At some point this week, Gordon Brown will ignore the advice of his own scientists, and reclassify cannabis as a Class B rather than a Class C drug. In doing so, he plans to demonstrate that he is tough on drugs, and in touch with Daily Mail readers. What he'll actually demonstrate is that he has no interest in creating a meaningful policy that tackles the problems of drugs, based on scientific evidence. Instead, he wants to continue to use a discredited system designed for posturing rather than action, in an attempt to win a few more votes.

In July last year, the Home Secretary sent a request to the Advisory Council on the Misuse of Drugs, headed by Professor Sir Michael Rawlins, asking them to reevaluate their position on the classification of cannabis, and in particular to address mental health concerns that had been raised in the press. The results were apparently released to the home office last month, but have yet to be formally published. Embarassingly, it's been reported that the study will contradict Brown, suggesting that cannabis should remain a Class C drug, and pointing out that the latest research has "found nothing to support a theory that rising cannabis use in the 1970s, 1980s and early 1990s led to increases in the incidence of schizophrenia later on".

The public debate about drugs has been horribly skewed by newspapers such as the Daily Mail, who are more interesting in promoting less-than accurate scare stories than presenting any sort of useful, contextualised information to their readers. Hence we are left with clearly balanced articles such as "Cannabis: The Scientific Evidence Against It", and the ridiculous "Cannabis 'kills 30,000 a year'".

Why are these articles unhelpful? Well the first because of its lack of context: I could find all sorts of evidence that chocolate is dangerous, but publishing an article called "Chocolate: the evidence against" would obviously be a distortion. The second article, quoting 30,000 deaths a year from cannabis, is based not on any actual, erm, evidence, but instead actually refers to the potential number of deaths caused by the tobacco cannabis is smoked with, and assumes ridiculously that all 3.2 million cannabis users are smoking several joints every day.

Now, I'm not suggesting for one moment that cannabis doesn't have side-effects, sometimes serious. But is it really such a big and serious issue that an already heavily-burdened police should be spending vast amounts of their time and money chasing stoners? To paraphrase Bill Hicks, go into any town centre on a Saturday night, and watch all the crime and disorder taking place - are the people involved a) stoned, or b) drunk? The answer should give some clue as to where policy really needs to be directed, and why so many people are alarmed by this obsession with tackling a problem that exists largely in the imagination of a few newspaper editors.

These newspaper editors may have rather a big influence in government circles as well. The Science Select Committee's amusingly-titled 2006 report - "Drug classification: making a hash of it?" includes a transcript of the Committee's grilling of a government advisor, the same Professor Sir Michael Rawlins: "You seem to be giving the impression—and I would not want the Committee to be unfair—that this is a very ad hoc sort of organisation, where there is a lack of transparency about where you get advice from. You have loose conversations with ministers which may or may not change policy. The Daily Mail, or some other organ, may exert undue influence. You may or may not have conversations with the Department for Education and Skills, even though drugs policy in schools is a massive issue. Are we being unfair here?" Very, replied Rawlins, pointing out that "you have only got to read the Daily Mail and read what they say about me and Professor Nutt to realise we are not influenced by them." The latest report from Rawlins' group backs him up, but it's impossible to escape the suspicion that the government are reacting more to media stories than to the advise of their own advisors.

In the face of all this, it would make sense for government policy to be driven by science. You would expect the drugs classification system to have more harmful drugs in the 'A' category, with a typical drug in 'B' and relatively harmless drugs in 'C'. In fact though, as as Science Select Committee revealed to the media in 2006 when it studied the issue, current classifications make virtually no sense. The problem is neatly summarized in the following graph:

In compiling the data, the Science Select Committee complained that the present system was based on historical assumptions, not scientific assessment. Their analysis factored in the actual harm caused by drugs, both to individuals and society, and placed Alcohol at 5th, and Tobacco at 9th. Cannabis lay 2 places behind tobacco, and drugs such as ecstacy - the target of major media campaigns - were shown to be far safer than alarmist media coverage suggested.

So the drugs classification system makes basically no sense from a scientific perspective. This was pointed out and scientists and drug experts called for reform, but none came. Instead, 2 years later we have Brown upgrading cannabis to a class B drug, against all medical reason. His justification for this is that it will "send a message" to drug users. Will it? What actual effect will this have?

In terms of procedure, not a lot. The only practical difference the law makes is to double the maximum sentence for possession from 2.5 years to 5 years. Since this sentence is virtually never used - as imprisoning such people simply occupies resources and criminalizes them further - the net effect on police activity is likely to be roughly zero.

That's the policing, but what about the actual usage? Well, Cannabis use has been declining anyway, and this decline actually accelerated after the reclassification of cannabis to Class C in 2004. "British Crime Survey statistics showed that the proportion of 16- to 24-year-olds using cannabis slumped from 28% a decade ago to 21% now, with its declining popularity accelerating after the decision to downgrade the drug to class C was announced in January 2004." The trend led to the following comment from Dame Ruth Runciman of the UK Drugs Policy Commission: "We do not believe the credibility of the current system or the clarity of message has been enhanced when, in just the space of seven years, five home secretaries have sought one way or another to address the classification of cannabis."

So the Home Office's own figures showed that Cannabis use fell more rapidly when it was downgraded to a Class C drug. Whether there's any link there is debatable though. In fact, research suggests that changes in the classification of a drug have little impact on its use. As Jenkins points out (although frustratingly I've not yet found the original reports): "a Mori poll for the Police Foundation found this [the impact of classification] to be near zero in the case of cannabis. Another survey, for the charity Rethink, found just 3% of young people knew what classification meant." Indeed, the 2006 Science Select Committee report stated in its conclusions that: "We have found no convincing evidence for the deterrent effect, which is widely seen as underpinning the Government’s classification policy."

But we're overlooking one critical factor. Actually the law will send a message - not to drug users, but to Daily Mail and Sun readers. It allows a Prime Minister who's just been absolutely battered in his party's worst election results for some 40 years to regain the initiative and make a policy announcement that will score cheap political points with tabloid editors and their readership.

And so the history of drugs policy continues, as the law is manipulated not to tackle any of society's real problems, but to allow a politician to cling to power a few days more.