Five things to ask your electoral candidate

In the run-up to the recent EU elections I received a letter from my local MP, Ruth Cadbury, to her constituents. It included a survey (also available online) canvassing our views on knife crime, specifically whether any of us had been affected by it, what we thought the causes were and what should be done about it.

Nobody would disagree that knife crime is a serious and growing problem in London, and on the face of it it is reassuring to hear that she is taking it seriously. But think about this for a moment. Shouldn’t we be the ones asking her about knife crime? After all, she is the one in a position to do something about it, and as the electorate it is up to us to vote for the candidate whose policies we agree with.

If she is serious about tackling knife crime then are her constituents really the best people to ask? Personally, if I had a patient come through the doors with a stab wound I would know enough to treat the blood loss while I was waiting for the surgeons to arrive, and I might be able to put in a chest drain, but there my expertise ends. Surely she should be seeking the advice of criminologists and other professionals who have investigated the problem and have some specialist expertise. She should be looking at other cities around the world which have had similar problems, how they have dealt with them, and with what success (for instance Glasgow, who took the unorthodox step of treating knife crime as a public health problem, and where it was once the murder capital of Europe, in the past 15 years hospital admissions for knife wounds have reduced by two-thirds).

I rather fear that she might be basing policies on what is popular with her constituents rather than on what works. That isn’t the way to win my vote. Instead, I have a few questions that I would like to put to anybody knocking on my door with a view to representing me in Parliament.

1. What is 60% of 40?
   • Politicians are quoting percentages all the time. I just want to check that they understand them, as a lot of people don’t (the correct answer is 24).
2. I have just paid £1,000 for a television. How much of that was VAT?
   • This is a simple test to gauge their understanding of the taxes that they will be imposing on us (most of which are a lot more complicated than VAT). The standard rate of VAT on most goods and services is 20% so in this case the cost of the television was £833 and the VAT was £167 (which is 20% of £833 rounded to the nearest pound).
3. Would you prefer a diesel or a petrol car?
   • There has been a lot of discussion recently about the risks to health of pollution from diesel exhaust, which (unlike petrol) contains fine particles and nitrogen oxides, and having a lot of diesel-powered vehicles together in a city clearly poses a public health problem. On the other hand, diesel engines are more efficient than petrol engines, so their carbon emissions are lower. They are also much simpler, more reliable and last longer, which in turn has some bearing on their carbon footprint, so this question touches on how they might balance local issues (which would favour petrol) against global issues (favouring diesel).
   • Of course it is a lot complicated than that, especially taking into account factors such as cutting down rainforests in order to grow palm oil to make biodiesel etc. (I gather that this is an unintended consequence of EU policies encouraging “green” fuels).
   • Electric-powered vehicles are arguably better than either, depending on where the electricity comes from and emissions from building the vehicle in the first place and from disposing of it at the end of its life.
   • Much better is public transport, which is something else you could ask about.
4. What is the most effective way to approach the problems arising from illicit drug use?
   • While I would not in any way want to deny the dangers to health of taking drugs, most of the problems arise not from the drugs themselves, but from how they are supplied. There is a lot of money to be made by criminal gangs at every point in the supply chain. Then the users themselves often resort to crime to fund their habit. Inconsistent quality-control leads to accidental overdoses…
   • Legalising the drugs, at least to the extent of regulating their supply, would get rid of most of this at a stroke. But most politicians feel that such a policy wouldn’t be a good way of getting re-elected. Indeed, when the Government Drugs Adviser and chair of the Advisory Council on the Misuse of Drugs (ACMD), Professor David Nutt, provided research evidence that conflicted with the Government’s view of things they sacked him.
   • So what I am really asking here is: Does the candidate favour evidence or ideology?
5. According to the Office for National Statistics, unemployment has gone up by 1,000 between the 2nd and 4th quarters of 2018. Would you care to comment on that?
   • I got these figures from the Office for National Statistics Web site. They are updated every month and there is a wealth of fascinating data available there, including details of their methods. The Government relies on the ONS to provide the figures that policies are based on, so how well does the candidate understand those figures?
   • If they start going on about how this is an example of the success or failure of this or that Government policy, then they are missing the point entirely, and they are also wrong.
   • In 2018 Q2 the number unemployed is given as 1,362,000 and in Q4 it is 1,363,000. In Q3 the figure is 1,377,000 and in 2019 Q1 it was 1,298,000, so you can see that I have been slightly devious in choosing my figures. However, what is more interesting is how these numbers are arrived at in the first place. The ONS does not, in fact, keep track of each and every one of us with regard to our employment, but instead takes what it hopes is a representative sample (interviewing about 100,000 people from 40,000 households) and then scales up the numbers. Even if their sample is a reasonably good representation of the coutry as a whole, there will always be a random element, and if you repeat the exercise several times you will get a slightly different result each time.
   • There are standard mathematical techniques for dealing with this uncertainty and these allow you to calculate a range of values (based on the sample size, the size of the population, the number of unemployed in the sample, and how certain you want to be of the result), within which the true value lies. The ONS explains this on their Web site and even gives an example for the three months from August to October 2017 (follow the link and scroll down to Sampling Variability) when the number unemployed was estimated to be 1,429,000, plus or minus 72,000. In other words, the true number unemployed was (probably) somewhere between 1,357,000 and 1,501,000. (Note that I say probably. I can say with certainty that the number unemployed is somewhere between zero and the total number of people in the UK, but this isn’t very useful. As soon as you start to narrow it down, a little uncertainty creeps in…)
   • So coming back to the original question, an apparent rise in unemployment of 1,000 could really be a rise of the order of 70,000, or for that matter a fall of 70,000. That’s why I didn’t choose the most recent quarter as the figures give a fall in unemployment of 74,000 between the last quarter of 2018 and the first quarter of 2019, which really is a fall however you calculate it.
   • I wouldn’t expect a parliamentary candidate to have all the figures at their fingertips – once in post they have researchers and advisers for that. But I would want them to have some idea of where the figures come from, and a feel for how they behave. I would at least hope that they would realise that you can’t be accurate to within 1,000 when you are dealing with numbers in the millions (this probably doesn’t apply to tax inspectors).

So there you have it – a little quiz to see if the person after your vote is competent in basic arithmetic, whether they understand anything about taxes, whether they favour local or global issues, whether they can take advice from experts and whether they have any understanding at all of the statistics that they will use to lie to you once they get elected.

Good luck!