Head for figures

Statistics, like facts, are neither good nor bad. If they’re accurate, they reflect the world as it is: some will be agreeable, others unwelcome. Many people profess to dislike statistics on principle, as if that tired comparison with lies, and damned lies, were some profound truth. But many others – and not only men, I suspect – would feel lost without a copy of Wisden to flick through, a price list for rare stamps, or a star-gazer’s charts.

It’s not just nerdy hobbyists who obsess over statistics. In business and in public life statistics enjoy an importance that grows greater every week. In heavy industry, the unit cost of widgets always has been a fundamental measurement of the manufacturer’s health. In broadcasting and new media, ratings are everything. Service industries such as banks are lost without some index of customer satisfaction – and now the same goes for schools and prisons. And performance indicators have sprung up across every government department too.

The man responsible for all government statistics is Len Cook, whose name recently appeared at the bottom of the census form sent to households across the UK. He does not wear a cape, or fangs, but Cook is the closest a civil servant can get to the vampiric character on Sesame Street they call The Count: whatever the subject, Cook supplies the numbers. And he does it with “masses of energy and infectious enthusiasm,” as one fellow statistician puts it. “Len is a little dynamo.”

Cook tells me about a famous statistician: Florence Nightingale. Though better known for nursing soldiers in the Crimea, she revolutionised the idea that social phenomena – such as deaths caused by unsanitary conditions – could be measured objecively.

Nightingale introduced new ideas in the collection, tabulation, interpretation and graphical display of descriptive analysis. (One particular innovation was the “polar-area diagram” – imagine a pie chart made up with slices from several different pies, some of them short and others extending way off your plate.) She also expressed her ideas vividly. As Cook puts it, she argued that poor hygeine in British barracks was a killer, “so bad it was like lining up British men on Salisbury plain and shooting them… a brilliant piece of evangelism,” he says.

Sitting beneath an a more recent specimen of evangelism – a photograph of the TV chef Ainsley Harriot helping to promote the census – Cook shows me graphics from a presentation he recently did in New Zealand. They represent the population as a whole, and the Maori population. As he scrolls down the page, the Maori chart widens like a tall pyramid, while the other flares like an irregularly stuffed sausage. “It’s not in the Florence Nightingale league, but it gives you some idea of what we can do…”

One advantage Cook has over Nightingale is his distribution system. “Without the internet, the work of this office [has historically been] delivered to the world in books that maybe [only] 500 people pay £40 for. Now, anybody can see it. Our website is the public face of statistics relating to the fourth largest economy in the world.”

Cook was born in Dunedin, at the southern end of New Zealand’s south island, where his father worked in a brewery. His mother had traditional aspirations for her only male child (a quarter of her entire offspring, as a statistician might put it). She wanted him to be a doctor. But that wasn’t Cook’s plan.

It turned out he was good at maths. Not the higher maths of dreamy philosophers, but the lower maths: adding up, multiplication, that kind of thing. For several years, he was determined to be an engineer – civil or chemical – but in the sixth form he became interested in economics; and a careers adviser, plainly no prophet, told him that economics and engineering would be a useless combination.

After university Cook joined New Zealand’s statistical office and he stayed there for 30 years, for the last eight as chief executive. In 1999, friends on the international statistics circuit urged him to apply for the equivalent job in Britain.

It’s easy to see why Cook might be interested. British statistics were already respected – the Economist, in the mid-90s, ranked them second only to Canada’s for accuracy, and second to the US for the speed at which they’re produced. But in the run-up to the 1997 election the Labour government had promised an overhaul to make them better still. Various number-crunching bodies were to be brought under the control of a single individual, who would take the grand title of National Statistician and enjoy free access to Gordon Brown and Tony Blair.

Cook got the job and moved to England early last year. He’s set up home in Victoria, close to his office in Pimlico. He lives with his partner – who is not, strictly speaking, his wife, as he carefully points out. (Under one of his other titles, Registrar General, Cook controls births, marriages and deaths, and technically is allowed to conduct weddings.)

Professor Denise Lievesley – director of statistics at Unesco and until recently president of the Royal Statistical Society (RSS) – says Cook’s job is no stroll in the park. “You have to be careful. Statistics often tell uncomfortable stories, and the easiest thing to do is to blame the messenger, or find fault with how they have put together the message.”

Tim Holt, Cook’s predecessor, paid with his job for a mistaken prediction, in the summer of 1998, that earnings would soar; causing the Bank of England to raise interest rates against howls of outrage from homeowners and industry, and to lower them again, soon afterwards, when revised figures were released.

So Cook must tread carefully, particularly around the potential for conflict. Government departments are both users and producers statistics, and inevitably there is pressure on statisticians to produce statistics to suit policies, rather than the other way around.

In the 80s and early 90s the Conservative government was constantly accused of manipulating employment figures. Labour, in 1997, promised to protect the integrity of official statistics. As well as consolidating various functions under the National Statistician, it established a regulator, the Statistical Commission.

But there’s some way to go. The treasury still controls the scope and definition of the retail price index despite criticism from the treasury select committee and the RSS. “Given that the chancellor of the exchequer is also minister for national statistics,” says Lievesley, “surely he should set an example by relinquishing ministerial control over the RPI.”

Statisticians are spread among the policy making ministries, making it difficult for Cook to oversee their work. That’s why the Royal Statistical Society wants a code of practice. “Who has prior access to data, and when?” asks Lievesley. “A lot of statistics are supposed to be produced by the Office for National Statistics or the government department with a neutral interpretation. But in practice what tends to happen is that they’re scheduled to come out at 11.30am and before that the Today programme interviews the relevant minister, who has obviously seen the data. So the statistics never come out in a neutral form. There is a political spin from the beginning.

“In Canada the minister receives data in a closed room and can’t give interviews to the press before publication. And for that reason Canadian statistics are percevied to be of high quality. I don’t believe that British statistics are of less integrity, but they’re perceived to be.”

Cook’s fierce commitment to integrity was highlighted in New Zealand when another public servant, the head of the audit office, was caught with his hand in the till. As Dennis Trewin, Cook’s Australian counterpart, remembers it, many people sympathised with the guilty man. But Cook was robustly critical, on the grounds that the public sector as a whole might otherwise lose people’s trust. For this unpopular stance, says Trewin, Cook “got a lot of stick”. He also stood up to politicians, which ultimately earned their respect. From the minister in charge of statistics, Cook received praise with a distinctly antipodean flavour. “Cook, you’re a cheeky bastard, but don’t change.”

Cook is aware the limitations of statistics. He acknowledges that they tend towards stereotypes: it’s entirely possible that no single member of a group possesses the characteristics of the group as a whole. “I don’t think that statistics alone give you wisdom. They’re not a substitute for other pieces of information. You should not take any statistical measure and try to use it without a broader contextual knowledge.

“But one of the real values of statistics is that they shift the debate out of anecdote, intuition and perception to the real nature of the changes we are seeing all about us. In public life you frequently find politicians using anecdote, and I think that statisticians should not get toffee-nosed about the value of anecdote, but if public policy is driven entirely by that – if you’re operating on hunches and intuition – you can only make broad changes.”

Cook’s own special interest is demographics. Within five years, he believes, the average age of women having their first child will rise above 30 for the first time. And the working population of London will change significantly. “There is going to be a huge shift in the ethnic mix. Already around half of the births in London are not white, and in a decade and the half of non-white new job entrants will be predominantly employed by an ageing white population. This reinforces the importance of policies for managing diversity, and actually doing something now.”