Pity Greece’s statistician taking blame

Posted on August 5, 2016

Farmers demonstrate outside the Greek parliament during a protest against pension reform and tax issues, on February 12, 2016.
Fears that Greece will exit the eurozone, a "Grexit", could revive if Greek authorities do not come up with "credible" reforms, notably on pensions, a senior IMF official said February 11. / AFP / ARIS MESSINIS (Photo credit should read ARIS MESSINIS/AFP/Getty Images)©AFP

In the early 2000s, I lived in Paris and got to observe President Jacques Chirac’s approach to EU deficit rules. Every year, France would run up a larger-than-allowed deficit and Mr Chirac would dispatch his prime minister off to Brussels with a memo that I imagined consisted of a single word, which so beautifully expresses Gallic exasperation: bof. The European Commission would back off and Mr Chirac could place a celebratory call to the Elysée’s sommelier.

Knowing this history of European budgetary chicanery, Andreas Georgiou, former head of Greece’s national statical office, can only feel thoroughly kippered. In 2010, as Greece’s economic problems snowballed, he was recruited from the International Monetary Fund to bring some sense to the country’s discredited statistical office. The numbers were ugly, and he reported as much. The 2009 deficit was higher than imagined — 15.8 per cent versus 12.8 per cent. The commission and the European Central Bank accepted his numbers and this allowed Greece’s creditors to apply severe bailout terms.

    If you ever thought statistics a dull profession, think again. For the alleged crime of exaggerating the deficit in favour of the creditors, Greece’s supreme court ruled this week that Mr Georgiou must face trial on charges which carry a prison term of up to 10 years. His political enemies on the left are delighted. Rather than examining Greece’s years of fiscal mismanagement, they would rather pin their budgetary debacle on a single statistician.

    It is part of Mr Georgiou’s misfortune that his profession has been so debased, often through no fault of its own. Mark Twain liked to repeat Disraeli’s quip about “lies, damned lies and statistics”, but in his memoir Life on the Mississippi later reflected: “There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.”

    We saw this in the recent Chilcot report into Britain’s decision to go to war in Iraq. This detailed how time and again factual and statistical trifles were whipped up into hard and fast statements of fact to bolster the political case for war. Did it matter if the child mortality rates in Iraq under Saddam Hussein were exceedingly difficult to estimate? Not if you were out to depict him as a monster. While the politicians railed about Saddam letting his country’s children die, investigators and statisticians were left to mumble from the wings, “that’s not exactly what we said . . .”

    The legacy of public suspicion about “dodgy dossiers” haunted the recent Brexit referendum. An army of Treasury statisticians could not persuade voters to believe the claim that every household would be £4,300 worse off by 2030. It seemed too precise a number attached to too distant a date, especially given that the government’s 12-month estimates do not always hold true.

    To the despair of serious-minded statisticians, the most widely read book in the history of their profession was written in 1954 by a freelance journalist named Darrell Huff. How to Lie With Statistics is a larky guide, with chapters titles like “The Well-Chosen Average”, “Much Ado About Practically Nothing” and “How to Talk Back to a Statistic”.

    One of the very first admonitions I received as a journalist was never to propose a story based on a survey. They almost always fall under the Huff category “much ado about practically nothing”, are usually statistically irrelevant and being pushed for some commercial or political end: scientists say coffee improves brain function; identity theft on the rise, so buy more insurance.

    In a chapter called “The Semi Attached Figure”, Huff writes of a judge in India advising a young British civil servant. “When you are a bit older, you will not quote Indian statistics with that assurance. The government are very keen on amassing statistics — they collect them, add them, raise them to the nth power, take the cube root and prepare wonderful diagrams. But what you must never forget is that every one of those figures comes in the first instance from the chowty dar [village watchman], who just puts down what he damn pleases.” This is the lesson imparted to every first year banking analyst or management consultant building their first cash flow model or strategic plan: garbage in, garbage out.

    But my favourite term in Huff’s book is “statisticulation”, the manipulation of statistics. A classic example of this was a 1986 Newsweek article which claimed that a single woman over 40 in America was more likely to be killed by a terrorist than to find a husband. The statistics underlying the claim were unpublished at the time, later debunked and the main claims of the piece eventually retracted. But the statistical gravity with which the original claim was made lodged it in the public consciousness longer than it deserved.

    Following the financial crisis, many of those in the thick of it excused themselves by saying it was no good pointing fingers. Everyone was to blame and it was time to work together for a solution. Amazingly, most of them were allowed to do so. It is Mr Georgiou’s rotten luck that of all the people up to their necks in Greece’s economic fiasco, it is the statistician who risks jail.

    The writer is author of ‘What They Teach You at Harvard Business School’

    You must be logged in to post a comment Login