Efficient Advocacy

It’s incredible how simple and yet revolutionary the principles are behind effective altruism as well as the ideas behind GiveWell and the Open Philanthropy Project; if you want to help people, don’t just donate to a charity that is looking to cure a rare disease, donate in a way that can do the maximum amount of “good” per dollar.  That often means donating to a problem that affects many people, that has known, measurable, positive solutions, and that has lots of room for additional resources to combat the problem.   If you don’t know about those organizations, you should definitely check them out.

Of course, there is an obvious elephant in the room when it comes to effective altruism: politics is complex, unscientific, and unpopular. In fact, GiveWell largely sidesteps the political sphere, ignoring a big swath of human activity which has tremendous impacts on society.  Of course, they have good reason to do this; it allows them to focus on doing good things without harming anyone’s tribal identities or alienating their donor base. Moreover, it’s hard to get good unbiased data on what political policies would actually provide benefits; if there was, politics wouldn’t be so divisive.

However, I don’t have a donor base, and I have slightly different feelings on which policies would be most effective than the average American or even the average effective altruist.  I wanted to see what would happen if we could assume away some of the unknowns about political policy.  Let’s assume that the postlibertarian philosophy this blog espouses is correct: markets are pretty good at allocating resources efficiently, government policy can help address some economic areas where markets might not work (inequality, externalities), giving the state power is generally a bad thing and must be justified, and individuals should have robust protections from their government. We aren’t assuming away the current political landscape of the US, we’re just assuming we’re right.

So what would a libertarian trying to maximize efficiency in advocacy do? Do you try and emulate the Koch brothers and create or fund political organizations that change policy outcomes? Do you focus on viable candidates? How much do you accept the political process as given? Do you focus on political reforms (proportional representation), education (IHS, Economics of Library and Liberty), or do you try to work on making your own rules (crypto, seasteading, space exploration)? Let’s leave those hard questions for another time, and focus on perhaps the most mainstream approach to politics: how should you prioritize the importance of various political issues? People usually have specific issues they care about that determine which candidate they’d like to back, and the Open Philanthropy Project even has a U.S. policies page.   But which issues are actually the most important?

Free Trade

One clear starting point is free trade. Trade has done more to pull people out of poverty than almost anything else, and it’s sustainable both in bringing capital to poverty stricken areas, and in keeping the poor out of poverty. Despite this, it’s conspicuously absent from the Open Philanthropy Project’s policy page.  There’s an argument to be made that as a domestic policy, free trade is not all good news; the people displaced by trade suffer real pain that is largely outside of their control to prevent. This isn’t to say that the benefits of trade aren’t worth it, but rather we should promote free trade due to its overwhelming benefits, and separately compensate people whose income is negatively impacted.

Despite the benefits from trade, there remains questions as to whether similar strides in development can be made with similar levels of effort; much of the poverty-ending growth in the past 20 years has come from China, which while economically underdeveloped, always had strong state institutions.  China is likely to continue to grow, though perhaps not at the same high rate as the last decade. Other underdeveloped nations (e.g. sub-Saharan African nations) do not have the same strength of institutions. Even unilateral free trade policies by developed countries (ending of all tariffs on imports from underdeveloped nations) might not pull them out of poverty; without stable governments, property rights, and some banking, there is no way for investors to focus on goods than can be exported cheaply to developed nations.  Nonetheless, freeing up trade between developed and developing countries only risks that trade could happen.  If it does, everyone benefits.

Politically, promoting unilateral free trade is not hugely popular, but is nonetheless already done to some extent.  Since the approach is an attempt to help developing countries, it may not be as politically difficult as selling a large comprehensive free trade agreement. Moreover, even if free trade isn’t having a great moment in 2016, there is a strong history of American promotion of free trade. This political fight quite doable.


So while free trade has caused impressive gains in human well-being recently, pulling the next billion people out of poverty might be harder due to a lack of good institutions in underdeveloped countries. But what if instead of trying to solve the problem of bad institutions (hard), you just move people to areas with good institutions? A person in a developing country is not particularly productive and thus might not be able to produce enough value to purchase shelter, food, water, etc.  Moreover, necessities are more expensive to bring to this individual given the lack of institutions (e.g. bad infrastructure means water has to be retrieved rather than delivered, and electricity isn’t usually available). But if you brought that person to the United States, they would suddenly be vastly more productive due to the strong institutional structure they find themselves in. Even if they were given zero access to the welfare state, even working below minimum wage by helping with construction labor or crop harvesting, they would earn far more and be able to spend it on far higher quality goods than in an underdeveloped nation.

The arguments for more liberalized borders are broad and compelling, and can be approached from a myriad of ethical frameworks, or simply by discussing the economic benefits. Immigration happens to be one of the areas identified by the Open Philanthropy Project, but for a real breakdown of arguments for broadly liberalizing immigration, check out openborders.info. The website not only discusses the strong case for the allowing of people to move across borders, but also addresses the best counterarguments to liberalized immigration. Seriously, it’s an incredible site, and if you are at all skeptical of the benefits of immigration (today’s politics indicate many are), definitely see what it has to say.

We’ve already discussed the benefits to immigrants, but we shouldn’t underestimate the benefits from expanding the economy by adding more labor market fluidity. This means an increase in specialization and thus higher productivity, greater selection among products, and higher quality goods. There is also additional consumption by new immigrants who can now afford much more than they could in their home country. In some sense, rather than exporting jobs overseas, immigration allows those jobs to grow or remain in the same market where they are now; this isn’t a nationalism argument, but rather an acknowledgement that immigration destinations are obviously better places than immigration sources, so why not add more jobs in the nicer places?

There are many other non-economic discussions, including the possibility of immigrants voting to change the government in bad ways (political externalities), lack of cultural assimilation, and welfare state funding problems. Some of these have counter arguments that may or may not be convincing (could immigrants really have worse political views than social justice warriors and/or Trump?), and some have obvious ways to mitigate them (don’t let immigrants use welfare programs for a long time). But no matter what, they have to be balanced by the benefits of expanding the free market, and the benefits of reducing global poverty. Politically, this issue is not easy…but even if the pro-immigration tradeoff is only slightly positive, this is an issue with a potentially huge payoff.


The economy is one of the most important issues to voters, but it’s quite a complex issue.  I was going to say it was too complex for anyone to agree on, but then I took a look at Hillary Clinton’s stated economic plan. Surprisingly, I found it pretty acceptable.  I expected to disagree strongly with everything she had to say, but actually most of the reforms make some sense; she advocates tax cuts for the middle class, investing in infrastructure, and easing the regulatory burden on small businesses.  There’s also a mention of a long-term capital gains tax reduction, which could have some long term positive growth impact. Of course, it also has some less exciting things like increasing the minimum wage, subsidizing college even more, and pushing corporations to share more profits. These aren’t horrible ideas (the minimum wage is at least a little negative), but will they promote long term growth? Maybe not.  But is there a fundamental disagreement between me as a moderate libertarian and Hillary Clinton? I think the gap isn’t too far.

Out of curiosity, I looked at Trump’s tax plan, since I figured here, at least, was a man I knew had no understanding of free markets. Interestingly, Trump’s tax plan is quite solid; it simplifies the tax code, reduces the number of brackets, eliminates any taxes for people making less than $25,000, and reduces the corporate tax to 15%. The caveat is that Trump claims it will be paid for by closing loopholes on the rich, which wouldn’t remotely cover the tax cuts. But that’s beside the point, as his plan demonstrates he does have some understanding of what causes economic growth. His views on trade and immigration still demonstrate his understanding is still greatly flawed, but economics isn’t the huge gulf of disagreement we might think it is.

Of course, Gary Johnson has some good radical ideas for the economy as well, including a replacement of income taxes with a consumption tax.  The Open Philanthropy Project also identifies “macroeconomic stabilization policy” as an important area, but while everyone does seem to agree macroeconomics is key and there is some common ground, there’s a lack of empirical evidence for any single policy.  It’s very tough to say what works and what doesn’t, perhaps more so than in the cases of free trade and immigration. That alone makes this issue harder to justify advocacy hours and money; will any policies actually have much effect on the economy, and will these translate to benefits to poverty stricken parts of the globe on the scale of free trade or immigration? On the other hand, the difference between 2 and 3 percent annual growth compounded over 20 years is a 49% increase in the size of the economy versus an 81% increase. For the US economy, that’s something like $5 trillion in annual productivity. That’s a pretty big deal given the interwoven globalized economy; poverty will be much harder to stamp out if the developed world is floundering.

Further complicating considerations, none of this has even covered monetary policy, which is a vital component of macroeconomic policy. Does the Fed know what it’s doing? Can the open market committee do better with prediction markets? How much better, and how hard would implement them? There’s currently very little political interest in doing so, and no major presidential candidates or members of Congress has talked about prediction markets. Overall, despite my interest in economics, it seems like effective advocates should be wary of spending too many resources in this area where there is so much difficulty in measuring results.

The War on Drugs

Apart from development, another method of choosing a policy would be to see how many lives can be directly saved rather than indirectly improved through economics. This is generally the method taken by GiveWell when they recommend various health charities in sub-Saharan Africa. They target health problems that are cheaply solved and under-addressed, and then direct all money into those areas. For example, they calculate that for every $500 donated to the Against Malaria Foundation, the bednets which are distributed from those funds save, on average, a single person from dying of malaria. That’s huge. It also turns out that there is a policy which the U.S. government engages in which causes the deaths of thousands of people a year: The War on Drugs.  The Open Philanthropy Project discusses this more generally under criminal justice reform.

It’s hard to say exactly what the toll has been in the War on Drugs. Even in just the most recent wave of drug violence in Mexico (roughly the last decade), estimates range from above 60,000 to 130,000, not counting those who are “missing”. Of course, some of these deaths would have occurred even if all drugs were sold legally in the United States, but many of the worst criminal contingents can only be funded due to the illegal nature of the drug trade; less violent groups (like transportation businesses) won’t move drugs because they have too much legitimate business to lose, while criminal groups, whose existences are already illegal, lose nothing additional by transporting highly valuable illegal narcotics. Thus, American drug policy directly funds the most violent criminal groups in Mexico.

Of course, the damage is not limited to a single country; the Colombian drug trade is also famously brutal. It also helps to fund many of the paramilitary groups which have been locked in continuous combat with the Colombian government for decades.  The drug trade also has major presence in Bolivia, Peru, Brazil, and other countries around the world. Most significantly to voters, the United States has also seen its share of drug violence, though it’s on the decline.  Nevertheless, FBI statistics indicate a few thousand homicides a year linked to drug violence, not to mention many thousands of inmates found guilty on drug charges, or involved in violence due to the profitability of the drug trade.

Many of these problems could be addressed by making the selling of drugs legal in the United States; criminal groups’ monopoly on drug transportation and sale would be broken, and drug cartels would be at a significant disadvantage profiting on narcotics. Indeed, the cartels would have real financial incentives to legitimize their activities (e.g. using court arbitration rather than murder). The benefits of simply making these activities not criminal cannot be understated. Criminal justice reform is key, but without cutting funding for criminal organizations, much of the bad policy effects will remain.

There is a big problem though; legalizing the sale of hard drugs is a radical notion and a non-starter for most voters. There’s also the real problem that an increased availability of these drugs will almost certainly lead to an increase in overdoses. Will the decrease in criminal behavior (and its associated poor outcomes) be outweighed by the increase in overdoses and drug abuse? I’d guess so. Today, without any of these drugs being made legal, overdoses are a big problem in the United States, mostly due to prescription drugs. I’d hypothesize that if someone wants to get their hands on dangerous drugs, they can already do it one way or another.  Nonetheless, most voters won’t see it that way, and the chances of legalizing most hard drugs seems even less likely than open borders.


Drugs aren’t the only way the government helps in the death of lots of people, sometimes the government starts conflict directly. The 2003 Iraq War, started preemptively by the Bush administration, ended with the deaths of about 60,000 combatants: coalition forces, contractors, Iraqi security forces, insurgents, and Iraq’s army under Saddam, with many thousands more wounded. But that doesn’t include civilian death estimates which range from at least 113,000, to 600,000.  This dwarfs anything from the drug wars.  Given the rise of the Islamic State in the unstable Iraqi power vacuum after the U.S. left, a significant fraction of the blame for the Islamic State’s existence falls to the Iraq War as well. The War in Afghanistan, while not a preemptive war, has continued long past the removal of the Taliban from power, and indeed continues today (in a limited capacity) despite Osama Bin Laden’s death 5 years ago.

This is tricky though; one could make a good case that Obama made the anti-war decision to not get involved in Libya and Syria, yet the results are mixed. Libya is an unstable state with an ongoing second civil war and casualties in the thousands. Syria is one of the most violent places on the planet today, with the ongoing, multi-sided civil war having resulted in hundreds of thousands of deaths and the displacement of millions of people both within and outside Syria. Certainly if there was a worst case scenario, it is what we have now, suggesting perhaps a stronger intervention could have resulted in, at worst, less instability.

But at the very least, avoiding broad conflicts that are started preemptively by the U.S. government seems to make obvious pragmatic sense, both though the lenses of utilitarianism and politics. Fortunately, the opportunity to avoid these conflicts does not occur often; a mistake as big as the 2003 Iraq War was unusually massive.  Depending on how you count military interventions and what is comparable to Iraq, a war caused in part by U.S. military involvement had not been witnessed on this scale since, the Iran-Iraq War (where Saddam was largely backed by the United States indirectly), the Lebanese Civil War (where a western peacekeeping mission failed miserably), or the Vietnam War (largely escalated under the Gulf of Tonkin Resolution).  Whichever is most comparable, it’s clear that not every election will have an Iraq War level impact on international stability and conflict. This is good in that huge conflicts with hundreds of thousands of casualties are not continually happening, but it means our method for selecting and campaigning for policies might better be served in other capacities most of the time. But in those highly vulnerable times, it’s vital that as much effort as possible is put behind stopping preemptive or escalating military interventions.

Catastrophic Risk

Another way of looking at policy is the minimization of risk of a calamitous event in the future, rather than implementing policies that save lives right now. This is part of the approach of the Open Philanthropy Project entitled “Global Catastrophic Risk”. The first and perhaps most obvious risk is a consequence of the increased globalization of the modern world: pandemics. Whether the origin is viral or a multi-drug resistant bacteria, the connected web of modern trade and business means that a single super disease could spread rapidly and cause immense harm. I don’t know what “optimal” funding would be for the CDC, but it currently gets about $7 billion a year, which is less than the TSA.  This seems concerning.

The other catastrophic risk the Open Philanthropy Project mentoins is the risk of AI. The problem, as discussed by Scott Alexander, is as follows:

1. If humanity doesn’t blow itself up, eventually we will create human-level AI.

2. If humanity creates human-level AI, technological progress will continue and eventually reach far-above-human-level AI

3. If far-above-human-level AI comes into existence, eventually it will so overpower humanity that our existence will depend on its goals being aligned with ours

4. It is possible to do useful research now which will improve our chances of getting the AI goal alignment problem right

5. Given that we can start research now we probably should, since leaving it until there is a clear and present need for it is unwise

Statements (1) and (2) make a lot of sense to me. I can see (3) happening as well, but it’s hard to know how easily a smart intelligence will have real power without a physical presence. But eventually? Sure. And once you’ve acknowledged (1), (2), and (3), if there’s even a risk that (4) is true, the possible benefits of it make (5) a certainty. Yes, AI risk sounds crazy to worry about, but AI could more effectively end humanity than any super virus, and we can probably do something about it right now.  But getting government funding for it? That seems like a tough sell at the moment, but with more and more famous researchers discussing it, I think small amounts of funding are possible.

The one area I didn’t see in the Open Philanthropy Project is climate change, which is very confusing. Perhaps they think enough people are already talking about it? I’m not sure. But clearly a small carbon tax (and accompanying reduction in other taxes) would be a nice risk mitigation policy, even if you’re not totally convinced climate change is going to be a big problem.  Politically, a carbon tax is challenging, but not insurmountable. At the very least, it’s worth discussing openly, especially in relation to reducing other taxes.


There are other areas where I haven’t had time to touch on, especially anything that could be considered a second order policy consideration: rule of law, free speech, electoral reform, campaign finance, etc. These are elements that are necessary for a functional dialogue on viable policies, but accomplishing them doesn’t directly improve lives in the same material ways that previously discussed policies have. They are worth addressing, and can’t be dismissed, but it is hard to know exactly the optimal weighting that should be done between effective direct policy and indirect institutional reforms. Likely future blog posts will discuss some of these ideas.

There are, however, several areas which are direct policy areas that I specifically didn’t discuss because they are really not important even though people think they are. Areas that are often ranked highly in importance by Americans but didn’t make this list include terrorism, income inequality, and even race relations. Terrorism just isn’t the major risk that people believe it to be, income inequality is bad insofar as it effects economic growth (which we have addressed), and race relations, while concerning and with some possible indirect effects, just don’t compare in importance to global poverty or horrific international conflicts. Healthcare policy is also often ranked highly, and the argument can certainly be made for its importance, but ultimately the returns on domestic health improvement are dwarfed by the possible returns in international health outcomes and economic development. By all means, let’s fix the American health system, but quick fixes and obvious choices are few and far between.

Unsurprisingly, politicians who favor libertarian policies tend to do well in this analysis. Republicans and Democrats tend to be more mixed. Donald Trump specifically has a lot of the worst of everything. He’s horrific on the free trade and immigration discussion, apparently changed his mind for the worse on drugs, and I am completely unable to trust his judgment on markets.  At least his alleged foreign policy is less hawkish than Clinton, but he’s such a wild cannon, and who knows what sort of conflicts he’ll bring the U.S. into.

Ultimately, this analytic framework is valuable even to people who disagree on the impacts of different policies; many will figure out ways to frame their most valued policy area as having more importance than those discussed here, likely myself included.  Nonetheless, forcing people to ask exactly how their policy will solve the biggest problems and have the highest probability of success would be enough to radically change political discourse. So next time your Twitter or Facebook are awash in the latest culture war dust-up, feel free to share some outrage with your side, but also frame it in the context of the real impact compared to other active policy areas which are often under-discussed.