In 2001 the Nobel Prize in economics was awarded for research on the consequences of asymmetric information. I don’t really care about the used car market but I do care about urban governance. A necessary condition for an urban mayor to be up to the job of providing local public goods and internalizing externalities is that he have kosher real time information on the “State of the City”.
Today’s New York Times highlights Mayor Bloomberg’s cutting edge data creation. As an empiricist what shocks me here is that his approach is viewed as so “revolutionary”. It makes me wonder what information previous Mayor’s staffs collected before they made key policy decisions.
Bloomberg Lives by Statistics and Gives Aides a Free Hand By JIM RUTENBERG
October 18, 2005
Barely 12 hours after winning the 2001 mayoral election, Michael R. Bloomberg, still dazed and hoarse from late-night celebrations and early-morning television interviews, wandered into his Midtown campaign office, turned to a key member of his transition team and asked only half-jokingly, "Now what do we do?"
But perhaps more significant, he reshaped the way the mayor's office runs New York, applying a results-based approach to almost every area of city government, and largely appointing his commissioners based on expertise and giving them nearly free rein to determine policy regardless of political consequences.
Wall Street Ways at City Hall
Mr. Bloomberg exercises control over the city much like Mel Karmazin, the former Viacom chief, famously did at his company: by closely monitoring the numbers produced by a team of star department heads who are free to run their agencies as they see fit so long as they meet strict production targets.
It is a strategy grounded in his experience in the business world. Mr. Bloomberg spent 15 years on Wall Street with Salomon Brothers, working his way from a $9,000-a-year position in its bank vault to a partner in charge of running the firm's information systems department. Pushed out of the firm in 1981, he then earned billions by using his $10 million severance to create a machine that compiled and analyzed real-time data from the financial markets to help traders gain an edge.
As mayor he has tried create a similar system from which to govern. Data analysis is religion for Mr. Bloomberg, and numbers are the lifeblood of his administration. They drive policy rather than just track progress.
It was in large part in the pursuit of more city data that Mr. Bloomberg created the 311 help line. It provides one-stop shopping for people seeking information about everything from parking rules to trash pickups. But perhaps more significant, residents' grievances on the line are also stored in a database so the city can immediately identify a festering problem area, and react.
Linda Gibbs, Mr. Bloomberg's homeless services commissioner, said she began the first citywide census of homeless people on the streets this year guided by the mayor's results-based approach. The census provoked serious debate within the administration, she said, because by creating a new set of numbers "you're taking the risk it could go in the wrong direction."
"His reaction was," she said, "you're not going to be able to overcome an issue unless you really understand it."
Ms. Gibbs said that at the height of a homeless crisis in 2003, she presented Mr. Bloomberg with a chart showing some good news: an alarming rise in the number of people in shelters was finally stabilizing. But the mayor was not content, and drew a line pointing downward from the plateau on the chart, saying, "Come back when it looks like that."
Such an approach has helped drive nearly all of the city's major indices in positive directions, just in time for Mr. Bloomberg's re-election campaign. But the approach has often drawn criticism, particularly on the schools front, with educators saying the administration is obsessed with test scores at the expense of a more holistic approach to improving student performance.”
AN INTERESTING incentives issue here is whether Bloomberg’s technical people “fudge” the statistics because they know that the boss will reward them for progress. Look at the test scores and education fact that Houston schools tell low scoring kids to stay home on testing day because this raises the school’s average score. For a quantitative incentive system to work there need to be accountants who can verify that the data is a representative sample of what the Mayor had hoped to sample and not a select sample.