STOP AND THINK

Increasing Police Searches Is Unlikely To Bring Down The Street Crime
Now Fuelling Political Panic

Police searches have to go up to get crime down. That, at least, is
becoming the received wisdom. The argument has largely centred on
London where searches fell by 88% in the five years to 2001, although
nationally they rose by 3%.

Meanwhile, crime overall rose slightly from a low in the mid-90s, but
this included a dramatic increase in recorded street crime. Driven
largely by school-age children taking each others' mobile phones, the
problem is proving as intractable as it was unforeseen, and has begun
to induce symptoms of political panic.

Research suggests that sudden drops in police searches tend to be
followed by a rise in recorded crime. These falls, however, may often
be a proxy measure for a more general withdrawal from proactive
policing. It cannot simply be assumed that a fall in searches will
cause a corresponding rise in crime, although there is an obvious
appeal in the idea that swapping the two sides of this spurious
equation will provide an answer to any given type of crime.

This idea is not only simplistic, but dangerous. It risks
short-circuiting informed debate about how to resolve serious
tensions between the use of search powers and their potential for
collateral damage. It is generally accepted that searches are
essential for the "prevention and detection of crime", in the words
of the Macpherson report. Yet searches have long been recognised,
both in the research literature and on the streets, as a running sore
in the police's relations with young people generally and black
people in particular.

It is essential to pinpoint how searches can contribute to tackling
crime and whether some could be dispensed with because their
potential for damage outweighs their positive contribution.

Most arrests for carrying offensive weapons (including firearms) and
going equipped (to burgle, break into cars, etc) come from searches.
It is obvious that arrests of this kind may prevent offences, but
they account for only about a quarter of the total. Searches for
stolen property are much more numerous, but, contrary to received
wisdom, they detect only a minority of street crime offences.

By far the largest number of searches in London are for drugs. They
made up nearly half the total in 2000/1, a higher proportion than
five years earlier. They dropped by 66% over the same period, but
searches for stolen property fell by 115%, and for "going equipped"
by 131%. When I looked at them in 1999, the vast majority of drugs
arrests from searches were for possession of small amounts of
cannabis. The Lambeth experiment seems to indicate that using police
resources for this purpose may actually detract from the ability to
tackle crime.

There are other ways that searches can, and do, contribute to
tackling crime and these partly explain the value the police place on
them. Information from searches can be an important source of
intelligence on suspects' whereabouts, appearance and associates.

The search power mainly used requires officers to have "reasonable
grounds" for suspicion that the search will uncover stolen or illegal
articles. However, a high proportion of arrests from these searches
are for other reasons, for example when the person turns out to be in
breach of their bail conditions or (like the Yorkshire Ripper) to be
wanted for another offence.

To use the power strategically for these purposes seems questionable
at best; and creative interpretation of the statute has tended too
easily to shade into harassment of people "known" to the police. It
is arguable that officers could produce the same results without
needing to search people.

So there are strong arguments for increasing the minority of searches
for offensive weapons (including firearms) and for "going equipped".
The trade-off between delivering on crime and collateral damage to
the community, though, is more finely balanced in the case of
searches for stolen property. And there is probably still
considerable scope for cutting back on drugs searches.

A lot of the "added value" produced by searches might be realised,
and possibly increased, if officers simply spoke to more people.

Ironically, nothing seems better calculated to undermine this
approach than proposals which would require officers, when they stop
and question members of the public without any intention of searching
them, to ask them their personal details, including their ethnic
origin. This would increase the potential for conflict and produce a
mass of probably inaccurate statistics which it would be impossible
to interpret.

We need a more systematic approach to gathering intelligence. It is
also important to be able to monitor the activities of individual
police officers, who are unsupervised in their encounters with
members of the public on the streets. But the most that it is
sensible to expect is that officers will offer those they stop a card
with their name and number on it and the place and time of the stop.