← More or Less
More or Less
Numbers of 2012
00:27:23
A guide to 2012 in numbers - the most informative, interesting and idiosyncratic of by...
Loading summary
Transcript51 lines
- [00:00]
Pippa Wells
Thank you for downloading this edition of
- [00:02]
Tim Harford
More or less from BBC Radio 4.
- [00:04]
Bill Edgar
Here's Tim Harford.
- [00:06]
Tim Harford
Hello and welcome to More or Less. This week, our special review of the year through the excellent, extraordinary, eccentric numbers of 2012. And who better to start the programme than one of the BBC's most excellent, extraordinary and some might say, eccentric presenters?
- [00:23]
Robert Peston
I'm Robert Paston, the BBC's business editor. My statistic of the year was given to me by Jean Claude Trichet. He's the former president of the European Central bank and he told me about the 80:20 ratio for Europe and the 20:80 ratio for the United States. In Europe, roughly Speaking, banks provide 80% of the finance needed by businesses and households, whereas in the States, banks provide only 20% of the finance required by businesses and households. Why does this matter? Well, it tells us a huge amount about the crisis in the Eurozone and it also tells us a huge amount about why the economy of Britain and many of the economies of Europe in general have singularly failed to recover since the great crash of 2008. Whereas, although the American economy isn't off to the races, it has grown rather more strongly than the European economies. So let's explain the significance of being so dependent on banks. When banks are weak, when they have a shortage of capital, which is palpably true of many Eurozone banks, and arguably, according to the governor of the bank of England, true of British banks, they are reluctant to lend. What they try and do is they actually try and shrink. They try to use the ghastly phrase, they deleverage. And when banks deleverage, when money is not going to households and businesses, frankly, households and businesses can't spend, they can't invest and the economy can't grow. And also, when you have banks playing such an important role, it also turns out banks are enormous. They're so big that when they get into trouble, even governments like the Spanish government find it very difficult to afford to bail them out. So the 80:20 ratio tells you two two fantastically important things. It tells you why the European economy has been so sluggish and it also tells you why governments like Spain and Ireland have been on the brink of bankruptcy, because their banks are so big they can't afford to rescue them.
- [02:38]
Pippa Wells
My name is Pippa Wells and I'm a physicist working at CERN in Geneva. At the end of last year, we had a tantalizing hint that there might be a new signal emerging in the data which that we'd been recording at the Large Hadron Collider. And we're trying to answer various questions about how matter works. Why do elementary particles have mass? In the 1960s, several theorists proposed a mechanism to give mass to fundamental particles. And Peter Higgs realized that this new field that they needed would actually produce itself a new particle. The Higgs boson. Over the last decades, people have been searching for the Higgs boson and, and we were really hoping that this year we'd confirm it and be able to announce the discovery of the Higgs boson. So my number for this year was five. That's the significance we had to reach in order to be able to say that we'd really found something. If you make a measurement, the data tends to spread around the central value. And the width of that spread is characterised by the standard deviation, which we call sigma, the Greek letter sigma. About two thirds of the data will be within one sigma of the central value. But we need to get five sigma away from the central value to be able to say we've actually made a significant discovery. This means that the events we're seeing from well known processes could not have fluctuated up to give a fake signal with a probability of more than 1 in 3.5 million. To everyone's astonishment, both experiments had reached the five sigma level. And both with the same mass for the Higgs boson of around 125GV. So that's about 125 times heavier than than a hydrogen atom. So the fact both experiments had seen it, and what's more, with consistent mass in the two experiments, we were absolutely confident that we'd found something new.
- [04:18]
John Rodder
Conclude by saying that we have observed a new boson with a mass of 125.3 plus or minus 0.6 GeV.
- [04:28]
Tim Harford
And as the experiments at CERN continue, Pippa and the team have found an even stronger signal that a form of the Higgs boson exists.
- [04:38]
Bill Edgar
I am Bill Edgar. I cover statistics in football for the Times and I'm the author of Back of the net. 100 golden goals. We're just reaching the 20th anniversary of the last time an openly gay footballer played in the league. In England, that's the top four divisions. Justin Fashioner was the last one and his career in England ended at Torquay United. So I wanted to work out what the chances were that no league footballer in that time has been gay, whether openly so or not. Assuming there was no reason for the proportion of gay footballers to differ from the proportion of gay people in society as a whole, many surveys have tried to decide what proportion of British Men are gay. Answers have ranged wildly from 1 1/2% to 6%. So I took the lowest figure of 1 1/2 percent. In that 20 years, 13,600 players have appeared in the league. Now, the chance of picking 13,600 men randomly from the whole population and none of those being gay is one chance in five times 10 to the power of 90. That's one in five with 90 noughts. To make a slightly more understandable number, I then looked at Premier League players, of which there have been 3,200 in the past 20 years. The chances of there being no gay footballer among those is 1 in 10 to the power of 21, which is 1 and 21 naughts. Now, to illustrate exactly how unlikely that is, if the entire Earth's surface was covered with drawing pins edge to edge, the chance of you picking up one specific drawing pin at random would be 200 times higher than the chance of there being no gay player in the Premier League. Over the past 20 years, no doubt there have been many gay players in the Premier League, but perhaps not as many proportionately as in the country. Overall, we don't really know. Now, this situation is not confined to England. There is no apparent record anywhere in the world of a top class footballer who is openly gay. The nearest example would seem to be Anton Hussain, a part time player in the Swedish third division, whose father Glenn once played for Liverpool. All in all, it's a rather sad commentary on prejudice in the sporting world.
- [07:01]
Tim Harford
Bill Edgar from the Times. In women's football, the record is better with a number of American players and the American and England team's coaches being openly gay. Not all sports seem to have this problem. There were, for example, 21 openly gay competitors in this little sporting occasion.
- [07:24]
Gabriela Labresche
Hello, my name is Gabriela labresche and I work for Decision Technology. I'm a sports analyst. We decided to look into the home advantage, partly because of looking at Greece and Australia's performance in the recent Olympics, but also because everybody's interested to see how Team GP will be doing in Rio 2016. As soon as a country is awarded the Olympics, they start to pump funding into sports. And this means that they have athletes that are better prepared and better funded, have better equipment and therefore have a better chance of doing well. Additionally, there's a home advantage effect. The home Olympics is really going to be the peak of a nation's performance. We wanted to see what the rate of decay was after the Olympics and to see if after the Olympics these teams were dropping off faster than they should have been, or if in fact they were doing better and managing to maintain that Olympic glow that little bit longer. So what this involves is looking at the time from when they held an Olympics and and games afterwards and trying to fit an exponential function to it. And then we can compare the different rates. What this gives us is a total medal decay rate and additionally a gold medal decay rate. And then you can compare how different countries have performed after their Olympics. The standout appalling performance has been Greece since Athens 2004. They have a decay rate of -1.04, which is double the largest of any of the previous medal decay rates. They've really fallen through the floor. There's no other way to describe it. The interesting thing is when you look at Australia, because the Australian media got very upset over the summer about perceived poor performance. And the answer is that in terms of total medals, they haven't done that much worse than we would expect. It's perfectly in line with the average decay rate. However, in gold medals, they're slightly below. The question which I found most interesting is one of the reasons that the London 2012 Olympics were sold on was the legacy it would leave. But can you really have a legacy? Does every host nation, after hosting an Olympics, suddenly become a nation of much better sports people? Is that in any way possible or statistically proven? The answer is no. Twelve years on, the probability of them winning that many medals is significant. But if you look 16 years on, it's not significant, meaning that statistically there's no significant legacy. Sixteen years or more afterwards, you can't tell the difference between a nation that's hosted the Olympics and hasn't hosted the Olympics in their recent past. So what the legacy of Team GB will be is definitely unclear for the future.
- [10:23]
Helen Joyce
I'm Helen Joyce. I'm the Economist's Brazil correspondent, based in so Paulo, and I moved here in July 2010. My number of 2012, the number that surprised me most during this year is the average age of retirement of a Brazilian working in the private sector. It's 53. Now, Brazilians don't die at 55 or 58 or 60. They live well into their 70s. So that means that now Brazilians live about 20 years in their retirement on average. You can imagine how much their pension system cost. So the system I'm talking about, it's for private sector workers, but it's actually run by the government. It's like national insurance in the uk. You pay in, it's compulsory, it's pay as you go. So it's current pensioners who are living off your contributions. And the hope is that in the future you're going to live off other workers contributions. There's a similar system for public sector workers, but the point is these are all run by the government and backed by the government. I looked into the numbers. The only country, the only big country in the world that spends more on pensions than Brazil is Italy. And Italy is well known to be having a pensions crisis. So that has two big implications. One of them is current. It means that all other sorts of government spending are starved and it's children who lose out in Brazil. The second implication is for the future. Every year Brazil is spending more on pensions. And calculations suggest that if they were to keep the current System by about 2050, workers will be having to put nearly their entire salaries into pensions. Not their pensions, mind you. It's a pay as you go system. Other people's pensions workers would be working solely to support old people and keeping nothing for themselves. Now, obviously that can't happen. Things that can't go on don't go on. So this is going to come to an end. But when? So Brazil has certainly got less than 10 years, I would say, to reform its system in a sensible manner. You know, allowing people who are already in the system and keep the benefits that they have accrued already do it gradually, bit by bit, raise the retirement age a year at a time, otherwise they're going to get to a stage of complete deadlock. Nothing will happen until it nearly collapses.
- [12:30]
Tim Harford
Helen Joyce, you're listening to More or less. I'm Tim Harford and we'd love to hear your numbers of the year. Do email them in and any numerical musings you may have over your cold turkey and mince pies. Our email address is more or lessbc.co.uk.
- [12:48]
Jack Straw
jack Straw here. I'm the Member of Parliament for Blackburn. My number is 1901 and 1901 is the number of road fatalities last year, 2011. It's a really interesting number in my opinion, because it illustrates the dramatic improvement in road safety that has happened in the last 40 years. And the last 80 years, there was one fatality for every 280 vehicles on the road 80 years ago. It's now one fatality for every 1780 vehicles. So six times better chance of not having a fatality in your vehicle than you had. The really interesting thing from my point of view is how, as a result of controversial measures like the breathalyser, seat belts and improvements in the safety of vehicles, and the safety of junctions. The number of fatalities has dropped from a peak of touching 8,000 forty years ago down to below 2,000, even though the number of vehicles has shot up
- [13:51]
Pippa Wells
now.
- [13:51]
Tim Harford
You might wonder why we've invited Jack Straw, a politician, into our hallowed studio to talk about his numbers of the year. There is a good reason for this. He's thumbing through a table of data sitting right in front of me. When we heard that Jack Straw was willing to come on, more or less, I was tempted to say, we should get back to the Today program where you belong. But then we read his autobiography, Last Man Standing, published in September of 2012, where he confesses to being obsessed with mathematics. An anorak, a man who in his spare time, as he swims up and down the pool, will calculate prime numbers just to give himself something to do. Jack, welcome to the programme.
- [14:31]
Jack Straw
Thank you very much. And welcome to all the anoraks who are listening to the programme.
- [14:35]
Tim Harford
We embrace the anorak here on more or less calculating prime numbers. That's a bit pure maths for us. We prefer the rough and tumble of statistics. There can't be many people who do that while they're swimming.
- [14:44]
Jack Straw
Well, it's amazing how quickly the 40 lengths go past. So that's why I do it. And it keeps your brain active as well, presumably.
- [14:50]
Tim Harford
It also helps in the clash and refutation of political debate, if you can just pull out the statistics rather than just making them up, as most politicians would do, to actually remember the relevant statistics. Presumably this actually starts to become a relevant political skill.
- [15:03]
Jack Straw
There are many occasions when I'd be looking through a set of data, for example, for a written parliamentary answer, and quickly scan it and realise it didn't add up. And I can remember one slightly terrifying occasion when I was presented with a set of figures about the numbers of police officers in training. I looked at these figures, I thought, these don't add up. And they are different from a set of figures that I gave about two months ago. And sure enough, they had left off all the trainees in the Metropolitan Police District because they were not part of national police training for these purposes, so was able to send them back. It would have been a really bad story if I then issued, and then they would have had to send an apology. It would have been worse if the data had exaggerated the truth rather than underestimated it. But this kind of thing could happen all the time.
- [15:51]
Tim Harford
I mean, we should give you credit for being an evangelist for statistical independence. You are, I believe, an honorary fellow of the Royal Statistical Society.
- [15:59]
Jack Straw
I'm an honorary fellow of the Royal Statistical Society, so I feel like I
- [16:01]
Tim Harford
should be tremendously kind to you, but I still, I have to push a little. As I've admitted, we haven't really been able to catch you out in any major statistical blunders.
- [16:10]
Jack Straw
Hardly at all, actually.
- [16:11]
Tim Harford
Listeners, if you can remember any, do write in and tell us. But you were part of a government where there were statistical errors and messes and so on. I mean, one thing that really springs to mind is the immigration data after an EU accession where the borders were opened to Eastern European workers and many of them came up. There were forecasts made. This was not your department, you were in the Foreign Office at the time. But there were forecasts made that were woefully, woefully. And when you see the data doesn't add up, the forecasts don't add up. You know, you go into the Cabinet meeting and you think these numbers are important, what do you say?
- [16:44]
Jack Straw
That raises some very interesting issues about research methodology rather than the venality of politicians. Because this was not a policy area where we plucked figures from the air. There were really serious work which was then published. The prevailing sense, not just amongst the politicians but amongst, more importantly amongst statisticians, was that these predictions were about the best available. In fact, they turned out to be woefully off beam. So there's a PhD there for somebody to look at why that went wrong. But so on that occasion it wasn't the politicians who made the initial mistake.
- [17:17]
Tim Harford
There's a story in your book about your introduction to the Poisson distribution. I love this sort of mix of politics and. Tell us about that.
- [17:24]
Jack Straw
Well, I had a charmed life for the first 20 months as same secretary and people were writing me up as somebody kind of walk on water at the time. I sent a fax back before emails to my special advisor to say the gods are going to get me. And then they did. And at a whole series of events just bunched together one after another, big and small. People couldn't get a passport, the immigration system computer collapsed, I had to release a terrorist. All these things bunched together and Paul Wiles, who was a very good statistician and head of stats in the Home Office, came to see me and talked to me about the Poisson distribution and how this helped to explain how events which were random might be bunched together. And he said, I think you're the victim or an illustration of the Poisson distribution.
- [18:13]
Tim Harford
So very reassuring because it means none of it was Your fault?
- [18:16]
Jack Straw
Oh no, well, sort of. Some of it was my fault, some of it was just out the blue. But it was a nice light hearted interlude from the fact that these exocets were just coming in the window from a clear blue sky. Bang. And another one would come in. Bang. One week I had to make three emergency statements in the House of Commons apologising for yet another mess up.
- [18:37]
Tim Harford
Wonderful statistics. They're so powerful and yet they're also so consoling.
- [18:40]
Jack Straw
They are consoling and they're good fun.
- [18:42]
Tim Harford
And our next number is from someone with even more statistical eminence than the self proclaimed anorak Jack Straw, our national statistician, Jill Matheson.
- [18:52]
Jill Matheson
My number of the year has to be 56.1 million. That's the population of England and Wales announced in July from the latest census. That's nearly 4 million people more than a decade earlier. And we know, we've known for some time that we're an aging population with one in six of us now aged 65 or over. But what's also interesting is the other end of the age spectrum with 10.6 million children in England and Wales, that's more than there have ever been in the country before. And in particular there's been a big increase in the number of very young children under the age of five, over 400,000 more than 10 years ago. One of the interesting findings from the census release in December that got us all talking in the office was the increase in the proportion of the population who said they had no religion. That was 15% in 2001 and it's gone up to 25% a quarter of us in 2011. In fact, the place in the country that had the highest proportion of people saying they had no religion was Norwich. So 4 out of 10 people in Norwich said they had no religion.
- [20:10]
John Rodder
My name is John Rodder, I'm a hydrologist. It's been remarkable. 2012 and the first months of the year saw the drought which had continued from 2010 onwards. And it was a drought, a remarkable drought covering quite a lot of the country. And by March it has intensified and was covering much of the southern half of the uk. And then of course, hose pipe bands were introduced on the 5th of April across much of southern England. And then within days then it started to rain and it was an absolute deluge which turned the summer into the wettest for a hundred years. The rainfall for the month of April was 229% of average, which is an enormous amount of water. It was 134 millimetres and floods replaced the drought. And in some places where there was even a hose pipe ban still on, it was remarkable. Turning back to the end of March, I estimated at that time there was a shortfall in the rainfall across the country of about 120 millimeters. That was what was needed to bring us up to the sort of average figure for the time of year. But in fact, the amount of rain in April and the succeeding months totaled about 600 millimeters. So the shortfall of 120 was readily and rapidly made up by these large amounts of rain.
- [21:49]
Tim Harford
And for our next number of the year, I'm joined in the Studio by Dr. James Grime of the Millennium Mathematics Project at the University of Cambridge. James, what's your number of the year?
- [21:59]
James Grime
My number of the year goes back all the way to the first of when a paper was released which solved one of the unsolved problems, which was, how many clues do you need to solve a Sudoku?
- [22:09]
Tim Harford
And I've seen these problems, and you get a few numbers as seeds that give you hints and get you started. And if it's an easy Sudoku problem, you'll get a lot. And if it's a fiendish Sudoku problem, you won't get very many. And so the question is, what is the least number of seeds you have to be given to actually solve this problem? And what's the answer?
- [22:29]
James Grime
So it turns out that the answer is 17, which means not. Not all Sudokus can be solved with just 17 clues. But there are no Sudoku grids that can be solved with 16 or fewer clues. And when I say solved, I mean solved uniquely. So there's only one solution at the end.
- [22:48]
Tim Harford
So 17. It doesn't get any more fiendish than 17.
- [22:51]
James Grime
It doesn't. The Sudoku fans had found puzzles that could be solved with 17 clues, but they weren't able to find any that could be solved with 16 or fewer. So they started to conjecture that maybe 17 is the minimum number. Maybe that's as good as it gets. But to solve that would involve checking every Sudoku solution and then check that for every possible 16 clue.
- [23:17]
Tim Harford
Starting position doesn't sound that hard. How many combinations are there?
- [23:20]
James Grime
The total number of Sudoku Solutions is 6700 million million million. The total number of 16 clues. Then on top of that, that you would have to check for each one would be another 33 million billion. So it's a massive problem, and even with a computer, it was thought it would take 300,000 years to check all the possible solutions and check them for all the possible 16 clues to show that none of them work. A mathematician called Gary Maguire at the University College Dublin used a clever idea so that he didn't have to check every possible Sudoku solution and he didn't have to check every possible 16 clue starting position and he reduced the problem mathematically. And then what he was left with, he checked with a computer program which ran for the whole of 2011 and then he released his results first of January 2012.
- [24:17]
Gillian Tett
In my view, the most eye popping statistic in 2012 was the amount of money that the Americans spent on their election. They spent about $2.5 billion on the direction presidential campaign. But when you throw in all the money for the Senate races and the House of Representatives, you end up with about $6 billion worth of total spending on the campaigns. And that number is absolutely huge compared to other Western countries. I mean if you work it out per person in America, it's about $18. If you compare it to the last general election in the UK was 50p per person and the last Canadian election was more like eight Canadian dollars a person. However, there's a rather interesting catch. If you look at the sheer size of the US economy and you look at the amount that's invested in elections, then actually the American number may not be so large after all. And my other eye popping statistic of the year was that the Americans actually spent $7 billion on potato chips this year and they spend about $8 billion on the Halloween celebrations. So there are some people who say, well actually spending $6 billion on an election is not that bad, at least when you compare it to the other things that Americans could spend money on in a consumer culture.
- [25:35]
Tim Harford
Gillian Tett, columnist and assistant editor of the Financial Times. And now David Spiegelhalter, loyal contributor to More or less professor for the Public Understanding of Risk at Cambridge University.
- [25:47]
David Spiegelhalter
My notable number, not really my favourite is 50,000. According to the Panorama program that was on 10th of September. This is the estimated number of deaths of pensioners in England that would be avoided over the next 10 years if a minimum alcohol price of 50p per unit was set. Now, even before the program was aired, there were bloggers saying this figure seemed really completely implausible. Yeah, it's about 80% of all the deaths of pensioners that according to the Office for National Statistics were even remotely associated with alcohol. Then a few weeks later the BBC put up a little notice on their website saying that, well, unfortunately due to human error of the statistical team at Sheffield University that did their analysis. The figure shouldn't have been 50,000. It should have been 11,500. And they re edited the program now. In fact, this number still seems rather large. Estimates for lives saved due to minimum alcohol pricing are generally based on a model built at Sheffield university. But this 50,000 figure didn't seem to fit even with their previous analyses.
- [26:51]
Tim Harford
Professor David Spiegelhalter and that's all we have time for this week, and indeed this year. If you've spotted any statistical sins or you have your own number of the year to contribute, please email more or lessbc.co.uk and if you want to know more about the programme, please go to BBC.co.uk more or less happy New Year.
- [27:14]
Bill Edgar
More or Less was presented by Tim
- [27:16]
Tim Harford
Harford, the Financial Times undercover economist.
- [27:19]
Pippa Wells
It was produced by Charlotte Pritchard and the editor was Richard Varden.