Stop and Search code of practice: twelve month review - quantitative report

2. Changes following implementation of the Code

2.1 Change in police use of search and seizure

During the twelve month period following the introduction of the new Code of Practice (CoP) on Stop and Search (June 2017 to May 2018), there were 32,307 encounters involving either a search or a seizure in Scotland. This compares to 44,249 during the equivalent twelve month period of the previous year and represents a 27% reduction in the use of these police tactics. The number of searches and seizures did not drop as a direct consequence of the introduction of the Code of Practice on 11^th May 2017. Rather, this reduction reflects an ongoing steady decline which can be traced back to the publication of a critical HMICS report in March 2015. Figure 2.1 shows the decline in recorded searches and seizures between June 2016 and May 2018, and highlights the year before the introduction of the CoP and the year after. From the start to the end of this two year period, the number of encounters decreased by 36%; however, the difference between its highest level (in June 2016) and lowest level (in February 2018) represents a reduction of 52%.

Figure 2.1: Number of searches and seizures carried out in Scotland, June 2016 to May 2018

The number of searches and seizures conducted during the twelve months following the introduction of the CoP is clearly smaller in comparison to the equivalent twelve months of the previous year. In addition, there are distinctively different trends, with a more ongoing decline after the introduction of the COP than is evident in the 12 months before. There is some evidence of a seasonal trend in both periods, with higher numbers in the spring and summer months, although there was no similarity in seasonal trend over the winter months. Interestingly, however, the sharp rise in encounters during April and May 2018 brought them back in line with the number in April and May 2017, which may be an early indicator that usage of these tactics is starting to 'level out'.

There was a distinct reduction in all types of searches and seizures following implementation of the CoP. Table 2.1 shows the breakdown of policing encounters by number and percentage in the twelve months after the CoP came into force compared with the equivalent twelve months of the previous year. Statutory searches reduced by 19% overall, although they increased in terms of the relative share of all encounters by 9%. This is partly due to the phasing out of consensual searches following the introduction of the CoP (although the number of consensual searches had already diminished to tiny numbers during the year prior to the CoP). However, it is mainly due to a 63% decline in seizures, which fell as a proportion of all encounters from 16% to 8%. As the analysis compares the same time period in each year, this difference is not accounted for by any seasonal variation and, instead, suggests that police use of seizure reduced at a far greater rate than the use of stop and search.

Table 2.1: Number of searches and seizures pre and post-implementation of the CoP

	June 2016 to May 2017		June 2017 to May 2018		% Change in N	Difference in % share
	N	% share	N	% share
Statutory searches	36,627	83%	29,773	92%	-19%	+9%
Consensual searches	707	2%	0	-	-100%	-2%
Seizures	6,915	16%	2,534	8%	-63%	-8%

Note: Column percentages may not total 100% due to rounding.

This differential change in the use of search and seizure is confirmed by the indexed trends in Figure 2.2. This chart shows that the percentage change in seizures declined far more steeply than that for searches during the year prior to the introduction of the CoP and, despite an initial increase in seizures immediately prior to the CoP introduction (which is most likely a seasonal effect) they continued to decline more steeply afterwards. Indeed, the number of seizures declined by 90% between June 2016 and its lowest point in December 2017, although it showed signs of increasing again in the early months of 2018 (which could, again, be a seasonal effect). This compares with a 38% drop in searches between June 2016 and its lowest point in February 2018, although there are also signs of some increase in subsequent months.

Figure 2.2: Indexed trends in police use of search and seizure, June 2016 to May 2018

The trend in seizures will be discussed further in section 3 in relation to the use of search and seizure as a policing tactic for dealing with underage drinking; however, it is clear from Figure 2.2 that the police use of search and seizure changed before and after implementation of the CoP but it did not change consistently for both tactics.

2.2 Change in positive detection rate for stop and search

Despite the decline in the overall number of searches, the success rate in terms of positive detection increased in the year following the introduction of the CoP. Table 2.2 shows that in the 12 months prior to the introduction of the CoP there were 37,325 searches, of which 11,689 (31%) were positive. In the year after the CoP was introduced, there were 29,768 searches, of which 11,236 (38%) were positive. This represents a statistically significant increase of 7% in positive outcomes following the introduction of the CoP, which suggests that officers were applying a higher threshold of reasonable suspicion when using stop and search. However, given the very large reduction in the total number of searches, it is important to examine the absolute change in the number of positive searches. Taking statutory and voluntary searches together, Table 2.1 shows that there were 7,561 fewer searches (a decline of 20%) following the introduction of the CoP, but Table 2.2 shows that there were only 453 fewer positive searches (a decline of only 4%). This indicates a net reduction in searches resulting in the recovery of an illegal or dangerous item.

There was an increase in the positive search rate across all search types, as shown in Table 2.2. Searches for stolen property saw the largest increase in positive detection rate (9%). There was a shallower increase in the rate of positive searches for drugs (6%) and, especially, for offensive weapons (4%). Indeed, the rate of positive searches remained lowest for weapon searches, with only 26% of such searches resulting in an item being recovered. A new category for recording searches of people carried out in accordance with a warrant was added to the Police Scotland database following the introduction of the CoP, and these had a higher than average positive rate.

Table 2.2: Number and percentage of positive police searches pre and post-implementation of the CoP

	June 2016 to May 2017		June 2017 to May 2018		Difference in % of positive searches
	N	% of all searches	N	% of all searches
All searches	11,689	31%	11,236	38%	+7%
Searches by type:
Drugs	10,252	32%	9,293	38%	+6%
Stolen property	895	35%	1,011	44%	+9%
Offensive weapons	422	22%	380	26%	+4%
Warrant	13	28%	456	41%	+13%
Other reason	107	21%	96	28%	+7%

Note: There were seven searches on the Stop and Search Database relating to terrorism for which no outcome was published.

2.3 Geographical change in the use of search and seizure

2.3.1 Overall change by Command Area

Prior research (Murray 2014; McVie and Murray, 2017) demonstrated that the use of search and seizure in Scotland varied significantly by geographic area, and this was also found in the six month review of the CoP (McVie 2018). Police Scotland consists of three large Command Areas: East, North and West. Table 2.3 shows the change in the number of searches and seizures during the twelve months before and after the implementation of the CoP within these three Command Areas. It also shows the percentage change in the total number of searches and seizures, and the shift in the relative share of all encounters within each of the three areas. Table 2.3 shows that the very large reduction in searches and seizures at a national level was predominantly driven by falling numbers in the West Command Area, which is evident in terms of change in absolute numbers and relative share of all encounters.

It is clear from Table 2.3 that the majority of all searches took place in the West Command Area. However, there was a 41% decline in searches in the West during the twelve months after the introduction of the CoP. This contrasts sharply with an 8% increase in searches in the North and a 23% rise in the East. As a consequence, the relative share of all searches in Scotland shifted away from a predominance in the West, although almost half (45%) of all searches still occurred in this Command Area following the introduction of the CoP.

The number of seizures conducted in the West Command area declined significantly, by 68%, in the year following the introduction of the CoP. There were also reductions in the use of seizures in the North (-12%) and the East (-26%) Command Areas, albeit to a far lesser extent. Again there was a shift in the relative share of all seizures, with a net loss in the West Command Area and net gains in the North and East. Nevertheless, unlike searches, the vast majority (79%) of all seizures continued to take place in the West.

Table 2.3: Number of searches and seizures pre and post-implementation of the CoP by Command Area

	June 2016 to May 2017		June 2017 to May 2018		% Change in N	Difference in % share
	N	% share	N	% share
Searches
North	8,488	23%	9, 201	31%	+8%	+8%
East	5,727	15%	7, 032	24%	+23%	+9%
West	23,119	62%	13,540	45%	-41%	-17%
Seizures
North	221	3%	194	8%	-12%	+5%
East	467	7%	345	14%	-26%	+7%
West	6,227	90%	1,995	79%	-68%	-11%

Note: Column percentages within searches and seizures may not total 100% due to rounding.

2.3.2 Monthly change by Command Area

The overall figures tell only a partial story and do not account for change over time. Figures 2.3 and 2.4 show how the number of searches and seizures changed on a month by month basis prior to and after the introduction of the CoP in each of the three Command Areas. These figures are presented as indexed trends, fixed at zero in June 2016, so they show the percentage change in overall encounters from this time point onwards.

Figure 2.3 shows that there was a steady and consistent fall in the number of searches in the West Command Area from around August 2016 to December 2017, although in the early months of 2018 numbers started to rise again (the trend line suggests more of a gradual increase than a seasonal effect). Clearly, therefore, the decline in the number of encounters in the West Command Area started well before the introduction of the CoP. In the North and East Command Areas, the indexed trends showed far greater month to month fluctuation (which is partly a result of smaller absolute numbers of searches). The trend in searches in the North Command Area varied from month to month, but it showed no consistent rise or fall over time. Whereas, in the East Command Area there was a slight upward trend in searches between October 2016 and January 2017, but no consistent pattern beyond that. Therefore, Figure 2.3 does not suggest that the higher number of encounters in the North and East in the twelve month period following the introduction of the CoP was due to a marked upward trend from May 2017. Furthermore, it does not show any distinct seasonal trend in numbers of searches in any of the three Command Areas.

Figure 2.3: Indexed trends in searches by Command Area, June 2016 to May 2018

The trend in seizures is shown in Figure 2.4. The number of seizures was much smaller, so the trends are subject to an even greater degree of fluctuation than for searches. Nevertheless, Figure 2.4 shows a gradual decline in seizures that started well before the introduction of the CoP within all three Areas. Between June 2016 and March 2017, seizures fell by between 63% and 85% across the Command Areas. There appeared to be a slight increase in the number of seizures just prior to the introduction of the CoP, which may well have been a seasonal effect; however, there was another distinct period of decline in numbers across all three Command Areas after the CoP was introduced. There does appear to have been some increase in seizures in the early months of 2018, with a very large spike in the East in April 2018. However, the overall pattern of decline in the number of seizures over the two years is starkly different to the changing pattern of searches, shown in Figure 2.3.

Figure 2.4: Indexed trends in seizures by Command Area, June 2016 to May 2018

Looking at the total number of searches and seizures in each Command Area makes an assumption that the level of demand for policing is similar in each area; however, this is unlikely to be the case. There are many factors that can impact on the level of demand for policing, but one of the most common factors is the population size which varies from area to area. In order to compare Command Areas on a like-for-like basis, the rate per capita of search and seizure was calculated using mid-year population estimates from the National Records of Scotland. The rates were calculated per 10,000 people aged 12 to 59, since the vast majority (over 97%) of all searches and seizures involved people within this age range.

Figure 2.5 shows the difference between the search and seizure rates in each Command Area in the twelve months prior to the introduction of the CoP and the equivalent period after the CoP was introduced. Overall, search rates per capita were far higher than seizure rates across all three Command Areas. Prior to the introduction of the CoP, the West Command Area had a far higher rate of search and, especially, seizure than the North and East. Following implementation of the CoP, the seizure rate in the West Command Area reduced dramatically from 40.9 to 13.1 per 10,000 people aged 12-59, although it continued to be four times higher than that in the East (3.2) and five times higher than in the North (2.4). Seizure rates declined only marginally in the East and North.

The rate of search in the West Command Area also declined substantially, from 152.0 to 89.1 per 10,000 people aged 12 to 59; whereas it increased from 54.2 to 66.2 in the East and from 105.1 to 114.7 in the North. Taking population size into account, therefore, the rate of search actually became smaller in the West than the North Command Area following the introduction of the CoP.

Figure 2.5: Rates of search and seizure per capita pre and post-implementation of the CoP by Command Area

Note: Rates per capita based on Mid Year Population Estimates for 2016 (June 2016 to May 2017) and 2017 (June 2017 to May 2018) for people age 12-59.

2.3.3 Overall change by Division

Within the three Command Areas there are a 13 Police Divisions. Table 2.4 shows the change in number of searches by Division. There was substantial change at a Divisional level, but no clear pattern of change within Command Areas which suggests that change in the use of search as a tactic was influenced predominantly by factors at a local level.

The Division with the largest number of searches during both periods was Greater Glasgow. Nevertheless, there was a 59% reduction in searches in this Division in the twelve months following the introduction of the CoP compared with the previous year. The relative share of all searches attributable to Greater Glasgow reduced from 39% to 20%, which represented the greatest overall change. There were fairly sizable reductions in the number of searches in each of the other West Divisions, with the exception of Dumfries & Galloway which increased by 69% over this period. Nevertheless, the relative share of the other West Divisions did not change substantially.

In the North Command Area, there was a 19% increase in searches within the North East (increasing its relative share of all searches by 6%) and a 7% increase in the Highlands and Islands, but a 7% decline in Tayside. Whereas, in the East Command Areas there were large increases in Edinburgh (32%) and the Lothians & Scottish Borders (62%) but only a minimal increase in Fife (6%), and a small decline in the Forth Valley (-5%). Overall, the relative share of all Northern and Eastern Divisions increased over this period, largely due to the large reduction in searches in Greater Glasgow.

Table 2.4: Number of searches pre and post-implementation of the CoP by Police Division

	June 2016 to May 2017		June 2017 to May 2018		% Change in N	Difference in % share
	N	% share	N	% share
West Command Area:
Greater Glasgow	14,606	39%	5,919	20%	-59%	-21%
Renfrewshire & Inverclyde	2,219	6%	1,630	6%	-27%	0%
Argyll & West Dunbartonshire	1,593	4%	1,364	5%	-14%	+1%
Lanarkshire	2,283	6%	1,750	6%	-23%	-1%
Ayrshire	1,613	4%	1,520	5%	-6%	+1%
Dumfries & Galloway	805	2%	1,357	5%	+69%	+2%
North Command Area:
North East	3,852	10%	4,567	15%	+19%	+6%
Tayside	2,268	6%	2,112	7%	-7%	+1%
Highlands & Islands	2,368	6%	2,522	9%	+7%	+2%
East Command Area:
Edinburgh	1,854	5%	2,445	8%	+32%	+3%
Forth Valley	1,546	4%	1,466	5%	-5%	+1%
Lothians & Scottish Borders	1,169	3%	1,896	6%	+62%	+3%
Fife	1,158	3%	1,225	4%	+6%	+1%

Note: Column percentages and % change may not total 100% due to rounding.

Turning to seizures, Table 2.5 shows that the numbers were much smaller but the percentage changes were much larger. The North East Division was the only one to see an increase in the number of seizures in the twelve months after the introduction of the CoP, rising by 39%. Seizures in all of the other Divisions fell, ranging from a 15% drop in the Lothians & Scottish Borders to a substantial 76% drop in Greater Glasgow, which resulted in a 21% fall in relative share of all seizures. The most substantial declines occurred in the Divisions of the West Command Area, but it is clear that this was a pattern that was replicated across the country.

Table 2.5: Number of seizures pre and post-implementation of the CoP by Police Division

	June 2016 to May 2017		June 2017 to May 2018		% Change in N	Difference in % share
	N	% share	N	% share
West Command Area:
Greater Glasgow	4,217	61%	1,019	40%	-76%	-21%
Renfrewshire & Inverclyde	438	6%	228	9%	-48%	0%
Argyll & West Dunbartonshire	174	3%	108	4%	-38%	+1%
Lanarkshire	859	12%	308	12%	-64%	-1%
Ayrshire	498	7%	310	12%	-38%	+1%
Dumfries & Galloway	41	1%	22	1%	-46%	+2%
North Command Area:
North East	77	1%	107	4%	+39%	+6%
Tayside	48	1%	28	1%	-42%	+1%
Highlands & Islands	96	1%	59	2%	-39%	+2%
East Command Area:
Edinburgh	88	1%	63	3%	-28%	+3%
Forth Valley	93	1%	43	2%	-54%	+1%
Lothians & Scottish Borders	185	3%	158	6%	-15%	+3%
Fife	101	2%	81	3%	-20%	+1%

Note: Column percentages and % change may not total 100% due to rounding.

2.3.4 Monthly change by Division

It is not possible to look in detail at change in the number of seizures by Division on a month by month basis as the numbers are too small. However, focusing on the number of searches there is evidence of considerable geographical variation in shifting police practice as a result of the implementation of the CoP. Looking at indexed trends from June 2016, three Divisions recorded a lower number of searches in the twelve months prior to the CoP being introduced than in the twelve months after. Figure 2.6, below, shows that in the months following the introduction of the CoP (represented by the dotted vertical line), the number of searches and seizures in the North East, Lothians & Scottish Borders, and Dumfries & Galloway were consistently higher in the period following the introduction of the CoP than in the previous year. It is worth noting that the change seems to have been precipitated in the month or so immediately prior to the implementation of the CoP and to have been sustained in the period afterwards.

On the contrary, five Divisions – all in the West Command Area - displayed a large and significant decline in the number of searches, which started well before the implementation of the CoP and continued after it. Figure 2.7 shows that the number of searches in Greater Glasgow, Ayrshire, Lanarkshire, Argyll & West Dunbartonshire, and Renfrewshire & Inverclyde all declined significantly during the twelve months preceding the introduction of the CoP, albeit at different rates and times, and with a certain degree of month to month fluctuation.

Figure 2.6: Divisions showing an increase in search following implementation of the CoP

Figure 2.7: Divisions showing on ongoing decline in searches pre- and post-implementation of the CoP

Renfrewshire & Inverclyde showed the greatest decline in encounters during the pre-CoP period, falling by 77%, although this was closely followed by Lanarkshire and Argyll & West Dunbartonshire which fell by 48% and 46%, respectively. The number of searches in Ayrshire fell by around a third. In all four of these Divisions, the number of searches and seizures did increase again in the months immediately prior to the introduction of the CoP. This is consistent with anecdotal reports of an increase in confidence in using stop and search following the roll out of the national training programme by Police Scotland. Following the introduction of the CoP in May 2017, there was a further reduction in the number of encounters in all four Divisions until the end of 2017. In three of the four Divisions, the number of encounters increased again in the early months of 2018. It is difficult to say whether this is indicative of greater confidence in using stop and search or just an emerging seasonal trend.

In Greater Glasgow, the number of searches remained fairly consistent between June and November 2016, but declined significantly afterwards and continued to do so until reaching its lowest point in December 2017, representing a 77% fall in searches since June 2016. If there was an increase in confidence amongst officers in the West that was attributable to the Police Scotland training programme, it did not appear to be evident amongst those working in Greater Glasgow. Like some other West Divisions, there was a slight increase in encounters in the early months of 2018; however, the rate of change was far slower than in Ayrshire, Argyll & West Dunbartonshire, and Lanarkshire.

Figure 2.8: Divisions showing a stable pattern in searches pre and post-implementation of the CoP

Within the other Police Scotland Divisions, it was more difficult to identify distinctive trends in the number of searches conducted in the period before and after the introduction of the CoP. Figure 2.8 shows that in Tayside and the Highlands and Islands (both Northern Divisions) the trend in encounters was fairly stable over time with relatively small fluctuations in activity. The two exceptions to this were a large spike in July 2016 in Tayside which is explained by an increase in searches during the T in the Park music festival in Kinross, and a spike in the Highlands and Islands in August 2017 which coincides with the Groove Loch Ness concert in Inverness. However, there is no clear evidence of a direct effect of the introduction of the CoP in these two Divisions.

In the remaining three East Divisions - Fife, Forth Valley and Edinburgh - there was no clear observable pattern of change before or after the introduction of the CoP. Figure 2.9 suggests that each of the three Divisions had a fairly stable pattern of searches and seizures which was interspersed with some large peaks in activity. There is no consistent seasonal trend evident, which suggests that the peaks are probably attributable to local initiatives or activities. For example, the large spike in Edinburgh in August 2017 is almost certainly attributable to the Edinburgh International Festival during which there is a high policing presence. It is notable that there was not such a large spike during the equivalent period of the previous year, however. As with Figure 2.8, there is no obvious evidence that the introduction of the CoP had any direct effect on policing practice in these three Divisions. Or at least any impact of the new legislation was most likely to have been set in motion well before the actual implementation period.

Figure 2.9: Divisions showing no consistent trend in searches pre or post-implementation of the CoP

2.3.5 Change in positive detection by Division

Section 2.2 noted that the overall reduction in the number of police searches in the period following the implementation of the CoP coincided with an increase in the positive detection rate (regardless of the reason for the search) compared to the equivalent twelve month period of the previous year. Given that there was such a large degree of change and variation in use of search at a Divisional level, it is helpful to examine how the likelihood of a positive outcome changed within Police Divisions. Figure 2.10 shows the percentage of searches which resulted in a positive outcome for each Division, and for Scotland as a whole, in the pre- and post-CoP periods.

There was fairly wide variation in successful outcomes across Divisions both before and after the CoP was introduced; although it is notable that the degree of variation reduced in the post-CoP period. For example, the gap between the Divisions with the largest and smallest detection rates declined from 27% in the pre-CoP period to 22% in the post-CoP period. Some Divisions showed little or no change in positive outcome between the two periods; for example, Lanarkshire, Lothian & Borders, Tayside, Dumfries & Galloway and North East saw very minor changes in outcome (2% or less). Whereas others had experienced a large increase in positive search outcomes, such as Greater Glasgow, Edinburgh, Fife and the Highlands and Islands. Only two Divisions - Tayside and Forth Valley - had experienced a decline in the positive search rate.

Figure 2.10: Percentage of positive police searches pre- and post-implementation of the CoP, by Division

There appeared to be no consistent relationship between trends in searches and change in positive detection rates across Divisions. Four of the Divisions that had a large reduction in search rates (Greater Glasgow, Ayrshire, Argyll & West Dunbartonshire, and Renfrewshire & Inverclyde) all witnessed a significant increase in positive search rates. However, Lanarkshire, which also had a large decline in searches, saw no change in the rate of positive detections (although this Division did have the highest detection rate overall in the pre-CoP period). The three Divisions that experienced an increase in search rates after the introduction of the CoP (North East, Lothians & Scottish Borders, and Dumfries & Galloway) witnessed little or no change in the positive detection rate. Meanwhile, amongst those Divisions that had no change in the use of search pre or post-CoP, one (Tayside) saw a slight decline in positive detection rates and the other (Highlands and Islands) increased its positive search rates. Equally, amongst those Divisions with no consistent trend in searches before or after the CoP, two (Edinburgh and Fife) had a large increase in detection rates, while the other (Forth Valley) had a slight decline.

Based on these results, it is likely that whatever influenced the pattern of stability or change in the rate of positive outcomes for searches across different Scottish Divisions, it cannot simply be explained by changes in the number of searches carried out. It is likely that other factors, including proportionate use of stop and search and the application of reasonable suspicion, played a key role in the changing pattern of successful searches.

2.4 New information on stop and search

2.4.1 Strip and intimate searches

Since the introduction of the CoP, information about the use of strip searches and intimate searches has been recorded on the NSSU Stop and Search Database. According to Annex C of the CoP: "A strip search is a search involving the removal of more than outer coat, jacket, gloves, headgear or footwear". It is permitted only in circumstances where the officer has reasonable suspicion that a person has concealed an article (such as drugs or a weapon) under their clothing, and it should be conducted in a relatively private place so that the person detained cannot be seen by others. Annex C further states: "An intimate search consists of the physical examination of a detainee's body orifices other than the mouth". Intimate searches are clearly much more intrusive than a strip search. They can be conducted only on issue of a warrant by a Sheriff and must be carried out by an authorised health care professional. The use of strip and intimate searches was not recorded by Police Scotland prior to the introduction of the CoP, so activity can only be reported for the period from June 2017 to May 2018.

There were 1,537 strip searches conducted across Scotland during the twelve months following the introduction of the CoP (an average of 128 per month). This represents 5% of all searches carried out during the twelve months following the introduction of the CoP. Of these, 66% were conducted inside a police station and 34% were conducted out of public view somewhere other than a police station. The vast majority of strip searches (69%) were conducted for drugs, while 30% were conducted during the execution of a Warrant (very few were carried out for other reasons). Almost half (49%) of all strip searches resulted in an item being found compared with 37% of non-strip searches.

Looking at protected characteristics, the majority of strip searches involved a person aged 18 or over, with only 4% (54 in total) involving a young person under the age of 18. The positive search rate for strip searches of young people (49%) was not significantly different to that for adults (48%). There was no significant difference between ethnic groups in terms of whether a strip search was carried out or not.

There were some sex differences in relation to strip searches. Searches involving women were significantly more likely to involve a strip search than those involving men (8% of all searches versus 5%, respectively). In terms of the location of the search, males were most likely to be searched inside a police station (70%), while females were almost equally likely to be searched somewhere other than a police station (52%) or in a police station (48%). The vast majority of strip searches for both men and women were conducted in relation to drugs; however, females were more likely to be strip searched under Warrant (44% compared to 26% for males). Since drug searches under Warrant are typically conducted at the locus, this largely explains why women were less likely to be searched at a police station.

Notably, 42% of strip searches amongst women resulted in a positive detection, which was significantly lower than the positive detection rate for men (50%). Standard drug searches involving men were also more likely to be positive than those involving women, although the extent of the difference was lower (35% versus 37%, respectively).

There were no intimate searches recorded on the NSSU Stop and Search Database during this twelve month period.

2.4.2 Issue of receipts

The NSSU database also records whether receipts were issued by police officers following a search. In section 6 of the CoP, it is stipulated that:

"The person who has been searched should be given a receipt (see Note 12). The receipt should include the following information:-

• police powers of stop and search;
• the right of a person searched to obtain a copy of the record of the search;
• the right of a person searched to complain, including how to go about making a complaint".

While officers are obliged to offer a receipt following a search, which includes placing it with the detainee's property in the event that the individual is subsequently taken into custody, the individual may refuse to accept the receipt. In these circumstances, the advice offered during stop and search training is that the officer should record in their notebook that the receipt has been offered and refused. The CoP also makes it clear that where the officer conducting the search is called to an urgent incident, it may be impracticable to provide a receipt. In this event, the individual should be informed that they can obtain a copy of their search record by calling at any police station (although, only if they have provided their details). Receipts are not issued for seizures.

In the twelve months following the introduction of the CoP, 25,444 receipts were issued following a search. This represents 86% of all statutory searches conducted during this period. In the remaining 4,329 encounters (14%), no receipt was issued. In the vast majority of these encounters (98%) the individual being searched either refused to accept the receipt or left the locus before the receipt could be issued. It was rare for an officer to either forget or fail to issue a receipt (this occurred in only 58 encounters, which represents only 1% of those involving non-issue of a receipt). There was a slight drop in the proportion of searches where a receipt was issued over time, from 87% in June 2017 to 82% in May 2018, which suggests some decline in tolerance for accepting a receipt amongst those being searched. However, there was an increase in the number of receipts that were issued retrospectively towards the end of the review period which suggests that Police Scotland were making efforts to rectify this situation.

There was some variation in the likelihood of a receipt being issued by time of the day (this was slightly less likely between midnight and 6am compared to other times of the day); and by day of the week (they were slightly less likely to be issued on a Friday or Saturday compared to other days of the week). Searches conducted under Warrant or for other reasons were more likely to result in a receipt being issued than for searches involving drugs, stolen property or weapons. Negative searches were more likely to result in non-issuance of a receipt than positive searches. Searches involving men were less likely to result in a receipt being issued, as were searches involving young people under age 18. There was no significant ethnic disparity in the issuance of receipts, however.

Looking at geographical variation, searches in the East Command Area were more likely to result in a receipt being issued (91%) than in the North and West Command Areas (both 84%). There was also considerable variation across Divisions. As shown in Figure 2.11, at least 90% of searches resulted in a receipt being issued in Forth Valley, Fife, Lothians & Scottish Borders, Ayrshire, Highlands and Islands, Dumfries & Galloway, Renfrewshire & Inverclyde, and Tayside; whereas less than 80% of searches in Lanarkshire, Greater Glasgow and the North East resulted in a receipt being issued.

Looking at the reason for non-issue of receipts by Police Division, there was very little geographical variation. As noted above, the overwhelming reason for non-issue of a receipt was that the individual either refused to accept it or left the scene before it could be issued. This explained 97-100% of non-issuance of receipts across most of the Police Divisions. The figure was lowest in the Lothians & Scottish Borders (93%) and Fife (89%), although the number of encounters where a receipt was not issued in these Divisions was very small so these differences could be an artefact due to tiny numbers.

Figure 2.11: Percentage of searches resulting in a receipt being issued by police Division, post-implementation of the CoP

It is expected that officers explain to any person being searched that they are entitled to obtain a copy of the record of the encounter within 6 months of the date of the search, although this only applies if the individual has chosen to give their name, address and date of birth, which they are not obliged to do if the search is negative. Information on the number of people who choose to obtain a copy of their record is not published on the Stop and Search Database; however, data provided by the NSSU showed that there were only 11 such requests in the twelve months following the introduction of the CoP. This included ten individual requests and one request made by a solicitor on behalf of their client. Individuals making requests included both men and women across a range of ages (from 20 to 70) and from across a variety of Police Divisions.

2.5 Summary of Section 2

There was a significant reduction in the number of searches and seizures conducted within Scotland following the introduction of the Stop and Search Code of Practice (CoP) in May 2017. However, trend data shows that this reduction was part of an ongoing decline in encounters that started well before the introduction of CoP and continued afterwards until early 2018, at which point numbers started to increase again. The data suggests, therefore, that the impact of the CoP on policing practice began long before its actual implementation and also continued afterwards. It is almost certain that the criticism targeted at Police Scotland about the over-use of stop and search as a tactic by HMICS and the media in 2015, followed by an intense period of political and public scrutiny and the decision to abolish consensual searching and introduce a CoP, influenced the large scale reduction in encounters well before the CoP finally came into force.

There was a far greater proportionate reduction in police use of seizures compared to searches over the two years studied in this report. The number of seizures of alcohol and tobacco products fell by around two thirds in the twelve months following the introduction of the CoP compared to the preceding year. This was around three times larger than the percentage reduction in statutory searches, although the absolute number of searches continued to be far greater. The period of decline was very similar for both searches and seizures, however; and both started to rise again from around March 2018. The large decline in seizures was surprising, especially given the concerns expressed by policing representatives and other organisations in the public consultation period about the lack of a legal power to search for alcohol.

The decline in searches coincided with an increase in the relative productivity of search encounters, which suggests that they are being used more effectively and with a greater standard of evidence in terms of reasonable suspicion. In the twelve months following implementation of the CoP, the success rate was 38% compared with 31% in the equivalent period of the preceding year. Positive detections had increased across all search types, although it was proportionately greater for searches conducted as part of the issuance of a warrant and for stolen property, and lowest for searches to find offensive weapons. Despite a 20% decline in the overall number of searches, the number of encounters in which an item was recovered had fallen by only 4%.

There was considerable geographical variation in the changing use of search and seizure across Scotland. The number of searches and seizures was highest in the West Command Area during the twelve months prior to the introduction of the CoP, and this continued to be the case in the twelve months after. However, there was a far higher proportionate decline in both searches and, especially, seizures in the West compared to the East and North Command Areas. There was greater geographical similarity in the changing use of seizures, which declined across all three Command Areas and within all but one of the thirteen Police Scotland Divisions (albeit to different extents).

Trends in the use of stop and search were considerably different, however. Overall, the number of searches in the West Command Area declined in the twelve months following the introduction of the CoP, while the number increased in the North and the East Command Areas. Searches within all Divisions in the West had declined with the exception of Dumfries & Galloway which showed a substantial increase. Similarly, in the North and East, most Divisions had increased the number of searches but two (Tayside and Forth Valley) had reduced. Patterns over time revealed substantial variation between Divisions in changing use and success of stop and search. This suggests that the new legislation and policy around stop and search may have been interpreted and adopted differently across Divisions; however, there was little evidence that the introduction of the CoP in May 2017 had a direct effect on policing practice.

New information provided on the Stop and Search Database about strip searches revealed an average of 128 per month, representing around one in twenty searches overall. Strip searches had a higher detection rate than average, at 49%, which suggests that a higher threshold of reasonable suspicion is used when deciding to conduct a strip search. The vast majority of these involved people aged 18 or older, and there was no difference in positive detection rate between those under 18 and those aged 18 or over. There was also no significant difference in the use of strip search by ethnic group. Strip searches for both men and women were overwhelmingly for drugs; although, there were some sex differences. Women were more likely to be strip searched than men, especially through the execution of a Warrant, but detection rates were lower for women. No intimate searches were recorded in Scotland in the twelve months after the introduction of the CoP.

Receipts were issued after 86% of recorded searches. By far the most common reason given for non-issue of a receipt was that the individual being searched either refused to accept the receipt or left the locus before the receipt could be issued. It was rare for an officer to forget or fail to issue a receipt. Whether a receipt was issued or not varied according to a range of factors, including time of the day, day of the week, reason for the search and the characteristics of the individual being searched. There was also some Divisional variation in the likelihood of a receipt being issued, although very little overall difference in the reasons for non-issue of receipts. There was a small reduction over time in the issuance of receipts, and some evidence of an increase in receipts being issued retrospectively. Finally, although individuals have the right to obtain a copy of the record of their search encounter within 6 months, there were only 11 such requests in the twelve months following the introduction of the CoP.