Contents

Improving our political system: An overview

by Tobias Baumann. First published in 2020.

An audio version of this article is available. It may not include all details and links available in the original article. For accuracy and completeness, please refer to the full text below.

Given our vast uncertainty and nearsightedness about the future, we should perhaps focus on putting society in as good a position as possible to address the challenges of the future. Or, put differently, we should optimise for a proxy along the lines of “how well is society working in 2100”.1 One important dimension of this is a functional political system. Efforts to avoid harmful political and social dynamics, and to strengthen democratic governance, are therefore of interest to effective altruists.2 

This post outlines possible reforms to improve our democracy3 and discusses whether those are a cost-effective intervention from a longtermist perspective (with an emphasis on s-risks). While specific reform proposals have been discussed in effective altruism (e.g. 1, 2, 3), I’m not aware of any systematic review of this cause area. 

Problem analysis

The idea that our political processes leave something to be desired is quite commonplace. In fact, dissatisfaction with Western democracies is at a record high. But what exactly is the problem? The following (not necessarily mutually exclusive) dynamics are some plausible candidates:

  • Excessive political polarisation, especially party polarisation in the US (Mason, 2018), makes it harder to reach consensus or a fair compromise, and undermines trust in public institutions. (See 1, 2 for previous discussion.) The available data suggests that affective polarisation has increased in some countries (including the US) but not others (Boxell et al., 2020). Excessive polarisation and tribalism also exacerbate conflicts and therefore constitute a risk factor for s-risks. Conversely, more charitable and thoughtful political discourse (deliberation in particular) makes severe conflicts less likely. 
  • Governments or public institutions might systematically make poor decisions, either due to incompetence or due to bad incentives. This could be related to a general deterioration of public discourse (cf. fake news, post-truth politics). Poor institutional decision-making would hamper an effective response to transformative AI or any other future challenges.4 
  • Our political processes tend to focus on short-term issues and disregard the interests of future generations. This could be due to epistemic biases or because present-day people actually do not value future generations much. (See here for more details.)

Of course, not every narrative about how “politics is broken” is accurate, or rises to the level of posing serious long-term risks. Despite the many complaints about politics, it seems plausible that existing democratic systems are quite good in the grand scheme of things. So perhaps we should instead consider risks of a serious deterioration of existing systems, resulting in very bad long-term outcomes. 

If history is any guide, one plausible scenario is that the “wrong” individuals and ideologies may become dominant, potentially resulting in a permanent lock-in of a totalitarian power structure.5 Historical examples of such totalitarian regimes were temporary and localised, but a stable global dictatorship may become possible in the future.6 This is particularly worrisome in combination with malevolent personality traits in leaders.  

Interventions

A plethora of interventions have been suggested to address the above-mentioned issues. The following candidates seem most interesting: 

We should, however, be careful to not be biased towards salient and tangible interventions such as voting reform. (See also this comment.) Many aspects of the policy process, such as political culture or norms for what is acceptable speech, are harder to grasp but equally important. (For instance, it is currently considered illegitimate to use violence against people you disagree with – which can’t be taken for granted – but there is not yet a norm to be charitable towards opponents, to do thorough research before making factual claims, or to have balanced rather than extreme or one-sided views.)

Also, it is often unclear whether these suggested reforms are actually improvements. (For instance, there is widespread disagreement about which voting method is best.) This is in part because of a lack of high-quality studies on whether a certain intervention achieves a certain outcome. But more fundamentally, it’s because the political process is complex, so it’s often impossible to foresee how a change might possibly backfire. 

Is the area too crowded?

One might argue that efforts to improve our political system are generally too crowded and intractable. Political dysfunction is a well-known problem, so if there were any easy solutions, one would expect that they would have been found and implemented already. (This is the case for many fundamental principles of democracy such as checks and balances.)

A related concern is that parties or individuals in power often profit from the status quo and will therefore block reform attempts. Due to inertia and such vested interests, countries rarely make dramatic changes to their political system once it is established (bracketing major disruptions). For example, in the UK, voting reform has a centuries-long history but Westminster elections still use First-past-the-Post. 

This is true to some degree but it still seems plausible that we can do useful work. While quite a few people work on this in a broad sense, it is still possible that their incentives are not well-aligned with actually solving problems (in particular, the problems we’re most concerned about). For instance, my impression is that discussions of voting reform often suffer from the problem that people tend to pick their favourite method and then cherry-pick one-sided arguments in favour of it. (Conversely, perhaps a lot of thoughtful academic work on this lacks a political platform to implement proposals.)

Conclusion

Improving our political system is a plausible priority area and should be on our radar. In particular, I think preventing excessive polarisation or malevolent leaders is among the most important priority areas for s-risk reduction. However, the area is only a top priority if we can identify convincing reform proposals that are both tractable and robustly positive.

For more details on this topic, I recommend Magnus Vinding’s book Reasoned Politics and Chapter 9 in Avoiding the Worst: How to Prevent a Moral Catastrophe.

  1. This isn’t to say that this perspective is a no-brainer. For instance, one could disagree on the grounds that pivotal events such as transformative AGI might happen soon, so we should focus on those.[↩]
  2. Another candidate that fits well into this framework is moral circle expansion, which increases the degree to which society (or certain parts thereof) will be motivated to avert future suffering.[↩]
  3. I think we should largely focus on liberal democracies rather than, say, trying to improve China’s political system, because we aren’t in a good position to influence the latter.[↩]
  4. One could object that what would count as an “effective response” by the lights of average values may still be bad from a suffering-focused perspective, if most value systems do not entail concern for the suffering of all sentient beings. Perhaps this is a reason to focus not on political efficiency but more on changing the values and focus in politics.[↩]
  5. It is worth noting, though, that totalitarian regimes do not necessarily constitute a large s-risk; so it is not entirely clear if suffering reducers have a differential interest in avoiding such an outcome. It could be argued that the danger arises mostly from extremist ideologies which exacerbate conflict; this is conceptually distinct (though perhaps correlated in practice) from the degree of control of the government over citizens’ lives, or whether a country uses a democratic process to determine its leader.[↩]
  6. However, the rise of a totalitarian dictator is merely the most salient version of such risks. For instance, perhaps the Middle Ages were close to a lock-in of feudalism: individual kings come and go, but the broad power structure was quite stable for a long time.[↩]