Representation for Future Generations in Sweden – a summary of our work so far

Written with Signe Savén, who has been especially involved in the full write-up.



Future generations don’t make it to the voting station. Because of this, their interests are poorly represented in current political systems. This post summarizes some work we have done in EA Sweden to better represent future generations in the Swedish political system. The idea is to find ways in which the political system can be tweaked or affected to better take the interests of future generations into account. This can be done by things such as: making adjustments to the legislative process, setting up an Ombudsman for future generations, giving rights to future generations or having a Minister for the future in government.


The idea is based on and inspired by work that has been done primarily in the UK starting in 2017 by among others FUSE (Future of Sentience, now a part of Effective Altruism: Cambridge) and CSER (Centre for the Study of Existential Risks), which for example lead to the formation of the All-Party Parliamentary Group on Future Generations in November 2017. The group consists of MPs interested in issues related to future generations, has a secretariat hosted by CSER and e.g. puts on seminars highlighting specific global catastrophic risks.


So what have we been up to? Following a similar project plan to the UK group, we have done research looking at different methods to giving representation to future generations in countries around the world, e.g. by setting up Ombudsmen, granting rights to future people and having committees in the legislature tasked with protecting the interests of future people. We have also researched the Swedish context, including interviews with over 20 relevant stakeholders. Based on the above, we've decided on a number of recommendations, which we are currently doing work to implement.


Below, we’ll describe the rationale behind the project, the progress we’ve made so far, a brief description of the results of our research (full write-up here), our plans going forward and some advice to other groups interested in doing similar work.



1. Project rationale

2. Progress so far

3. Summary of our write-up

3.1. Criteria for suitable institutional tweaks

3.2. Recommmendations

4. Plans going forward

5. Should other groups try to do the same thing?

6. Kudos


1. Project rationale

There is a large emphasis on basic research in the global catastrophic risk-space, with e.g. ALLFED being a notable exception. This is plausibly a good strategy in a space where we have little knowledge about what policies will make a difference and being wary that some of them might even be harmful. However, as we find out more about how to counteract global catastrophic risks, there needs to be a way for the research to affect the behaviour of relevant actors. So far, this has mainly been done by raising awareness among those who are likely to develop the potentially harmful technology – i.e. researchers and companies. Another approach, aimed at in this project, is finding ways for the research to affect policy. In particular, there is a lot of value to capacity-building in this area: laying the groundworks to be able to affect policy further down the line. Given the above, a project of this kind could have several benefits: direct impact by positively affecting policy, capacity building by developing networks and skills in the area, in addition to exploration value by learning more about how to have effective altruism and research on global catastrophic risks affect policy.


Projects of this sort come with risks of actual harm that one should be wary of. Firstly, the project could be directly harmful if it caused harmful policies to be put in place. This could e.g. happen given high degrees of uncertainty about what policies will decrease existential risk. Secondly, a project of this kind could have negative reputational effects. If carried out poorly, effective altruism or work on global catastrophic risks could come to be viewed negatively by important stakeholders such as politicians. In short, it could lead to worse opportunities to positively affect policy.


In short, the rationale for going ahead with the project was firstly that it seemed plausibly valuable, especially in terms of exploration value and capacity building. Secondly, we thought that we had the resources to carry it out, both in terms of committed members and in terms of connections to organisations and individuals. Many x-risk researchers are Swedish (e.g. Anders Sandberg, Max Tegmark, Nick Bostrom and Olle Häggström) and there are at least two Swedish organisations working actively with global catastrophic risks willing to support us with advice, expertise and networks: the Institute for Future Studies and the Global Challenges Foundation. Thirdly, the reputational and brand risks to the effective altruism movement in engaging with politics are likely less severe in a small country (only ~10 million inhabitants) where English is not the first language.


2. Progress so far

Since starting the project in January, some of the main things we have done are:

  • Researching what ways of representing future generations are suitable in a Swedish context, looking at e.g. how future generations have been given representation in other countries and previous attempts to do so in Sweden. Based on research and interviews with dozens of stakeholders, ranging from politicians, to researchers and campaigners. See the abstract of the write-up of this work below and the write-up itself here.
  • Found and acted on an opportunity to do object-level work, publishing this article – with Anders Sandberg, Carin Ism, Max Tegmark and Olle Häggström – in a Swedish newspaper arguing for a lethal autonomous weapons-ban when we saw that there would be a vote on the topic in Parliament.
  • Started work to communicate and implement our proposed changes. This has been done by putting on a launch event, organising and participating in seminars, writing an opinion piece about our recommendations in one of Sweden’s main newspapers, and reaching out to and meeting with members of the Swedish Parliament.


3. Summary of our write-up

Below, we’ve cut out some relevant parts of the write-up: criteria used for deciding on what institutional changes to recommend and our recommendations. For details about what has been tried in other countries, a list of all the different changes we considered and a history of these kinds of institutional tweaks in Sweden, have a look at sections 4-6 of our write-up.


3.1. Criteria for suitable institutional tweaks

When thinking about what changes to the legislative system to recommend, we have taken the following criteria into account.



Achievability concerns the likelihood that a system is successfully established. Institutions and mechanisms that are difficult to implement are less likely to be established than those that are more easy to implement. Difficulties can manifest themselves as complicated legislative changes (e.g. changes to the constitution) that require a lot of political support over time to be implemented, or costly initiatives that demand that resources be transferred from already established institutions to new ones. Establishing a new institution can create a lot of opposition from already existing institutions, because they may fear that their resources or mandate are threatened.



Apart from being established, a system must be maintained in order to be able to have beneficial effects on policy-making. Sustainability is the property to remain operational over time. In order for institutions and other mechanisms of a system to be sustainable, wide support, both publicly and across the political spectrum, or constitutional entrenchment, are important factors.


Important factors to consider are whether the institution manages to create support for itself or if it might cause its own destruction (Caney 2016). Further, legitimacy, and power in proportion to perceived legitimacy, are important factors. Institutions that have received much power early on have not lasted long, suggesting that soft-power mechanisms are preferable, at least initially.



Achievability and sustainability are necessary conditions for a system to be able to have any effect on policy over time, but they are far from sufficient. Once they are in place, it is crucial that the system is effective, that is, that it affects policy positively. In order for a system to be effective, it must have the power to influence policy. This power may have been formally delegated to an institution (e.g. the power to veto new legislation) or may be have been obtained informally (e.g. by the media).


To ensure effectiveness, the proposed system should address the causes of political short-termism. Such causes include ignorance of the future, electoral and economic dependence, type of performance indicators and auditing duration, and media coverage. Short time-frames within these areas tend to spill over and make politics short-sighted as well. In addition, there are several aspects of human nature that causes us to be biased in policy-making. For instance, we often fail to detect creeping problems, we are more likely to take action if a victim is identifiable, we respond more to vivid risks and tend to ignore things that are not in front of us, and we tend to be prone to positive illusions. In addition, we have a tendency to procrastinate and may fall for temptation, weakness of will or self-interest (Caney 2016, in Institutions for Future Generations). A system that aims to represent future generations should include mechanisms that corrects for these causes of short-termism.


Access to necessary resources

In addition, for a system to deliver beneficial policy, it must have access to necessary resources. This includes sufficient funding as well as relevant information and expertise. In order to create beneficial policy, different policy-areas must be sufficiently understood, which makes access to subject-matter experts and high-quality research essential. Given the low maturity of research into GCRs, this criteria becomes particularly important to us. This access could be obtained by empowering a specific institution to carry out and collate relevant research. Further, it is important that new policy is able to respond to changes and is made relevant. To facilitate this, it is important that the system is transparent and allows input from all stakeholders to be taken into consideration.


3.2. Recommendations

The initiatives we propose are presented below, each followed by a brief explanation. The recommendation is not that one initiative be implemented before the next one is started. Rather, we would expect that several recommendations be pursued in parallel.


A. Ensure that there are sufficient means to implement and achieve the Climate Act and Agenda 2030 (the Sustainable Development Goals)

To achieve the ambitious goals, leadership from government, parliamentary anchoring and concrete actions are required. To support this, we propose a Future Commission in the Swedish Parliament (Riksdag), whose first task is to monitor the implementation of both the Climate Act and Agenda 2030. The hope would be that the body would be used going forward doing other work as well.


This proposal seems like it can be impactful by supporting the implementation of the Swedish Climate Act and Agenda 2030. However, we also think it is a good proposal even if one is primarily concerned with GCRs. This is because such a Future Commission could potentially be used to put technological risks on the political agenda, and because it allows us to get support from a wider range of stakeholders who are more interested in Agenda 2030 and climate change than technological risks.


B. Establish an Ombudsman for future generations

The Children's Ombudsman was established because children lack the opportunity to protect their own interests. The same applies to future generations. An Ombudsman for future generations is needed to monitor the work of government, parliament and all authorities to take into account the interests of future generations and to propose new legislation that protects their interests.


A State Ombudsman for Future Generations would help ensure that the interests of future generations are properly taken into consideration, both in the legislative process and when it comes to implementing and enforcing new policy. The role of the Ombudsman would be to investigate, present opinions, and propose legislation, not deliver binding rulings, which makes it a soft power institution. In carrying out its work, the Ombudsman would help with continuous agenda-setting, thereby helping the system to become self-sustaining.


An Ombudsman needs a legal base to work from. Such a base could be made up of international treaties or conventions, legally recognized rights of future generations, or other legislation aimed at protecting future generations (Beckman & Uggla 2016). We are currently unsure about the strength of the mandate required and whether mandates with this strength currently exist in Sweden. The Sustainable Development Goals might form such a basis. Another candidate is the Swedish constitution which holds that “the public institutions shall promote sustainable development leading to a good environment for present and future generations” (The Constitution of Sweden, Ch. 1, §2). We expect to become clearer on this point over time.


In general, we think the argument in favour of an Ombudsman for future generations has public appeal, mainly since the rhetoric seems effective and because people have a good understanding of the function of Ombudsmen. It is therefore easy to campaign on. Furthermore, we think that an Ombudsman would be beneficial in implementing policy related to future generations. However, creating a new Ombudsman is likely politically difficult. They tend to last for a long time, but seldom be created. All in all, the Ombudsman scores low on achievability, but high on sustainability, effectiveness and access to knowledge.


C. Create a cross-political network for future generations in the Swedish Parliament

To move important future issues onto the political agenda, we propose that a cross-political network is established. Such a network would help create wide political support, which is key for both the achievability and sustainability of other institutions and mechanisms that we recommend should be developed and established later. It would also have positive ongoing impact on agenda-setting, aiding in ensuring that future issues remain a relevant topic of consideration.


Cross-political networks are not regulated by law and can take different forms. Some are organisations with their own statutes and internal rules, whereas others are more loosely arranged. What unites them is that their members come from different political parties. This means that there is a lot of freedom when it comes to establishing such a network, which leaves it open to us to consider what kind of role we want to have. Based on our interviews, these networks are common, but since they are not regulated it is difficult to know how many there are, what they do and what impact they have.


In the United Kingdom, a cross-political network – the All-Party Parliamentary Group on Future Generations – has been up and running since the fall of 2017. From what we have heard, that group is doing well. It has, however, been pointed out to us that there is a difference in the amount of power that the parliamentarians in the United Kingdom have compared to the parliamentarians in Sweden. But since it is up to the networks how they are organised and how they operate, we feel that a cross-political network is a good way to have impact at a low cost.


Further, we propose that the Institute for Future Studies takes on increasing responsibilities with regard to future issues, and take on a role similar to CSER in regard to the All-Party Parliamentary Group on Future Generations. The institute already carries out research on future issues. With increasing resources it could take on a broader role and serve as a think tank to the parliament, carrying out and collating research on a wide variety of future issues. We also think that it would be valuable if the institute could become a permanent body considering proposed legislation.


In summary, a cross-political network seems to score high on achievability, medium on effectiveness and access to knowledge, and potentially low on sustainability.


D. Increase the focus on future issues throughout the legislative process

Throughout the legislative process, consideration should be given to the effect new legislation has on future generations. This can be done e.g. by requiring committees to take into account the interests of future generations when certain kinds of decisions are made or in other ways amending the procedures of parliamentary committees. We think doing so could be impactful, since it would mainstream thinking about future generations. However, determining what changes are appropriate is very difficult, given the complexity of the legislative process and the constraints (mainly in terms of time in the Swedish context) of parliamentary committees. This can be done by requiring that the investigation (utredning) that precedes a new law takes a future perspective and evaluates the consequences for future generations of the issue at hand, by establishing a new body to routinely review proposed legislation (remissinstans) with regard to the interests of future generations, or by requiring that the interests of future generations be taken into account in all reports from the parliamentary committees (utskottsbetänkanden).


E. Support from civil society

Politicians need support from citizens who want politicians to take responsibility for long-term challenges. We need to show support not only for political promises to take certain actions, but also for the implementation that follows from such promises.


We believe that there is a gap that we can fill here. There are discussions about policy issues that quite closely align with the cause areas or issues that long-termist effective altruists might be particularly interested in that are currently not pushed for by civil society. For example, noticing that a ban on lethal autonomous weapons was being voted on in the Parliament, we could with quite small means write an article for Dagens Samhälle one of the more wonky Swedish newspapers with Max Tegmark (FLI), Carin Ism (Global Challenges Foundation), Anders Sandberg (FHI) and Olle Häggström (Chalmers Technical University). We will likely keep doing similar things in the future.


4. Plans going forward

Going forward, we are planning to:

Keep pushing our message 

  • Continue finding MPs that are sympathetic to our cause and that potentially could become part of a cross-political network.
  • Hold seminars and host events (e.g. a debate about the future with relevant youth politicians).
  • Get noticed more in the media (e.g. continue to publish in the big newspapers).

Put ourselves in a better position to do more good in the future

  • Build a coalition with important actors (organisations and individuals) to ensure that this becomes more than an effective altruism project. We don't think we'll be able to succeed in this campaign alone.
  • Collect signatures, e.g. from public figures.
  • Attract more people to the project core team to increase capacity (focusing on recruiting for specific roles). 
  • Update our communications and messaging strategies, along with our brand (mainly visual). 

Constantly evaluating our work to find ways to improve

  • Set specific goals against which our progress can be evaluated.


5. Should other groups try to do the same thing?

Before carrying out your own project of this sort, there’s a number of factors to consider. Firstly, there’s been quite a bit of discussion recently about the usefulness of direct work among local groups (e.g. here, here and here). Secondly, there are the risks mentioned above that need to be taken into account.


In summary, if you’re considering embarking on a similar project, we would give the following advice:

  • Don’t get started without talking to experts within relevant communities (e.g. x-risk researchers and people at effective altruist organizations). Carrying out a project like this, and especially reaching out to politicians, could turn out to be harmful.
  • Make sure you have the capacity to carry out the project. This includes having a group of people capable of carrying out the project (e.g. with strong social skills, project management, understanding of the issues and good judgment), having access to experts (much like the project in the UK would have been difficult to carry out without this, ours would have been difficult without access to e.g. the Institute for Future Studies) and having political connections.
  • Conduct cheap tests. If we were to do the project again, for example, it may have been a good first step to gauge the interest among MPs more before getting started.
  • Local context is really important. Though there seems to be some generalisable conclusions – e.g. that being adversarial often proves counterproductive – local context will have a large impact. We think we underestimated the importance of the local context.


6. Kudos


Kudos to the project team, Signe Savén, Robert Shepherd, Emil Wasteson and Denise Ferreras. Kudos especially to the people at CSER and FUSE who set up the All-Party Parliamentary Group on Future Generations, among others Beth Barnes, Natalie Jones, Lord Martin Rees, Tildy Stokes, Simon Beard, Julius Weitzdörfer and Mark O'Brien. Thanks also for especially helpful comments and feedback from Robert Höglund, Carin Ism and Stefan Schubert. Thanks also to our expert group Anders Sandberg, Olle Häggström, Kristina Persson, Birgitta Englin, Stefan Einhorn and Karim Atahualpa Jebari.


Comments (6)

Comment author: ole_koksvik 09 August 2018 01:00:23AM 3 points [-]

This was a really good, clear write-up; I found it very interesting. Thank you.

I have a couple of questions. If you want to reply by just telling me to read the full write-up: fair enough.

  1. There seems to be some potential overlap between A, B, and D. Could the Future Commission, or the Ombudsman, be used for oversight on other parliamentary committees, i.e. that part of their remit be to check that other committees discharged their responsibility to take future generations into account?

  2. Do you have thoughts on how you would go about pushing for a Future Commission?

  3. Do you think there'd be value to do some opinion polling on support for a Future Commission, and an Ombudsman?

Comment author: MarkusAnderljung 09 August 2018 08:29:57AM 1 point [-]

Thanks, Ole!

  1. Yep, I would say there is definitely overlap. At the very least between the Ombudsman and the Future Commission. They would both serve as checks. I think (and am not sure whether the others in the project would agree with this), we do not need to think about all of these as a package. If I could wave a magic wand and implement a set of institutional changes, I would likely be in favour of a different package. Instead, we've recommended things that fit well together where we would be incredibly pleased if one or two would be implemented.

  2. One of the main things we're doing now is thinking about how to get to implementation of these recommendations. My view is that we need to keep building a coalition with other organisations (e.g. environment-focused think tanks). On top of this, we need to keep finding allies among the MPs.

  3. I think that's a great idea! We'll likely spend the next few months taking a step back and deciding on how we're gonna be pushing for our recommendations. One really simple communications tactic would likely be to do some polling.

Comment author: ole_koksvik 09 August 2018 11:50:32PM 0 points [-]

Thank you for your replies.

[Warning: musings ahead.]

Re your reply on 2, I found it interesting that overall you seem focused on working with *insiders8 --- people already quite involved and invested in the political system --- rather than aiming for uptake in the population at large, and then to use that as leverage on politicians (which is what I had been musing about when thinking about how to cause political change). I wonder whether there's data on the effectiveness on the two different approaches. Most likely it's too dependent on context and question.

Comment author: MarkusAnderljung 14 August 2018 09:18:31AM 0 points [-]

Yeah, I agree that we've gone for quite a top-down rather than bottom-up approach. Though less so than in the UK case. There, my understanding is that the APPG was started only after having arranged 1-on-1 meetings with the relevant MPs.

My intuition is that it will be difficult to cause change in this area by driving popular opinion, but you might need some public opinion behind it to make it stick. To be crass, I would expect politicians to be able to wrangle very few votes by spear-heading initiatives such as these.

Comment author: DavidMoss 08 August 2018 02:14:47PM 2 points [-]

Very interesting project! Have you engaged with the work of Rupert Read at all? He's a philosopher in the UK, Green Party politician, think tank chair and sometime critic of EA. He's written a paper (2012) arguing for a body, building on the Hungarian ombudsman example, who would aim to act as a proxy for future generations, review and potentially veto legislation deemed harmful to their interests.

Comment author: MarkusAnderljung 09 August 2018 08:47:59AM 2 points [-]

Thanks, David!

I have not, though Jones et al 2018 has (this is the paper written by the same people who worked to set up the APPG in the UK). They say of the Read proposal that "We encountered several ideas which we do not include above, for various reasons. One of these is the proposal for a “third house of parliament”, or “Guardians”, made by Rupert Read (Read 2012). Under Read’s proposal these Guardians, appointed randomly amongst citizens on the same principle as juries, would have the power to (a) veto new legislation that threatened the basic needs and fundamental interests of future people, and (b) force a review of any existing legislation that threatens such needs and interests. He also suggests similar structures within local governments. As we found previously, institutions with veto powers did not last long, and as such we do not think this “third house of parliament” would be workable. In addition, we share concerns raised by Michael Bartlet about the proposed method of selection by lot (Bartlet, 2012)."

To give some context on veto powers not lasting long, the Hungarian Commissioner was removed with four years and the Israeli commissioner lasted about five years. Both of these had quite strong powers that they actually used, which seems to have made them politically controversial and had them removed after the next election.

Another crucial point here is what types of policies you think are going to be most impactful to work on. If you want to focus on environmental issues, I can see the allure of institutions that are given a lot of power. The bottleneck is not agenda-setting but rather closer to the right bills not being voted through. If you're more concerned about other GCRs, institutions or tweaks focusing on agenda-setting will be more interesting, such as the APPG.