LTC1: A Virtual Discussion with Prof. Matt Andrews (March 27, 2020)

On March 27th, 2020, the Building State Capability program at Harvard University hosted a virtual discussion with Prof. Matt Andrews, who answered audience questions on Leading Through Crisis.

Read BSC's Public Leadership Through Crisis blog series

Listen to Prof. Matt Andrews podcast with Shruti Mehrotra of Open Society Foundation on Applying Lessons from Experience to COVID19. 

Read Peter Harrington's blog post on Coronavirus and behaviour: Why leaders need better ‘risk communication’. 

Read Peter Harrington's blog post on Seeing Pandemics as Complex Adaptive Problems. 


Katya Gonzalez-Willette: Hello and Welcome to Building State Capability at Harvard University's podcast series. On Friday, March 27, 2020, the Building State Capability Program at Harvard University hosted a virtual discussion with Professor Matt Andrews, who answered audience questions on public leadership through crisis. 

Matt Andrews: Just really glad to have you all on board today so that we can talk about leading through crisis. I think it's a really important topic right now. It's an important topic more generally for many of the countries that we are from, where crises are things that happen more regularly than people give them credit for. So it's great to great to be on. 

Salimah Samji: Okay, great. We can get started. So, Matt, I'll ask you the first question that was posted. And the question is, "What in your opinion are the capabilities required for public leadership through a crisis? And what to you are the biggest challenges to building these capabilities?"

Matt Andrews: Sure. It's really the question. You can see where on in the blog posts we're thinking about exactly this issue. I think my estimate is we're going to probably have about 22, 23 posts going through these things. I think that when you're going through crisis, you need to think about what is the role of leadership. And I do think that you need to remember that there's a threat that you face, and that's usually what a crisis involves. And you as the leader, are going to have to firstly help your people manage the fear that they have in the face of their threats. I think it's a very important leadership role. That is a role that only a leader can play. Other people can't do that. I think the other thing is that you need to remember the crisis is about making decisions. So you need to set up your system so that your system can reflect on the decisions that need to be taken. Gather the information that is required to make those decisions, make those decisions and with confidence, know that those decisions will be acted upon. So you need to set up the mechanisms that ensure that decisions can be taken as well as possible under very, very difficult situations and that you can mobilize those decisions into action. Probably the next capability that you need is the capability to monitor what is happening in response to the decisions and being able to adapt and change and to be able to do that in a way that is both fast and legitimate. It's a very, very trying thing in the public space. Most governments that we work on, they are very risk averse and they want to know what the answer is, and they want to act on only the perfect answer so that they are not at risk of looking like they made a mistake. That cannot be your approach in the face of crisis, in the face of a crisis. You are absolutely going to be responding in ways that are imperfect. And you need to recognize that. Therefore, you need to monitor, you need to get information quickly and you need to be able to change in front of your people what it is that you are saying and doing. That is a very, very trying thing. It is a big, big leadership role in this context. And it takes people out of managing government systems like what most people in authority do, to leading people through things, which is about adaptation. So I think those are some of the key capabilities. Built into those capabilities, is the capability to, one, be able to build a team and again, to build a team that is flat, that is fast, that is flexible. To be able to mobilize leadership from people below you in your hierarchy. And when I talk about below, I'm invoking the idea that most public leaders are in hierarchies. You need to be able to mobilize leadership from people below you and on the sides of you. And you need to be able to give them a lot of space and create an environment in which they can they can do that. It's what we call that the holding environment. Amy Edmondson would speak about creating psychological safety for the people. That is the job of the leader in this case. The other thing that I would say is when you're talking about particularly people who are on that hot seat, on that deciding seat, another role, another key capability is to keep calm and keep still and keep away from the action. It is very important and very understated how important it is. Your leader needs to be the one who is actually saying, "I'm not doing enough. I'm not active enough. Why am I not actually there? Because this is going to be a marathon, not a sprint, and people need to be preserving their energy." They need to be mobilizing other people to do the work. They need to have those people doing the work on loop's so that they can refresh themselves. But the person who is making the decision needs to be the one who is available to manage his or her brain as much as they possibly can. And for that, they need calmness and they need to almost keep themselves above the fray. Again, I'll invoke Ron Heifetz here. The idea of getting off the dance floor and getting onto the balcony is very important during a crisis. It's an important capability that you need to have in yourself. So those are some of the key things that I think really matter here. The product that you're looking for is trust. You need to have trust and you need to build trust in your people, in you. And you need to build trust in your system so that your system can actually get through this. And that is something that has to do with messaging. It also has to do with action and it also has to do with being consistant. So a long answer for a very tough question. And the difficult thing is that many of the people who are leading right now have got to learn how to develop these capabilities -some that are within themselves, some that are within their team, some that are within their broader systems and some that are within their countries very, very quickly, because we have no time to waste. 

Salimah Samji: Great. Thank you. The next question that we have is, "In a time of crisis like this, no matter the amount of resources a government has, it will never be enough. How does a government gain public trust and be able to mobilize all of the community resources they have?" 

Matt Andrews: So for this one, if Peter is on, I'd love Peter to actually take the lead on this. So for people who don't know, Peter Harrington worked in Liberia on the Ebola crisis with the Liberian government. He also is a communications expert. And so I'm going to ask that, Pete, share a little bit about how the government engaged to build trust and also to mobilize people to change their behaviors in the ways that are required in a crisis situation. 

Peter Harrington: Thanks. Hi, everyone. Salimah, could you just read out the question again so everyone has it fresh? 

Salimah Samji: Sure. In a time of crisis like this, no matter the amount of resources the government has, it will never be enough. How does a government gain public trust and be able to mobilize the community resources that they have? 

Peter Harrington: That's a really good question. So as Matt said, I worked for about six months in Liberia advising the president and the head of the response during Ebola. The first and most obvious thing to say about creating trust from a government point  of view is tell the truth. That's, you know, that's first and foremost. And so I think it's essential not to spin and not to and not to try and kind of gain some political mileage, but just be very clear and truthful with the public. I think that's the first point. The second point, I would say is be transparent and be honest about the extent of the crisis, but don't try to sugarcoat the crisis and try to kind of put a positive inflection on it. There's a balance to be struck between being truthful and transparent and sort of scaremongering. I think Matt made some very good points in his public leadership series over the last week about using language which doesn't create anxiety, having a very calm body language when speaking to the public and tricks and tips of kind of that personal communication. But as a government, I think the second point is about being transparent. Thirdly, being very clear, if you're going to say, "This is bad and it's gonna get worse," then it's very important to follow that by saying this is what we are doing. So I think people can handle it, if you level with them. Publics can handle it if you level with them and say "this is going to get a lot worse," if you also make it very clear that "this is an evolving situation, but we are working to do the following things." And then the fourth thing I would say is really, really to make sure that you invest in your data management, because the data in an evolving crisis, in an epidemic, you're always two to three weeks behind the true picture with your data. But the more accurate you can get and the better you can get that up and running. So in Liberia, I worked alongside a guy called Hans Rosling who came to Liberia to work on this. Some of you might know him. He was a very well-known Swedish statistician. And him and the team that he had in the Liberian government did amazing work to have a sit rep every day that could then be communicated. And then the second part of the question is about mobilizing people. I think the behavior change social dimensions of this are so important to remember. This isn't just a medical phenomenon, it's a social phenomenon. You have to engage with people's sense of responsibility and their values, not just with information, not just appealing to the head and people's rational decision making with information, but to appeal to everybody's values. I saw the other day Dr. Tedros, the head of the W.H.O., used the term - "Beating coronavirus is everybody's business." And that is a direct quote of the slogan that Liberia used during Ebola. That's where that concept, this idea that this is everybody's business, that everybody has a role to play. Everybody has a responsibility. And I think particularly leaders who are addressing the public can be very powerful in connecting to that more values, hearts, emotional space, where people feel that they have a responsibility to act. I think Andrew Cuomo, is a very good example with communication in the U.S. I think there are leaders elsewhere who've done that. In my country, in the U.K., 8:00 p.m. last night, everybody in the country leaned out their window and did a minute's of applause for the health workers in the country as an appreciation. I think bringing home to people that that your loved ones are at risk, that there are health workers who are risking their life for you, that people are taking on very difficult things, and we all have to make sacrifices, helps to mobilize that behavior change, which is so essential to preventing the overwhelming of health systems you have. I'll leave it there. 

Salimah Samji: Great. Thank you, Peter. I'm going to move on to the next question. And the question that we had was - "Roles can be populated by different actors. Capabilities denote capacity plus ability. So the systems in the countries (health, police) can somehow immediately increase capacity by non-expert population and fastforward ability by trainings, example self swab test. Where else can this be done? Any ideas? 

Matt Andrews: Let me give you my sense of the work that we do when we do PDIA in countries. We work with the idea that capability is about not just the abilities you have. And I think if you don't have the abilities to do these things, you're in a lot of trouble. But it's not just about ability. I think that many countries have more ability than they think. Capability to me, we assume is about your empowered abilities. So we actually find that there are a lot of people in developing countries and I think this probably of the team in Liberia who were mobilized to do the data work. There are a lot of people who actually have latent abilities. They have been taught. They have been on courses. They do know how to analyze data. They do know how to do basic things. And one of the issues we find in many countries is that the organizational structures they find themselves in do not empower those abilities enough. And capability is about the empowered abilities that people have. So it's about to me ability plus authorization plus acceptance or motivation. What are people accepting as their responsibility to do in a given situation? What are they authorized to do in a given situation? And does it match up to the abilities that they have? So one of the things that I think is important is to work out where do you actually have people who are already trained beyond what you are using them to do? Where do you have people who maybe have better ideas? Because this is not just about mobilizing the medical community to do swabs. This is also about mobilizing people to think differently about what your country is going to look like. It's also about mobilizing teams to collect data. It's also about tracing in patients. It's also about having people manage the mental health of your community. This is a compound crisis that does not just have one dimension to it. So what I would say is the first thing to do is don't just think about where can we do a quick training. I think you also need to think about where can we reorganize our organizations so we authorize people in a different way. Where can we motivate people in a different way so that they are willing to take more risks to try the things that they have learned but have never really exercised in their jobs. And I think you can do that all over the place. Where we work in countries doing the PDIA, I think this is pretty much what we do is we go in and we work only with teams of people in the context. And we say to them, here's the problem that you guys think is impossible and let's create an environment in which you can experiment and do things really, really rapidly and where you can learn really quickly. And we find that people have more ability than they are displaying. It's just authorizing them a different way. Now, I think that this is something that is required in the face of a crisis. And it is the one thing that I think you can do fairly fast. I think ramping up your abilities and ramping up your muscle memory to respond is something that is very hard. And that's why I think most people who work on crisis management emphasize the readiness for crisis management, probably even more than the responsible crisis management. But if you haven't been doing that in your governments, I think there is a way that you can say, let's start to pull teams together. Let's start to give them tasks. Let's start to inspire them in different ways and mobilize them to do things that they thought that they couldn't do. I think this is a key part of the story in Liberia. I think it's a key part of the story in response to Ebola, where people thought this could not be done. And really, it was done and it it was done with people in the context. It's a key part of the work that we've seen when we've done PDIA. And, you know, the strange thing about PDIA is we almost try to create a crisis situation for the people we are working with. We try to say to them, "This is important. You need to work now." And within short periods of time, they generally manage to find new ways of working. What it requires for leadership, though, is it requires mobilizing your people and trusting them, giving them clear tasks, but then giving them the space to work those tasks out. I think it also requires a different form of delegation. So if you're going to go to your organizations and say, "Look, we don't know how to do this, but we think that you guys can probably work this out. Here's the task that we're going to put in front of you. We want you to give us some ideas in two or three days, in two or three hours, whatever it may be, because parts of this crisis are moving very fast and parts of this crisis are only going to materialize in a little while. And what we want to have is we want to have a response at that point." But the delegation is given the clear task, let them work on the clear task and then come back to them and make sure that you actually listen to what they say. Try it out and learn. It's a different form of delegation. And that's one of the ways that I think we can build capability - not just focusing on the ability side, not just focusing on how do we get more expertise immediately, but thinking, "How can we adjust the way we authorize, how can we mobilize and motivate people in a different way"? I think that those things are things that we can work on immediately, especially if we are in positions of leadership. I also very much would suggest reaching out to other folks. I think it's very important. Yesterday we put up a podcast with Shruti Mehrotora, who worked in Sudan. And one of the questions I asked her was, "When you were in Darfur and you had the refugee crisis after the war, where did you get all the expertise from?" And she said, "We got a lot, a lot of people on the phone. There were a lot of people who were not there". And, you know, there's a lot of former students who are listening right now to this, and, you know, I want to say to you, if you guys are available to help, please send me an email, because we are putting together coaches with people all over the world who can coach on specific things. It doesn't mean that you know their context perfectly, but it means that you could be a call for them and you could be a resource for them. And she said, "You know, there's actually a lot of people outside the situation can do in terms of expertise, in terms of advice, in terms of sharing stories from other countries. And I think that there is a way that if you do not have that expertise where you are now, you can start to tap international networks to try to get people on the telephone to help you. It is obviously not the first best solution, but there are a lot of people out there who are willing to help. We have already connected four coaches who are gonna be tele-coaches with people who are on the hot seat right now. And I think that they can help in a number of ways. They can help just provide a voice that is outside of the weeds, outside of the noise, which is important for people in the situation. They can also provide expertise, and they can also be the ones who are doing a lot of homework for you. And they can connect you to their networks as well. So I think that these are ways in which we can try to accelerate this process. It certainly is not the first best solution, but that's not what we're looking for right now. 

Salimah Samji: Great. Thank you. And a related question that we had is - "How do public leaders build multiagent teams during a crisis, such as COVID-19? 

Matt Andrews: So I think this is really important. And I like very much the work of a colleague of mine called Marshall Ganz. He does work on organizing. And in organizing you create what he calls snowflake organizational structures. They're flat.They are fast and flexible. Essentially, you're going to think about a snowflake. You have a nucleus in the middle where you have your team, and that's going to be a decision maker. And then they are going to decide where are the capabilities that we need, what what are the areas that we need to address. And they can create teams that are around them, but that are coordinated to them. And the coordination is going to come through information flows and through personal connections. It's one of the ways in which you can essentially distribute the task, because this is a gigantic task to respond to this, given all of the dimensions that it has. How do you mobilize those people in a quick way when we haven't worked in these kind of situations before? I think my my first piece of advice would be for the people who are in those decision-making situations to get a small team around you and map out the crisis as you see it. This is going to be identifying where you think you are going to need people, generating information, generating ideas and sharing ideas. Once you've mapped those out, the thing that I would suggest for you is firstly get some cross-functional top people, people who are good at doing a lot of things and who can be creative in advising you across a lot of things. Second, get some people who are functionally very strong in areas that you need them at that moment. In this case, you want to get somebody on that core team who is going to be a medical expert, somebody who can not only advise you on pandemics, but somebody who is trusted to provide almost the final word on a decision on a pandemic. If you don't have them in your team, you need to find a way to get on a telephone call, etc.. I think that you need to have somebody in that team who is, and listen to Shruti Mehrotra on this, who is very good at mobilizing people. Who actually can tell you who are the ministries that are meant to be doing this, who are the nonprofits that are in the space, who are the private sector people in this space and how do we mobilized them. And then you need a communications person that would be mobilizing your core team beyond their core team once you've identified the areas in which you think you need to work. I think you then want to start to identify lead contacts in those areas, get hold of them and start to deputize them to build those teams together for you. An example of this would be, I think Peter spoke about the information team in Liberia. You want to say, "Who do we know who has skills in this area?" Let them build their team together. Give them the clear task and say to them, "This is where we need the team built. This is how we need the answers for that." I think that this idea of kind of distributed teams that receive your tasks, that respond to your tasks, is one model of doing this. And I think it can be done fairly quickly. I think that what you as the central person need to do here is you need to trust a lot the advice of the people around you, not only in the areas you need to work, but also in who they think needs to be deputized. You need to also remember that you may not even be engaging directly with these people in this period of time. It may be people on your team who engage with them. And if you really like to control people, this might be one of the first times that you have to give that away and you have to trust people in that process. But you need to give them clear tasks. You need to get product from them as quickly as you can so that you can see that they are working so that you can build that trust as you move along. But I think that this is one way in which you can start to mobilize people across the system. Again, I think the two tools that you have are, one, you can tweak authorizing structures. And remember, when you do this, you need to clearly authorize people. I've already heard some of my former students and colleagues who are in this situation saying, "Yes, but people are at home. But, you know, they work in the Ministry of Education. We want them to be on this team. And they keep telling us, has the minister of education authorized to do that?" You need to create a clear authorizing mechanism so that those people can be put onto these teams clearly so that they know that this is part of their job description. And then you need to, through your communications, think of ways of motivating and mobilizing those people as well, so that that kind of personal acceptance of the task is something that rises up within them. So those are some of the ideas of things that we have worked with when we have done the PDIA work. They are, again, hardly first best solutions in this situation, but I think that they can work. 

Salimah Samji: Great, thank you, Matt. Another related question is - "How do we influence people to change norms?"

Matt Andrews: That's a tough one as well. I'm gonna ask again, Peter, to weigh in over here. How do you influence people to change norms? And I'm going to do that because in the situation with regards to Ebola, I think there are some very interesting tactics that were used and this is significantly going to be about communication. Peter, can we ask you to weigh in on this? 

Peter Harrington: Sure. So the headline answer is - it's difficult to do really quickly and that's what you're trying to do in this situation, is do this really quickly. So the best teacher of all is, is when the crisis or the epidemic comes at their doorstep. So I think, you know, a lot of behavior change happened or some behavior change happened in, you know, in the case of Ebola, it happened when people who at first hadn't believed in Ebola or thought it was a conspiracy,they saw people they knew or close to them get ill and in some cases die. That's a very powerful teacher. But I think as policymakers or as public leaders, it's not possible to simply rely on that. That's you know, that's not good enough. How do we accelerate that process and try and bring things to the public's doorstep so they will change behavior before it gets to that point? So I think the tools of what's known as Risk Communication are really important here. I think we have to acknowledge a few things about how people change their behavior, that there's a process people go through from awareness to then contemplating changing their behavior and then doing it and then potentially influencing others. What we're really trying to do, I think here is build new norms and very new taboos very, very quickly. The key point I would make there is that the way people respond and who they listen to and who they trust is changed. So the most important thing to remember is, I think in this context is, the message is really important. But who is delivering the message is really important, if not more important. So people listen in this day and age to different people than they used to listen to. So people are influenced much more, it's cliche to say social media, but it's true. They're influenced by their peers. People don't necessarily change their behavior based on what a leader on TV standing behind a lectern says. They're much more likely to listen to their peer groups, their social network, their family, other influences in society and get that information on those cues to change or update their norms from different places. So in Liberia, the behavior change communications only started to work when it stopped being national leaders standing up and saying it, and the behavior changed communications is the part of the response that I worked mainly on. It only started to work through local leaders, whether it's church leaders or village leaders or through influential celebrities, influential people in communities. What we did was we mobilized an army of about 30,000 people to go door to door. People who are from that community, who could be advocates and spokespeople in their community, who could influence the community to change behavior and adopt key behavior changes that would stop the epidemic. That's one country, one context. This has to be adapted for different countries and different cultures. But the key thing is I think we need to be thinking about who needs to be communicating these measures, whether it's health workers or people in communities or people on social media or people who have influence and using peer groups as much as possible to try and reinforce change in behavior. Then in terms of the message itself, as I said earlier, it's really important to connect this to people's values and their emotions much more than the kind of cerebral level. So actually showing people the impact, if necessary, showing people what's going on in overwhelmed hospitals, hearing from health workers who are on the front line. There was a nurse who broke down on video recently because she came off a 20 hour shift and found there was no food on the shelves. And things like that have a huge impact because people connect emotionally with it. So I think making sure the message connects with emotions and values and reframes values in terms of those norms, but also really actually being creative in who is providing that message and mobilizing a very, very distributed group of actors to deliver that message. That's the way you influence people, find people that the public listen to. 

Salimah Samji: Thank you, Peter. I particularly like the story idea. I had a visual of this poor nurse, and we all, I think, know what empty shelves in grocery stores look like. And we know how we feel when we see that. And I'm going to pivot now to take questions from people who've been putting them in the chat window. And I wanted to call on Abigail Bellows. If you could ask your question, as opposed to me reading what you put in the chat window. That would be great. 

Abigail Bellows: Sure. Hi to several of my professors. So good that you guys are doing this. So the question I asked is that this conversation seems to be focused on well-intentioned public leaders who are truly committed to that public mission. But in many countries and many developing countries in particular, those leaders need to contain spoilers among their own ranks within the government who are interested in instead exploiting the crisis for their own personal gain. As we've seen in the corruption spike that often accompanies disasters in many countries. So any lessons about how to manage, contain, counteract those spoiler elements within the government as public leaders are trying to exercise their role? 

Matt Andrews: Well, Abby that's a great question. One of my senses just out of this is that I think that corruption is an issue. And I think you want to make sure that when resources are coming into your country or the resources that you have, that you're using them as well as you can for the things that you need to. I was engaging with a former student who's managing his country's response, and it's one of the countries that is actually doing really well and hope to have a case up on a blog next week. And he was saying to me that one of the issues that they have is just the amount of resources the government is finding itself spending in a very, very short amount of time requires that they are very careful about how that money is spent and also about how people are perceiving that money to be spent, because a key thing that they need to do is maintain the public trust. So he said, you know, one of the things that they have done is they have brought in people from their justice sector and people from the province accounting sector and the Professional Internal Audits Association to come into kind of like a working group. And they've said, "Let's put something together that is a process that we can use to move this money quicker than we have before. And recognizing that there are going to be challenges with doing this. But we need to have multiple eyes on it and we need to have a new mechanism." Also, just communicating what a new set of rules are because in some places, corruption is also breaking the rules of procedure. And actually, during this period of time, governments are going to break the rules of procedure. We want them to break some of those rules of procedure because we want them to go quicker. But in this case, what they've said is, "We want to rewrite the rules of procedure as they exist." Now, in some countries, you would have emergency rules of procedure for disbursement, for procurement, etc. and people would automatically move towards that. If your country doesn't have that, you need to do that because you have to have some basis of accountability, some basis of saying, are we still operating according to rules? Right. Because if you move into a no rule situation with your public financial management system, your procurement systems or whatever, you could fall into a lot of trouble. You also said that one of the issues, and I don't know if you would call it corruption, but he said when we're looking at behavioral change and this was a country that invoked quite heavy suppression mechanisms very early on, he said people who violated the rules needed to be prosecuted. So they needed to be seen to be prosecuted. So this was something where they the police force, the prosecutor's office was put into high alert, and they prosecuted people very, very fast. He said there were two or three people who within the first day of suppression found themselves in a courthouse and were fined the equivalant of 20-25 thousand dollars. This is in a middle income country, so that is a significant amount of money. And they made it very, very public that this was happening because their goal was not to prosecute people who take their money. Their goal was to make people realize that they were serious about doing this. And again, this is about building the public trust and building a clear message and having people abide by that message. So I think these are huge issues. And we need to work out what kind of rule systems do we have in place? Are they clearly communicated? Do we have people who are watching them quickly? I remember speaking with General McMaster about the early days in the Iraq war when they found villages and they tried to move money very quickly to those villages. And I remember one of the things he said is, "You know, people were just willing to give them cash." But he said "We have to have accountability mechanisms because the perception of corruption is going to kill you, even if there's no corruption, even if you're giving the cash and it's all above board, people are going to say, what were the rules? Where was this? Where was this?" So what they did was and, you know, this is my rough memory of it, they did something like create like flying squad systems where they had a military lawyer working together with, I think, an auditer and working together with a village person. And they developed kind of like a system whereby they moved the money in each place. In each place the system was different. And in each place, the accountability was first and foremost people who were receiving the money so that they perceived that there was fairness and that there was transparency, but also so that the military could say, here are a set of rules we abided by here with the people who signed off on things. So it's going to be different. And I think that if you don't have those things, you need to swiftly develop those things. Even if you look at them and you say, "Well, in normal circumstances, this would not be acceptable." You need to have something. 

Salimah Samji: Great. Thank you. And thank you, Abigail, for asking the question. We have one last question that we're going to take, because I do want to keep us within the hour that we had set out for this. And this is from someone again, it's in the chat window, but she doesn't have a webcam and her connection is not great. She's calling in from Iran. The name is Maran Bakun."I wanted to know how much should we be worried about the transparency of the data which governments share about COVID-19 specifically in less developed countries in the Middle East?"

Matt Andrews: So let me give my answer and then I'm going to ask Peter how they thought about not only transparency but accuracy as well. I think that when you're probably asking about both of those things together, I think that we need to realize that all of the data that we have about this is going to be tricky, but at the same time, especially with this kind of crisis, but I think most crisis' data is very, very important. It's very important so that we can understand what the problem is. We can understand where the problem is. We can understand where the hotspots are. It's also very important so that we can communicate because, you know, you want your data to give you some kind of subjective basis to communicating. And it's also important so that when you try things out, you can actually monitor how those things are working. Remembering that nothing we do here is going to be perfect and that we are going to be trying things out. We have to monitor them. So data is really, really, really important here from the point of view of the teams that are going to be generating and using the data. I think you want to generate as good of data as possible as quickly as possible, and you want to try and improve it as you move along. So, you know, I would say in situations that I've worked in like this, you're basically saying - where is the data that we have? Bring the data in. Make some kind of reflection on how accurate it is, and then make a reflection very quickly about where we think the problems with that data are. And as a team starts to use the data, they should start to work to improve the data at the same time. I think that that's something that happened in the Ebola crisis as well. And I think that's that's something that you need to think about in terms of transparency. If you're a consumer of the data and you're not actually developing it, I think that you need to ask some of the questions. You need to ask where it's coming from. You need to ask if you've seen it before. You need to ask what you think the weaknesses of that data are. You need to ask those about both, you know, who is providing the data and what you think the political interests of those people might be. But also, who's providing the data and what you think the technical limitations of those people might be? My sense is that, you know, oftentimes data is something that we are building as we move along. And we need to be very aware of what we think are the strengths and the weaknesses of the data when we use it. That should be the case with data use at all times anyway. And if we think that there are political motives behind not releasing data, I think that we need to take steps to engage with that and to use that data in that way. If we think there are technical limitations, I think exactly the same. But Peter, how did you think about this? Because I know that data in the Ebola crisis was a big challenge. And I know the team was mobilized from the beginning and the team then started to develop the data. And you were then a consumer of the data using it on the communications side. So how did you think about the accuracy of the data, the transparency of the data and the legitimacy of the data as you were sharing it with the population, given that, you know, when you said earlier, one of the tools you use for behavioral change was actually kind of communicating where things were, sopeople needed to trust the data. 

Peter Harrington: Yeah, I think we worried a lot about the accuracy of the data, so let's focus on accuracy for a minute. I think at first I think it's really important what Matt said is that - you're never going to have perfect data. You're always gonna be weeks behind the real epidemic. So official cases will be well behind true cases. So you're never gonna have perfect data. But especially at the beginning, as you're trying to get that kind of data management team or system or unit, whatever it is, you know, particularly on epidemic data I'm talking about not financial data, but epidemic data. It's going to take time for it to get better and become more reliable. So we did worry a lot about the reliability of it. I go back to what I said at the beginning. You have to tell the truth and say this is the best data that we have available and kind of provide that kind of caveat and be honest about your level of confidence that you have. And I think just work as hard as possible to get people on the ground who will improve the quality of that data. So not just raw modeling. Try and diversify the sources and the kind of types of information that's flowing into the people looking at this. It's not just coding from certain sources, but blending many sources of information as you can to try and improve that. So there was a lot of really helpful information that came through from very rapid field surveys that were done in different areas of the country. That was really useful data that came in from field workers who were all over the country. It wasn't just from treatment units reporting their cases every day or their admissions or hospital admissions and so on. So I think really think of it as all sorts of types and sources on transparency. I guess the only thing I would say is it's really difficult in different context. But in a crisis, everyone understands that governments have a sort of a little bit more leeway to take rapid and decisive action, but it doesn't mean that accountability should be suspended. You know accountability still really, really matters in a crisis like this. I think, you know, in the UK, there's been a lot of controversy around certain modeling that was done and then, you know, major policies around whether to lock down or not were based on modeling. That modeling was then challenged. So I think it's really important for people outside government, consumers of data to say, "Well, what is this based on? What are the assumptions that you're making here?" And to ask questions of it and to hold the government to account, not to, you know, cripple and kind of paralyze everything, but to maintain that scrutiny. That's easier or harder in different contexts, but remains very important. 

Salimah Samji: Great. Thank you very much. 

Matt Andrews: Thank you very much, everybody, for participating. 

Salimah Samji: Follow our blog for more information on the next session because we'll post the registration links and question links up there. Thank you very much. 

Katya Gonzalez-Willette: To learn more about the Building State Capability Program's "Public Leadership Through Crisis" blog series, visit Thank you for listening.