I want to analyze the effectiveness of creating society-wide structures of incentives to motivate and reward people to work on social goods, particularly creating public goods markets (PGMs) (as exemplified by Vitalik Buterin, Zoe Hitzig and Glen Weyl in their paper on Quadratic Funding). My analysis suggests that this is promising in terms of importance, tractability, and neglectedness. I’ll also raise and respond to some potential objections to public goods markets as a cause area.
What is a Public Goods Market?
A public goods market (PGM for short) is a decentralized, highly responsive process for allocating capital (for example, from a government, a group of people, a philanthropist, or any other context) according to the collective desires of a population to support the creation of public goods. I see public goods markets as a mechanism to shift influence, resources (in the long-term, not directly), and responsibility from governments and the private sector, towards the “third sector” (non-governmental public sector), which will play a larger role in a PGM-using society. At the moment, Quadratic Funding is the only process I know of which meets these criteria, although it’s possible that other processes might exist. For more details on how Quadratic Funding works, see Vitalik Buterin’s primer on QF or listen to Glen Weyl on the 80,000 Hours Podcast.
Currently our societies have two main sectors: The private market, which maximizes outcomes for individuals but is incapable of producing public goods, and government, which is supposed to create public goods, but in practice lacks the flexibility and diversity of the free market, inhibiting it from being as effective as would be desired. This combination causes the private market to be in control, even in places where it shouldn’t, causing negative externalities, such as pollution and global warming, and leading people (both as consumers and employees) to be treated poorly by corporations which are required to maximize profits, among other popular criticisms of modern capitalism. The combination of traditional governments and traditional markets does little to promote the production and development of public goods that can help save lives, reduce poverty, increase people’s life satisfaction, and help societies better navigate existential threats. There does exist the ‘third sector’ today, which represents causes that are neglected by both the private market and government, however the third sector is under-funded, and outside of EA’s circle of influence, the resources in the third sector are often poorly allocated.
This is where public goods markets enter the picture. Governments are, in fact, a crude form of a public goods market - they have a revenue stream that receives money from society at large, they have a mechanism which asks the society (albeit with very low resolution) what the money should be spent on, and the income is spent according to the information that is received. It’s like walking into a store because you want chocolate milk, and you see that you can either get an orange juice or a bottle of water - but you don’t want water or orange juice, you want chocolate milk. It’s not a very flexible store, but it is technically a market nonetheless. (I suspect that much resistance from certain subcultures towards paying taxes will evaporate as this problem is resolved, which will be a valuable side-effect of PGMs becoming mainstream)
A public goods market will, over time, radically change the very nature of the societies that adopt them. What are currently private endeavors will become public endeavors, and government projects will become (public) market projects. The non-profit sector of society will flourish, compared to the relative stagnation of the sector that currently exists, and combined with recent innovations in technology, it seems reasonable to expect that public goods markets will usher in a new era of what it means to be a person who lives in a prosperous society: instead of people existing primarily to serve society, society will exist primarily to serve people.
If you look at the things a government does, you can categorize them into two groups: The first is exercising power and authority, such as creating and enforcing laws, collecting taxes, or regulating businesses, while the second is serving as a patron for socially important projects, such as education, public spaces, and utilities. The first category is what government is designed to do, but the second part’s association with government is a simple consequence of the fact that government collects public finance, and is therefore, historically, the spender of public finance. It’s a government’s duty to compel people to contribute money to the social good, but that doesn’t automatically give government the mandate to decide how it gets spent.
One could argue that the allocation of government spending is already decided in a democratic manner - we elect our representatives in congress / parliament, and our representatives are the ones who decide how to spend the money. But there are some problems with this - minority needs are under-served by this setup, there isn’t a direct link between individuals and the decisions that get made, and people don’t always trust their representatives to make decisions that help the population at large. While majority rule may make sense for making binary legislative decisions, when dividing up resources earmarked for the social good, it makes sense that minorities should decide where a portion of the resources go; quadratic funding ensures that a project which benefits a minority of the people gets a sizable minority of the resources - but will get even more funding if it can synthesize the needs of a minority with the needs of the majority.
Whether or not there is corruption in countries’ legislatures, it is undeniable that many people feel that there is corruption. This creates a feeling of alienation between the public and the government, and leads to individuals being reluctant to pay taxes. This reluctance will then make people lobby against taxes, and make people feel justified in engaging in tax evasion, even looking positively towards people who don’t pay their taxes. Furthermore, this alienation can make voters object strongly when they see their legislature funding projects that they didn’t personally ask for.
Public Goods Markets and current EA causes
A society that uses public goods markets (especially if it is implemented on a global scale - something which I hope will one day will be implemented, but do not intend to imply here is currently tractable) will naturally focus on issues that people care about. While the traditional causes EAs focus on don’t overlap perfectly with what one should expect a global public goods market to promote, there is a large overlap, since people want to be healthy (and the poorest people would be represented in a global PGM), nobody wants to die in a nuclear war, or be subject to any other existential risks, and care about their family and descendants, and are therefore motivated to consider how the world will be in the future. It seems reasonable to expect that if a global PGM were implemented with a budget of 1% of global wealth per annum, it would lead to projects currently highlighted by GiveWell, the EA Funds platform, and 80,000 Hours to be bolstered much more than the individual members of the EA movement can accomplish by contributing to the projects directly, simply by virtue of the relative scale of a (admittedly hypothetical and ideal) global PGM in comparison to the EA movement.
Currently, families are given a choice between going to a free publicly-funded school, or can pay to go to a non-publicly-funded school. A family that goes to a public school has minimal choice in what kind of school they go to, since they belong to a certain school district, and the district’s schools have no reason to compete with each other - but many families still choose public schools because they are free. Meanwhile, private schools are incentivized to compete with other schools, and therefore have to hold themselves to a high standard to continue to attract new students, which ensures a high quality of education - but only families which are well-off can afford to send their children to private schools.
In a society with a government-funded PGM, it seems reasonable that many people will vote for certain non-government schools to get funded on the condition that they provide free education. Since there will be several different schools competing for schooling funds, the schools which are most effective at educating students (at least in ways that the public can verify) will receive the most funding from the public, while also allowing people to support alternative schooling methods (within reasonable bounds, similar to what we expect from private and charter schools) if they think that an alternative method is superior.
PGMs may positively change our relationship with news. It's desirable to have a reliable source of information that is independent from both the government and wealth, and many news sources end up getting skewed and biased in ways that help ensure the agency keeps money flowing, rather than reporting what actually would be best to have reported. A PGM allows society at large to support the creation of media, without arousing suspicion of propaganda, since the government has no actual control over the news agencies or whether they get funded. With a PGM funded news agency, it's in their best interest to report information that the public finds valuable, and to do so in a way that people find tasteful.
Other potential impacts
I was originally planning on including more case studies of how PGMs may make society better, but in the interest of actually publishing this essay, I’ll just include this list of areas I expect PGMs to improve the status quo (feel free to ask about my thoughts on these in the comments):
- What is currently private will become public
- Healthcare (in US)
- What is currently government will become part of the market
- Healthcare (in Europe)
- Public spaces (parks, squares)
- The rise of things which don’t fit in either government or the private market
- The cultivation of knowledge outside of “education”
- Litigation / lobbying
- Reviews (e.g. Yelp)
- AI alignment
- The citizenry will transition from serving society to being served by society
- Domestic poverty
- The relationship between survival and labour
- Labour goes from being an obligation to rewarded community service
- What it means to be a consumer
Implementing public goods markets is quite tractable. There are two main ways public goods markets can be implemented: one through convincing governments to set aside some of their budget to be distributed by public goods markets, according to the voice of their constituents, and the other as a grassroots movement to create public goods markets within communities.
The tractability of Public Goods Markets as grassroots movement
Regardless of the tractability of the government route, the grassroots route seems like a sure winner. The basic logic is that given a choice between a society that cooperates, and a society that doesn't, one would always rather live in a society that does (if one uses Causal Decision Theory as opposed to Timeless Decision Theory, then richer members of society may not want to cooperate in the same pool as poorer members, though should still prefer to cooperate with members of society from the same strata); and when a society that has good cooperation competes with one that doesn't, the cooperative society wins.
"Kickstarters" can be run for different cities, with a funding goal that is proportional to the total wealth of the city. In the long-run, the goal will probably be around 1% per year of the city's wealth, but towards the start of the project, the goal might be much less. If the funding goal is met, then the money is gradually distributed over the course of a year according to the voice of those who live in the cities that funded the project, and if the funding goal isn't met, then nothing happens (and everybody keeps their money). Since most people would rather live in a society that works together, everybody has a certain value (which is greater than 0 for most individuals) they would get from seeing the project succeed, and it would be in their best interests to contribute that amount to ensure that the kickstarter doesn't fail.
(I’m not aware of similar projects having been done before, aside from Kickstarter itself and IndieGoGo, which operate on the same principles, and I know that Kickstarter has restrictions that would prevent such a project from being launched there. However, the success of Kickstarter and IndieGoGo [which are fundamentally aimed at achieving similar cooperation that requires an investment from each member which would be against the individual’s interests in the absence of such an assurance mechanism, but is in their interests given that their participation is required to ensure success] suggests to me that such an approach should be viable)
While not every locality might cooperate to make the kickstarter succeed, those that don't will lose (on the margin) jobs, people, and reputation to the localities that do create public goods markets for themselves, and in the long run (50 - 250 years), it would be reasonable to expect that a large proportion of societies will adopt public goods markets as a result of such a grassroots effort.
I'm particularly interested in the possibility of this grassroots effort, since it requires relatively few resources to lay the foundations for, and will (if it works) more or less drive its own success once it gets going. The most obvious problem with this is that people will find it weird to both pay taxes to their government and pay money to the public goods market, however we should still expect people will do so, since the cost-benefit analysis above still holds, and once the public goods market establishes itself, the government will adapt accordingly based on public pressure.
Can the grassroots movement approach lead to a global PGM?
While the thesis of this post does not require establishing a global PGM to be tractable, here are some thoughts on this possibility: The idea of a global PGM may seem like a naïve idealist’s fantasy in light of the fact that despite democratic government being widespread over the globe, there is no global democratic government, with the largest democracy covering no more than 2% of the world’s land area (Canada) or 18% of the world’s population (India). However, it’s worth noting that the way democracy is currently practiced places an effective ceiling on the size of such a nation: since almost all functions of a government are decided by a binary vote (either in the legislature or referendum, and with presidential elections (in non-parliamentary republics) being usually done through a mechanism that degrades into effectively a binary vote), and since people vary continuously across many dimensions of values and beliefs, a republic will alienate a large portion of its population if it grows to become too big (Federalism goes a long way to increase the effective ceiling by making only the most important and largest-scale decisions to be the responsibility of the complete population). However, the Quadratic Funding mechanism is designed specifically to scale gracefully in ways that should (in theory, at least) allow a Quadratic Funding-based PGM to cover an arbitarily large geographic area.
This leaves open the question of whether PGMs which serve a relatively affluent population will extend to benefit less privileged populations, which won’t be discussed in this post.
The Tractability of Government-funded Public Goods Markets
Going down the government route, it will be important to engage voters in a dialogue about the benefits of public goods markets. We can communicate to people that public goods markets will allow for more diversity, flexibility, and accountability in publicly-funded projects, and that it allows for the creation of public goods that the free market fails to create. Public goods markets are an idea that can excite many on both sides of the political debate, so it's very much imaginable that it can get enough momentum to get implemented as public policy.
It is true that there are vested interests that will be resistant to governments adopting public goods markets, particularly in institutions that currently already receive funds from the government, and stand to lose from the increase in competition for public funding that would be caused by a government funding public goods markets. There are plenty of examples of policies being adapted against the wishes of vested interests, it really is a matter of whether or not enough public support can be rallied to counteract the strength of vested interests; at the end of the day, in a republic, the public spirit wins the fight, but only if the public makes a clear stand. I'm optimistic that it's feasible to rally enough public support for public goods markets, but I'd like to hear someone else's analysis of this.
The concept of public goods markets is very new- as far as I know, the concept was first introduced in 2018 in the paper by Buterin, Hitzig, and Weyl which introduced Quadratic Funding. There exists RadicalxChange, Collective Decision Engines, and Gitcoin Grants, but the work being done to promote public goods markets is very small compared to its importance, and in particular, the grassroots strategy I outline above is something that, to the extent of my knowledge, has never been attempted.
As a result of the recency of the concept of public goods markets, and its having not yet expanded widely, it is quite neglected, but even when it becomes less neglected, I still expect that it will continue to have very promising returns on investment (If I had more time, I would delve more into whether PGMs are likely to still be a high-impact area when they become less neglected, but in any case, the current state of affairs is that PGMs are quite neglected, so additional resources and minds working on promoting them will make a large difference in their outcomes)
Some arguments against focusing on public goods markets, and responses
We can't trust the population at large to make effective decisions
[I no longer endorse what I previously had written here, and will leave this currently blank in the name of not letting perfection get in the way of good enough; I have thoughts on this issue and hope to respond to this later]
Giving decision-making power to the people leads to selfish decision making; We can't expect such methods to lead to altruistic decisions
The Quadratic Funding mechanism encourages people to support projects that benefit both themselves and a wide range of people; since the amount of funding a project gets is proportional to the square of the number of people who support it, projects that benefit only a few will not see much success. This means that Quadratic Funding will naturally promote projects that are beneficial to a large number of people.
There is a legitimate concern that if a certain subset of people use QF, but another set of people don’t, then the group will make decisions that benefit itself, at the expense of those not in the group. For this reason, it will be vital to do what we can to ensure there are a few big public goods markets, which represent a representative sample of humanity, particularly in regard to poverty, as opposed to many small public goods markets centered on priviliged groups of people. I think there is room for discussion in the EA community how to ensure this happens.
Democratic decision making cannot lead to good outcomes for animals or people who are not yet alive; people who are alive today will systematically undervalue existential risks relative to a long-term frame
Regarding animals, I think that there’s a good chance this is a good argument for EAs who are primarily concerned about animal welfare to focus on interventions other than public goods markets. That being said, typical people do care about animals, and I expect that public goods markets will help promote projects that will help animals. Particularly in the case of factory farming, I expect that it can help non-animal based meats, as well as more humane methods of farming, to be economically competitive against high-efficiency, low-morality factory farming.
Regarding future people, however, improving the way we make decisions should be an effective way to help people in the future, because we’re setting a precedent for how they will make decisions - if they inherit poor frameworks for making decisions, they will do things that hurt themselves, and if they inherit good frameworks, they will make good decisions.
Further, it is worth observing that things that are bad for future people are often also bad for the people alive today. By aligning society in ways that helps avoid global warming, nuclear war, and bad AI, and that reward sustainability, we help both ourselves and future people. It is also worth observing that future people will be the offspring of the people who are alive today, and since people are motivated to do what’s right by their family, it should be expected that people would be motivated to make sure their descendants inherit a good world.
More democratic decision making only improves outcomes in the long-run; EAs interested in maximizing outcomes in the short-term should focus on the problem directly
Working to introduce public goods markets now should vastly improve the outcomes we see (within regions where PGMs are implemented) within the lifetimes of many EAs, so I don’t really buy this argument.
Public goods markets enable the creation of public goods, but don’t prevent public bads (e.g. overuse of shared depletable resources)
It is true that PGMs don’t have any direct regulatory power, so they may not be able to prevent all public bads. That being said, many public bads that arise in our current system are a result of there being no incentives not to partake in abusive behavior. In a society with a strong public goods market, companies competing on the private market can collaborate with projects which receive funds from the public goods market. If one company does something which is bad, but would otherwise be profitable, the PGM can give funds to a project that will reward rival companies which don’t partake in the bad behaviour, preventing the bad company from becoming dominant, as would happen in current society.
Public goods markets may end up funding rival projects (e.g. national militaries, or political parties) that cause the funds to more or less cancel out and be wasted
I see two main workarounds for this. One would be to simply prohibit projects which are rivalrous in nature. Another approach would be to have a mechanism to gauge consensus as to if a rivalrous project is acceptable for a broad swath of the population, and if it is, to allow it to be funded by the public goods market, while prohibiting rivalrous projects which don’t have a broad consensus behind them.
Creating good outcomes for people who are well off has less marginal value than creating good outcomes for people who need the most help, which conflicts with the principle of one person, one voice, which is implicit in a public goods market
There may be some legitimacy to this concern. One solution would be to give people who aren’t well off a bigger voice, however I think this would be inadvisable, at least within the mainstream public goods market, for many reasons - among them, it sets a precedent for giving different people different levels of influence, which would create room for abuse once the precedent is set, and it may make wealthier societies less willing to participate in a global public goods market.
Ultimately, I don’t think public goods markets should be viewed as a silver bullet to solve all the world’s problems. A global public goods market will help the people who are worst off, and it will also allow already wealthy societies to solve the problems that face them. It won’t magically solve inequality overnight, but it will help put people from drastically different levels of privilege on a more even playing field, while helping everybody in some form or another.
However, this does make me think about the possibility of distributing philanthropic aid using public goods markets - this becomes a separate subject from what I set out to discuss in this post, but I do think it’s worth thinking more deeply about the potential benefits and drawbacks of giving recipients a say in how philanthropic aid is used. (Compare this to Give Directly, which also lets recipients decide how to use funds. This gets into the same problems of private vs. public incentives that I’ve discussed above)