Are Evidence-Based Decisions Impossible in Politics?

Yves here. Even though the author unwittingly demonstrates the very syndrome he is lamenting in his set-up regarding Trump (a President actually has to have done something bad in office before he can be impeached, and you need enough Congresscritters in the House and Senate to be of that view for it to take place), once he gets past that initial stumble, he provides insight on how officials in Canada make decisions and what gets in the way of sound processes.

By David Moscrop , the author of Too Dumb for Democracy? Originally published at openDemocracy

It is astounding that human beings continue to be astounded by human behaviour. The Enlightenment ideal of the rational, calculative, dispassionate reasoner is so deeply embedded in our self-narrative that despite routine violations of our expectations and explanations of why people make the bad decisions they do, we scratch our heads every time we live up to who we really are.

How could the United States elect Donald Trump as president, and how could they not immediately remove him from office? How could the United Kingdom choose Brexit and then allow the high and costly drama of its realization to drag on for years? How could citizens in Ontario elect the inept Doug Ford as premier? Why aren’t we moving faster on climate policy when we know that the threat of climate change is unprecedented and existential?

Well, meet us. Right? There’s your answer.

Making good political decisions is difficult because human beings spend so much time in what economist and psychologist Daniel Kahneman calls “System One” mode. This is a way of thinking that is fast, emotional, easy and efficient. It’s heavily reliant on cognitive shortcuts and fast judgments, and prone to bias. Much of our politics exists in this space – the immediate, the gut, the familiar and emotional. And many of our structural incentives – our institutions, rules, laws and habits – keep us locked into this mode.

It’s not that we don’t want to do better. We often do. Plenty of us like evidence, rigorous reasoning and careful thought. But in a busy, distracted, speedy and complex world, we often take the path of cognitive least-resistance and get stuck battling the psychological tendencies that lead us towards poor reasoning. And given that we are often rewarded for doing so, at least in the short term, we’re encouraged to keep it up.

For example, we pick a party or position to support (for whatever reason), and we dig in. We are now part of a team. When our side wins, we feel good. We feel connected. We feel like winners. In some cases, this connection becomes part of our identity – like the sort of partisan attachment that is so common in the UK and USA right now. After that point, little is likely to move us.

You might expect politicians to be better. After all, they’re professional political decision makers. But you’d be mistaken. Like us, many try to do better but run up against their own psychology – and the structures that shape their profession.

In November 2019, the not-for-profit, science-boosting organization Evidence for Democracy released their new report Evidence in Action. They interviewed a sample of members of Parliament in Canada about the ways in which they gather and use information, wanting to know what role data and evidence plays in shaping policy and decision making among members of the House of Commons.

The researchers found that these processes varied widely, though a few findings were common: Google searches are used frequently; most MPs (94 percent) trust the Library of Parliament; and many like to talk to constituents (71 percent), experts (71 percent), and external organizations (88 percent).

While this shows that MPs value some kinds of data and evidence, the group also found that evaluations of what counts as ‘credible’ are wide-ranging. Most importantly, perhaps, respondents flagged a handful of barriers to using data and evidence in their work. What’s fascinating about the challenges they cite is that they’re mostly structural or institutional in nature. While making good, evidence-based political decisions requires us to consider and manage human psychology, doing so is made much more difficult when the institutions, rules, habits and practices around us constantly undermine our efforts.

For example, nearly 60 percent of MPs interviewed for the report cited “navigating information that had spin or bias” as a key challenge to using evidence for decision-making. Observers of politics will recognize that in some cases, political actors will view this sort of information as a necessary evil. Partisan politics incentivizes truthiness in service of winning. The 2019 election campaign in Canada was nasty– and yet the Liberals (who partook in the nastiness) managed to hold on to government, though their power is diminished with a minority in parliament.

The most common challenge MPs cited, however, was limited resources, especially too little time and the information overload that ensues. Good political decision making takes time. In an environment in which you’re constantly bombarded with information – often conflicting, sometimes suspicious – it is difficult to sort through and understand it all. The stresses of insufficient resources undermine even good faith efforts to make rational decisions.

In a related finding, MPs mentioned that they didn’t always have the resources to manage information, which is to say not enough money or staff. Democratic politics is often cut-rate politics. Citizens expect politicians to be frugal; and while they want good policies, they are hesitant to pay for them. Therein is the paradox of the democratic miser.

Finally, as the Evidence in Action report summary notes “There was a discussion of how politics is never fully ‘evidence based’ because so many other factors have to work into decisions.” Returning to politics and psychology, that’s the heart of the matter. “Other factors” include, no doubt, constituents who may not be receptive to data and evidence, or certain data and evidence. It also includes instances where, for partisan reasons, a party has no interest in deploying good research – for example, if it undermines a core goal that’s important to their political fortunes.

Addressing the challenge of such partisan goals is central to better politics. The defense of such skullduggery is as old as politics itself. “We’re the best option,” says the party. “But if we don’t win, we can’t govern. So there are some unsavoury things we may have to do to win, but that’s the cost of doing business.” In other words, the ends justify the means, and in a strategic political environment – like in a legislature or during an election, each of which is partisan, competitive, and often zero-sum – that may well be true. The trick, then, for securing better evidence-based political decisions is to change the incentives that are built into the structure of our politics. How do we do that?

Deliberative and participatory approaches to politics such as citizens’ assemblies and participatory budgeting are essential for transforming politics and securing better political decisions. By bringing day-to-day people together, giving them access to resources like time and expertise, and asking them to give reasons back and forth in service of reaching the best political decision, we can shift the balance of power towards political decisions that are made first and foremost in the public interest and without the distorting effects of partisan considerations.

For instance, imagine an assembly tasked with deciding climate policy. Rather than asking individuals to list priorities and preferences on the fly – as would a poll – or encouraging them to repeat talking points from the news or strategic actors – which they come across day-to-day – deliberative assemblies provide people with an island of calm in the middle of a roaring sea. Here, people can take their time, consult experts, and engage with their opposites as interlocutors rather than as opponents. So instead of repeating common solutions off the top of their heads (for instance, carbon taxes), or adopting the partisan positions of political parties (which might be their only frame of reference), individuals can engage with evidence and with each other, trading reasons back and forth in the service of developing a range of policy options.

These deliberative bodies could then become guides for other citizens and reliable resources for politicians (and in some cases, let’s be honest, political cover for them). As a bonus, deliberative assemblies that are well designed and executed can also help manage some of the cognitive gremlins that sneak into and upset good political decision making – for example by filtering relevant information from irrelevant information and managing cognitive overload, while giving people time to think through options, thus sparing them from the need to use heuristics to produce quick judgments.

Better political decisions are possible. Given the challenges we face now and those we’ll come to face in the future, good decisions are also essential. To make such decisions, we must address both the psychological and structural barriers that impede rational, truth-based politics among politicians and the general public.

Better decisions will require data and evidence. They’ll also require an environment buttressed by institutions that support and incentivize the use of these things. As challenging as these transformations may be, we can at least be encouraged by the fact that the best research and experience is showing us ways forward.

Print Friendly, PDF & Email

22 comments

  1. kimyo

    evidence might matter if governments were comprised of elected officials who represent their constituents rather than corporate sponsors. then you’d never see trudeau building his pipeline or macron cutting back on worker rights. boeing would never have been able to sell a single 787max. merkel would have thrown all of vw’s top executives in jail.

    the author doesn’t seem to understand how government works today. the problem is not due to a lack of data or evidence.

    Reply
  2. Steve H.

    From the linked Kahneman article:

    “we can be blind to the obvious, and we are also blind to our blindness.”

    Understanding biases are one of the great advancements of the past generation. That’s System 1, and they are a mountain to climb. Perhaps insurmountable, as a species. But the difficulties in System 2 are just as engaging.

    We can kind of get uncertainty, through System 2. Heisenberg, Schrodinger, Godel, Bayes [b. 1701!]… (The double slit experiment was in 1801!) We handle it by choosing our orientation, our angle of inquiry – so logic focuses on being consistent, and lets completeness take care of itself.

    Paradox also refers to behaviors stemming from punishments. Someone smacks the dog for being on the couch, behavior suppressed in smackers presence, but dog is more likely to get on that couch when alone. Paradox becomes perverse when the outcome is opposite of intention. Sources to note include:

    Merton, 1936: The Law of Unintended Consequences.
    Lipsey & Lancaster, 1956: The Theory of the Second Best.

    and Kahnemann, 2002:

    “because we tend to reward others when they do well and punish them when they do badly, and because there is regression to the mean, it is part of the human condition that we are statistically punished for rewarding others and rewarded for punishing them.”

    Great. Agency becomes difficult in such circumstances. “[A]ccept the things I cannot change” swamps the probability space. “It’s a dangerous business, Frodo, going out your door.” Within such an unintelligible [Roget 549] universe, response time gets selected for, quarterly reports drive behavior, and System 1 wins.

    That’s the field that politics operates in. Politics comes from polis, the walled city, who’s in whom’s out. Cui bono, cui malo. Nowak’s work indicates that the roi necessary for large-scale cooperation is enormous, but if a minority (10% or less keeps popping up) can get 90% of the goods, it’s worth their while. But that elite (cadre? class? cohort?) needs to not bump into any other large groups, so divide-and-conquer rules. Thus, “there’s no such thing as society. There are individual men and women and there are families.”

    There are different ways of handling this roi issue. American democracy solves it by reducing the necessary investment – to vote for President takes only hours every four years. (Unless suppression.) Homesteaders keep the number of cooperators low. Religion solves it by making the rewards infinite. The MMT lens can clarify the rewards-per-capita of economies of scale, if it motivates sufficient voters to redistribute those rewards from the elites.

    Reply
  3. Jane

    Better decisions will require data and evidence. They’ll also require an environment buttressed by institutions that support and incentivize the use of these things. As challenging as these transformations may be, we can at least be encouraged by the fact that the best research and experience is showing us ways forward.

    Which is exactly why Harper tampered with the census and shuttered 100s of research projects in Canada and why the GOP are doing much the same to the EPA and other bodies in the US today.

    Reply
  4. eg

    Mostly Kahneman’s “system 2” thinking is employed to provide a narrative to defend “system 1” thinking.

    Good luck escaping millions of years of evolution — there’s a reason “system 1” gets priority …

    Reply
  5. Livius Drusus

    Truth, rationality and evidence are all important, but it is true that we are not the rational actors of Enlightenment thought. It might be better to just admit that we are all biased to some degree and that politicians represent different interests and that often these interests conflict. No politician can be this honest but it would probably help if they were.

    I wonder if this problem is a product of the neoliberal era where politicians are supposed to be like apolitical corporate managers, something that is not possible and perhaps not even desirable. Political decisions will likely always be a mishmash of ideology, interest group politics and even naked personal ambition on the part of political actors.

    Reply
  6. PlutoniumKun

    On the subject of citizens assemblies, they’ve been in operation for several years now in Ireland – I was deeply sceptical at first, I saw them as a way for politicians to avoid hard questions, but they have been a resounding success. They’ve been used to address changes to the Irish constitution, especially to permit same-sex marriage and abortion. They took the sting out of very difficult and contentious issues and came up with very sensible solutions which attracted very broad support right across the political spectrum. There is currently one working on climate change and they seem to be coming up with very good solutions which politicians will find very hard to resist.

    Although I expect pressure to wind them up once they start coming up with solutions that threaten the economic status quo.

    Reply
    1. NotTimothyGeithner

      I think politicians make better decisions when faced with constituent pressure. “Liberals” bemoan Chuck Schumer’s latest betrayal or Inane statement, but they never demand he do anything in the mean time.

      Separate citizen assemblies might be useful as they do represent a clear alternative, but democracy requires unrelenting pressure on our betters. They are not friends. They are employees, and every citizen should be ready to speak to the manager at the slightest issue.

      Trumka didn’t stop the government shut down with his 12 million members, but the boss of the flight attendants threatened to have her 38000 (? I have to check) members withhold peanuts and the shut down was done. Pressure not praying to a Saint Mueller type works.

      Reply
    2. Susan the Other

      Citizen assemblies are very effective. And good local radio helps too. I don’t know how to sort evidence out, distill it, and keep everyone on the same page without people getting together. Proximity is the best conveyance for understanding. For some it seems like a chore, but there are people who were born to it. They have energy and curiosity and like discussions. It is group learning. And the outcome is often a cascade of good ideas and responses. When a certain threshold its reached, of understanding based on the facts, everybody seems to get it pretty fast. And I think there is something important about smaller groups, who begin the discussion, and then spread out to larger ones. It reminds me always of Rupert Sheldrake’s argument that human intelligence and new knowledge floats around in the ether and everyone is benefited. It only takes one person with a realization who talks to another and so on.

      Reply
    3. False Solace

      Citizens assemblies would probably work better in countries where the inhabitants are less thoroughly propagandized. I wonder about the outcome of an American citizens assembly on whether to bomb Iran.

      It all sounds like a liberal’s wet dream. Take a bunch of know-nothing deplorables and educate them until they produce the correct opinions. I wonder what options the assemblies would be allowed to consider and who gets to determine the information put before them. If the only possibilities are neoliberal ones it’s not like they’d help us to break out of the dead-end capitalism now destroying the planet at a record pace. The citizens assemblies become just another hollow exercise in manufacturing consent — getting the sheep to decide what the wolves will have for dinner. Perhaps they’ve worked a few times, but that’s only because the wealthy didn’t bother to subvert them.

      Reply
  7. The Historian

    Interesting article. I think much of what Moscrop says is true, particularly this:

    It’s not that we don’t want to do better. We often do. Plenty of us like evidence, rigorous reasoning and careful thought. But in a busy, distracted, speedy and complex world, we often take the path of cognitive least-resistance and get stuck battling the psychological tendencies that lead us towards poor reasoning. And given that we are often rewarded for doing so, at least in the short term, we’re encouraged to keep it up.

    And I do like his solution of:

    By bringing day-to-day people together, giving them access to resources like time and expertise, and asking them to give reasons back and forth in service of reaching the best political decision, we can shift the balance of power towards political decisions that are made first and foremost in the public interest and without the distorting effects of partisan considerations.

    But in a “busy, distracted, speedy and complex world”, how do we do this? It is going to take major changes in how we work and live, not to mention how we are educated, to allow us the time and ability to carry out his solution.

    Reply
  8. PKMKII

    The problem with “Evidence-Based Policy” is that evidence, in and of itself, is not a policy. Evidence is always going to be filtered through ideology, as we all eat from the trash can. Climate change is a good example; ecofascism may be an evidence-based response to climate change science (the Jackpot is coming, so we need to ensure that it is only the pure that survive it), but that doesn’t make it good. This is why technocratic politics always come off as incomplete, it tries to pretend that an essential part of politics can be ignored.

    Reply
  9. Carla

    “In a related finding, MPs mentioned that they didn’t always have the resources to manage information, which is to say not enough money or staff. ”

    That is the only mention of money in the post, and neither “donation” nor “contribution” appears. Furthermore, the author claims:

    “By bringing day-to-day people together, giving them access to resources like time and expertise, and asking them to give reasons back and forth in service of reaching the best political decision, we can shift the balance of power towards political decisions that are made first and foremost in the public interest and without the distorting effects of partisan considerations.”

    Really? Honest to God? That’s all it takes to “shift the balance of power”? Well, hot damn!

    Does the author mean to say that in Canada, there are no rich corporations or individuals willing to provide “money or staff” to the poor MPs tasked with making policy decisions?

    While I doubt that Canada’s Constitution has been interpreted to mean that “corporations are persons” and “money is speech,” somehow I suspect that big corporations and wealthy individuals are not wholly without power in Canada’s democratic system.

    Reply
    1. Susan the Other

      I think corporations have changed. They are still prisoners of their “bottom line” so we’ve (our system) put them in a position to always defend their own survival. If they didn’t have to worry about losing their market share things would probably change pretty fast. Corporations now realize they have damn near killed the consumer with all their 20th century nonsense. Not to mention the planet. The big sticking point here in the US is the medical industry and pharma. They are going down kicking and screaming the whole way. But look at finance – it even ate itself. 2008. It looks to have actually been tamed.

      Reply
      1. Carla

        So… Ohio and Pennsylvania are being laid waste by new plastics plants to consume fracked oil, while we have no solution for the waste plastic polluting fresh and ocean water and destroying wildlife, not to mention all the negative consequences of fracking… and somehow corporations have changed? Finance has been tamed? I’m just mystified by your conclusions.

        Reply
  10. m sam

    No matter how “rational” someone claims they are in their decision making, they still bring many assumptions along with it, most of which are unexexamined, unacknowledged, or even unnoticed. That those assumptions are loaded with biases and illogic is a given.

    That we mere humans may attain some perfect state of pure rationality and enlightenment is nothing but a Utopia: unobtainable. We are not robots, however much some people might want it to be the case. It is strange how people cling to the notion of achieving perfect rationality in our politics (as if we even have that in our sciences!) I think it is even a bit naive. And dangerous; I would submit those who claim to be the most rational are the least likely to examine their underlying assumptions, biases, and motivations, and are therefore the least trustworthy.

    Reply
  11. anarcissie

    I am surprised no one mentioned Hume’s A Treatise of Human Nature wherein he famously says ‘Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them.’
    (Argument at https://www.pitt.edu/~mthompso/readings/hume.influencing.pdf)
    By passions he seems to mean not only emotions but intuitions as well. The emotions and intuitions emanate from not just from the skeletal language of logic, but the whole being and the social and physical world in which the subject is embedded. Emotion and intuition are obviously faster than reason, more comprehensive and robust, and are often more accurate.

    Besides the fatal inefficiency of reason, when talking about socially-acceptable evidence, we have to note that what is considered good evidence — fact — is socially determined and hence subject to power. For instance, there are people who actually believe that the content of the New York Times is factual, because, like the old-time Party readers of old Soviet Pravda, they are expected to believe in it.

    I suppose a really strong mind might develop a passion for reality, and give poor reason a helping hand; but there do not seem to be a lot of such minds, and they are often troubled by those around them. It is a hard problem in any case.

    Reply
  12. laudyms

    When I was studying anthropology, humans were generally described as Toolmakers ….. that’s been revised over the years as other animals have also been observed using tools.

    My own view has settled on humans as Mythmakers… who, when freed from the constraints of basic survival, sit around speculating and making up stories and fantastic narratives to entertain and gain power over each other. I include all religion in that web.

    Reply
  13. blackerman

    This is sort of a weird conversation. I’m with Kimyo here who posted the first comment: politicians are serving a single master, and in most cases that master is not us. The problem with their decisions is not that they’re not evidence based, but that the range of acceptable decisions–to their bosses–is simply way too narrow and doesn’t include us, particularly if they’re about the allocation of money and power. Once politicians step outside that range, the whole edifice of institutional power–money, media, and all their connections–will work to bring them down or back into the fold. There’s little space for simply sitting back and trying to figure out ‘best solutions’.

    Reply
  14. Off The Street

    Journalism, among other industries, would benefit from adoption of ISO 9001ish concepts. Provide and follow documented procedures that pass independent certification audits. Those work for many industries whose participants like to assure their counterparties that they really are aware of, and following consistently, reasonable and specific procedures, and to continue to earn and retain their trust.

    Visualize standards that identify and call out rhetorical devices, hearsay, weasel words, anonymous source attribution and such. Instead of a Fog Index, indicate a BS Index. Provide some type of Epistemic Certainty, (hat tip for publicizing that notion to me to SlateStarCodex, among others), or Scientific Method-like metric regarding the viability of the article and theses, as an anti-tabloidization and anti-Silicon Valley black box factor.

    Loosely related, look into ISO 18091 for local government operations, too. It may cheer you up if you find that your local elected, appointed and staff people are following that.

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *