No I wouldn't create a person who would spend their entire life in agony. But I think the reason many people including myself hold the PAV despite the procreation asymmetry is because we recognise that, in real life, two thing are separate: (1) creating a person; (2) making that person happy. I disagree that (1) alone is good. At best, it is neutral. I only think that (2) is good. 

If I were to create a child and abandon it, I do not think that is better than not creating the child in the first place. That is true even if the child ends up being happy for whatever reason (e.g. it ends up being adopted by a great parent).  

In contrast, it is indeed possible to create a child who would spend their entire life in agony. In fact, if I created a child and did nothing more, that child's life would likely be miserable and short. So I see any asymmetric preference to avoid creating unhappy lives, without wanting to create happy lives, as entirely reasonable. 

Moreover, I do not think moral realism is correct and see different views of population ethics as being subjective. They depend on each person's intrinsic values. And no intrinsic values are logical. Logic can help you find ways to achieve your intrinsic values. But it cannot tell you what your intrinsic values should be. Logic is a powerful tool, but it has limits. I think it is important to recognise where logic can help—and where it can't. 

Thanks for the considered response. You're right that the Total View is not the only view on which future good lives has moral value (though that does seem to be the main one bandied about). Perhaps I should have written "I don't subscribe to the idea that adding happy people is intrinsically good in itself" as I think that better reflects my position — I subscribe to the Person-Affecting View (PAV). 

The reason I prefer the PAV is not because of the repugnant conclusion (which I don't actually find "repugnant") but more the problem of existence comparativism — I don't think that, for a given person, existing can be better or worse than not existing. 

Given my PAV, I agree with your last point that there is some moral value to ensuring happy people in the future, if that would satisfy the preferences of current people. But in my experience, most people seem to have very weak preferences for the continued existence of "humanity" as a whole. Most people seem very concerned about the immediate impacts on those within their moral circle (i.e. themselves and their children, maybe grandchildren), but not that much beyond that. So on that basis, I don't think reducing extinction risk will beat out increasing the value of futures where we survive. 

To be clear, I don't have an objection to the extinction risk work EA endorses that is robustly good on a variety of worldviews (e.g. preventing all-out nuclear war is great on the PAV, too). But I don't have a problem with humans or digital minds going extinct per se. For example, if humans went extinct because of declining fertility rates (which I don't think is likely), I wouldn't see that as a big moral catastrophe that requires intervention. 

I think humans will go extinct at some point, so reducing extinction risk just kicks the can down the road. 

On a selfish level, I don't want humans to go extinct anytime soon. But on an impartial level, I don't care really care whether humans go extinct, say, 500 years from now vs 600. I don't subscribe to the Total View of population ethics so I don't place moral value on the "possible lives that could have existed" in those extra 100 years. 

I agree that "systemic change" (broadly speaking) seems highly intractable for EA and that it does not seem to be that neglected outside of EA. 

However, one possible advantage to this lack of neglectedness is that there are probably plenty of people already working on, or interested in, some forms of systemic change who are already sympathetic to EA ideas but don't know how to translate that into action. A lot of those people may also be mid-career, like me. 

EA might be able to play a useful coordination role between such practitioners and publicise ideas that could improve their effectiveness with relatively little resources. There are plenty of common "best practices" like red-teaming, Delphi method, or reference class forecasting that can help decision-making but are still not widely used in much of government (at least in my experience, though I'm sure it will depend on the country involved). Of course, there is already a lot of material out there already on "how to be more effective" but much of it is either (1) low-quality; (2) focused on business and personal success; and/or (3) rather expensive (for courses aimed at business leaders). 

Even something like mildly improving the epistemics (and perhaps morals) of people who are already in positions to shape important policies could be quite impactful, though very hard to quantify. I personally know many people in policy who genuinely want to do good but would not be willing to switch to one of EA's cause areas. It seems like EA's message to such people is basically just "earn to give", which feels like a waste. That's not to denigrate earning to give at all - it just feels like there is more potential there that EA is not tapping into.  

My comments are more about EA turning to focus on system-level interventions in general, rather than to address the problems arising in this particular time of flux. But, you know, even if the best time to plant a tree was 20 years ago, it could still be worth planting today.

Hi James, thanks for your comment. A couple of points in response:

OECD Model and residence-based taxation

I disagree that the OECD Model has led to a predominantly source country-based tax system. Quite the opposite — relative to the UN Model, the OECD Model favors residence-based taxation. 

In broad terms, under the OECD Model, residence States have taxing rights over an enterprise's business profits (Article 7) unless the DTA provides otherwise. The key exceptions are:

• Article 5 (Permanent Establishment (PE)). To the extent those business profits are attributable to a PE in another State, the residence State gives up taxing rights over those profits. While virtually all countries agree with this general principle, source States and residence States have different negotiating positions on this, with source States preferring a broad definition of "PE" and residence States preferring a narrow definition. You can see this by comparing the OECD Model's relatively narrow PE definition and the UN Model's broader definition. 
  
  This is what I meant when I said above that "taxing rights are primarily based on physical presence". The concept of a PE has historically been based on physical presence. Over the years, the definition has broadened to include things like dependent agents but not all of those changes will be fully picked up by States that favor residence taxation. 
   
• Article 6 (Immovable property - i.e. land and natural resources). Apart from the PE Article, this is probably the main article giving a broad taxing right to source states. It is a longstanding and widely accepted principle that source States have primary taxing rights over immovable property.
   
• Articles 10 and 11 (Dividends and Interest). Under the OECD Model, residence States give up some taxing rights over dividends and interest income, but they limit the source States' rights to apply withholding tax to some fixed percentage of the dividend or interest (e.g. 10% on the gross interest payment). The exact rates will be key issues in any treaty negotiation. Residence States tend to seek lower rates; source States tend to want higher rates. 
   
• Article 12 (Royalties). This is an important difference between the UN and OECD Models. The OECD Model does not provide for any source State withholding tax on royalties. The UN Model does provide for source State taxation but leaves the rate to be determined in bilateral negotiations. The lack of source taxation on royalties can lead to double-non-taxation because if the residence State doesn't tax foreign-sourced royalties, the royalty income escapes taxation altogether. So this creates an incentive for multinationals to shift their intellectual property to countries that won't tax foreign-sourced royalties (or only tax them lightly). (I should note that, despite the OECD Model, many OECD countries do agree to some source taxation on royalties in their DTAs. But that will be decided in bilateral treaty negotiations.)
   
• Article 13 (Capital gains). Residence States get taxing rights with some limited exceptions such as if the property is immovable property or part of a PE in the other State. Put simply, this means residence States get sole taxing rights over capital gains from most share sales. 

Just to clarify, when I say poorer/developing countries tend to prefer source country taxation, that is a simplification and generalisation. There are some developed countries like Canada that favor strong source taxing rights (I think this is because of its heavy reliance on natural resources and possibly because so many more US businesses operate in Canada than the other way around - but this is speculation on my part). 

Consumer-country based taxes

I haven't read Piketty's book but I generally like the idea of 'destination-based taxes' which are those based on the destination/location of consumers like you describe. Consumption/sales taxes tend to be destination-based whereas income taxes tend to be 'origin-based'. 

My understanding is quite a lot of academics and economists like destination-based taxes, particularly the 'destination-based cash flow tax' and some prominent economists have written some detailed papers on it (see e.g. Auerbach et al, 2017). In 2016, the US Republican Party even included a proposal for it, but that did not go ahead. 

I am broadly sympathetic to the idea but I have not looked at it in detail and it's not my area of expertise. I might look into it more at some point in the future, especially if the idea gains traction in the US (I suspect it would be hard for other countries to implement it unilaterally). 

Hope that is helpful and happy to clarify if there's anything I haven't explained well. I'll also flick you a PM.

Hi Johannes, thanks for your comment. Glad to hear you're working in this area too and thanks for providing that additional context for the global minimum tax. 

One difficulty in international tax policy is that it can be really hard to work out what good tax policy looks like, apart from any national interests. I've only been loosely following the global minimum tax and I understand there are competing views as to whether the minimum tax is a good idea and what level it should be set at (i.e. a higher rate is not necessarily always better). 

Personally I'm agnostic on this because I simply don't know enough about the various arguments and counterarguments, which is why my original post focused at a higher level on international tax policy being a relatively neglected cause area and on how international tax policy development is dominated by developed countries focusing on their own national interest (two points I feel more confident about). 

But I'd be keen to discuss this and other tax issues with globally-minded people like yourself. I'll send you a private message :)