Today was dubbed “Police Super Thursday”. The latest crime figures came out just as Her Majesty’s Inspectorate of Constabulary (HMIC) gave us an update on how forces are dealing with the cuts.

The police watchdog said most of the 43 constabularies in England and Wales are coping well after taking a budget cut equivalent to 20 per cent in real terms by 2015.

And there was good news for the government in the crime figures, with overall crime down by 9 per cent or 7 per cent, depending on how you measure it.

But can we believe these figures?

Not everyone trusts the police to record crime figures accurately, or even make an honest attempt to do so.

Dr Rodger Patrick, a retired detective chief inspector with the West Midlands force, is one of a number of former police officers to lift the lid on dubious practices designed to “game” the figures.

These including under-reporting crimes, getting suspects to admit to large numbers of unsolved offences in return for a shorter sentence, throwing resources at areas where performance is being monitored, and even “stitching up” defendants by fabricating evidence against them.

Allegations like these are so serious that Her Majesty’s Inspector of Constabulary Tom Winsor has announced a review of how the police record crime.

But they are only allegations: we don’t actually have hard evidence of persistent or serious inaccuracy in crime recorded by the police.

What we do have is the Crime Survey for England and Wales (CSEW) – formerly the British Crime Survey – a completely separate set of statistics based on interviews with a representative sample of people.

The survey has its own methodological flaws, but it cuts the police out of the loop and crucially, it has remained consistent over the last few decades, allowing us to follow long-term trends in crime.

If the police were fiddling the numbers we might expect the crime survey to show radically different trends, like crime going up instead of down. In fact, despite the huge differences in the volume of crimes picked up by the two systems –  the trends are fairly consistent.

The CSEW suggests that there were 8.6 million crimes in England and Wales in the year ending in March, a 9 per cent drop on the previous year. Police recorded 3.7 million crimes, a decrease of 7 per cent.

Both measures have been falling over the last 10 years and are now at their lowest levels since comparable records began.

Is Britain unusual?

No. Many developed countries have seen big falls in crime over the last 20 years, according to UN statistics.

Comparisons between different countries can be problematic because each has its own way of defining and recording crime, but the homicide rate is a fairly good measure of violent crime.

We looked at figures for the number of intentional homicides in various developed countries. Most (15 out of 20 OECD members who provided complete data ) showed clear evidence of a declining homicide rate since 1995.

This, along with other indicators of falling crime in America and western Europe, is another reason why we shouldn’t be too sceptical about a real fall in crime in Britain.

What about the recession?

Many of us expect to see more crime in hard times, but the economics of crime is pretty complicated. Lower wages ought to tempt people to turn to crime, but criminologists also point out that lower wages mean people spend less money on goods that can be targeted by thieves.

Some researchers have claimed to find a strong correlation between unemployment and crime. If that’s right, the fact that the recent downturn has caused lower levels of joblessness than previous recessions may be significant.

Most experts conclude that the causes of crime are so complex that changes in the economy alone won’t necessarily outweigh other factors.

So why is crime falling?

There are a number of theories, none of them proven.

The US economist Steven Levitt, co-author of Freakonomics, has suggested that the legalisation of abortion in America in the 1970s meant there were fewer unwanted babies to grow up into maladjusted young miscreants.

The abortion theory fits the figures in the US: crime rates started to drop 18 years after the legislation. But when you transpose the same theory to Britain the timescale doesn’t quite work.

Another leftfield theory is that high levels of lead in petrol were responsible for a post-war boom in crime, because lead poisoning makes people more aggressive. As catalytic convertors became the norm and lead emissions were reduced, crime fell dramatically.

Some researchers have claimed to find compelling evidence of a link between lead in the atmosphere and crime rates in many different countries, and even at the level of individual neighbourhoods in cities.

A more common-sense explanation for falling crime rates is Britain’s growing prison population, which has almost doubled since the early 1990s.

The LSE’s Centre for Economic Performance produced this graph which appears to show a strong historical association between changes in the number of people behind bars and crime.

This fits with the experience in America, where an explosion in the prison population matched a drop in crime rates. But neighbouring Canada has also seen crime fall without locking up significantly more people.

There are any number of alternative theories: the ageing population, changes in technology, better education, improved policing practices.

Perhaps the most pressing question is whether more police on the beat cut crime. It sounds like common sense, but it’s fairly difficult to prove cause and effect.

Researchers at the LSE found that ramping up officer numbers in problem areas did cut crime. And they found that there was less crime in the wake of the 2005 London bombings, when there was a higher visible police presence in the city.

The academics suggested that a 10 per cent increase in police could be expected to cut crime by 3 per cent.

Does that mean we should be worried now, with police officer numbers at their lowest level in England and Wales since 2002?

Not necessarily, according to HMIC. The watchdog has found that forces will increase the proportion of officers in frontline roles from 74 per cent in 2010 to 78 per cent in 2015, meaning the frontline will be protected from most of the cuts and should only shrink by some 5 per cent.

Only time will tell whether that has a measurable effect on Britain’s crime figures.

By Patrick Worrall