some distance from the downtown area in place that was difficult to find. This resulted in rather few passengers using the night buses. More important, however, was the location of the taxi stand, which lay at a considerable distance from the downtown area where the gypsy cabs operated (see Figure 2). ![img-1.jpeg](https://popdatasets.blob.core.windows.net/popdatasetmdimgs/04-35%28F%29/img-1.jpeg) Figure 2. Downtown Tønsberg prior to the introduction of the measures. Clearly, the gypsy cabs were much more accessible than the legal means of transportation. The customers It is not an offense to use illegal taxi services. That it might be dangerous to do so is probably not something that most passengers think about, but they are in fact placing themselves at considerable risk of victimization. This is particularly true for women by themselves who may be raped. By using gypsy cab services, the passengers are also supporting an illegal activity, which may in the long run undermine the legal alternatives. The fourth factor is thus a lack of awareness as to the consequences of using gypsy cabs. ## RESPONSE The measures introduced to solve the problem were chosen to affect the contributory causal factors identified, namely the imbalance in the availability and supply of legal means of transportation by comparison with the illegal alternative, and low thresholds in relation to both the supply and utilization of gypsy cab services. The proposed measures were: 1. Blocking off the pick up areas used by gypsy cabs. 2. Making buses and taxis more accessible. 3. Introducing stiffer penalties for those caught driving gypsy cabs. 4. Informing suspected gypsy cab drivers about these consequences. 5. Informing the public about the risks involved and about legal alternatives. Discussions were initiated in October 2003 with the following partners: - Tønsberg City Council - The National Road Authority - The Tønsberg Engineers' Service - The Tønsberg Taxi Association - Norway Bus and Nettbus - Tønsberg Car Parking and Europark - The Tønsberg Social Welfare and Immigrant Offices. - Vestfold County Council Blocking off pick up areas for gypsy cabs A longer section of the highway had to be closed in order to block access to two of the parking lots and the street used by the gypsy cab drivers to pick up passengers. It was also proposed that two parking lots located outside of this area - one already in use and the other a potential location to which gypsy cab drivers could relocate when their former pick up locations were put out of commission - ought also to be closed off during the critical hours. This last location was identified as a possible displacement area. Both the City Council and the Road Authority responded favourably to the proposal involving the extension of the closed-off section of road. The National Road Authority provided the equivalent of US\$ 8,000 to help pay for this work, and the Tønsberg Engineers' Service contributed by providing manpower and materials equivalent to the sum of US\$ 15,000 . The companies responsible for the parking lots, Tønsberg Car Parking and Europark, agreed to make the lots in question inaccessible during the critical hours. If only these measures were put into effect, however, they might be expected to create a worsened situation involving prolonged waiting periods. Furthermore, if more people were to stay longer in the downtown area waiting to get home, this could give rise to an increase in violence and disorder. It was therefore important to make legal means of transportation more accessible. Making buses and taxis more accessible Moving the bus stand to a more central location and providing a new taxi stand closer to the downtown area, would make buses and cabs more easily accessible. Cooperation between the different cab companies could also bring about a more efficient use of their collective passenger capacity, shortening queuing times for customers. It was decided to establish a new taxi stand at Tønsberg square. In order to utilize the total passenger capacity more efficiently, this stand was made available for use not only by local licensed cabs but also by others. The bus stand was relocated to the closed section of the highway near the kiosk, which removed concerns about traffic safety (see Figure 3). ![img-2.jpeg](https://popdatasets.blob.core.windows.net/popdatasetmdimgs/04-35%28F%29/img-2.jpeg) Figure 3. Downtown Tønsberg following implementation of the measures. Following proposals made by the police, the bus companies also started to provide a night bus service on Friday nights. The Vestfold police force and the bus companies requested that the County Council subsidize this service. The county decided to provide compensation for possible losses up to a sum of US\$ 40,000 for the period from December 2003 until the end of April 2004. Stiffer consequences for persons caught driving gypsy cabs The Occupational Transport Act makes it possible for the police to award immediate and tougher sanctions than those previously employed. Besides the usual fine (equivalent to US\$ 1,500), use of the car could be prohibited for three months and it was decided to put this particular regulation into use. In the past, other authorities had not been informed of the gypsy cab drivers' offences, which might result in further consequences, particularly for the nonNorwegian drivers. It was agreed that information about persons sanctioned for providing illegal taxi services should be passed on to the relevant authorities for further action. Informing suspected gypsy cab drivers about consequences A decision was taken to produce informational leaflets; one set for suspected gypsy cab drivers and another for actual and potential customers. In the leaflets to be handed to suspected gypsy cab drivers who were stopped by police officers, the possible consequences of the offence are explained. The leaflet was printed in Norwegian, Arabic, Kurdish and Albanian. Informing the public about risks and the legal alternatives The leaflet for passengers in suspected gypsy cabs stopped by the police contained information about the risks associated with travelling in gypsy cabs. It also contained details of the night bus service and phone numbers for legal taxis. For potential passengers, two types of material were created - posters and small cards. The posters were placed in downtown locations on Friday afternoons and then removed on Monday mornings. These posters presented information about the risks of riding in gypsy cabs and about legal alternatives. The cards, which contained information about the bus and taxi services, were to be handed out by police officers to members of the public in response to queries about how to get home. The leaflets for customers were financed jointly by the bus companies, the Tønsberg Taxi Association and the police. All measures were implemented by December 2003. ## ASSESSMENT The assessment consisted of four elements. 1. A follow-up meeting with the Tønsberg Taxi Association. 2. . Follow-up interviews with people affiliated with gypsy cab drivers 3. . Repeat observations 4. A check of recorded acts of public disorder and assaults. The meeting with the Tønsberg Taxi Association Approximately 40 cab drivers and owners participated in a meeting to discuss the new situation. The vast majority claimed that the problem had improved considerably. A few had seen suspect activities at a location outside the closedoff zone, but agreed that it was hard to know whether it was gypsy cabs or waiting parents/friends. Interviews with persons affiliated to gypsy cab drivers There were simply no gypsy cab driver's affiliates to be found. Interviews with the staff of hotels and other establishments confirmed the changed situation. Even if some drivers still were active, the activity was not open and blatant as it was before. Observations The observations were carried out over the course of two weekends. The aim of the observations in the first weekend was to detect possible displacement areas and new locations being used by gypsy cabs. It was found that a few gypsy cabs were still around, although their activities were much less intense and obvious than before. These activities were noted at three locations; two at the northern end of the closed-off area and one at the other end (see figure 4). ![img-3.jpeg](https://popdatasets.blob.core.windows.net/popdatasetmdimgs/04-35%28F%29/img-3.jpeg) Figure 4. Observed locations. The observations in the second weekend were restricted to these three locations and to the most intense period of activity - i.e. 1.30 to 4.30 am . These observations are not directly comparable with the preintervention observations because only two of the three locations were observed in both periods (the third was now blocked off and nothing was happening there) and the time span of observations was more restricted in the follow-up period. ${ }^{3}$ Unfortunately, it was not possible to separate the location data so that a comparison could be made for just the two locations observed in both periods. However, it was possible to separate data by times of observation. In Table 3 below (see Appendix A), one weekend from the pre-intervention August observations was therefore selected and the same time interval (i.e. 1.30 to 4.30 am ) was used as in the post-intervention observation. In view of the positive feedback from the Taxi Association and the inability to find affiliates to interview, the findings in Table 3 are surprising, with more gypsy cabs apparently being observed during the post- intervention Saturday-Sunday observation period. There are a number of possible explanations for this apparent anomaly: 1. It could be the result of using police officers, untrained and inexperienced in research, to make the observations. Some of the observed cars may well not have been gypsy cabs, but legitimate persons fetching children or friends. One reason for thinking that this might be the case is the much lower rate of observed journeys per car on SaturdaySunday. 2. As noted the locations observed were not identical in the two periods being compared, and it is quite possible that the third area observed in the follow-up period was a "displacement" area. 3. It could be, as claimed by one of the observing police officers, that "summer birds" from Lithuania (i.e. seasonal visitors) were trying to establish gypsy cab operations at the third, newly-observed location. There seems to be some validity to all these explanations, but it is impossible to determine the weight to place on each of them. The public order situation One possible unintended negative effect of the intervention might have involved an increase in public disorder and violence. As can be seen from Figure 5 (see Appendix A), there are substantial fluctuations in levels of both disorderly conduct and assaults, but it is difficult to discern any obvious trend. To check whether there has been an increase following the implementation of the measures, significance tests have been conducted for the periods January to April for the two proceeding years, as compared to 2004 (see Table 4, Appendix A). Even if some differences may be noted, these are not significant, indicating that the public order situation has not in fact changed. ## CONCLUSION Despite some anomalies in the data, specifically in the observation data, there are good reasons to believe that the measures introduced to deal with the gypsy cab problem have been effective. Gypsy cabs have been excluded from the prime pick-up locations in the downtown area and access to legal cabs and buses has been greatly improved. The local taxi cab drivers and the police believe that far fewer gypsy cabs are operating. Even if there has been some displacement of the gypsy cab operations, these are now much less blatant than they formerly were. There are no indications of a feared deterioration in the public order situation. The implementation of the problem-oriented approach has thus successfully helped the local police to reduce the gypsy cab problem in Tønsberg. Project contact person: Johannes Knutsson Position: Professor Address: National Police Academy, Pb 5027 Majorstua City/State: 0301 Oslo, Norway Phone: + 4799491255 Fax: +47 23199901 E-mail: johannes.knutsson@phs.no # APPENDIX A Table 3. Observations of three pick-up areas, 1.30-4.30 am. One weekend in August 2003 (pre-intervention) and in one May 2004 (post-intervention). | | Friday - Saturday | | | Saturday-Sunday | | | | :-- | :--: | :--: | :--: | :--: | :--: | :--: | | | cars | Journeys | journeys/car | cars | journeys | journeys/car | | August 2003 | 18 | 28 | 1.6 | 6 | 12 | 2.0 | | May 2004 | 8 | 12 | 1.5 | 16 | 18 | 1.1 | Figure 5. Monthly number of assaults and cases of disorderly conduct reported in the downtown area of Tønsberg, January 2002 to April 2004. ![img-4.jpeg](https://popdatasets.blob.core.windows.net/popdatasetmdimgs/04-35%28F%29/img-4.jpeg) Table 4. Average number of assaults and cases of disorderly conduct in downtown area of Tønsberg, January to April 2002, 2003 and 2004. | | 2002 | 2003 | 2004 | Test of sign. 02 vs. 04 | Test of sign. 03 vs. 04 | | :-- | :--: | :--: | :--: | :--: | :--: | | Assault | 6.75 | 16.50 | 12.25 | n.s. | n.s. | | Disorderly conduct | 7.25 | 9.75 | 9.75 | n.s. | n.s. | n.s. $=$ not significant [^0] [^0]: ${ }^{1}$ The picture of cars and drivers that emerged from the record checks was substantialy the same as that which emerged from checks made on the 14 gypsy cabs identified during the systematic observations. ${ }^{2}$ A second wave of post intervention observations will be carried out in August 2004 to examine long term effects over the more intensive summer period.