Wednesday 11 May 2011

Prison sentencing and the media

This morning, a story is circulating in the media which apparently once and for all proves that longer sentences cut crime, so we can all just stop thinking about it and start locking people up for as long as possible. Huzzah! It has appeared in the Star, the Express, the Guardian and the BBC, alongside pretty much every other news outlet.

The news is a huge boon to conservative thinkers who have long advocated tougher prison sentences, so it was no surprise to see the likes of unstoppable Tory gobshite Philip Davies MP crowing about it on Twitter:
We now know for sure that the longer people spend in prison the less likely they are to re-offend! Some of us have said this for years!

So, what do the findings say? Well, according to pretty much all papers, they say things like "The longer the prison sentence the less likely an offender is to commit a further crime, according to research" (the BBC). The Guardian, while saying much the same thing, helpfully links to a PDF of the figures so we can have a look for ourselves (as all online news outlets should do in 2011). Here we find, among other things, a couple of very important quotes that are missing from Davies' crowing, and most of the media reporting:
The findings are not conclusive on whether the deterrent effect of longer custodial sentences is effective at reducing re-offending
So yeah, the findings are not...wait, what? I thought Philip Davies MP said that now we know FOR SURE? How can this be?
Despite higher re-offending rates, offenders receiving sentences of less than 12 months do not have access to offender management programmes and are not subject to supervision by the Probation Service upon release. This latter factor is also likely to explain some of the difference between community sentences/suspended sentence orders and short prison sentences.
Oh, right. So there's a fundamental difference between the 'short' sentences and the longer ones which means that factors other than simply the length of sentence itself could be responsible for a discrepancy. That is really a quite major difference, as it implies that effective managment programmes and post-release supervision are possibly having a big effect, not just the actual banging-up of people for as long as possible.

Indeed;
Custodial sentences of less than twelve months were less effective at reducing re-offending than both community orders and suspended sentence orders
That's another nuance somewhat left alone in the media coverage today, which all seems startlingly similar. (Although, if you read the Express' frankly childish attempt to cover it you might come out a tad stupider than if you'd read one of the other, real newspapers). While the Guardian mentions the importance of community sentences for minor crimes, the Mail's effort, somewhat unsurprisingly, doesn't.

The only thing that's really clear from this study is that, like most reports, you can spin it how you want, and that newspapers will spin it in a way that reflects their politics. Or, at least, that newspapers will copy other newspapers' spin. There's a lot of depth and complexity to the figures, but the bottom line is this - it seems ludicrous that you could have, for example, a BBC headline that says;
Longer prison sentences cut reoffending, study suggests
...referring to a report that says;
The findings are not conclusive on whether the deterrent effect of longer custodial sentences is effective at reducing re-offending
Or so you would think!

Wednesday 4 May 2011

On AV and the Daily Mail

Tomorrow sees many of us head to the polls to vote on the alternative vote (AV) system, a system which, if implemented, would sort of actually change some shit a bit. Predictably, the Daily Mail doesn't really like it.

Under the typically restrained and understated heading of Vote No tomorrow to stop Clegg and his cronies destroying democracy in Britain - forever, the Mail's leader column argues that putting candidates in order of preference rather than just voting for one is "fiendishly-complicated". Because putting candidates in order of preference (if you feel like doing so), is presumably incredibly taxing to its readers. Why on earth would you want a system which risks encouraging voters to think about it, when you can just stick your customary X next to whoever your local Tory is?

The piece goes on to refer to "the lies, cynicism and personal insults of the desperate Yes camp", a particularly laughable charge to anyone who's paid even the slightest bit of attention to the No campaign.

For this paper passionately believes that the arguments against the arcane AV system, in which candidates are marked in order of preference, rather than with a simple ‘X’, are overwhelming.
The simple 'X', there! Nice and simple. Not like those complicated 'numbers'. What are they all about? We don't know, and we don't want to know! Let's just do an X, please, so we can be back in time for Emmerdale. It's not all frivolity though, the Mail has actually thought about this shit. In the next paragraph they drop their big fact bombs:

The reallocation of losing votes, until somebody gets 50 per cent, wrecks the historic principle that every citizen has one vote of equal value, which can be counted only once.
This is, broadly speaking, horseshit. Or at least a distraction from the issue. Winning votes remain the most important. If a candidate gets 50% of first preference votes, they win! If not, they don't have such a convincing mandate. AV then starts to count up the second preferences, then third, and so on. If the candidate who didn't quite win is popular as a second choice, then he will win. What AV does is attempt to seek the candidate who meets with the approval of most voters. The Mail prefers the system whereby a candidate with the approval of 30% of the electorate, in a low turnout, would still win even if the other 70% absolutely hated the bastard, simply because their votes were split between the other much nicer candidates.

Again, the counting of second and third preference votes only comes into play if the 'winner' doesn't have a majority. Under first past the post, your vote isn't really as equal as you think. If you don't vote for the winning party, your vote and your opinions count for precisely jack shit. You don't get to influence the election one bit.

Next we get to my favourite bit of the article:

Votes initially given to fringe parties, such as the BNP, will be counted two, three or even four times — and prove decisive in some constituencies.
Now, in the very next paragraph, we get this:

Overwhelmingly, AV is a system which — by requiring candidates to campaign for second, third and fourth preferences — favours bland, common denominator politicians over bold, decisive leaders. It rewards those who cause minimal offence — rather than those who have the courage of their convictions.
So, there you have it. AV is a system which rewards the most inoffensive candidates. But it also rewards the most offensive candidates, such as the BNP! I'm pretty sure you can only make one of these arguments, though perhaps the Mail is putting the "the BNP will win!" argument as their first choice, and expressing a second preference for the contradictory "no offensive politicians will be able to win!" argument. Either that or the Mail doesn't actually consider the BNP offensive, which I suppose is always a possibility.

A moment of decision in the polling booth is replaced by a process of relative judgment, as voters try to decide who they dislike least.
Doesn't that just fill you with terror? Voters would be largely unable to just vote on the spur of the moment by tossing a coin, or voting Tory on a whim because they were given a blue pen and their favourite colour is blue. They'd have to have some actual preferences! Nuances to their views! Imagine a world in which a voter who wants to vote Green, but would also rather keep the Tories out and is painfully aware that the Greens are unlikely to win, was given the ability to express his or her preferences in a simple numerical order? It'd be fucking insane!

Much of the rest of the article is devoted to detailing the pant-soiling nightmare scenario AV might bring, of hung parliaments and their resultant coalitions, with leaders who didn't win the popular vote colluding to form uneasy alliances and breaking manifesto pledges. I don't really feel it's necessary for me to write a clunking punchline to that, is it? Let's just sound the IT'S OBVIOUS WHAT I'M DRIVING AT HERE klaxon and move on.

The article continues to moan about the Coalition government, which obviously could have only happened under the AV we don't have:

The replacement of Trident has been delayed . . . counter-terrorism powers have been weakened . . .  the promise to reduce the number of non-EU migrants to the tens of thousands has been downgraded . . . reform of Labour’s insidious Human Rights Act has been kicked into the long grass.
And the reason the Tories couldn't force through all these promises? Because they didn't have a mandate. There was a hung parliament. The Tories failed to convince the majority of people that these policies were important, and so they had to compromise.

Indeed, the messiest compromise of them all is the referendum itself — an expensive distraction which is taking place for no reason other than Mr Clegg insisted upon it as part of the price of his support.
Of course, the fact that it's only now that we the public get to actually vote on AV is a demonstration of one of the limitations of the first past the post system. We would never have had the option of doing this if the Tories had been in complete control, even if they only had a low percentage of the vote. AV is not a perfect system, but because of the brutally black-and-white nature of FPTP, we're most likely not going to get the choice of alternatives like the single transferable vote or full proportional representation unless we get this, because it's usually not in the interests of parties who rule under FPTP to implement. Only the hung parliament has afforded us this opportunity for now, and we'd probably need another to get a similar chance in future.

The latest estimate is that, of those certain to go to the polls tomorrow, around two-thirds will vote No.

But, alarmingly, more than half of those asked say they may not bother to take part at all. This is where the danger to our democracy truly lies.

For it is certain that the luvvies and political anoraks who support AV — if only as a step to full proportional representation — will turn out in their droves to cast their ballots tomorrow.
Ah, the political anoraks! They'll be out there, voting. With their bloody considered political opinions, the big fucking nerds. Get a life! Just vote for who your dad voted for, or for whoever's promising the most frequent wheelie bin collections.

And, thanks to a disgraceful agreement between Mr Cameron and Mr Clegg, no minimum turnout is required for the referendum to be binding.
...just like no minimum turnout was required for the current election's result to be binding. You know, the one that brought us here. The irony here is something else; the Mail is arguing against AV, a system that tries to appoint a candidate with the broadest majority appeal, while defending a system which actually gave us the no-overall-mandate situation it's complaining about, a system in which the Conservatives failed to get an overall majority on a relatively low turnout.

Of course, no lazy No-to-AV article would be complete without "We will be stuck with a system used by only three countries in the world", and sure enough that appears at the end of the article, enabling you to cross off the last bit of your No-to-AV bingo card. It's just a half-arsed argument that plays into people's fear of change; it adds nothing to the debate about how well AV might actually work and just replaces it with "You don't want to look like Fiji, do you? They're probably fucking MENTAL in Fiji!".

So anyway, there you have it, the AV debate, laid out in idiot's terms by the Mail. To summarise: Vote no to AV, because it's waaaaaaaaay complicated and you couldn't possibly understand it. And it'll bring boring, safe, bland, do-nothing candidates who are also extremist and offensive. Also, NICK CLEGG LIKES IT AND HE IS A DICK!

Actually, that last argument is reasonably compelling.