jueves, 9 de febrero de 2017

When does telling the truth become lying? (you could even ask the BBC)

This article is a bit of a detour from my usual themes, as it does not directly relate to international politics, although the underlying theme has had, in my opinion, a huge effect on political decisions we have seen electorates take in the last few months.

The article is prompted by the main story run by BBC Radio 5 Live, in the UK, in the news items during the breakfast program on Wednesday 8th of February. The headline was:
UK Police forces have spent £22Mn paying informants over the last 5 years. During the same period, police budgets have been cut by 18%. This was followed by the question, by the presenter, of ‘Is this good use of public money?’.

The first statement is a news item, as the information has become available through the publication of a Freedom Of Information Request appertaining to the financial years 2011-2015 (for an example of what these look like, you can read the FOI reply by the Met Police, the biggest spender, here. An internet search will give you access to the replies of the other police forces, if you have the time and energy to combe through them).  The second statement is not news, and it is completely irrelevant to the news in the first statement. It is clearly designed to generate an emotional response in listeners. A lot of money is being paid out to informants by the police while their total budget is being reduced. The program anchors seemed quite outraged by this use of public money.

The news item offered very little further detail, other than the biggest spender was the Met, with around £5Mn, with the Northern Ireland Police force in second, with £2Mn.

All the statements in this news item are correct, or close to correct (I have found different figures for the reduction of police budgets, but all between 18% and 21%, so let us accept that the data are primarily correct).

So, the message seems to be police are giving a lot of money to informants, money they don’t have. The second statement, about the 18% reduction, must be there to elicit a feeling in the audience that police don’t have that money to pay out. Some may argue that it is there to provide context, but it is not, it is there to take out of context. It cannot provide context because it is in no way comparable or related to the figure in the first statement, absolute monetary values cannot be compared to percentages in the absence of any other information.

Let’s now look at what may have been appropriate context and an appropriate framing of the reporting of the results of the Freedom Of Information Request, had the aim of the BBC been to inform the public:

First question is, is the money being spent on paying informants a large sum? There are 45 Police Forces in the UK, of which 43 replied to the FOIR. The period is 5 years. Therefore, £4.4Mn are being paid to informants per year, an average of £102,326 per police force (£8,527 per month per police force). The number looks a lot smaller when presented like this.

To put it further into context, we could look at total police budgets, and whether the money paid to informants is significant within those budgets. The budget for police forces in the UK for 2015 was (you can find these numbers in many places, this PDF is a police funding report to the UK Parliament):
  • England: £7.4Bn
  • Scotland: £1,064Mn
  • Wales: £743Mn
  • Northern Ireland: £661.5Mn
  • Total: £9,868Bn per year.

We can now get some understanding of what the number actually means. The money paid out to informants is 0.0446% of the total police budget. To translate that to numbers most people can understand, for every £1,000 the police spend, 44pence are paid out to informants. It is now starting to be difficult to understand how this item makes it into the major headline of a mainstream news program. The equivalent in household terms would be a member of a couple sending repeated emails, every half hour, to their partner to explain that they are going to spend £1.20 of their £3,000 monthly salary on something necessary but unpopular. Or the headteacher of a small rural school convening an urgent meeting of the governors to inform them that he intends to spend £22 of the £50,000 budget in extracurricular education, and the governors starting a consultation to see whether that expense is sensible.

Another way to look at this is, if the police did not spend this money in informants, what could the police do with it. Could they police better? Let’s look at this, and for this, we can focus on the commonly used metric of number of policemen on the street. The UK has just over 202,868 personnel measured in Full Time Equivalents (basically, this means that two part-time people working half time count as one Full Time Equivalent). The average cost per FTE, dividing the total budget by the number of FTEs is £48,494.59 per annum (don’t jump at this number, this is not the salary, is the full cost of providing policing, per head).

So, let’s imagine that police stopped paying informants and added more policemen to do the policing. The money saved on informant payments would allow to pay for 90 more policemen across the country, taking the total number from 202,868 to 202,958 (it is not only you who has to look twice at the number to tell the difference, in practical terms, there is no difference).

Now that we have established that the police are paying a very small and irrelevant amount of money to informants, let us look at the naming and shaming. The biggest spender is the Met, over £5Mn over 5 years. Northern Ireland Police £2Mn. No more information was given out, the numbers get small, so not worth reporting. Let me tell you, for example, that West Midlands has paid £974,953 over 4 years. So, the Met is paying 5 times more than West Midlands. The Met, as the information is presented by the BBC, is paying 2.5 times more than the second biggest spender! How profligate!
Now let us try to put that into context, which is actually very easy. It is obvious that the amount of policing relates to the size of the police force and to the population policed. Every item of policing would be expected to be more or less proportional in amount to these context metrics. If we work on the basis that population is the key metric, it turns out that the Met is not the highest spending, Northern Ireland spends a lot more (to be expected I think in light of the previous troubles in the territory and the fact that some of these remain under the surface).


Finally, let us look at the claim that these amounts have been spent while budgets for policing were reduced by 18%. Putting these two facts in the same headline generates the impression that police are throwing money at informants even when they have less. But, how is the money paid to informants evolving? If we look at the numbers, once again in the FOIRs, we find:


  • Met Police have reduced the money paid to informants by 15% since 2012, and by 49.5% since 2009.
  • Northern Ireland Police have reduced the money paid to informants by 31.5% since 2011.
  • West Midlands Police reduced the money paid to informants by 46% between 2011 and 2014.
The police force are actually reducing the amount paid to informants faster than their budgets are reducing.

It turns out that the BBC could have reported: ‘The Freedom Of Information Requests from UK Police Forces have been received. 43 out of 45 forces replied, which is fantastic. The figures show that the police are spending very little amounts in paying informants, and that they are successfully reducing these amounts faster than their overall budget is reducing. However, since the BBC asked the wrong question in the FOIR, we still don’t know how the value of the results produced by those informants actually relates to the money spent’ (that, by the way, is the question I would really have liked to be able to answer, it is the only one that matters, although looking at the figures I am quite happy to assume that probably the value generated by the informants is bigger than the money spent, who better than the police to evaluate that?).

This is as far as I feel it is worth carrying on with this analysis of BBC reporting style. The question is: why does the BBC present this information in the way they do, and not in the suggested alternative way. And this is not a problem on this specific issue. The way this news was reported is consistent with the way most news are reported. Every day, there are several examples of so call reporting that I could take to task in a similar way. There are extensive studies about the reason for this. The statement is ‘Good news doesn’t sell’, in several variations (if you want to read some research on this, you can look at this Columbian Journalistic Review of a Pew Research Center Study for People and the Press, on this subject, and that is just one example, there is a lot of academic research on this subject).

I am not going to go now into the consequences of the systematic misrepresentation of information by mainstream media. In fact, when you look at the messages behind Brexit and the Donald Trump campaign (of course I had to bring these 2 into this article!), they are anchored on misperceptions generated by these systematic bad news. Neither of them would have happened without them. This blog will aim to illustrate that, amongst other many things, but this is enough for today.

I leave you with the scary thought that the BBC has a very good reputation for neutral news reporting and, due to its funding model, has no real need to sell or prove audience figures. If the BBC is systematically doing this, what can we expect from its commercial counterparts?