The death of ten-year-old Damilola Taylor on a housing estate in North Peckham, London has highlighted the problems of urban violence and crime. Home secretary Jack Straw, responding to the stabbing of the Nigerian schoolboy, described it as a “wake-up call”.

Peckham is currently in the middle of a £270m regeneration plan that includes the demolition of the 1960s tenement blocks, linked by 16 miles of unlit walkways. Low-rise housing has already replaced half the blocks, but the local residents – many from ethnic minorities – still suffer nearly 7,500 violent crimes each year.

Despite the 1997 Knives Act, introduced after the fatal stabbing of the London headmaster Philip Lawrence, teenage gangs are routinely carrying knives as “fashion accessories”. In the past two years, over 28,000 crimes involved a knife or other sharp weapon in the London Metropolitan Police area alone. Peter Gammon, president of the Police Superintendents' Association admitted, “There are more knives being carried and there is more of a readiness to use them.” Some 30% of all homicides in England and Wales involve a knife making it the most common method of killing.

Following Damilola's death the press is now giving greater publicity to similar incidents such as the murder of another Nigerian boy, 17 -year-old Shola Agoro, who died after being stabbed on the same North Peckham estate just nine days earlier.

Scotland Yard reported a 30% increase in London street crime, which Tory leader William Hague blamed on falling police numbers and a drop in the use of stop-and-search powers following the Macpherson report into the death of black teenager, Stephen Lawrence.

According to the Recorded Crime Statistics – which comprise notifiable offences recorded by the police – crime is on the increase. Total offences in the 12 months to March 2000 increased by 3.8% to 5.3 million, violent crimes rose by 16%, sexual offences by 4% and robberies by a massive 26%. Not surprisingly, cities fared worst with Manchester, London and Glasgow having the highest rates of homicides and robberies. A further large increase in recorded robberies is expected when figures up to the end of September are published.

Victims of crime tend to be the young, the unemployed, single parents, private renters, those living in inner cities and in areas of “high physical disorder”. According to Loughborough University's Professor of Criminology Philip Bean, these precisely match the categories from which most criminals come. “The 1960s housing estates – that's where everything awful resides,” he said. “You find large numbers of criminals there. They burgle each other and they assault each other.”

The image portrayed is of our cities descending into anarchy with both property crime and violent crime rates now higher than in the US. But is this grim picture accurate?

The Recorded Crime Statistics are notoriously unreliable. Firstly, they rely on crimes being reported to the police. The majority of crimes though, go unreported. According to the Home Office's biennial British Crime Survey a major reason for this is the public's lack of confidence in the police.

Secondly, not all crimes reported to the police are recorded. Since records began in 1860, desk sergeants have massaged the figures to eliminate complainants they feel to be time-wasters, and to produce for their superiors results that show their force in a good light. Conviction rates are easily improved by not recording crimes where the culprit cannot be found or by persuading convicted criminals to admit to crimes they did not commit. The British Crime Survey reveals that nearly half of all offences reported to the police vanish from the records. The police record the majority of vehicle thefts because of insurance company requirements but they ignore half of all woundings, two-thirds of attempted burglaries and almost three-quarters of common assaults.

ABI figures

The British Crime Survey is a far more accurate guide to crime trends as it records the public's experience of crime. It shows that while in 1999 there were a staggering 14.7 million offences committed this represented a 23% reduction since crime figures peaked in 1995 and is now lower than the crime levels of 1991.

Virtually all categories of crime have reduced since 1995. Vandalism is down by 17%, burglary by 27%, vehicle thefts by 32% and violent crime by 20%. The ABI's own figures support this overall drop in crime. In 1999 insured theft losses totalled £708m, representing a massive 30% reduction since 1992.

Official recorded crime statistics may indicate that crime is worse in the UK than it is in the US but comparisons are meaningless unless incidents are reported and recorded in the same way. On average 33,800 people are killed by guns each year in the US compared to a total of 750 homicides in England and Wales. This suggests that the US is rather more dangerous than the UK.

Fear of crime

Despite the substantial fall in crime, the British Crime Survey reveals that most people are now more worried by crime than they were when crime was at its peak. A recent Mori poll found that public concern about crime had doubled since Damilola's death. The greatest increases in concern were from professionals, managers and high income families – who actually face the lowest risk of crime. All this suggests that fear of crime is more due to negative press reporting and the use by politicians of spurious statistics to support a political agenda than it is to individuals' own experiences.

Forecasting future crime trends is notoriously difficult. A 1999 Home Office research study suggested that property crime is linked to consumer spending and the number of young males in the population. The study reckoned that for every 1% increase in consumer spending, burglary and theft would increase by about 2%, and that for every 1% increase in the number of males aged between 15 and 20, it would increase by a further 1%.

The British Crime Survey acknowledges the contribution that improved security has made to reducing property and vehicle crime. Compared to 1992, twice as many homes now have burglar alarms and 44% more homes have window locks. In the same period, the number of car alarms and immobilisers have more than doubled. Local crime reduction initiatives such as Neighbourhood Watch and the government's Crime Reduction Programme have also played a valuable part.

The jury is out though on the effect the economic cycle has on crime levels. The links are tenuous, but falling unemployment and relatively high economic growth may depress levels of property crime, perhaps by reducing the need for the proceeds. Other factors such as wage rates (including the minimum wage) the availability of overtime and the buoyancy of the black economy may also be important.

There is no shortage of government funding and incentives aimed at reducing urban crime. Schemes such as On Track and Youth Inclusion aim to help children at risk of getting involved in crime, while the Neighbourhood Renewal Fund provides an £800m package to help rejuvenate run down inner cities. The Targeted Policing Initiative provides £15m of additional funding for specific crime reduction projects. Out of this, Merseyside has just been awarded £1m for schemes to tackle car crime and burglaries against small businesses.

Crime in some parts of our cities is clearly still a major problem – for the individuals affected and for their insurers. But hard facts suggest though that the situation is not as bad as it is painted in the press. The success of the government's various initiatives to combat city crime and reintroduce a sense of normality to the worst-hit areas will depend on a collaborative approach between government departments, industry and the local community. Insurers have a key part to play. The prize for success is high – not only in terms of reduced crime and improved underwriting results but in preventing the deaths of young children like Damilola Taylor.