EAF’s ballot initiative doubled Zurich’s development aid

post by Jonas Vollmer · 2020-01-13T11:32:35.397Z · score: 240 (104 votes) · EA · GW · 20 comments

Contents

  Summary
  Initial proposal and signature collection
  Counterproposal
  Campaign and vote
  Policy implementation
  Cost-benefit analysis
  Potential replications
  Acknowledgments
  Appendix
    Counterproposal by the city council
      Unofficial English translation
    Original proposal (ballot initiative)
      Unofficial English translation
    Media coverage
None
20 comments

Summary

Initial proposal and signature collection

In spring 2016, the Effective Altruism Foundation (EAF), then still based in Basel, Switzerland, launched a ballot initiative asking for the city of Zurich’s development cooperation budget to be increased and to be allocated more effectively. (For information on EAF’s current focus, see this article [EA · GW].) We chose Zurich due to its large budget and leftist/centrist majority. I published an EA Forum post introducing the initiative [EA · GW] and a corresponding policy paper (see English translation). (Note: In the EA Forum post, I overestimated the publicity/movement-building benefits and the probability that the original proposal would pass. I overemphasized the quantitative estimates, especially the point estimates, which don’t adequately represent the uncertainty. I underestimated the success probability of a favorable counterproposal. Also, the policy paper should have had a greater focus on hits-based, policy-oriented interventions because I think these have a chance of being even more cost-effective than more “straightforward” approaches and also tend to be viewed more favorably by professionals.)

We hired people and coordinated volunteers (mostly animal rights activists we had interacted with before) to collect the required 3,000 signatures (plus 20% safety margin) over six months to get a binding ballot vote. Signatures had to be collected in person in handwritten form. For city-level initiatives, people usually collect about 10 signatures per hour, and paying people to collect signatures costs about $3 per signature on average.

Picture: Start of signature collection on 25 May 2016.

Picture: Submission of the initiative at Zurich’s city hall on 22 November 2016.

The legislation we proposed (see the appendix) focused too strongly on Randomized Controlled Trials (RCTs) and demanded too much of a budget increase (from $3 million to $87 million per year). We made these mistakes because we had internal disagreements about the proposal and did not dedicate enough time to resolving them. This led to negative initial responses from the city council and influential charities (who thought the budget increase was too extreme, were pessimistic about the odds of success, and disliked the RCT focus), implying a <1% success probability at the ballot because public opinion tends to be heavily influenced by the city council’s official vote recommendation. At that point, we planned to retract the initiative before the vote to prevent negative PR for EA, while still aiming for a favorable counterproposal.

Counterproposal

As is common for Swiss ballot initiatives, the city decided to develop a counterproposal. Their initial draft did not say anything about effectiveness and focused on other topics that we did not consider important. On 28 June 2018, I was able to present the initiative to the financial commission of the city council, carefully preempting objections and misconceptions (see the slides). I managed to convince them to make significant changes to the counterproposal in return for promising to retract the original proposal (which is also standard practice in Switzerland). The city council voted to bring a counterproposal to the ballot that increases the development cooperation budget from $3 million to $5–17 million per year (0.3–1% tax percentage points (calculated as 0.3–1% of the federal tax amount), with exceptions in case of budget deficits) and states that grants should be made “based on the available scientific research on effectiveness and cost-effectiveness.” The counterproposal was supported by all parties (SP, glp, GP, AL, EVP) except for the right-wing ones (FDP, SVP), with 79 in favor, 38 against, and no abstentions (see the minutes (p. 883)). We supported the counterproposal because it contained the key points of our original initiative and had a high chance of success. As promised, we then retracted the original proposal.

Historically, 75% of counterproposals had passed in the city of Zurich. According to public opinion polling (“Sicherheit 2018” p. 147), 59% of the Swiss electorate and 85% of left-wing voters (a ~70% majority in Zurich) agree with the statement “Switzerland should provide more foreign aid.” Other polls had found similar results. For these reasons, I assigned a 77% probability that the counterproposal would succeed at the ballot.

Campaign and vote

As is usual, the parties in favor of the counterproposal invited us to their campaigning committee. On the EAF side, the campaign was coordinated by Noémie Zurlinden, a development economics Ph.D. student at the University of St. Gallen, and Patrick Stadler, who previously worked at a Swiss aid agency and New Incentives (a charity which went through Y Combinator and received several GiveWell incubation grants). They worked with the committee on billboards, media interviews, a press conference, and a press release. The parties contributed substantially to the budget (around $20,000, of which we contributed $5,000) and administration of the vote campaign.

Picture: Vote campaign billboard. Translation: “One tax percent against global poverty – YES on 17 November 2019. Zurich for a good life, also elsewhere. For effective development cooperation.”

The vote took place on November 17th, 2019. The counterproposal passed with 69.7% in favor, which is considered a very strong majority. Voting turnout was 36%.

Picture: Vote results.

See also:

Policy implementation

This seems to be the first time that Swiss legislation on development cooperation mentions effectiveness requirements. We are very excited about this. At the same time, this means it’s pretty unclear how the additional funding will be allocated in practice. Much of it will go to large local charities, but parts of it might also be allocated to projects that are considered particularly effective by EAs (e.g., the Swiss branch of TamTam, a charity that primarily conducts fundraising for the Against Malaria Foundation).

Well-known development economist Prof. Dina Pomeranz of the University of Zurich offered to advocate for the counterproposal publicly and participated in several media interviews. (We had been in touch for many years, she’s a member of our advisory board.) Hopefully, the local administration and members of parliament will make use of her expertise in professionalizing the allocation of funds to development projects. Plans for such a professionalization announced by city officials could constitute another positive outcome of the ballot initiative.

We have the relevant contacts to encourage proper implementation of the initiative, but our organization now has a different focus and so won’t be able to spend much time on this. If any Swiss effective altruists interested in international development would like to help improve implementation (on a volunteer basis or paid), I encourage them to get in touch (jonas.vollmer at ea-foundation.org). I think this could be a highly effective use of their time, likely worth hundreds of dollars per hour invested.

Cost-benefit analysis

I quickly prepared a post hoc cost-benefit analysis of the initiative. For details, see the Guesstimate model. In the bullet points below, the numbers sometimes do not add up precisely due to rounding and small imprecisions introduced through the randomness of Monte Carlo simulations.

Costs:

Benefits:

Costs vs. benefits:

This compares favorably to most conventional fundraising methods, which usually raise $1.50–10 per dollar invested. It seems similar to other EA fundraising projects, though it looks less attractive from a longtermist perspective (because these funds are specifically going towards global development).

That said, some substantial indirect benefits aren’t accounted for:

Potential replications

Given the clear success of the initiatives, effective altruists should likely consider replicating similar ballot initiatives elsewhere. Ideas include:

Depending on the type of initiative, this could be a promising activity for EA groups (consisting of volunteers and some part-time staff), aspiring EA politicians, or organizations with several full-time staff.

If you are interested in launching another initiative in Switzerland, please let us know in the comments and email jonas.vollmer at ea-foundation.org. I won’t contribute myself, but I can put you in touch with funders and others who are interested and share additional information. Those interested in U.S. ballot initiatives can reach out to Rethink Priorities.

For an extensive report, see Rethink Priorities’ Intervention Profile: Ballot initiatives [EA · GW].

Acknowledgments

I would like to thank Kaspar Etter, Aaron Gertler, Zachary Robinson, Jason Schukraft, Patrick Stadler, Pascal Zimmer, and Noémie Zurlinden for giving feedback on this article.

Appendix

Counterproposal by the city council

Gegenvorschlag des Gemeinderats zur Volksinitiative «Ein Prozent gegen die globale Armut (1%-Initiative)» (Gemeindebeschluss)

AS 856.100

Beiträge für die internationale Zusammenarbeit

Art. 1 Die Stadt gewährt jährlich Beiträge für die internationale Zusammenarbeit. Der Umfang der Beiträge entspricht mindestens 0,3 und höchstens 1 Steuerprozent.

Art. 2 Wenn die Stadt einen Bilanzfehlbetrag aufweist oder wenn die letzten drei Rechnungsjahre insgesamt mit einem Defizit von mehr als 30 Millionen Franken abgeschlossen haben, können die jährlichen Beiträge tiefer ausfallen oder ganz entfallen.

Art. 3 Die Stadt strebt für das Vergabeverfahren möglichst tiefe Kosten und, wo sinnvoll, eine Koordination mit dem Bund an. Die Vergabepraxis orientiert sich an der vorhandenen wissenschaftlichen Forschung über Wirksamkeit und Wirtschaftlichkeit sowie an den Aspekten der Transparenz und der Ökologie.

Art. 4 Der Gemeindebeschluss vom 5. März 1972 betreffend Entwicklungshilfe im In- und Ausland (AS 856.100) wird aufgehoben.

Art. 5 Der Stadtrat setzt diesen Beschluss in Kraft.

Unofficial English translation

Counterproposal of the municipal council to the popular initiative "One percent against global poverty (1% initiative)" (municipal resolution)

AS 856.100

Contributions to international cooperation

Art. 1 The City shall grant annual contributions for international cooperation. The amount of the contributions shall be a minimum of 0.3 and a maximum of 1 tax percentage point.

Art. 2 If the City shows a balance sheet deficit or if the last three financial years have, in total, concluded with a deficit of more than 30 million Swiss francs, the annual contributions may be reduced or omitted altogether.

Art. 3 The City shall aim to keep the cost of the grant award procedure as low as possible and, where appropriate, to coordinate it with the Federal Government. The award practice shall be based on the available scientific research on effectiveness and cost-effectiveness as well as on the aspects of transparency and ecology.

Art. 4 The municipal resolution of 5 March 1972 on domestic and foreign development aid (AS 856.100) is repealed.

Art. 5 The City Council puts this resolution into effect.

Original proposal (ballot initiative)

See also the initiative sheet.

Städtische Volksinitiative

Ein Prozent gegen die globale Armut (1%-Initiative)

Gestützt auf Art. 15 ff. der Gemeindeordnung der Stadt Zürich und das Gesetz über die politischen Rechte stellen die unterzeichnenden Stimmberechtigten der Stadt Zürich folgendes Begehren:

Die Gemeindeordnung der Stadt Zürich wird wie folgt geändert:

Art. 2 septies

1 Die Stadt unterstützt hochwirksame Hilfswerke im Bereich der Internationalen Zusammenarbeit mit einem Prozent ihres Budgets.

2 Mit den zusätzlichen Mitteln soll eine möglichst grosse Wirkung erzielt werden, insbesondere im Bereich der globalen Armut und Gesundheit. Die Stadt unterstützt daher Hilfsprojekte, welche durch unabhängige wissenschaftliche Forschung, insbesondere randomisiert-kontrollierte Studien aus der Entwicklungsökonomie, als kosteneffektiv eingestuft wurden.

3 Die Stadt setzt sich im Rahmen ihrer Möglichkeiten beim Bund und beim Kanton Zürich dafür ein, die Öffentliche Entwicklungshilfe (APD) auf ein Prozent des Bruttonationaleinkommens zu erhöhen.

Begründung:

Die globale Armut ist eines der wichtigsten ethischen Probleme unserer Zeit: Nach wie vor sterben in Entwicklungsländern jeden Tag 20’000 Kinder – ein Elend, das wir nur ertragen können, weil wir es nicht selbst täglich vor Augen sehen.

Über zwei Drittel der Schweizerinnen und Schweizer wollen, dass die Schweiz mehr Entwicklungshilfe leistet (repräsentative ETH-Umfrage «Sicherheit 2015»). Trotzdem investierte die Schweiz 2015 nur 0.52 Prozent des Bruttonationaleinkommens und erreichte damit nicht einmal das UNO-Mindestziel von 0.7 Prozent. Weniger wohlhabende Länder (z. B. Niederlande, Grossbritannien, Dänemark und Schweden) spenden deutlich mehr als die Schweiz. Als einflussreichste Gemeinde der Schweiz kann sich die Stadt Zürich besonders gut auf Bundesebene dafür einsetzen, die humanitäre Tradition der Schweiz fortzusetzen.

Der Nutzen der Entwicklungshilfe wird immer wieder pauschal infrage gestellt, vor allem medial. In den letzten 10 Jahren wurde jedoch intensiv geforscht, und wir wissen heute bedeutend mehr über erfolgreiche Entwicklungszusammenarbeit: Unabhängige Forschungsinstitute haben Hilfsprojekte identifiziert, deren hohe Wirksamkeit durch wissenschaftliche Forschung und insbesondere randomisiert-kontrollierte Experimente («RCT») mehrfach nachgewiesen wurde. Selbst skeptische Expertinnen und Experten anerkennen die Wirksamkeit dieser Projekte. Die Stadt kann sich auf diese unabhängigen Evaluationen verlassen und einen wissenschaftlichen Beirat beiziehen.

Mit dieser Initiative setzen wir hohe Qualitätsansprüche an Hilfsprojekte und nehmen unsere globale Verantwortung wahr.

Unofficial English translation

Municipal Popular Initiative

One percent against global poverty (1% initiative)

Based on Art. 15 et seq. of the Municipal Code of the City of Zurich and the Law on Political Rights, the undersigned voters submit the following proposal to the City of Zurich:

The Municipal Code of the City of Zurich is amended as follows:

Art. 2 septies

1 The City shall support highly effective aid organizations in the area of international development cooperation with one percent of its budget.

2 The additional funds shall be used to achieve as large an impact as possible, especially in the area of global poverty and health. The City shall, therefore, support aid projects that have been deemed cost-effective by independent scientific research, in particular, randomized controlled trials from development economics.

3 Within the scope of its possibilities, the City shall commit the Federal Government and the Canton of Zurich to increase official development assistance (ODA) to one percent of gross national income.

Rationale:

Global poverty is one of the most important ethical problems of our time: 20,000 children continue to die every day in developing countries - a misery that we can only bear because we do not see it before our eyes daily.

Over two thirds of the Swiss people want Switzerland to provide more development aid (representative ETH survey "Security 2015"). Despite this, Switzerland invested only 0.52 percent of its gross national income in 2015 and thus did not even reach the UN minimum target of 0.7 percent. Less prosperous countries (e.g., the Netherlands, the United Kingdom, Denmark, and Sweden) donate significantly more than Switzerland. As Switzerland's most influential municipality, the City of Zurich is particularly well-placed to promote the continuation of Switzerland's humanitarian tradition at the federal level.

The effectiveness of development aid is often questioned across the board, especially in the media. In the last ten years, however, there has been intensive research, and today we know considerably more about successful development cooperation: independent research institutes have identified aid projects whose high effectiveness has been demonstrated several times through scientific research and, in particular, randomized controlled trials ("RCT"). Even skeptical experts acknowledge the effectiveness of these projects. The City can rely on these independent evaluations and consult a scientific advisory board.

With this initiative, we set high quality standards for aid projects and meet our global responsibility.

Media coverage

Note: All media coverage is in German. If you do not speak German, I recommend using DeepL to translate the reports.

When the initiative was launched, media coverage was mostly positive:

When the city found issues with the initiative and recommended against it, reports were more critical:

During the vote campaign and after the vote, media coverage was mostly positive. Some reports were critical in a way that appeared reasonable to me.

After the vote:

20 comments

Comments sorted by top scores.

comment by Jonas Vollmer · 2020-01-13T11:37:15.448Z · score: 39 (14 votes) · EA(p) · GW(p)

EAF also launched the following ballot initiatives (through its now spun-off project Sentience Politics):

  • Ban on factory farming (see also this), federal initiative in Switzerland, signatures collected, vote expected in ~2023
  • Basic rights for primates [EA · GW] (see also this), Canton of Basel-City, signatures collected, initiative first deemed invalid then valid upon appeal, not exactly sure what the status is but I think the vote is expected in ~2021
  • Sustainable nutrition, Lucerne, counterproposal passed (60% in favor) at the ballot in September 2018
  • Sustainable nutrition, Basel, rejected at the ballot (67% against) on 4 March 2018
  • Sustainable nutrition, Zurich, counterproposal passed (60% in favor) at the ballot in November 2017
  • Sustainable nutrition, Berlin Kreuzberg/Friedrichshain, implemented in a weakened form by the city without a vote in ~2018

Edit: Note that the main idea behind "sustainable nutrition" initiatives was to reduce meat consumption and promote veganism and animal welfare. There are also significant environmental benefits, but those weren't the main reason for launching the initiatives.

comment by JanBrauner · 2020-01-13T20:05:14.464Z · score: 3 (2 votes) · EA(p) · GW(p)

I first thought that "counterproposal passed" means that a proposal very different to the one you suggested passed the ballot. But skimming the links, it seems that the counterproposals were actually similar to your original proposals?

comment by Jonas Vollmer · 2020-01-13T21:38:03.467Z · score: 5 (3 votes) · EA(p) · GW(p)

That's correct. The original proposals for sustainable nutrition explicitly mentioned "plant-based" and "animal-friendly" food, but then the counterproposals only said "sustainable" or "environmentally friendly." So I'd say overall, from an animal welfare perspective, they were moderately successful. We didn't have the time to evaluate their actual impact, though I think this would be a worthwhile project for EAs, especially if it results in an EA Forum article similar to this one.

comment by Andreas Schneider · 2020-01-14T12:38:46.146Z · score: 14 (7 votes) · EA(p) · GW(p)

A larger development cooperation budget implies additional taxes or cuts from other budget items in Zurich.

I notice that the post does not attempt to compare this cost to the benefits of the ballot initiative.

The EA common sense analysis is that a marginal dollar donated to global development does more good than a marginal dollar donated to charity in wealthy countries. But this intervention is pretty different from a typical donation decision (because it potentially relies on increasing taxes, and so depends on one's views on taxation). Does the EAF team have any thoughts on this issue?

comment by Jonas Vollmer · 2020-01-14T17:26:56.413Z · score: 11 (4 votes) · EA(p) · GW(p)

Thanks for the input!

Because modeling this involves several judgment calls and would make the analysis much more complex (and harder to understand), we decided it's better not to include it in the quantitative model and instead just mention it in the text.

I also think this is unlikely to change the numbers by more than 10%. I think it would take several fairly strong assumptions to change that, such as you think that Zurich's marginal budget is used effectively in an important cause area such as global catastrophic risk research funding.

Some brainstorming ideas for how to model this cost:

  • You could model a tax increase as a reduction in income for Zurich residents (using data on per-capita GDP in the city of Zurich, this is available) and compare that to an increase in the income for the average development cooperation recipient (taking into account that some funding is used for Swiss development cooperation staff compensation). The line of reasoning from this article (also linked above) could be helpful to then translate this into welfare changes.
  • You could try to better understand spending cuts by looking at the budget items and which ones tended to be cut during past cuts, then try to estimate how they compare to development cooperation (or the things Zurich residents usually spend money on).
comment by Andreas Schneider · 2020-01-15T01:12:39.209Z · score: 12 (6 votes) · EA(p) · GW(p)

Hi Jonas, I appreciate the response.

You could model a tax increase as a reduction in income for Zurich residents (using data on per-capita GDP in the city of Zurich, this is available) and compare that to an increase in the income for the average development cooperation recipient

This would make sense if we model the result of the ballot initiative as a move by nature and treat all humans symmetrically. However, when the decision is made by a mechanism that can be influenced (such as a legislative body), then under a variety of moral views, it matters very much how the decision was made. A voluntary donation, a donation resulting from social pressure, a donation using stolen money, and a donation funded via taxation can all be treated differently.

For example, under views particularly critical of taxation, taxation is equated to theft and one might consider foreign aid via increased taxation unacceptable regardless of the effectiveness of aid. Philosopher Michael Huemer has defended such a view starting from common sense morality.

A similar issue exists for budget cuts (when the recipient of public expenditure is no longer taxed).

comment by Jonas Vollmer · 2020-01-15T09:35:25.247Z · score: 19 (10 votes) · EA(p) · GW(p)

Right, a non-consequentialist analysis might lead to different conclusions in this case. Thanks for pointing that out!

I think there's still a pretty strong case to be made that in the case of development cooperation, it's not quite as straightforward because developed countries have harmed developing countries in many ways (colonialism, tax havens, agricultural export subsidies, etc.). Thomas Pogge has argued along these lines IIRC, so one could look at his views on this.

More generally, we live in a highly globalized world, we routinely interact with these countries through trade, etc., such that it seems plausible that we do have some responsibilities towards them. And we're talking about one of the very wealthiest cities in the world (with a per-capita GDP of $180,000!) giving a relatively small additional amount. So if there is one particular case where Huemer's arguments appear particularly implausible, it is probably this one.

Overall, I don't think it's obvious whether the case for development cooperation becomes weaker or stronger if we take into account various non-consequentialist perspectives.

comment by Benjamin_Todd · 2020-01-13T19:20:37.584Z · score: 13 (9 votes) · EA(p) · GW(p)

Great news! I'd be keen to hear where the money ends up being allocated.

comment by G Gordon Worley III (gworley3) · 2020-01-14T19:09:03.471Z · score: 10 (6 votes) · EA(p) · GW(p)
Die Vergabepraxis orientiert sich an der vorhandenen wissenschaftlichen Forschung über Wirksamkeit und Wirtschaftlichkeit sowie an den Aspekten der Transparenz und der Ökologie.
The award practice shall be based on the available scientific research on effectiveness and cost-effectiveness as well as on the aspects of transparency and ecology.

What is the likelihood of this sentence of the policy having teeth? For example, let's say people administering this money want to use it for a prototypical low-effectiveness intervention, like opening an art gallery in a poor country. Is there a mechanism in place to stop them? Who decides if a grant was chosen based on scientific research on effectiveness? Can, for example, a citizen sue the city for failing to follow this policy and have a judge rule they misallocated the funds, impose some penalty, and require they act differently in the future?

To me this language seems just vague enough that a motivated politician could use it to fund almost anything they wanted, so I'm wondering what evidence there is to believe this policy will do anything, as this has a great deal of impact on the measure of its effectiveness (so much so that it could flip the sign of your assessment and maybe all they money was spent to buy empty words).

Obviously we can't know for sure until after we have seen grants awarded and especially seen grants misawarded and what the response was to that, but I'm curious what information we have now about this since I'm unfamiliar enough with Swiss government that I can only make an estimation based on my outside view prior that governments tend to find a way to do whatever they want regardless of what the law says unless the law or popular sentiment can actually force them to do what a policy intended.

comment by Jonas Vollmer · 2020-01-15T09:53:33.993Z · score: 12 (7 votes) · EA(p) · GW(p)

They already have a committee allocating the grants which includes some academics, and they said they want to further improve the award practice. We have suggested specific academics they could work with. I'm not sure what it will end up looking like in practice. There are certainly some people in the administration who are eager to preserve the status quo, whereas others seemed quite excited about effectiveness improvements.

I don't think it's possible for citizens to sue the government for failing to implement a ballot initiative (or at least that's very uncommon). But there are many indirect ways to enforce an initiative, e.g., we could talk to the members of the city council who we know and work with them to submit motions to improve the implementation of the initiative. In general, referenda are taken very seriously in Switzerland.

As I wrote above, the bottleneck is likely EA-aligned people with development knowledge wanting to spend a couple of hours per year on this (rather than formal ways of suing/filing complaints if it's not implemented in the way we'd like). I think even a few small, friendly nudges would go a long way.

so much so that it could flip the sign of your assessment

That sounds like you think it might have been net negative, but I don't see how that follows from your points. Unless you think the entire budget has literally a zero impact, which I think is very unlikely for the following reason:

I think it's likely to have a significant positive impact if citizens of a city with a nominal per-capita GDP of $180,000 (source) give more money to people in developing countries (with a per-capita GDP which is ~2 orders of magnitude lower), even if that happens inefficiently. (There's a lot of EA and non-EA writing on the indirect effects of foreign aid, etc. so I'm not going to elaborate more on that here.)

comment by Jonas Vollmer · 2020-01-26T16:27:59.377Z · score: 9 (6 votes) · EA(p) · GW(p)

Update: We've hired someone part-time (a couple of hours per month) to help ensure good implementation of the initiative.

comment by evelynciara · 2020-01-14T23:29:49.726Z · score: 4 (3 votes) · EA(p) · GW(p)

This is awesome! I didn't know that municipalities had international development budgets. I wonder if something like this has been done in the United States.

comment by Matt_Lerner (mattlerner) · 2020-01-13T16:24:38.429Z · score: 4 (4 votes) · EA(p) · GW(p)

Fantastic work! In your post [EA · GW] introducing this initiative you wrote that the base rate for passage of ballot initiatives was 11%. A conservative reading of the data here (taking the low value of $20m for development funding raised) seems to indicate a 100:1 return on investment. Taking the base rate, this $10 in effective development aid for $1 spent on advocacy (in expectation). If the development aid is effectively spent, the implication here is that money spent on an initiative like this might be ten times as effective in expectation as money donated directly to a top-rated charity. This assumes, of course, that the base rate is accurate.

In that initial post, you had an exchange [EA · GW] with Stefan Schubert about the relevance of your assumed base rate. You discussed the importance of polling at that point but it's not clear to me where you left off.

This success really seems to highlight the importance of public opinion polling here. The value of information in this domain is very high, since you're trying to identify the avenue which will provide the greatest leverage. Choosing the wrong avenue has no value, and potentially even minor reputational costs for your organization or for EA in general. Choosing the right avenue has huge upsides.

Public opinion polling seems crucial to this end. In this scenario, prior polling might have allowed you to identify a reasonable figure beforehand (avoiding the $87 million overreach). More importantly, though (if I understand the procedure correctly), it might have enabled you to avoid the counterproposal process and to pinpoint an optimal figure to ask for-- perhaps one higher than the one you ultimately got.

I don't want to diminish the achievement here, which I think is huge; I just want to point out that extremely useful information for this effort can be retrieved from the public at relatively low cost. In the future, this information can be used to reduce the uncertainty around efforts to fund ballot proposals and increase the expected value of these efforts by lowering the probability of failure in expectation.

comment by Jonas Vollmer · 2020-01-13T16:41:46.517Z · score: 8 (7 votes) · EA(p) · GW(p)

I agree with the importance of "choosing the right avenue." I still don't think public opinion polling is very useful for that purpose (especially if some polling data is already available). In fact, I think public opinion polling would have been unlikely to clearly identify the key issues because the general public has much less pronounced and well-informed opinions than politicians and other stakeholders.

At least for Swiss initiatives, getting reactions/opinions from the responsible legislative body and the people they trust (like local charities in this case) seems much more useful because it shapes the legislative bodies' official recommendation to voters. I think it was a mistake not to do more of that type of stakeholder engagement in the early stages of the initiative, and that mistake almost led to a complete failure of the initiative.

Also noteworthy: Talking to local politicians is much cheaper still than doing public opinion polls (costs a couple of hours rather than thousands of dollars plus a lot of work to get the polling right).

That said, I think doing some polling before launching an initiative could also be somewhat helpful.

comment by Matt_Lerner (mattlerner) · 2020-01-16T19:02:20.275Z · score: 11 (3 votes) · EA(p) · GW(p)

Thanks for your response. I think I should make clear (as I really didn't do in my initial post) that I mean my comment more broadly: when EAs think about doing ballot initiatives, they should strongly consider doing public opinion polling. In a setting where an EA advocacy group is trying to select (a) which of X effective policies to advocate and (b) in which of Y locales to advocate it, it seems (to me, at least) that polling is cost-effective, since choosing between X*Y potentially large number of independent options is a nontrivial problem that requires a rigorous approach.

In your setting, however (making the binary choice of whether or not to advocate for policy P in location L), I understand why you chose the strategy you did. Your point about the relative cost-effectiveness of talking to local politicians versus conducting an (arguably) expensive poll is well-taken. I don't have any idea how Swiss referenda work and I conclude from your comment that voters largely follow the lead of their representatives.

I'm not sure how you're thinking about future efforts along these lines, but if you're planning on selecting from a longer list of policies and cantons, I think polling—in a cheap way—could challenge your legislative strategy for cost-effectiveness, at least as a guide for initial research investment.


comment by Jonas Vollmer · 2020-01-16T22:28:39.527Z · score: 2 (1 votes) · EA(p) · GW(p)

Fully agreed, thanks for the clarification!

comment by Jonas Vollmer · 2020-01-13T21:40:04.515Z · score: 3 (2 votes) · EA(p) · GW(p)

To further clarify: I think in many circumstances (e.g., for a ballot initiative in Switzerland on the federal level), public opinion polling would be crucial. But for this specific type of city-level initiative, I don't think it would help much.