PDIA in Practice 10: Building and Maintaining Authorization

Camila Lobo [00:00:02] Welcome to Part 10 of the practice of PDIA Building Capability by Delivering Results Podcast Series. This twelve part series based on a video series used for our PDIA online course. We’ll walk you through the PDIA or problem driven iterative adaptation approach to solving complex development problems. More than fifteen hundred development practitioners in 90 countries have used the PDIA approach. 

[00:00:26] Authorization is a necessary condition to build state capability. However, it is not easy to buit authorization to act. In today’s podcast, Professor Matt Andrews will discuss the challenges of gaining authorization and offer practical ideas to address them. 

Camila Lobo [00:00:46] Matt, can you share with our listeners, what do you mean by authority and why this is important for building state capability? 

Matt Andrews [00:00:54] The process of building state capability involves people, it involves people who are doing new things, trying to find new functionalities. One of the types of people that we often interested in is those who authorized this kind of work because you’re looking at governments and contexts where people are being authorizing all sorts of things for a long time. But those things haven’t been getting it done. So now you say, well, who’s going to authorize the kind of change that allows you to build new kinds of state capability that invariably are going to affect some of the political settlements? Invariably going to create some new winners and some new losers. Often times we think that this is the minister of finance or the president or the prime minister, someone at the top of government who rides in on a white horse, who gets the problem that you’re identifying, who identifies the solution and who’s willing to sign on the bottom line and say, I’m going for the change. The problem is that a few years down the line, we found that that person either didn’t survive very long or that that person survived and didn’t really mean what they were saying when they signed on the bottom line. Or perhaps that that person just didn’t come on the authority or the power that we thought that they did. Theoretically, we would say that the issue with authority is very, very complex because often the people who are authorized to facilitate the kind of policy or reform changes we are trying to introduce those who are at the center of government with power are those who are embedded in the system that already exists and they have the least interest to actually move that system along. So assuming that those people are the ones who really want change and they are going to give us the power to do it is probably not a great assumption to be making. Instead, we need to be looking around in a system and thinking, who is it in systems that identify problems? Who is it in systems that identify solutions and how do they move the people who have authority and power so that they willing to move behind a change program? In a lot of the research that we do, we find that this leads us to multi agents solutions for change, not to single heroic solutions to change. We need to have people who can envisage the need for change, who can construct a problem in a way that those who own power are forced to pay attention to it and are forced to say, if we don’t deal with this problem, we might find ourselves in a sticky situation. And then we also need to have people who get around the people with power and say, look, if you get behind this change agenda, we’ll support you. You’re not going to be on your own. We will resource you. We will give you ideas. We will help you so that you can survive this change process. Oftentimes these people aren’t around and you’ll see a change process that simply doesn’t get you where you want to go to. One of my favorite examples of this is the anti-corruption reforms in Malawi. After 1994 1994, Malawi became a brand new democracy and the donors pushed the government to accept all sorts of anti-corruption legislation and a brand new commission. The president was the one who signed off on this and everyone said, well, this has got such high level attention, it’s game to succeed. Well, a few years down the line, there was some major end to the major corruption crisis and the person who is at the center of it was: the president. So a new president came in and said, well, I’m going to commit like the previous one didn’t to make sure that corruption goes away. When that president went out of office, it was under a cloud of corruption. Now, in a place like Botswana, you see something very different. Botswana had a major crisis in the early 1990s where there was corruption in roads. There was corruption, the education sector. There was corruption in the agriculture sector. And the cabinet as a whole came out and said, we need to do something about this. The people at the top of government were saying, if if we don’t do something, this is going to get closer and closer to us and it’s going to create some political troubles for us in the future. At the same time, you had people who were in the parliament and people who in the civil service who were saying we need to do something about this, too. Over time, they gathered together. They created an anti-corruption unit. And this anti-corruption unit did pretty far reaching things. But it wasn’t just about one person. It was about multiple people coming together to deal with the problem that they could not ignore anymore. This is authority. It’s something that we cultivate and it’s something that we build in groups and not around individuals. 

Camila Lobo [00:05:26] Thanks, Matt. People often make assumptions about their own bureaucracy. Can you share more about the difference between an ideal and a real bureaucracy? 

Matt Andrews [00:05:37] In many cases in development, in any kind of bureaucratic setting. People think that authority is about the voice from the top, that someone from the top is going to tell people lower down in the bureaucracy what to do in those people are going to jump and respond. Now there’s a big literature on the ideal type of bureaucracy going back to Max Weber and a bunch of people like that and it assumes certain things about the bureaucracy. So if we think about it like this, this is your kind of bureaucratic shape. It’s just a top down hierarchical triangle. And this things that we assume the one thing is we assume that there’s someone at the top who actually commands respect and has authority. We assume that that person can send orders down through the hierarchy and is establish ways in which that can happen. Generally, we assume that they don’t go all the way down. But those kind of tiered effect so you have senior managers and then you have mid-level managers and then you have frontline folks. And we’re assuming that essentially the commands are going down. All of those levels, they are being captured and then they are being faithfully transmitted down the levels. Now, that’s the ideal type of bureaucracy. It’s an ideal top that also assumes that there is no political influence into the bureaucracy, but that essentially it comes in through the top and the rest of the bureaucracy is actually shield from political interest in the setting. Now, this is the ideal type. What’s the real type? Well, in most cases, the in real type you have a lot of political influence, basically puncturing the bureaucracy and coming in through the walls and it sometimes comes in at the bottom. It sometimes comes in at the side. And it oftentimes really makes it impossible for the people at the top to do very much. The other thing you often find is that the folks at the top don’t have the level of respect, the level of of authority that is required for their voices to actually make a huge difference. In many cases, the ones who are at the top or people who are appointed politically and they come and go and the ones who are here at that senior bureaucrat level, they actually the ones who’ve been there for their entire careers. And you find that the voices of it don’t necessarily have as much influences as the voices over here. Often times when we’re dealing with people who are lower level bureaucrats and we have the authority from the person at the top, the low level bureaucrats say yes, but they don’t hire us. These people are the ones who are in the middle, the ones who are going to be there in five years time and who were there five years ago.

[00:08:03] So you don’t necessarily have that clear approach. You also often don’t have these lines of authority that move downwards. There’s often a lot of breaks in those lines of authority. And it could happen there could happen lower down. It could happen at the front. Sometimes in bureaucracies, you actually find that these things are broken off completely. And the ones who are at the frontline are a completely different organization. They are listening to people on the ground. They are not listening to people at the top. So what we like to say to people is you can’t just assume that you have an ideal type bureaucracy because an ideal type bureaucracies are precisely that. They are ideal type. You need to understand what your bureaucracy looks like, what your authorising environment looks like, and you need to do that by essentially penetrating it, engaging with it, finding out who the people are, who really, really have the power, the authority and the influence. And that’s a process. 

Camila Lobo [00:08:56] Can you further explain what you mean by the reality of authority? What does it look like? 

Matt Andrews [00:09:03] Now, the reality is that most people aren’t in one organization, they are actually kind of in sectors where they are multiple authorities and multiple strands of authorization. And so we like to talk about the realities of authority when we say examine your authorization environment and make sure that you have enough authority to do things. We are very aware of the ways in which the authorizing environment can be dysfunctional. And I’m going to present some of those ways now not to tell you that you can’t do things, but to give you some idea of the assumptions you need to be careful that you are not making. 

[00:09:37] So let’s start over here and say the ideal type bureaucracy is basically a kind of a a a hierarchical triangle. And one of the assumptions that we make is we make the assumption that the person at the top essentially sends orders down the organization and that those things get done. The first dysfunction we often find is that there are breaks and orders and that’s number  A here where essentially something in the street in that organization gets in the way of the order going down. 

[00:10:05] So you’ll have people at the top who are telling others, this is what you need to do. But one or two levels down, that message gets lost and there is a completely different message that is going through. Now, you need to be careful that your policy idea, your change process doesn’t get caught in that dysfunction. 

[00:10:21] The second thing that we folks figure out is also that sometimes organizations only appear as this kind of hierarchy right at the top and then they take on a different shape altogether. They could be flat. They could be a big box. We don’t really know. It could be that there is a lot of kind of egalitarian sentiment beyond the the high levels. And so what happens is that you have the person at the top who is sending down the message. But then what happens is that there is a complete break, meaning that they actually don’t have any control over the rest of the organization. Even in a formal sense. Now, this is often what would happen in a sector where you would have a ministry of education and the minister of education is in control of a very small set of the people in the sector. And when you get down here, the people are principals, district superintendents, teachers, etc. And the authority mechanisms over them are on the whole, not the minister, it’s the head of the teachers union and it’s the governor, etc. So what happens is that you have a break and that’s number B, number C is essentially within the organisation that you have conflicting forms of authorization. So it’s not that you get as a message coming down, the thing just gets broken. It’s that the message that is coming down is actually interrupting messages coming from authorises at other levels. So you find that people who are senior bureaucrats are telling people lower down to do certain things. The minister is telling them to do other things. And essentially it becomes about who shouts the loudest and the longest. And that’s also dysfunctional. And it means that you don’t really get the formal ideal type bureaucracy that you thinking about. Now, if you’re assuming that you have that formal type of bureaucracy and this is what you really have, you’re probably not going to be developing a good authorisation model. Number D is essentially over here, which is similar where you have the minister telling people down here what they need to do. But the difference is that those people are also working in overlap with another ministry and that minister has told them to do something completely different. Now, this could happen if you are dealing with an agency and the agency is formally tied to one ministry or one sector. But actually the people working there have got a lot of formal or informal ties in another sector or in another part of government. And essentially they are being told to do two completely different things from very high levels that also can cause a lot of dysfunction. Number E is where within your authorization model you have competing interests from the people at the top. So essentially you have authorizer and maybe the education ministry and authorize it in the health ministry and they are trying to work out how you do a policy on an on sports or on kids education or kids healthcare and really they have conflicting ideas. And essentially what could happen in this dysfunction is that you never even get any messages coming down because they spend all their time fighting amongst themselves. If you get caught in that kind of dysfunction, it becomes very difficult for you to win over anyone because they compete with each other. The last one is the idea that you actually have a lot of authority that comes from the bottom of the system across all of these organisations, whether it’s citizens or whether it is just lifelong employees. And they are actually sending different messages as well to everyone that is in the system. And those messages are often coming from very different places in very different ways. So this is a tremendously messy diagram and I don’t apologise for that, because what we’re trying to say is building authority is a messy process. It is not perfect. And you should not be assuming that it’s ideal or perfect. You should be finding out just how tough it is and you should be dealing with the realities of it on the ground. So this gives you some idea about what to look at. But you now need to look at it. Talk to the people around you ask them what dysfunctions are really there and then get a real view of what you’re dealing with. 

Camila Lobo [00:14:10] You often say that maintaining support in a change process is a real challenge. How should our listeners grow and maintain their authority? 

Matt Andrews [00:14:19] In PDIA we don’t take for granted that, you know, the authorization is just going to stay with you for a period of time as you iterate and you try new things. You have to work hard to maintain that authorization and to grow it as you move along. So, for instance, we would say you need to try something that’s functional, but you can’t try that forever and ever and ever. You have to stop at one point in time and go back to the people who authorized you to do that in the first place and say to them, this is what we achieved. This is what we learned. This is what we think we can do the next time around. Now, that means that you stop and you build your legitimacy so that there are more people who look and say, gee, what they doing makes sense. We may not have tried this before. It may seem a little bit risky, but there’s actual results. The key result that comes out of PDIA that politicians and administrators are often very interested in is the learning. Is your organization is saying we tried something and we found out that it worked. When we did it this way or that way. Now, some people don’t see this in their organization because they don’t have a learning culture in their organization. But most theorists found that when you do have learning, the people at the top start to see the learning as an avenue to do new things. And then you come in, you say, well, now we’ve got that authorization. Why don’t we try a bigger step the next time around? But then you need to stop again. And you can’t just carry on forever and ever because your authorizers are going to be saying, well, how much are we giving you to do this and what are you doing in return? We have to create a political story to tell people why are we supporting this and what you’re doing for us and what you’re producing. So we stop again and we say, well, this is what we’re doing. This is what we’re learning. And this is what we can to do the next time around. And you do this all the way up, constantly building these vertical lines on the graph. Now, this helps you manage what people often talk about as the political cycle and the timing problem of many reform and policy programs, where it just seems longer to do something functional than the political environment allows you to do. What this is saying is you need to fit your steps into that cycle, but your cumulative set of steps can go over cycles because you are building enough support as you move along. The other thing is that it’s not just building legitimacy with the same actor over and over again. It’s expanding the number of actors who authorized your reform and your change as you move along. 

Camila Lobo [00:16:50] Thank you for listening to part 10 of the Practice of PDIA podcast series. Tune in to listen to part 11 where we discuss iteration to learn more about building and maintaining authorization in PDIA download our toolkit at BSC.CID.HARVARD.EDU.