It’s a great honor to be here at Brookings today. The people in this building have done some outstanding research and analysis on government performance, so it’s a pleasure to be here with you to talk about data-driven governing – an issue that is near and dear to my heart. And that’s because I see a world in which our creativity and imagination have expanded the outer bounds of human achievement and potential as never before. Government helped make so much of this progress possible. And yet creativity and imagination are not exactly the first words people today associate with our government.
What if they were?
2. What If?
What if we tackled our biggest problems by using data-driven strategies, instead of conventional wisdom, or the way we’ve always done it? What if we could make our communities safer by knowing, in real time, where crime was happening every day and then deploying police officers to those precise locations? What if we could put an end to lead-poisoning among children instead of ignoring it as if it were a problem that just cannot be solved?
What if we improved public safety by using big data and the experience of years of recidivism to identify probationers and parolees who are truly the greatest threats to society? What if—by sharing medical records and targeting personal interventions—we could cut avoidable hospital admissions by 10% in a single year?
Imagine if the overall performance of every school could be measured over time so that every parent and citizen could see? Imagine if one common platform not only measured the job skills in greatest demand in any given county or metro area, but also allowed employers to find the skilled workers they need?’
In Baltimore and in Maryland, we did all of these things – and more.
This is the new way of governing. It’s not about excuses, deflecting blame, or ignoring problems; it’s about transparency; it’s about accountability; it’s about performance management. It’s not about left or right; it’s about moving forward. It’s about setting clear goals, measuring progress, and getting things done. The old ways of governing–bureaucracy, hierarchy—are fading away. A new way of governing is emerging. It calls for a new kind of leadership at every level; leadership that embraces a culture of accountability, embraces entrepreneurial approaches to problem solving, embraces collaboration. Leadership, in other words, that understands the power of technologies like smart maps, and GIS, and the Internet to make the work of progress open and visible to all. This new way of governing has quietly taken root in cities and towns all across the country. In blue states and in red states. Pursued honestly and relentlessly, it holds the promise of a more effective way of governing at every level of our public life — local, state, and federal.
Our approach in Baltimore was actually born in the subway system of New York City. In the early 1990s, Jack Maple was a New York transit authority police lieutenant. And Jack believed there was a better way to deploy his officers. With nothing more sophisticated than some paper maps and colored markers, Jack Maple started plotting where and when robberies took place on his section of the subway. And then he sent his undercover detectives and transit officers to stop criminals where they were most likely to strike, at the times they were most likely to strike.
He put, in his own words, cops on the dots.
Jack Maple and his cops drove robberies down to record levels. The media came calling. And the new police commissioner came calling. And soon Jack wasn’t just plotting out a strategy for part of the subway. He was deputy police commissioner of the entire New York Police Department, developing a system that came to be known as CompStat. And the NYPD, under his and Commissioner Bratton’s leadership, went on to reduce violent crime to levels that few would have thought possible 20 years ago.
New York’s ongoing success in reducing crime and saving lives led to a revolution of performance-measured policing in cities and towns all across the United States. And one of the first of those major cities was my city — Baltimore. You see, when I was elected Mayor of Baltimore in 1999, we had allowed our city to become most violent, the most addicted, and the most abandoned City in America — with more population loss over the prior thirty years than any major City in our country.
At the beginning of our Administration we were able to put an additional 200 police officers onto the streets of Baltimore. An important question: where to send them? We could deploy them equally to each of the six council districts. Or if we wanted to be real political about it, we could deploy them to the council districts with the highest numbers of primary voters.If we wanted to be real, real political about it, we could deploy them to districts where the greatest number of people voted for me. Or, we could send them to the hotspots where the greatest numbers of our citizens were being shot, mugged, or robbed. And that’s exactly what we chose to do. And we repeated that Compstat process every day and every week—constantly searching for better tactics and better strategies. And over the next ten years our work, thanks to courageous police officers, drove down major crimes in Baltimore faster than any big city in America.
There’s a baseball equivalent of this CompStat strategy – some call it ‘moneyball,’ some call it ‘the shift.’ Put your fielders where the past performance of their hitters say they are most likely to hit the ball. Put your police where crime is most likely to happen. That’s the deployment of resources for maximum effect. That’s goal driven, data-driven thinking. It helps win baseball games. And it helps make a city safer. We brought this new way of governing and getting things done not only to our Police Department, but to the whole enterprise of City Government. And we became the first major city in America to do so.
We started to create a new culture of higher expectations in City Hall – one of accountability and transparency, centered around results – and a constant search for better ways to get things done. The leaders started to emerge, and were recognized by their colleagues. We set high goals, and we used data to tell us whether or not the things we were doing, every day, and every week, were working. Our CitiStat approach, like CompStat was built on four principals, four tenents: timely, accurate information shared by all, rapid deployment of resources,effective tactics and strategies, and relentless follow-up.
Every two weeks, if you can picture this, on a constant and rotating basis, my team and I held CitiStat meetings with agency heads and their leadership teams in the CitiStat room on the sixth floor, with the big boards of data they brought. Everything was mapped out and indexed to previous reporting periods for everyone to see. Ideas were shared, questions were fired back and forth. If we hit a goal, we wanted to know how. If we failed to hit a goal, we wanted to know why.
And it worked.
We brought crime down by 43 percent. We reduced the number of children poisoned by lead by 71 percent. And early on, when former Mayor William Donald Schafer accused our Administration of having no vision, we responded by issuing a 48 hour pothole guarantee. And our crews met that guarantee 97 percent of the time. The Kennedy School at Harvard, in 2004, gave us their Innovations in Government Award. Of course we didn’t implement Citistat in Baltimore to win awards; we did it to survive.
We did it to make our city safer, cleaner, and a better place for kids to grow up. For many years, it had seemed like drug dealers were more effective than our government. But thanks to Citistat, that reality was starting to change.
When I was elected governor of Maryland, in 2007, we took this approach statewide with StateStat. The goals were bigger, the measures more diverse, but the premise was the same: data-driven decision-making, collaboration, follow up, and results. And we shared those results – good or bad — with an online dashboard anyone could access to see where we stood and where we were going. With this approach we achieved something of a public safety Triple Crown, driving down crime to a 30-year low, incarceration to a 20-year low, while at the same time reducing recidivism by nearly 20 percent. With this approach, our teachers, principals, parents, and kids made our schools the best in the nation, five years in a row.
We cut in half the number of children in foster care, to the lowest levels on record.
We set a goal of reducing infant mortality by 10 percent, and when we met that goal we kept going – reducing infant mortality by more than 17 percent overall, and 25 percent among African American families.
We took on health care costs with a commitment to driving down preventable hospital readmissions. By creating a platform for health care providers to share patient information, by mapping the incidence of chronic conditions, and by aligning the profit incentives to wellness rather than sickness for the hospitals — we drove down avoidable hospital readmissions by more than 10 percent in the very first year of trying.
It used to be in Maryland that every Governor would set a 40 year hope for cleaning up the Chesapeake Bay. Instead, we started to measure actions and results. We created Baystat to identify and to map not only the sources of pollution, but the actions that we can take, in the right place, together to halt the flow of pollutants into the rivers and streams of the Chesapeake Bay. We set two-year milestones and took measurable actions to reduce storm-water runoff, to expand the number of acres planted with winter cover crops, and upgrade clean technology at sewage treatment plants.
We made it possible for citizens to click on any of the tributary basins where they live and see whether we were making progress and hitting our goals. And for all that effort, we reduced nitrogen, phosphorus, and sediment levels by 14, 15, and 18 percent, respectively. We’ve restored hundreds and hundreds of acres of stream buffers and natural wetlands, and doubled the number of native oysters now filtering the waters of the Chesapeake Bay.
Did we meet every goal we set? No, we did not. With true performance-measured governing, failure has to be an option. A temporary option. If we met every goal, then we probably weren’t setting our sights high enough.
One tragic example: after six years of steady progress, Maryland, like many other states, experienced a deadly spike in heroin overdoses. So we set a new goal of reducing drug overdose deaths by 20 percent. We made some progress reducing prescription drug abuse by mapping out facilities that were supplying the pills and then shutting them down. We got more people into treatment. But it wasn’t enough to prevent this tragic spike. As with any of these efforts, when what you are doing is no longer working, you come up with a new approaches. And so we did, and so we must.
What I’ve learned in 15 years of executive service, taking Compstat to Citistat—and taking Citistat to StateStat—is the larger the human organization, the more important performance management becomes. Our framers never set out to create a nation that muddles through, we came together to form a more perfect union. Data-driven decision making and performance management are essential to that mission in these modern times.
As you some of you know, the problem at our federal level isn’t the lack of goals or a lack of data. Agencies have dozens of goals and performance metrics and strategic objectives. But what are the big goals for our nation? And what are the actions that allow us to achieve those big goals together? Too many federal goals are about process, not about outcomes. Having meetings is not a goal. To the public, all of this process means very, very little to their lives. At our federal level, we must have a clearer view of the most important things our government is trying to accomplish and why.
This requires clear goals that reflect what “we the people” actually value. And the difference between a goal and a dream is a deadline. Without a doubt, there is no progress without jobs. And job creation should be our highest goal. But let me give you three other examples that speak to our values as a people. The infant mortality rate in the United States of America is the highest of all the developed countries in the world. If we value reducing infant mortality as a nation, then our goal at the federal level should be to do just that by a measurable amount by a certain time.
If we were to reduce infant mortality across the nation at the same rate we did in Maryland, we would save more than 4,000 babies each year. That’s 4,000 families that would be spared an unfathomable loss. It’s so easy to become lost in measuring everything from soup to nuts. We must measure what we value, and value what we measure. If we increased kindergarten readiness across the nation at the same rate we did in Maryland, we would have 825,000 more American children ready to learn on their very first day of school.
That’s 825,000 more children that don’t start out behind, 825,000 more children taking their first vital steps toward success in life. Final example – if we reduced preventable hospitalizations across the country at the same rate we did in Maryland, we would keep 600,000 more Americans out of the hospital each year. That’s 600,000 of us on their feet instead of flat on their backs in hospital beds.
In other words, Americans should know what our federal government’s top five objectives are. Job creation. Improving the security of our people. Improving the education and skills of our people. Improving the sustainability of our way of life. And improving the health and wellness of all Americans. Federal employees should know how their work and their agency contribute to the achievement of those objectives. And leaders, staff, and the public should all know whether we’re making progress together, and where work still remains.
Finally, coming to the table together at the federal level cannot be be just a box-checking exercise. What good is a lofty policy without follow up on the ground—in the “small places close to home,” where it really matters? We need a new method of executive management, a method that becomes central, every day, to the important work of our federal government.
Our federal government’s objectives should be a reflection of what we value most – and those critically important things that we can only accomplish together.
Early in my first term as mayor, as we held community meetings all across town, every week, to talk about our fears, our frustrations, and our hopes, I invited folks to come and ask me anything. At one of these meetings in East Baltimore, a 12 year-old girl named Amber waited her turn, and took the microphone, and told me: “Mr. Mayor, there are so many drug dealers and so many addicted people in my neighborhood that some people in the newspaper are starting to call it Zombieland.” “And I want to know: do you know that people call it Zombieland? And what are you going to do anything about it?”
The question she asked of me was really a questions she’s asking for all of us. Behind all of our data, there are people. Living their lives. Shouldering their struggles. They deserve a government that works.