Jump to:Page Content
Making government more efficient and more responsive to the needs of its constituents is central to the core mission of the Harvard Kennedy School. Herman “Dutch” Leonard is professor of public management whose research focuses on leadership, organizational strategy, crisis management and financial management.
Q: You have been working recently on a project called "Acting in Time" in the area of crisis management. What are the main challenges you have been looking at?
Leonard: The Acting in Time project was initiated by Dean David Ellwood. The basic idea is to examine a series of policy areas where leaders in all sectors can reliably spot a challenge or opportunity before it arrives and have a chance to work on it sooner rather than later. In order to be successful, in order to act in time, leaders first need to anticipate a challenge; then need to be able to figure out something useful to do about it sooner rather than later; and then need to mobilize resources to take action.
Our research team – Arn Howitt, Doug Ahlers, Graham Allison, Pete Zimmerman, and David Giles here at HKS, and several others from outside the School – has been focusing on acting in time for crisis management, trying to be prepared to respond to a large-scale disaster or to recover after the fact from a large-scale disaster. And what we’ve been finding is that there is a great deal to be learned from recent experiences – both in Banda Aceh and in New Orleans, two large-scale disasters that we’ve had a chance to examine fairly closely. What we’re finding is that the challenge of recovery in the aftermath of a large-scale disaster is something scholars and disaster response planners haven’t thought enough about – so we are beginning to devote some time and effort to understanding recovery better.
In Banda Aceh, where the tsunami struck in 2004, the recovery process was, in our view, too centrally controlled. A series of outside agencies came in and thought they knew what to do. They didn’t spend enough time working with local communities to try to figure out what their priorities were and how they wanted to proceed.
New Orleans demonstrated almost exactly the opposite situation. In the recovery phase following Hurricane Katrina, there were community groups that organized and tried to figure out how to repair and rebuild and reconstruct their neighborhoods, but they got almost no help from the larger surrounding authorities and structure. So Banda Aceh was too centralized, and New Orleans was too decentralized.
Recent events involving Cyclone Nagris in Myanmar and the earthquake in Sichuan Province in China are reinforcing these lessons. The military government in Myanmar – for its own reasons – appears to be making the classic mistake of operating in a highly centralized form – and many thousands, and perhaps tens of thousands of people, will die as a result in the “second wave” of disease and exposure following initial storm. In China, the response appears to be progressing more effectively – here, the problems seem to be more that China had in place before the earthquake a set of rules and programs for reinforcing or replacing dangerous buildings … but did not act in time in the years and months before the event to prevent thousands of people from being inside those buildings when the devastating earthquake struck.
The crisis management researchers of the Acting in Time project now are working with the city of San Francisco. The Bay Area faces a reasonably high likelihood of a significant earthquake sometime over the next 30 years. So we hope there is still time to put in place some mechanisms that would help San Francisco recover more quickly from a major disaster that might take place there. We are trying to work with the city and county of San Francisco to organize a network of leadership groups that would be in a better position to help the city recover quickly in the aftermath of a major disaster.
What we’re hoping to do is to work with people in the city and county of San Francisco to act in time, to be organized and ready and to have built a platform for accelerated recovery before they actually need it. It’s a very exciting project.
Q: You have written extensively on government disaster response, particularly in regards to Hurricane Katrina. What have public sector agencies learned from their successes and failures and how are they coordinating efforts to improve their execution in the future?
Leonard: All of us were horrified by what happened in New Orleans several years ago when Hurricane Katrina came ashore. But the Acting in Time group at HKS is concerned that we may be drawing some of the wrong conclusions from that experience.
When Katrina made landfall it was really only a Category Two hurricane. It had been a Category Five hurricane while it was offshore, but by the time it came ashore it was only a Category Two. So how did we manage to turn a Category Two hurricane into an international disgrace?
The answer to that is it took centuries to make the series of errors required to turn a relatively minor storm into such a catastrophe. Society began by spending centuries placing large amounts of value in harm's way and, in the decades before Katrina, did not spend adequate time and effort and resources to protect the value that had been placed in New Orleans by building more effective levee systems to prevent flooding.
Then, in the years and months before Katrina, governments didn’t do enough to build the apparatus and the procedures or to train people and to have the equipment for a rapid response in the event that we did have a significant disaster in that area. And then finally, in the days before Katrina, we didn’t do enough to mobilize in advance the security presence and other response elements that might have been effective in the immediate aftermath. So there were many mistakes made over very long periods of time that contributed to the significant challenges of Katrina.
What our research team is finding now is that many people walked away from that experience with the idea that what we need is a more centralized system. They look at some of the failures of the response and they conclude that what we need is a more military-style organization to handle disasters for the federal government and to coordinate the states and to act more like a military agency, with command and control systems and a central headquarters that makes decisions and gives instructions.
There are several things wrong with that conclusion. First of all, our constitutional structure doesn’t really allow for that. States are independent of the federal government. They are, in many areas – including creating and enforcing their own laws, ordnances, and regulations – sovereign entities. Thus, we need to coordinate and collaborate between states and the federal government rather than trying to create a single unified organization that is going to command everybody – and that is the first problem with the centralization idea.
The second problem is that if you create a centralized agency that is trying to manage everything, it would literally drown in the myriad details that have to be handled in the aftermath of a major disaster.
So what we’re trying to do is shift the philosophy. And this is really what we have found in our research. We need to change from a centralized mindset to recognizing that we will intrinsically operate on a decentralized basis. And the question is not whether to operate on a decentralized basis, but how to operate most effectively on a decentralized basis. What can we do that would create conditions in which that decentralized action could constitute a superior performance in terms of a disaster response? Now there are answers to that question. There are ways in which we could design the system so that it can operate successfully on the decentralized basis that we will inevitably and intrinsically have to have. So that’s the basis of the research that we are currently doing.
We refer to this as “fast and light.” We are trying to shift the philosophy from what we call a “big, heavy, slow and stupid” kind of response to a decentralized, rapid, and innovative response in which individual units are able to take action on their own in these circumstances, and to coordinate with one another through a centralized agency, but not to have to wait for central instructions about what to do.
There are some good examples of “fast and light” at work in New Orleans, including the Coast Guard. Two days before the storm came ashore, the Coast Guard began moving assets toward the region so that they were there and ready and prepared to move as soon as the hurricane passed. It came with units that were self-deployed; they didn’t wait for instructions; they knew what it was they were trying to do. They came with the equipment and supplies that they needed, so they were self contained. They didn’t get there and wait for somebody else to bring them food. They brought everything they needed.
So the Coast Guard movement into the Katrina impact zone is a good example of what can be achieved with a “fast and light” kind of philosophy. The Coast Guard operated pretty independently of most of the other organizations in terms of authorization. And on site they were very effective at coordinating with other organizations that were trying to be helpful in the aftermath. What we’re trying to do is to maximize the effectiveness of a decentralized, intelligent, adaptive response rather than to create a centralized response which we don’t think is either much of a possibility or very likely to succeed.
Q: You have recently been doing work on recovery efforts in Banda Aceh after the tsunami and in New Orleans after Katrina and now, prospectively, in San Francisco (before the next earthquake). What have you been finding in New Orleans that can help other cities prepare to be able to recover more quickly if they face a major disaster in the future?
Leonard: New Orleans offers many lessons on a variety of dimensions. Some areas of the city are responding and recovering much more quickly than others. And when you go and ask why that is – what is the underlying structure that permitted that – it’s not resources; it’s not that they had some privileged access to government or outside help. It’s that they had local leadership that brought people together and began to problem solve. Local leaders began to look at what the challenges were, to try to set priorities to figure out what they needed to do. And they developed skills as they went along.
One of the more inspiring examples that we have had the privilege to observe (and, to a modest degree, support) is in the community of Broadmoor. What we found is a local leadership team, many of whom had not met before Katrina, but who had each been involved in leadership in the neighborhood before the storm began. And there was the Broadmoor Improvement Association, which had some existing infrastructure for organizing people and meetings. Out of that grew a very extensive effort. By organizing people locally, that group was able to partner with some external sources, including major corporations that provided resources. And they got in touch with people here at the Kennedy School. So we were both doing research and providing advice to the community as they organized on their own behalf to try to figure out how to go forward and recover.
What we’ve learned from their example is that an essential ingredient in rapid recovery is existing local leadership and infrastructure of leadership. A key to effective response is the ability of local leaders to come together and figure out what the challenges are, to figure out how to make progress on them, and to reach out to outside organizations and get resources that they need in order to continue progress.
How does that apply to San Francisco? Well, in the aftermath of a major earthquake in San Francisco there will be widespread devastation but it won’t be uniform. Some areas will be more significantly affected than others. What we need in advance is neighborhood groups, leadership teams and other groups that would be able, very quickly after the event, to begin to do the local assessment. They would have some preexisting knowledge of the city agencies and the other resources that might be available, so that they’re able to reach out and connect as quickly and effectively as possible with those resources outside their neighborhoods. And of course those agencies will in turn need to know about and be prepared in advance to reach out quickly to state, federal, civil society, and private organizations outside the area for help and additional resources.
So what we’re doing is trying to take some of those lessons coming out of New Orleans and figuring out how to build the infrastructure in San Francisco in advance so there will be what we’re calling a platform for accelerated recovery – a set of elements that we put in place in advance, that will allow the city of San Francisco to move quickly toward recovery in the aftermath of a major event.
Q: You have also worked on social enterprise and social entrepreneurship. Are we seeing the emergence of new approaches to social change that are likely to significantly increase the impact of socially-oriented organizations?
Leonard: There is a great deal of discussion these days about social entrepreneurship and new models for creating social change. Many people are suggesting that new models of for-profit organizations with a social mission may transform the landscape of social change. It’s the early days in this, and so it’s hard to tell yet exactly how much activity there will be in this area and how much of a transformation that will occur. But we are seeing some really interesting trends and experiments.
One of the key things that people have focused on is the idea that traditional philanthropy and traditional forms of social activism in the form of non-profit organizations are not easily scalable. In order to take a traditional non-profit organization funded by donations or grants and make it bigger to have larger impact, you basically have to find more grants, more money from outside, more donations. You have to take everything that you’re doing and just do more of it.
But there are other models – the for-profit model, for example – that are more intrinsically scalable. So if you can find an organization that has a social mission, or build an organization that has at its heart a social mission and is able to turn a profit through its own direct activities, it may be a more scalable model.
The other possible important avenue is a major change in the way non-profits are working with one another to form networks of organizations to work on social problems. Individual non-profits that are working on issues with the homeless or education, for example, are not going to solve the entire problem. A homeless shelter is not going to solve the problem of homelessness by providing beds on a short-term basis. An individual homeless shelter can’t by itself do very much about the overall problem, but what it can do is build its part of the solution.
But now the question is – who is going to solve the problem of homelessness? Who will be working on the problem more generally? Here, we need to turn from solution-building to problem-solving, and that requires either a larger organization, or a network of organizations that are doing different things. So if we had a network of organizations that together were building a series of different elements that might be necessary for a solution, we are problem-solving by having constructed that network with the different solutions being built in tandem within it.
And what’s exciting about this as an idea is that it leads us, as researchers, to begin thinking about a different set of tools. We have been working on leadership and management issues within individual organizations, we’ve been figuring out how to help a particular homeless shelter do a better job of operating itself as a shelter. But now the question is: how do we help organizations form into networks that are actually going to work on the larger complex of issues and solve the larger problem? That’s a set of questions for which we don’t necessarily have performance tools or ideas yet about the best ways in which to manage or lead those kinds of organizations. So it’s an exciting new area in both of these dimensions – both “social entrepreneurship” and what we call “new social enterprise,” the for-profit models within social enterprise, and also within the traditional organizations forming into these larger networks.
Social mission-driven operations have a social mission at the heart of what they do, and the question is how do they get resources and how do they scale their impact? And we are seeing several new and interesting models for how to do that along these different dimensions.
Interviewed by Doug Gavel on April 23 and May 14.