Thursday, January 3, 2008


I had to rub my eyes when I read this article. It seems that France Inc. has hired Mars & Co., an international consulting firm, to grade each of the government's ministers on his or her performance. Without any expression of shock, dismay, or consternation, Le Monde calmly informs us that Xavier Darcos will be graded on the number of overtime hours worked by teachers and by the seniority of teachers in disadvantaged districts; Brice Hortefeux will be graded on the number of illegal immigrants expelled and the number of legal immigrants selected for their workplace skills; Christine Lagarde will be judged by a retail price index designed to measure the success of the reform of the Loi Galland; and so on. "All of M. Sarkozy's political objectives are numerically measurable," the consulting firm cheerfully reported (and one can of course be confident that this was an objective and rational judgment untainted by self-interest, since objective, rational, measurable judgment is the firm's stock-in-trade).

The last time I can remember a government possessed by the managerial ideology to this degree was when LBJ relied on Robert McNamara, with his expertise in operations research, WW II bombing evaluation, and manufacturing Fords, to measure American success in Vietnam. All the indicators were up, McNamara reported to his boss, who comforted himself with this objective, rational judgment and continued to drive straight off the cliff, while McNamara, who allowed sensible emotion to overcome his impeccably aligned figures, jumped to the World Bank, where green eye-shades were more sartorially appropriate. This is the first concrete indication I've had that Sarkozy will fail. Perhaps he will have the sense to ignore what his consultants tell him.


Anonymous said...

New York City's Mayor Bloomberg seems to have a similar yen for statistics. (Makes sense for a man who made the bulk of his vast fortune by designing a system to collect and crunch real-time market data to give stock traders an edge.) Most of his policy ideas spring out of numbers and their analysis; he justified his smoking ban in 2006 to Rolling Stone by saying only that heart disease deaths in the city had dropped by 1,200 in the year since the law had gone into effect, "which is almost exactly what the statistics predicted." His administration does censuses, follows test scores, and even indexes 311 calls by location so complaints about pot holes and garbage collection can be triaged for greatest efficiency of remedy.

Sometimes Bloomberg comes across as though the reliance on numbers has gotten him out of touch (in the same article, he chided New Yorkers for fearing another terrorist attack when "There is a much greater risk from lifestyles that hurt you - smoking, walking across the street without looking both ways, not putting bars in the window if you've got kids and you live above the first floor, those kinds of things." Which is technically true, but sounds myopic.)

But I'm still hoping to move to New York this year. The city seems to function with efficiency. At some level, there is a political math in which most ideas should be measurable, and it's not just election results.

Anonymous said...

"The only statistics you can trust are those you falsified yourself" [Winston Churchill]

Means and objectives?

Statistics are usefull to ensure that the means we mobilize are actually helping us towards the objective. That's was the point you made in your recent point regarding Discrimination and Wages.

Evaluating Darcos against the overtime hours worked by teachers is evaluating the mean; not the objective such as the proportion of children able to read at 10 yr old.

Still wouldn't it make sense some kind of independant organisation (which is neither the case of Mars & Co, nor of the Assemblée Nationale) does it against the missions as they have been defined in the LOLF? Or is it a "managerial ideology"?

Unknown said...

Thanks for your comments. I have nothing against honest efforts to establish objective measures of policy when used as a tool by those responsible for implementing the policies. The problem is that these "measures" are being used as whips to crack over the heads of ministers, not tools to improve outcomes. And the criteria discussed in the article seem remarkably ill-conceived. Of course the discussion is not comprehensive, so I may have been unfair. I am also told by a reliable source that there is nothing new about the practice of hiring consulting firms to monitor policy in France.