Sunday, 29 September 2013

Institutionalised Truth

I believe that Democracy has some form of intrinsic value. The concept that each individual within a country has an input into the systems that govern their lives has a powerful allure. However in order to maximise this value, a democratic state must ensure that the information required to make informed decisions is readily available to its members.

By casting my vote I am making a choice which I believe will most likely bring about the state of affairs that I wish to see in the world. This requires a degree of information which allows me to predict, of the choices available to me, which will most likely result in the sorts of outcomes I wish to see occur. As our information becomes poorer, there is a substantial risk that our choices become less related to the outcomes we wish to see – as this disparity grows, the value that democracy provides to us diminishes.

Information within our society is currently too expensive, in terms of both time and resources, and consequently damages our democracy. It is expensive because, while it is easy to obtain thanks to the internet and 24/7 news channels, it has become increasingly difficult to verify. The sheer number of news sources available means that for all but the most basic questions, you will be able to find at least a semi credible source to support most sides of an argument. In the face of information overload, with limited realistic ability to discern what is true and what isn’t, it is hard to believe that one’s ability to vote according to their outcome preferences are not harmed.

My proposal is that the state set up a series of independent, highly specialised institutions, each of which have a narrow remit of either producing, or assessing the veracity of, certain information. The value of these institutions would lie in creating a clear authority on certain questions, allowing the media and general public to be better informed.

The institutions must be specialised, otherwise they risk losing the knowledge which makes them an authority on the issues they talk about. They must also be funded by the state, firstly to ensure they receive sufficient funding, but more importantly so that their neutrality is ensured and they are seen as a credible source of information (compared to say news channels and think tanks). The state has a successful track record in creating respected neutral institutions (for example central banks).

I admit at this stage my argument sounds a bit like a naïve rant by a liberal (semi) intellectual who is frustrated at the state of political decision making in the west. However, while this may all be perfectly true (it is), it should not remove from the important point that this would produce clear value to society.

What I am proposing has already successfully been done in some countries to a limited extent. Fiscal councils around the world, such as the OBR in the UK, provide independent projections of government budgets. In the US and the Netherlands, fiscal councils also evaluate the proposed spending plans of opposition parties so that voters may know whether they are credible options. Politifact in the US assess the degree of veracity in the statements of politicians (however as it is a non-public institution it comes under attack for lack of neutrality – ironically by both sides of the political spectrum).

My area of knowledge is primarily economic – hence the examples given above, but I could easily see room for institutions in the areas of environmental science, medicine and law among others.

There are some obvious drawbacks of this proposal. Firstly there must be a limit of some kind imposed to the number of institutions and to their scope. What kind of questions are appropriately addressed by such institutions? How much information must the state provide the public with before it has absolved its duty? We clearly couldn’t have a thousand government owned and operated think tanks.

On a more fundamental level, I worry about the impact of having a state line on “the truth”. Many of the questions I would hope are assessed by such entities, are difficult and complex. A state line might stifle discussion on these areas and be a way to control debate in a country. It goes strongly against the Millian ideal of discovering truth through extensive public argument. However while this risk exists in the extreme, I do not feel it undermines a cautious move in the direction I advocate.

So in summary, my argument:

1) Democracy has intrinsic value.

2) This value is linked to the information citizens have access to.

3) Currently information is too expensive in our society and the state has an obligation to  lower this cost.
This cost is due to the difficulty in verifying the validity of arguments.

4) The state should set up focused independent institutions which seek to provide clear  guidance on issues of import to the public.

5) Institutions of this kind do exist to a limited extent currently, do work and are a positive  impact on society. We should expand on these.

Open questions:

1) How much information does the state need to provide to absolve its obligation to its  citizens?

2) What is the limit to the kind of questions we are comfortable providing a state researched  view on?

3) Why does publicly funded media fail to provide the information the public requires? / Does  it fail?

4) What are the precise questions we would ask these institutions to answer?

I’m sure there are many more beyond this initial list.


  1. Will, thanks for this post. I agree that the topic of information for citizens is hugely important. I only partially agree with your proposal of state-funded research institutions. I think what we should aim for is a pluralism of private and public research institutions, and also some pluralism within state-funded institutions. Your proposal sounds too "platonic" to me, in the sense of the state deciding about what counts as "good" or "bad" information. This not only narrows down the possibilities in the generation of knowledge, there is also huge potential of abuse in case a state turns less-than-liberal in any way (just think about the kind of research institutions a stoutly religious state would have...). I think that to some degree the complexity and plurality of information is the price we have to pay for having a liberal system with many different points of view - but as I said, I do agree that state funded research institutions can play an important role in this process.

  2. Hi Will, I haven't thought much about this before, thanks for the post. In response to Lisa's comment, don't we already have lots of private think tanks and institutes? I assume that one would need to evaluate on a case by case basis the legitimacy of state-sponsored guidance to processing information. But it seems safe to say that the closer it comes to assessing facts - rather than value-laden judgements - the less problematical.

    It occurred to me, while reading your post, that the same premises may work towards the conclusion that school curricula ought to include citizenship classes to help us sort out the enormous amount of information. And, in particular, to spot tendentious presentation of information; for instance, to teach people how to read statistics...

    1. I definitely support this proposal - after all a state must equip its citizens with the tools required to operate meaningfully within society.

      However - I do not think this can be a replacement to the institutions I propose as Andrew suggests it could be. While this lowers some of the costs to verifying information, I do not think it solves the principle issue which is that many debates have got to such a level of complexity that without being a specialist in these areas we cannot distinguish between sources of information.

      For example I do not think I could accurately determine the risk that nuclear energy presents as a source of long term energy supply to the UK and I would consider myself as having a fair degree of statistical understanding.

      Part of the problem is that for many important debates which political parties don't want to get involved with, there is a tendency for only individuals with more extremist tendencies to engage in them. (This is perhaps another advantage of these institutions - publicising issues which there is little political will to discuss).

  3. Will, thanks for the post. I have two queries for you, both partly leading from the comments offered by Lisa and Anca:

    1) With Lisa, I am worried about state-linked guidance on 'the facts'. One worry here would be the concern Lisa offers regarding what a state could do with that power (though I suppose a good system of checks and balances would reduce this worry somewhat). There is also a concern regarding the heuristic implication of state-sponsored facts. (I guess this speaks on the opposite side of Anca’s concern in a way; even with a plethora of non-government sources of research, what kind of counterpoint would they provide to the state’s authorised institution?) I would have a third worry too along the lines of ‘where facts exists’. In some cases perhaps there are clearer facts, maybe like the environmental issue. But one of the great functions of plurality of information is precisely that it helps us access information in many diverse corners which is actually fruitful for ascertaining what is correct and what is a good way to go about pursuing certain ends. For these reasons (and possibly others), I wonder whether it might be preferable to focus our attention on things like Anca’s thought about citizenship classes. If we could improve capacity for finding and verifying information among individuals, that, too, would cut their costs, but with probably less susceptibility to information control by the state. (In formal terms, I guess what I am asking is whether premise 4 of your argument follows from premise 3; might not alternative routes be preferable?)

    2) Building on the end of this comment, I wonder if I could ask you to expand on the following. You note that “2 – [democracy’s] value is linked to the information citizens have access to”. No question here. But I would guess it would also be true that it depends on the extent to which citizens are willing to access and utilise that information in certain ways. Thus, my query is: would the underlying thrust of your argument also point towards a duty (enforceable or otherwise) to engage in politics in a certain manner?

    1. Ok addressing your second point Andrew (which I’m sure is trying to make me look bad):

      Question: Do citizens have a duty to engage in politics in a certain manner?

      Inflammatory Answer: Only if they want to vote.

      Actual answer: Let me sit on the fence a while and think about it.

      My original argument is concerned with what is necessary for a citizen to extract the value from democracy should they want to. For me to accept democracy, I must have the ability to extract the value that it offers – without sufficient information I cannot do this.

      If 90% of votes are entirely random let’s say (cast by robots) and only 10% are active choices about how individuals would like to see the state work, do I again lose the value of democracy as these choices (and the process) which I value are being ‘diluted’ by random chaos?

      This could conceivably be an argument to insist that voters have a duty to engage in politics to a certain degree in order to not undermine the value of democracy – however I am not sure this minimum threshold would be high enough to be meaningful in practical terms. How high must a citizen’s engagement be to move from ‘dilution’ to ‘active choice’?

      I would agree in principle though that if a citizen really had 0 concerns about his vote, but just voted randomly for fun, he would be undermining the value of democracy to a degree.

  4. Thanks for the comments. In this response I'll try and elaborate a bit more on the risks this proposal creates and why I think they can be limited – also I’ll flesh out a bit more what I envisage these institutions producing.

    Andrew I'll reply to your second point in a separate comment.

    In short I think that these institutions would be part of a plural debate, but would also help to hold open debate to account and play some sort of an arbiter role (A neutral source which could knock down flagrantly untrue statements, without accusations of bias).

    In response to the ‘where facts exist’ problem, I think they will ultimately have to be asked to think about tough questions which don’t clearly have a factual answer, however I think that this is not necessarily a problem, so long as the work put into these questions is transparent and open for criticism.

    As a somewhat longer response:

    I agree that there is significant (theoretical) potential for abuse in this system, a fact that I acknowledge in the final paragraph of my post. Clearly there must be an open and transparent debate on issues to limit the risks created. I do not believe that such a state institution would eliminate the multitude of think tanks and institutes that currently exist. These public institutions would be held to account by such sources and as such be made more credible as a result. They are part of a plural system – not an alternative. Ideally these institutions provide a starting point for citizens to become informed on an issue, from which they can read numerous other sources to develop their understanding if they so wish.

    I would go so far to say that these institutions could only survive if they were seen to be neutral and applying ‘best practice’ in the work that they do. Fiscal councils for example are held in high esteem by the media on both the left and the right of the political divide - primarily because their methods are transparent and high quality.

    I am more concerned with the issue of what can actually be addressed by these institutions. If we limited the debate to purely factual issues there would be limited risk – however I also think there would, in reality, be limited issues this would actually help us with. To be truly effective, I think these institutions must wade into murkier territory and address questions which are less clear.

    I think a good way to describe these institutions is as permanent parliamentary commissions. When government has an important issue to address it tends to form a neutral body of experts to work on a problem and report back. Typically these reports, while not universally accepted, are considered to be of high quality with useful policy recommendations. At this point though the commission is dissolved and the reports gather dust in a draw. By turning these commissions into permanent specialised bodies, you firstly create an advocate to remind the public of this work, secondly you create a system which can respond much more rapidly to important questions – likely because it has been done in advance. You will also gain dynamic benefits due to the institution developing skills over time by focusing on one question.

    Examples of reports that have been produced: On banking standards, the environment, energy generation, reform of the house of lords, taxation in the UK, inequality etc. These are consistently thrown away and ignored because they are politically difficult to accept and have no institutionalised advocates.

    1. Will, a spontaneous thought on your last point: isn't it a different - and arguably as deep - problem that all the knowledge that is in fact there does not really have an impact on policy making? I happen to have read the UK report on improving banking, and it's a great piece of work - but so far, it hasn't had much impact, as far as I know!
      To play devil's advocate: let's have all the research institutions you suggest - it won't change policy making, so why should citizens bother with it?

    2. I think the Banking proposal is a perfect example - it took a long time to produce, meaning that by the time it was released much of the financial reform was already on its way and difficult to change.

      Furthermore it did not have a platform from which to publicise this information once it was produced - the commission was disolved. I would hope that with a permanent institution in place it would be better placed to promote it's research.

      But I do fear to a degree you are right - good information can only take us so far. It is a necessary, but not a sufficient condition of a well functioning society.

    3. (as an aside I also think the UK report on banking standards is an excellent document).

  5. Being a bit more flippant, I would kind of like to see a small civil service of experts designed to inform and advise the public rather than the government.

    1. I think that is a very slippery slope - government departments devoted solely to 'educating' the people that elect their bosses.

  6. Thanks for the post, Will. Like others here, I share many of the worries that you highlight, but I am little more reserved about endorsing your proposals for reform.

    First, I worry about how effective state-run information institutions would be. This is for a number of reasons. First, it seems as if they might be prohibitely expensive to finance - that is, in order to offer credible advice they would surely have to be generously financed so that they are able to match the kinds of research done by a wide range of other (private) institutions. Second, I worry about whether the general public would be inclined to listen to the advice of state-run information institutions. I think many people prefer to get their information from Newspapers and TV, which are conventient and can tell the audience exactly what they want to hear. As further evidence of this point, let us consider two cases. First, let us note that though '[t]he state has a successful track record in creating respected neutral institutions (for example central banks),' it is not the case that the general public are very engaged with these institutions - most people, that is, are not aware of the central banks reasoning, advice, etc. Second, let us note that there are already some information institutions that aim to be neutral (Media Standards Trust, Full Fact), but the impact that these institutions have is minimal when compared with the media.

    A second set of worries relates to the complexity of information and, more importantly, how many judgments seem to involve at least some normative element. The kind of information research by such an institution would surely be a political decision and thus not in any true sense 'neutral'. For example, the decision to scrutinise unemployment figures (rather than inflation figures, for example) might be seen as a political move designed to further the Left's employment-based agenda.

    To conclude, I'd like to throw a further proposal into the hat: tighten media regulation. It seems to me that it will be very difficult to resolve the problem you highlight without changing media practices. This could be done with the introduction of new legislation, or through the establishment of an ethos of responsibility in the media industry. These steps, it seems to me, are necessary for progress to me made.

    1. I am definitely in favour of a more highly regulated media.

      Two brief points though: Firstly - I think part of the role of these institutions would be informing the media as well as the public. The existence of the information these institutions produce would at least give the opportunity for journalists to be better informed. Secondly as bad as our media may be, they will still likely report the findings of these institutions and can't just dismiss the information even if they disagree with it. I believe that media bias typically is realised through the framing of a debate / advocates selected for issues etc. I would hope if these institutions existed, it would be a matter of credibility of a media institution to at least mention that institutions position.

      On the neutrality issue I would again point back to parliamentary commissions - these answer complex questions, which have normative elements, yet are rarely accused of significant bias.

      There is probably little further we can get with some of these questions until they are put to the test, but it is still useful to get a sense of what everyone's gut reactions are to the proposal - so thanks.

  7. Will, what about the role of public universities? Maybe there are some cultural differences in the ways in which they see their universities, but in some countries, in some disciplines, university professors regularly do take on such a role. It might be instructive (if you want to go deeper into this) to study the way in which academics serve as policy advisers. Some points that immediately come to mind:
    - there is still pluralism (sometimes with a bit of dogmatism mixed in), e.g. between Keynesian and mainstream economists
    - the incentive mechanisms in academic research are not the same as the ones for being a policy advisers (in Germany, one well-known ranking of economists now distinguishes between their role as researchers and their role as public intellectuals). Not quite sure what this implies, but one would have to pay attention to the mechanisms of selection and self-selection one would get between the research institutions you suggest and other academic employers
    - charisma and an ability to deal with the media seem quite influential. Not sure this could (should?) be blocked if you had government-run research institutions. Or would they get professional PR people to help get their message across? That seems worrying as well, in a way…

    1. Let me first say I'm really enjoying the array of alternatives people are proposing - education, media regulation and better use of existing institutions all sound like interesting proposals I'd support exploring.

      Some initial reactions to the use of universities: There would seem to be a clear incentive for universities to stay out of public debate - primarily due to their receiving funding from central governments.

      Secondly I think that individual academics will always be much more prones to accusations of bias than an institution (e.g. Paul Krugman).

      By its nature I would question whether a university can have a view or position. More than most other institutions it is an assembley of individuals with wildly different beliefs/backgrounds/ideologies/areas of expertise - from anecdotal stories I hear, creating anything resembling a 'university' position would be incredibly challenging.

      In general though I am very positive about academics getting involved in public debate and wish that it occured more often (despite the generally negative cloud this has cast over the economics profession). I also feel that many academics are highly averse to speaking out in such a way and that this is a great shame.

  8. Will, I definitely agree with the basic premise of your argument: information is necessary for the effective functioning of democracy.

    However I am a little worried about your proposals. There are practical concerns; the extent to which a state sponsored institution would have sufficient funding, will or means to provide a viable information checking service. These have been well covered in the other comments.

    It has been mentioned in other comments, but I would like to elaborate a little further on the difficulties of choosing which information to ‘fact check’ and how to check whether it is indeed factual and finally (and perhaps most controversially), what counts as ‘fact’.

    I am not trying to come down on either side of what ‘counts’ as knowledge/objective truth (and hopefully at some point on this blog we can get into this discussion), but rather try to highlight how one institution would find it very difficult to provide a ‘neutral’ fact checking service.

    To elaborate, take a suspected faulty statistical claim:

    90% of all those under 25 taking job seekers allowance are not actually seeking a job and are choosing to be without work.

    An economist would use economic methods to check this through. She may use current government data to conduct the calculations again, to find out that (for the sake of my own political views) only 0.1% of under 25s on jobseekers allowance are not seeking work. She may even suspect the original data was faulty, and collect new data to cross-check the findings.

    A sociologist may instead look at this claim and conduct a study to find that the majority of of those choosing to be without work are doing so for a ‘legitimate’ reason (be it they are physically unable to work, they are caring for their families at home, there is a freak illness amongst the recently unemployed which means when this survey was conducted none of them were capable of trying to find work etc) and that this statistic is misleading for what it doesn’t include.

    An anthropologist may conduct a ground level long term study of a few families on job seekers allowance (an ethnography) to understand how we define ‘seeking a job’ and whether the current economic definitions of ‘work’ are appropriate for the different productive activities of families.

    And a feminist may wish to conduct a totally different analysis on how this statistic misses important dimensions of productive activities, such as domestic work and reproduction, which without we wouldn’t have a functioning society or that maybe we should value.

    I am not trying to see any of these methods at checking the validity of the claim would be correct or incorrect. Perhaps only one way is truly valid. But I doubt one fact-checking institution would be able to give full credit to all these approaches (and many more) to checking the ‘validity’ of a ‘factual’ statement. The worry would be that if the institution chose one approach, the rest of approaches would not be seen as equally 'valid' and the debate, and even how we go about the debate, would be biased.

    To summarise: There are so many debates and arguments on what methods are valid for information collection and even which knowledge ‘counts’ for important decision making, I don’t believe one institution would ever be able to address these AND provide a ‘neutral’ fact checking service.

    Based on this, I am inclined to agree with Lisa’s comment that perhaps we should focus more on funding a diversity of public universities with an emphasis on public intellectualism – to encourage these important criticisms of ‘facts’ reported in the media and types of knowledge currently accepted.

  9. Katie, I think you have created a false problem.

    Of your economist, sociologist, anthropologist and feminist. Only one has chosen to answer the question at hand (what percentage of under 25 year olds receiving JSA actually are seeking jobs).

    The other three seem to have contributed valuable information to the wider debate but have not answered the specific question. As such I don't see this as an example of 'equally valid ways of answering the same question'.

    I do think though that maybe we've moved to much towards the idea that these institututions would be literal fact checking services, whereas in reality I intend for them to be a kind of 'civil service for the public' in select areas.

    If a parliamentary committee or civil service report did not consider multiple dimensions when answering complex questions or alternatively specifically highlight what they are not addressing - it would be a fairly poor report.

    I do think though that this presents a weakness though that tough questions will require caveats etc. that will likely remain unreported.

  10. I don't think its a false problem, I think its very real - as demonstrated by your preference for a particular approach to checking for the validity of the statement.

    I did not set up a question, rather I set up a potentially false statement. You seem to suggest that there was only one valid way of checking for its validity, whereas I believe that all four would be important ways of considering how accurate the statement itself or the assumptions behind the statement is. Now lets not get into that discussion now. But the fact we disagree over this - demonstrates that there are debates to be had about how to check if something is 'true' or false'.

    You suggest this service would not be a fact checking institution, if so when you say 'civil service for the public', it sounds like you mean conducting research for the public? Which is what current universities should be doing. And I feel more comfortable that it is universities, as they have sufficient separation from the government that the research produced isn't considered state sponsored 'truth' and due to multiple universities, there is a more of a plurality of perspectives.

    For me, I would rather address the paucity of information through increased funding for both universities, and most importantly, accessible academic research (where possible) or outreach and some form of education to enable citizens to navigate through the huge amount of information currently available (which is something I think we all agree with).

    1. So I still think your problem comes down to being specific about the question - but as you say maybe not worth getting into this discussion now.

      I really don't think universities are a good substitute for these institutions for many many reasons (although they could perhaps be linked like universities and research centres are).

      Many of the reasons I have listed above (lack of independence, bias of individual academics, difficulties reaching an institutional view). More importantly though is that academic questions are of a substantially different nature to policy questions. There is a reason that the civil services of most developed countries are not staffed by PhDs. True academic information will be useful, but the practical realities which face policy questions do not allow for the type of clear and neat way of thinking that is found in academia.

  11. Thanks Will. My concern with state-funded neutral and independent institutions is how neutral or independent they really are. Sure they might be neutral/independent of political parties but that does not mean they will not biased towards particular ways of viewing and understanding society. An institute set up to provide information on the causes of unemployment is very unlikely to produce a report titled "Unemployment: an inherent feature of capitalism", for example.

    The more general point is that these institutes will tend to produce information, suggestions, proposals etc that exist within the liberal, market based, political centre, rather than analysing the more fundamental shortcomings of society. Furthermore because they are set up politicians with little interest in more radical ideas and even less interest in being constantly criticised by a credible well-funded radical institute, it seems unlikely that the kind of information needed for a truly informed democratic debate will come from these kind of institutes. And in the rare cases that they do (e.g. drug councils recommending legalisation, climate change bodies arguing for major carbon reduction) they are ignored, or have their credibility attacked or have their funding withdrawn.

    My question is, assuming that we believe that the structure of present society is fundamentally unjust and that we need radical ways to understand and change it, how useful can we expect these independent bodies to be in changing the democratic debate when they tend to think (and are set up to think) along the standard lines of the politicians and the political class that created them?

    1. Bruno - I completely agree with this line of criticism. Ultimately these institutions will not in anyway really challenge the 'deep' structures in our society.

      I think an endorsement of this policy is really an attempt to improve policy in the short to medium term, while thinking of how we can build support for more radical change in the longer run.

      I do think a more informed electorate ceteris paribus is more likely to support more substantial change though.

  12. Will, thanks for the post and apologies for my tardy reply.

    I echo your sentiment that an informed electorate is essential to a functioning democracy. As if we needed proof, someone invented the Tea Party to demonstrate the dangers of misinformation being dished out to an uninformed constituency.

    I also echo many of the comments above about the difficulty of deciding what issues this institute should address. It would seem that this institute would probably be better at addressing issues of efficiency (such as economic debates) than ideological debates.

    Here my cynical concern is that even people who vote on, say...economic issues, won't necessarily do so from an objective perspective of "what will grow our economy". Rather, they will vote based upon "what will benefit my own economic interests". When looking at Obama's bailout of the auto industry, for instance, I imagine few voters cared about the macroeconomic implications (which would likely be the focus of your institute's study), but about what the bailout meant to them individually and what it would mean in terms of setting a precedent for future bailouts. I do not believe that the institute would be able to (or should try to) address these differing perspectives but unless it does so, it's recommendations will carry little weight with most voters

    Even if we agree that there are some issues which this institution could help us to better navigate, however, I think you are overestimating the extent to which policy-makers and the general public would utilize this resource. As Tom says above, the majority of people search for information which reinforces their own world view, rather than objective information which may challenge it. As such, I imagine that the practical effect on voters and policy-makers would be minimal.

    One measure which could be taken and which may help in moving us towards your overarching goal (which we all seem to share) would be to institute a great focus on Critical Thinking in our education system - not necessarily as an independent subject but by integrating it into the way we teach a broad range of subjects. This would not solve the issue of people having insufficient time to fact check everything, but it would encourage people from a young age to question their assumptions and those of others, rather than simply engaging in activities which reinforce them.

  13. I should have included that a benefit of developing Critical Thinking as a modus operandi within much of the electorate is that (as far as I can see) it is a tool of empowerment which is value-neutral. That is to say, it teaches us to challenge all points of view, thereby circumventing the problems of which issues is it appropriate for the institute to address and how can we do so from an objective standpoint.