Public Institutions

Baroness O'Neill of Bengarve Excerpts

Thursday 30th June 2016 Lords Chamber

[Baroness O'Neill of Bengarve (CB)]

My Lords, I, too, had taken this debate to be rather broader than the question of the declaration of interest, let alone one specifically about the sources of pensions. It is not irrelevant, but I think that this is, as the noble Lord, Lord Norton of Louth, has just shown us, something of very much broader concern. We are, I suppose, 30 years out from what is taken by some to be a revolution in accountability and transparency. I have no doubt that that revolution was, in many ways, made possible by the advent of forms of IT—it was much easier to make far more public. It is, I think, reasonable to ask ourselves: is it working? Are more accountability and transparency having the effects we hoped they would have—not just in public institutions or in Parliament, but much more generally, in all institutional life and perhaps beyond institutional life?

I do not think it is actually that new—we have had some forms of accountability and transparency for decades, perhaps centuries. After all, what is financial audit of a company, what is the publication of accounts, except accountability and transparency? Yes, we all know that it has a very limited focus and there is an awful lot you cannot learn by reading a company’s accounts, even if you are more of a forensic accountant than I am. However, we have to ask ourselves, many years out, whether this attempt to extend accountability and transparency has worked the way we thought it would. Has it increased public confidence? The evidence is very mixed and some of it is very discouraging.

It is a commonplace to say, “Oh, by the way, trust has declined”. I am cautious about saying that, because the empirical evidence we have is extremely mixed and people tend to rely rather too much on opinion pollsters, who ask questions about generic attitudes. They ask, “Do you trust doctors?”, “Do you trust politicians?”, Do you trust the media?”. People say, “No” or “Yes”, or they rate trust on a one to five scale, and we all know those scales. However, when you look at the time series—and we have it for a few, such as politicians and journalists—you discover that the people who are most mistrusted now were also most mistrusted 25 years ago. Conversely, judges and nurses, who were among the most trusted, remain among the most trusted. I shall come back to this but we might want to think a bit more critically about what we are trying to look at when we talk about trust.

What should we do if we hope to have greater public confidence? This may be the wrong question. Accountability and transparency might surely be expected to have more effects on trustworthiness than on trust. That seems intuitively the case. If I know that the information about my institution or my conduct is going to be public, one would hope that I would be a bit more careful about whether I behave in a trustworthy way. Whether other people trust me is, of course, up to them. Trust is something that others give, not something that I can control. There is no automatic read-across in my view from accountability and transparency to that magic goal of more trust, although there might be a better connection to trustworthiness.

I shall take accountability and transparency separately. A good question would be, “Does accountability always increase trustworthiness?”. The answer seems to be, “Sometimes it does, but not always”. We have quite a lot of let us call it, rather simply, unintelligent forms of accountability around, and this Parliament is responsible for increasing their number in a very generous way. I give noble Lords just one little example from schools. It is the classic example of forms of accountability that create perverse incentives. A performance indicator for schools is the number of A to C passes at GCSE, and there are other benchmarks for primaries, sixth forms and so on. What does a rational school do? Well, they want to get more A to C passes per pupil. So a very good thing to do would be to put the pupils in for less demanding subjects and to withdraw those who will not exactly shine from taking too many, or too difficult, subjects. A metric that was meant to improve children’s education thereby ends up damaging it. Perverse incentives are littered across our system of accountability. Of course, they do not improve trustworthiness—people feel that the schools are not doing a better job by children—but nor, probably, do they improve trust.

However—this is much lower profile but I suspect that it is more serious—the attempts to increase accountability have a deadening effect, even when there is not the high-stakes disaster of the perverse incentive, simply because of the burdens of compliance becoming too great for too many people who should be looking at the task they are doing and the people they are meant to be serving. I give noble Lords an example. A few years ago I chaired a little inquiry into the safety of maternity services in England and Wales. One of the midwives said in evidence, “The problem really is that it takes longer to do the paperwork than to deliver the baby”. There is one thing that a midwife is meant to be doing—namely, keeping an eye on the mother in labour and on the newborn, not doing the paperwork. That, I fear, is an example of a rather dangerous burden of compliance, but we all know the word “tick box” now. Tick-box compliance is with us for a reason—namely, we thought that more compliance and more accountability would always be better, so we ratcheted up and up. Should we wonder, then, that people have to spend so much of their time—and misspend their time—doing the compliance? If noble Lords read—as I did recently—Swimming with Sharks, which is about the 2008 crisis in the financial sector, they will discover something very interesting about compliance. When the author, a Dutch journalist called Joris Luyendijk, started asking people in banks, “What sort of animal are you?”, the traders said, “I’m a lion”, “I’m a wolf”, “I’m a fox”, “I’m a bird of prey—an eagle, perhaps”, but in compliance the people said, “I think I’m rather like a beaver. Beavers don’t have much time for servicing others, they’re too busy chewing”. So we have to be pretty cautious about stupid forms of accountability—and we have fantastic numbers of them.

Does increasing transparency increase trustworthiness, even if it does not increase trust? Not always. Consider how often your Lordships have encountered the minutes of an institution—a public institution or, for that matter, a commercial institution—and thought, “Well, these minutes are mightily bland and uninformative”. But we should ask why they are bland and uninformative. They are bland and uninformative because somebody told the person who was writing the minutes that they really should not be too sharp-edged or create any controversy and that they should damp it all down, forgetting that a minute is valuable only if it is an accurate document of record. That is where the transparency that is so commonly the case for organisations now—you have to publish the minutes—can be damaging.

However, sometimes more transparency does increase trustworthiness. Publishing a list of interests is quite a good idea, although I note that throughout the institutions many people confuse the question of whether people have interests with the question of whether they have conflicts of interest. They are quite different things. An interest is a standing interest; a conflict arises in a particular situation and has to be dealt with.

I will give your Lordships an example of where a lack of transparency does lead to a lowering of trust. Consider the state of Delaware. Delaware has almost as many corporations registered in it as citizens. Why? Because you do not have to declare the names of the directors—no transparency, it is not in the public domain. That is why any corporation that wishes to do something that its directors might not wish to have attributed to them will incorporate in Delaware. It is just a small example of offshoreness, albeit within the jurisdiction of the United States. In general, declarations of interest and the identification of conflicts of interest can be helpful because they help us know who is responsible for what.

Does transparency increase trust? The empirical evidence is extremely poor on this point. It looks as though transparency often reduces trust. Of course, there has been only sporadic research but it is not very convincing. I think we know why. The noble Lord, Lord Norton of Louth, touched on this. Very often transparency is just too little to increase trust or trustworthiness. Transparency can increase trust only if it is accompanied by, as the noble Lord said, engagement, dialogue and communication.

What is transparency? It may be fashionable but it is actually a matter of putting stuff in the public domain. That is easy in the age of IT. You can tip shedloads of stuff into the public domain, but guess what? It is not something that everybody reads. When I hear in an institutional context people saying, “We should be more transparent”, I think, “You want to let yourselves off lightly, don’t you? That won’t be enough”.

Why does it not work? First, it is often not accessible. People do not always find the stuff. Those of your Lordships who know how long it takes to become familiar with the website of any one of our government departments or any business that tries to do it well will realise that tipping it on to the website does not always mean that people find it. Worse, it is not merely that it is not in practice accessible to everybody; when they get there it is often not intelligible. That really does put people off from placing their trust. When they can understand roughly what it means, it is very frequently not assessable; that is, they cannot judge what it all comes down to.

A few years ago, I took part in a Royal Society working party on science as an open enterprise. Science depends immensely on communication and transparency, but we came through that thinking that mere transparency and openness are never enough. What we need is intelligent openness, which means taking account of which audiences there are and what they can follow—and following up on it.

If we were successful in being transparent, what could we achieve? Could we perhaps make it easier to make judgments of trustworthiness? I do not speak about trust but trustworthiness will need three things: evidence of honesty, evidence of competence and evidence of reliability. We are not going to get there by mere transparency. Too often, it is unintelligible and unassessable even if people find it. In short, transparency, that fashionable nostrum of the 1980s, has not delivered more than we should have expected. It might have delivered more or less what we expected but every time I hear someone say, “We should be more transparent and start to flood people with data”, that is just not enough. Neither stupid accountability nor stupid transparency will work.

Finally, to take one last kick at something, the question: “How we should rebuild trust?” is very unfortunate. It is for other people to give or refuse trust; we can be trustworthy but they give trust. So that question tells me that people already have their PR hats on and are thinking, “How can we persuade other people to trust us?”. It has a certain whiff of a conman’s question. If we think about the well-known Mr Madoff, who made off with so many people’s savings, can you not just hear him saying, “How can we rebuild trust?”, and what would he mean by that?