UCL Uncovering Politics

How to Run Public Administration

Episode Summary

This week we ask: How should you run the public administration? Should administration be close to or insulated from politics? And what should the role of private and other non-state actors be?

Episode Notes

We’re focusing this week on public administration. While mention of the word bureaucracy rarely lifts hearts, it’s incredibly important for the development of public policy, for the delivery of public services, and for all the other things that the state does. 

To consider how public administration can be run well, UCL Uncovering Politics is joined by Marc Esteve, Professor of International Public Management in the UCL Department of Political Science. 

Mentioned in this episode:

Episode Transcription

Season 7 Episode 2 

How to Run Public Administration

13 October 2022


services, public, corporations, collaborative governance, public sector, public administration, study, question, cost, government, private sector, operate, private sector actors, focusing, actors, citizens, organisations, public service, ucl, uncovering


Alan Renwick, Marc Esteve


Alan Renwick  00:08

Hello this is UCL Uncovering Politics. This week, we ask how should you run the public administration? Should administration be close to or insulated from politics? And what should the role of private and other non-state actors be? Hello, my name is Alan Renwick, and welcome to UCL Uncovering Politics, the podcast of the School of Public Policy, and Department of Political Science at University College London. We're focusing this week on public administration. That might not be the sexiest of topics mentioned of the word bureaucracy, rarely lifts hearts. But it's incredibly important for the development of public policy, and for the delivery of public services, and all the other things that the state does. So how can you run public administration? Well, we're going to be investigating that question by focusing on two recent studies. One of them looks at structures for the governance of public services. Are public agencies that are subject to tight political control better or worse than public corporations, which operate at some removed from government. And what's the impact of mixed public private ownership? And the other study looks at collaborative forms of governance where public private sector and nonprofit actors cooperate in developing policy and delivering outcomes. The common denominator between these two studies is that both are co-authored by my colleague, Mark Esteve. Mark is Professor of International Public Management here in the UCL Department of Political Science, and a prolific author on many aspects of public administration. And I'm delighted to say that he joins me now.  Welcome mark to UCL Uncovering Politics. And let's, let's start with a very broad question. We're dealing with a subject that as I hinted at the start there maybe has a reputation for being a tiny, tiny bit dull. So what makes you excited about studying public administration? 


Marc Esteve  02:06

Hello, and well, first of all, thank you very much for inviting me to this podcast, it's a real pleasure to be here discussing these topics with you. What makes public administration interesting for me? I guess, the best way to explain this is to start by saying that all my family, my father and my mother, are pretty much everyone in my family actually, even like my expose, they are public servants. So, and my father actually used to work as a politician. So I guess, that I've been able to see since I was a kid which is the impact that good public policies and good governance can have on our society. And I think that we live in, in in in a country, that actually most OECD countries, if you if you look at the percentage of GDP that is the spend through the public sector, it is usually more than 50%. Which means that if you're interested in making this world, a better world, I don't think that there is a bigger platform than the public sector to achieve that. And it might not sound as sexy as other ventures, I realise that. But I think that if we are able to improve the public sector, the impact that this can have, into our communities, into citizens, it is just huge. So that's the main reason why I decided to spend my career looking at how to improve or at least trying how to improve public services and governments.


Alan Renwick  03:38

And that sounds like a really big question, what what does it mean? What does it mean to improve public service? 


Marc Esteve  03:44

That is again, it's a very good question. At the beginning of your presentation, you said that my research basically focused on how to run an administration well. And I guess the first question that we should ask is, what does it mean for the administration to run well? How do we define 'well'? And I think here for for a long time, the focus was on how many services governments can offer, and which is the resources that are attached to each of these services? Now, over the last decade, the shift has focused on effectiveness and efficiency, whether we can achieve the so called value for money. So how can we get out of the system for each pound, each euro, each dollar that we include into this system. And I think this has really helped us to become better at managing public services. But over the recent years, we have realised that this is just a partial picture of the problem. We also have to consider the quality of the service because you can be extremely efficient at providing a very low quality service, but that's probably not what we want as a society. So I think that the first thing that we have to start discussing, or to try to find an agreement with is, which is the right balance between cost, efficiency, and quality. And then of course, the main question is how can we achieve the desired quality level for the number of services that we want with effectiveness? And I think this is something that is going to draw most of the research on policy implementation, and governments, in the next years.


Alan Renwick  05:36

So let's, that's really helpful as a starting point, let's move on to the first of the studies that you've recently published that we're focusing on here, because that focuses very much on cost, doesn't it? And thinks in terms of the cost implications of different ways of structuring public sector services.


Marc Esteve  05:51

Exactly. So in this first study, what we wanted to do is, we basically that the main reason why we started this is because we were able to get our hands into a very large database that had the variable cost for eight different public services. And it's very unusual to be able to compare costs among different organisational forms for multiple services. So we got very excited about this database. And then what we wanted to address with this was, well, let's have a look at what happens when you include the private sector in the delivery of public services. So let's look at how different organisational forms can actually have an impact on the cost of the service that is delivered. So, in particular, we were interested in what happens with the corporatization of public services. We know that now there are certain public debates about whether it makes sense to corporatize certain public services. Other views are more in favour of remunicipalisation, renationalising, certain public services. So we wanted to shed some light into, which is the particular effect of the corporatization of public services into the cost of these services. So what exactly does corporatization mean? So I attempted to sum up a little bit of this at the start, but I don't know if I got it right, in terms of the difference between public agency. So first, is there's a difference between public agencies and public corporations. And then there's some more stuff as well about mixed corporations. So there are these different styles. So sum up for us what is different between them. Yes, so, the first thing that I shouldn't say, especially if we have an international audience is that public corporations mean different things in different countries. Because they're very much, it very much depends on the law that each country has. But we define public corporations as single purpose organisations that operate under the private law, and they tend to distance themselves from the influence of elected officials. So public corporations are the way in which we can implement public services, mainly driven by managerial logics, not so much by the logic of elected officials. So politicians.


Alan Renwick  08:19

So they're still owned publicly.


Marc Esteve  08:20



Alan Renwick  08:21

But the kind of structure of governance is more akin to what you would find in the private sector. 


Marc Esteve  08:25

Exactly. And in the paper, we discuss it, there are three main characteristics that distinguish them. The first one is that public corporations do not operate under the classic administrative law. And this means that they are less subject to financial scrutiny. The second aspect is that precisely because they do not operate under administrative law, they have much more flexibility when implementing personnel management practices, okay, because they operate under a private or a commercial law. And the third aspect is the public corporations are not restrict to operate under the jurisdiction of administrative law. Which in practice means that they can expand their services according to their needs. So, in a nutshell, public corporations should have more flexibility than main governmental bodies, or, for instance, public agencies. Now, when we discuss in the paper, though, is that you can have two main types of public corporations, you can have public corporations, in which only the public sector operates, but you can also have public corporations in which you have the participation of the private or the nonprofit sector in which, in which case, we will be referring to mixed firms.


Alan Renwick  09:55

Okay, so those are the different kind of structures that you're looking at and I guess from what you've said there that the expectation would be that public corporations with like greater independence and being less subject to those constraints would be able to operate at lower cost than public agencies that are more tightly bound into the state. Is that the expectation that you have?


Marc Esteve  10:20

Exactly actually, this is the first hypothesis that we have in this study, we think that because they have more flexibility in their management practices, public corporations will operate with lower costs when compared to public agencies.


Alan Renwick  10:36

And then, and then mixed corporations with more private sector engagement, I guess the expectation would be that they would be even, even lower?


Marc Esteve  10:43

Well, you're actually the empirical evidence that exists on this particular topic is mixed. Because on the one hand, it's true that if you operate with the private sector, the private sector is known to be more efficient, or to be able to operate with lower costs. But at the same time, it's also true that if you mix the public and the private sector, you may as well get the worst of both worlds. So here, if you mix two different organisations, you're going to have higher transaction costs, you're going to have more principal XXX layers within that relationship, which means that, you know, we were not sure here, in our hypothesis, whether in a mixed corporation, we would be getting the best of both sectors, or the worst of both sectors. And we decided to hypothesise for the second option. So we actually thought that mix corporations were actually incurring higher costs than public corporations, because we thought that the coordination costs among both sectors would actually have an effect on the final costs of the service. So we think that here, cooperating probably has a cost.


Alan Renwick  11:56

So listeners will now be desperate to know what the answer to that is. But we do like a bit of methodology on this podcast. So before we get to the answer, we'd better just explore the methodology a little bit. So you mentioned that you've got this fantastic database from local public service delivery bodies in Spain?


Marc Esteve  12:13

Yes, Alan, thank you for your question. So in this study, what we have is eight local public services. So we have data from eight local public services. And these are solid waste collection, waste treatment, sewage, street cleaning, waste related environmental protection, water distribution, libraries, and social services. Probably, the audience will already be thinking, it's difficult to imagine a library that operates under a mixed public corporation. And they are right, we do not have cases of every organisational form for the eight local public services. But what we have here is information about the effective cost. Which means the cost incurred by local governments to deliver each of these eight services. And this is data that comes from the Ministry of Finance and Public Administration of the Spanish government. So what we have in this study is we are able to differentiate between four main organisational forms. The first one, are public agencies, which is the baseline group that we use. This will represent classic traditional public administration from within the public sector. In the next configuration, what we have are= public corporations in which the private sector does not participate. So 100% public corporations. The third category would be mixed public corporations, in which the private sector has a minority of the shares of the organisation. So there is a minority of the participation of the private sector. And then the fourth category would be mixed public corporations in which the private sector has a majority of the shares. So it means that the participation of the private sector here is much more important. These are the four categories that we are able to compare.


Alan Renwick  14:11

And so you've got public services being delivered by different versions of these in different local authority areas in Spain, basically. And that leads on to the crucial question then of what do you find? What what what does this variation lead to? 


Marc Esteve  14:27

So the results, quite surprisingly, overall, we've seen that there is no evidence of cost advantages from the corporatization of public services when compared to public agencies. So, are public corporations better in terms of costs than public agencies? The answer for these eight services is 'no, they are not'. Even more, when the public firm, when the public corporation has private participation, but retains government ownership majority. So when the private sector participate with a minority of shares, the service production tends to actually be more expensive. This might well reflect we think that the government's costs of mixing the public and the private sector, could be particularly acute in this organisational form.


Alan Renwick  15:20

So this is really interesting, because you said at the start, that we shouldn't just be interested in the cost of delivering public services, we should be interested in the quality of services and various things. But presumably, in this study, you're focusing on cost, because the argument in favour of corporatization is, it will deliver services at lower cost. So you're kind of focusing on the key argument for corporatization. And you're finding that that key argument doesn't hold up? 


Marc Esteve  15:47

Yeah, because the whole idea here is, well, if we are able to give more managerial flexibility to the public sector, this means that public managers will be able to bring the costs down of a public service. But funnily enough, what we find is that this is not the case, at least for the eight services that we have analysed in Spain. Now, we do not know whether this is something that would apply to other countries or to other services. But using a rather large database with multiple  services with multiple characteristics, what we find is that public corporations are not more cost efficient than public agencies, and that the collaboration with the private sector might work well, but when the private sector operates with a minority of shares in the in the joint venture, what we have is actually higher costs. Now, this is what the data is showing. But then we can, of course, try to interpret this data and try to provide reasons on why this might be the case. And what we try to delve into in the in the last section of the article is, why would this be the case? Why public corporations have higher costs than public agencies? And here we propose we suggest an argument. And again, this is just our thinking, the data doesn't allow us to see whether this is the case or not. But we think that the main reason why public corporations may incur higher costs is because the directors of these public corporations do not have a very strong incentive to bring the costs of the services down. For example, Alan, imagine that you were now the head of a public corporation, that was focusing, for example, in running the London tube. And your employees told you that you either increase their salaries, or they are gonna, they're gonna go on a strike. Now, the public doesn't really change, citizens in London doesn't really change, they don't even know, which is the overall cost or the efficiency of the service. What they want is a tube that runs. that runs well in time, right. And that brings them from point A to point B. Now, if you are threatened with a strike, I think that the incentive that you have as a as a director here is to make sure that the services keep running. Which means that what we have found in some of the cases that we've been able to analyse individually, is that the working conditions of employees in public corporations are actually better than those working in public agencies. In other words, it might be more expensive to run a public service through a public corporation. Because the fact that public corporations do not operate under the Public low is not really bringing the cost down, on the contrary, is allowing, for example, unions to have more power to improve the working conditions of employees. Now, I'm not suggesting this is something wrong, not at all. But we have to decide how we're going to make the best of each of these organisational forms.


Alan Renwick  19:11

And that's a that's a really interesting argument. And one that might surprise some listeners in the UK, at least who are used to the idea that contracting out, and, you know, removing public services from very close political control to something more remote is associated with harm to labour standards, lower pay for people involved in there. So you're suggesting that at least in some circumstances, we get a different dynamic going. And I guess the interesting question then is, well, what are the circumstances in which we get different dynamics?


Marc Esteve  19:44

Exactly? That's a very good point. So I think that here the difference is when you contract out the delivery of public service to the private sector, then the incentives of the private sector are certainly to bring the costs of the service down, because then they're basically going to make more money. Right? But when the corporation is fully public, they don't have this incentive. They don't have the incentive of bringing the cost down, which means that the employees of this public corporation may end up working with better working conditions, then those operating under Public Law. So under main governmental bodies, or in this case under public agency. So I think that, I guess the argument here would be, it's, you know, public corporations do not operate under public law, but their managers are still operating under political incentives. And this might not be a very good mix.


Alan Renwick  20:40

And we often make perhaps an overly simplistic distinction between privatisation and nationalisation. And there being basically two ways of doing things, but actually, you're suggesting that a lot of what goes on is, is more subtle than that. And there are variations somewhere in the middle here. And actually, we need to be paying close attention to the dynamics within that.


Marc Esteve  20:59

Absolutely, I think, obviously, that the devil is in the details. But for example, I tend to warn our students to use the word privatisation with caution, because to me, privatisation should refer to those public services that we have decided that they should not be considered public services anymore. Right? Those public services that are outside the scope of the implementation of government, in which case, the only thing that governments does, as with any other economic activity of the country is to basically regulate the rules of the of their game. Now, when we have private sector actors, helping in the implementation of public services, we have cases of indirect implementation of public services. Externalisation, contracting out mixed firms, etc. And I don't think we should call this privatisation, although we are extremely used to do it. But I think there is a huge difference between those services that are public services, whether they are implemented by the public or the private sector or a mixed combination of both. And those services that have been privatised, and therefore they are not considered, or they should not be considered public services anymore. And this is the first of a series of studies that we are starting to, to create, to make, to try to understand, which are the characteristics in which each organisational form would work best for our citizens? Would actually allow us to implement better public services? Which is the ultimate objective that we all have.


Alan Renwick  21:38

We better move on to the second paper, or we'll never have time properly to discuss it. So this is a paper again where you're looking at the involvement of non-state actors in in state activities if you like. And you're including her private sector actors, and also non-profit actors. And you refer to the notion of 'collaborative governance' to describe the sorts of relationships that you're exploring. So I guess one question to ask here is, what do we mean by, what do you mean by collaborative governance? What's going on here? And then the second question is, so we've been talking about a paper where you're focusing on the impact of these arrangements on costs. In this paper, you're looking at impact on public perceptions and public confidence in the arrangements. So we have quite a different focus in terms of the outputs that you're you're studying.


Marc Esteve  23:34

Yes, you're right, Alan. So before delving more into the characteristics of each organisational form, we wanted to, because we were discussing with some of our colleagues, what do citizens think about different organisational forms? Do they care? Do they even realise sometimes who is the owner of the service, the public service that they are using? Whether the hospital is public or is private, or is a mixed firm? Do they even care about that? So we decided to take a broad approach to service delivery. And here we talk about collaborative governance. And we understand collaborative governance as those situations in which governments do not implement a public service alone, but in collaboration with the help of other organisations. And these can either be public, private, or from non-profit initiatives. So in this paper, what we wanted to do is well, let's start with the first experiment, trying to see whether citizens care on who are the actors involved in the delivery of services. And then we also include another treatment, which is, let's tell them the performance of the policy to see whether this is what actually really matters because we thought in our minds basically that the underlying hypothesis that we had is, we don't think citizens care that much about who implements a service. We think that we're they really curious, they want a service that works. Obviously, as any time that you do a study, we were able to answer this question, but the answer is more nuanced than the one that we thought.


Alan Renwick  25:18

So before we get to the answer, we should as ever just explore the methodology quickly. So so this is based on a survey.


Marc Esteve  25:26

This is an experiment that we implemented the a survey on 1,470 US citizens. And here, what we wanted to propose to them or experiment was a classic two two two two two factorial design in which we had.


Alan Renwick  25:44

Now, that's a very complicated phrase, you have to explain what this means.


Marc Esteve  25:47

We are changing the survey that we are sending to respondents. And we have a control group. But then we also have what we call a treatment groups. And these treatment groups basically, is a set of different groups, a set of different surveys. And in this surveys, we play with three main experimental factors, which are representation, how many organisations are involved in the delivery of the service. The second one is performance information. And the third one is the issue [of] complexity. So according to this, you can have six scenarios. Right? One in which the representation is low, or one in which the representation is high. One in which the performance information, it's telling you that the policy is not working very well. And one that it's telling you that the policy is indeed working very well. And finally, issue complexity. So one that tells you the issue that this policy is trying to tackle is very simple. The issue that this policy is trying to tackle is very complex. And then obviously, we have a combination of all these different experimental factors. This is what we call technical the name of this is a between subject two per to per two design.


Alan Renwick  27:09

So you've got the representation feature there is basically varying the process of service delivery, who's involved in doing that. Whether it's basically just the government or it's the government, plus some private sector actors, plus some nonprofit actors.


Marc Esteve  27:26

Exactly. So we, yeah, we basically have your representation or we call of different stakeholders. The idea, basically, is to compare whether the public sector does it alone, or does it in collaboration with other stakeholders, other key actors in the field.


Alan Renwick  27:41

So the crucial question as ever is, what do you find? So do you find that people care about the process through which the service is delivered? Or do they just care about the performance of the site?


Marc Esteve  27:52

Well, so what we find is that the main explanatory variable on what do people perceive about this service is the performance of the service. Which is something that probably you know, the audience is gonna be thinking, 'well, you didn't have to go through all the trouble of implementing this talent to find this out'. It's true. But we also found a few more nuanced aspects that I think are worth considering. So, for instance, those individuals or. Well, let me explain this, I guess, with the right order. So first, the results that we find demonstrate that positive performance information has a very robust positive influence on participants perceptions of the collaborative governance. And this happens among the three aspects of collaborative governments. Performance information is by far the one that has a larger effect on the perceived legitimacy. And this effect, again, is consistent across the different outcomes variables of the study. But we also see that when we involve diverse stakeholders, specifically with a balance of actors between the public and the private and nonprofit sectors, citizens also perceive that the service has higher legitimacy. So while the main driving indicator here is performance, we also see that citizens feel more comfortable when more actors are involved in the delivery of public service. And there are differences according to some individual characteristics. In this case, we were able to ask participants, the level of trust that they have on public organisations. And what we find when we consider these is that those respondents that have high trust in public organisations do not really care about representation in the service delivery. On the contrary, those individuals that do not really trust, do not have high trust on the public sector are those that particularly welcome having other stakeholders from the private or nonprofit sectors involved in the delivery of the service.


Alan Renwick  30:10

And I guess the next question that comes to my mind, but I think it's one that's going to require another study, isn't it? Is, is it the presence of private sector actors that people like? Is it the presence of nonprofit actors that people like? Is it just anyone other than the politicians is good in some people's minds? But I think you can't get into that in this study can you?


Marc Esteve  30:31

No, I'm afraid we cannot.


Alan Renwick  30:34

We will just have you back on when you've managed to do a follow up study!


Marc Esteve  30:37

That will be terrific. And if anyone in the audience is interested in helping us to answer this question, please let them contact us. Because this is certainly one of the things that now we would like to study, which is, who the citizens want to see in the other end of the public service? So we've understood that the main, the main objective, the main interests that they have, obviously, is to have that tube running so that they can go from point A to point B. So what they really cared about is having a service that works well. But it's also true that some citizens in the case of our study, a pretty large percentage of those respondents that we had in this study, also care quite a lot about who do we have at the other end? Whether it's a combination of actors, or it just the public sector. Now, which is the right combination, which is your question, is something that we still cannot know. But I think it would be very interesting to see how some individual characteristics explain, which are the preferences that citizens have to say, you know, who they want to see, at the other end of the service?


Alan Renwick  31:53

We're gonna have to wrap up there, I'm afraid, which is a great shame, because they've got a zillion more questions, but we're just going to have to have you back on the podcast when you've done some further work on this. And you've definitely demonstrated that examining processes of public administration is not dull, and indeed, you've demonstrated that the general public care about it, it's not just we academics who care about it. So thank you so much, Marc, it's been really great. 


Marc Esteve  32:15

Thank you very much Alan. 


Alan Renwick  32:16

We've been looking at two papers, both co authored by Marc Esteve. The first 'The Costs of Corporatization: Analysing the Effects of Forms of Governance' was published in the journal Public Administration earlier this year. And the second, 'What Drives the Perceived Legitimacy of Collaborative Governance: an Experimental Study' was published in Public Management Review, also this year. And you will find links to both papers in the show notes for this episode. Next week, we're looking at automation and immigration and how the ways we talk about them can drive policy outcomes and patterns of power. And I'm very happy to say that that episode will be hosted by a new member of the UCL Uncovering Politics team, my colleague, Dr. Emily McTernan. Emily is an expert in political theory. You may have heard her as a podcast guest back last January, and she discussed her work then on taking offence. It'll be wonderful to hear her as a more frequent voice in future episodes. Remember to make sure you don't miss out on that or other future episodes of UCL Uncovering Politics all you need to do is subscribe. You can do so on Apple, Google Podcasts or whatever podcast provider you use.  I'm Alan Renwick. This episode was produced by Conor Kelly and Eleanor Kingwell-Banham. Our theme music is written and performed by John Mann. This has been UCL Uncovering Politics. Thank you for listening.