The governor of Maryland is a long shot for the White House—and the best manager in government today.
When attacking CompStat and CitiStat, however, O’Malley’s critics were often even more damning, objecting to what they saw as the twin dangers not only of those programs but of data-driven systems of performance in general. The first is that an administration can consciously manipulate the stats in order to make its performance look better than it is. The second is that the measures themselves can incentivize the bureaucracy to play games with the numbers in ways that management neither anticipates nor notices—to the detriment of citizens. For example, in surveys conducted by criminologists in 2010 and 2012, retired NYPD officers reported that their fellow officers would cite a lower value for stolen goods in order to reduce the charges from a felony to a misdemeanor.
As mayor, and since leaving the mayor’s office, O’Malley has stood accused of both—by political opponents, civil rights groups, labor advocates, and by another, more slippery force: Hollywood. In one episode of David Simon’s HBO series, The Wire, various agencies of a fictional Baltimore, scrambling to make it seem like their numbers had improved before a weekly “Comstat” meeting, resort to messing with the numbers. “Juking the stats,” one character says knowingly. “Making robberies into larcenies. Making rapes disappear. You juke the stats, and majors become colonels. I’ve been here before.” In 2002, O’Malley had been there, too, but in real life. During an internal audit, the BPD found that officers had been under-counting the number of rapes that had occurred in the city. The official total was 178, when it should have been 211. While the problem was, in fairness, discovered by the O’Malley administration itself, the revelation became a useful political bludgeon for O’Malley’s opponents, who later would call loudly for more external audits, casting shadows of doubt on O’Malley’s progress.
Civil rights advocates, for their part, have been less concerned with allegations that O’Malley was juking the stats and more concerned that his data-driven programs were creating damaging incentives for police officers to make their numbers look good. In 2006, the American Civil Liberties Union (ACLU) and the National Association for the Advancement of Colored People (NAACP) brought a lawsuit against Maryland on the grounds that O’Malley’s policies as mayor had caused the number of illegal arrests and instances of police harassment of law-abiding citizens to skyrocket during his tenure. Indeed, according to city records, in 2005, police made 108,000 arrests—that’s one for every six people in the city at the time.
In 2010, the state settled the cases out of court. The BPD has since rejected “zero-tolerance” policies, established new protocols for minor infractions, and agreed to allow independent observers to keep an eye on police actions. (In 2012, police made fewer than 53,000 arrests.) O’Malley has said that he does not see the ACLU/NAACP lawsuit, or the state’s decision to settle it, as a rebuke of CompStat, which, he says, never encouraged officers to be more aggressive in their arrests.
Of course, the issue, and O’Malley’s response to it, reflects a larger debate playing out in headlines across the nation today. In April, yet another instance of teachers cheating in order to raise their schools’ test scores came to light, drawing fire from critics not only of policies like No Child Left Behind, but of data-driven systems in general, which, they say, often encourage appalling behavior. Proponents of such systems, and O’Malley himself, argue that while dangers like cheating and juking stats are real, they can be mitigated, and the value of using data-driven techniques outweighs those costs. (After all, when companies like WorldCom and Enron juked their internal accounting numbers, we didn’t respond by eliminating accounting standards; we tightened them up with Sarbanes-Oxley.) Harry Hatry, a fellow at the Urban Institute long involved in data-driven governance efforts, says the problem of misaligned incentives “is real—it’s a major problem, and we don’t do enough about it.” But he suggested the solution is to collect more data, in the form of surveys, for example—not to scrap the metrics altogether.
That ongoing debate has characterized much of O’Malley’s recent career. On the campaign trail in 2006, for example, O’Malley’s Democratic rival, Montgomery County Executive Douglas Duncan, as well as the incumbent governor, Bob Ehrlich, who is Republican, tussled with O’Malley again over what they called his “fuzzy math.” In his stump speeches, O’Malley claimed that violent crime had dropped by 37 percent during his time as mayor, giving Baltimore the second-fastest crime rate reduction in the country. His opponents, however, argued that crime had dropped by 24 percent, giving Baltimore only the sixth-fastest crime reduction rate in the country.
Both were right—depending on what baseline you looked at. In 1999, an audit by Linder & Associates found that the BPD had been mis-categorizing assaults, thereby artificially reducing the violent crime rate. If you re-categorized those assaults, the overall instances of violent crime would appear to spike dramatically. When O’Malley became mayor in 2000, he did just that. As a result, both the crime rate and the percentage by which he was able to cut crime appeared substantially higher. If you used the new baseline, his numbers were right; if you used the old baseline, his opponents’ numbers were. While the controversy cast a pallor on the election, it didn’t change its course. In June 2006, Duncan dropped out of the race, leaving O’Malley to win the Democratic primary and sweep into the governor’s mansion.
Less than a month after taking the oath of office in January 2007, O’Malley held a press conference. During the half-hour presentation, in which the new governor “alternately sounded like a policy wonk and deadpan comic,” the Washington Post reported, O’Malley laid out his plan to implement CitiStat at the state level. The predictably named StateStat would span all state agencies, he said, and, like CitiStat, include issue-specific programs to focus on cross-agency problems. Gallagher, who had helped launch CitiStat, would launch StateStat, too. With Maryland facing more than a billion-dollar projected budget shortfall the following year and state agencies unaccustomed to stringent oversight, O’Malley predicted at the press conference that there would be “growing pains,” he said. That was, it turned out, a major understatement.
Ten months later, Maryland was slammed, along with the rest of the country, by the recession (which O’Malley invariably and loyally refers to as “the Bush Recession”). With unemployment and foreclosures skyrocketing, and businesses shuttering their doors left and right, critics questioned whether that was the time to roll out a fancy new managerial tool at the state level. O’Malley and his team, however, were committed to it. By mapping government services with data, they argued, they could be more efficient with the increasingly strapped resources they had.
Feed the Political AnimalDonate
Washington Monthly depends on donations from readers like you.