The politics of the second-best

by Chris Bertram on February 12, 2023

Harry mentioned the politics of the second-best in comments to his post on higher education the other day. I guess it falls into something of the same space as non-ideal theory or realism, as opposed to moralism. The basic idea is that we shouldn’t hold out for purity if doing so gets in the way of making the lives of many people, some of them with urgent needs, better. And that makes a lot of sense. Pursuing the ideal policy, refusing to compromise, only allowing for perfect justice can seem like a form of self-indulgence that has real costs for those who can least afford to bear them. We always have to start from where we are, with the resources that we have and making progress can involve messy compromises with people that we don’t much like in order to do the good that we can.

Moreover, the rejection of the second-best in favour of the pursuit of the ideal can have rather dire consequences. Take Lexit (the left-wing case for the UK to leave the European Union) for example. Lexit is premised on the notion that EU membership, by limiting UK sovereignty, would make it harder for a genuine socialist government to implement its economic programme, what, with the EU being “neoliberal” and all that. Some Lexiters, not all, were also in favour of no borders or open borders, and rightly made the argument that the free movement embodied in the European single market was a privilege for (mainly white) European citizens, and one that locked out people from beyond Europe’s borders. EU free movement was racist. But the effect of Brexit, supported on such grounds by Lexiters, has not been to advance the socialist programme but rather to reduce the protections enjoyed by British workers and protections in other areas such as the environment. Brexit has made the UK labour market a little more open to non-Europeans, albeit subject to restrictive visas that make exploitation more likely, but it has not made life easier for people trying to escape conflict and persecution and British refugee policy has become even more punitive with a plan to deport refugees to Rwanda and a threat to abandon the European Convention on Human Rights if it proves an obstacle to state cruelty. The second-best policy of staying in the EU looks a whole lot better from here than the allegedly first-best policy of enabling a sovereign socialist democracy in one country.

But when it comes to particular policy areas, I still have reservations, and none more so than in the general area of migration where my own first-best preferences are for almost open borders (see my book for the caveats). The difficulty with second-best policy in this area is that the compromises it requires mean going along with some pretty nasty state treatment of individuals. This is because any second-best policy involves limits on numbers, selectivity, denying people visas, perhaps breaking up families and certainly other ties, detaining some people, deporting others, co-ercing people onto deportation flights, and so forth, (a non-exhaustive list). A second-best policy involves recognizing the “right” of nation states to do these things to people, albeit fewer people that have these things done to them at present. Because of this, second-best policy will always generate cases where the person subject to state coercion can ask of us whether we stand in solidarity with them or with the state that is subjecting them to be “repressed, fined, vilified, harassed, hunted down, abused, clubbed, disarmed, bound, choked, imprisoned, judged, condemned, shot, deported” (to partially quote from Proudhon). My guess is that most people with a strong sense of justice, even if they recognize the case for having a second-best policy, will instinctively side with the victim and not the victimizer. Such, at any rate, is my own case.



Brett 02.12.23 at 6:05 pm

I think if you’re going to embrace the second-best, you always have to be explicit with yourself and others that it’s a temporary measure, and you’re going to keep trying for the full thing.


Jake Gibson 02.12.23 at 6:32 pm

“Don’t let the perfect be the enemy of the good.”
Sometimes in reality, it is more like not letting the good be the enemy of the incrementally better.


CP Norris 02.12.23 at 9:30 pm

If your country is planning to accept 10,000 refugees this year and your lobbying and protesting gets them to change it to 20,000, that seems like a good thing. But it is little consolation to the 20,001st person in the queue.

Does that invalidate the good of the change? Perhaps it does, if the small change this year seems to excuse future brutality. It is very hard to draw causal relationships between political outcomes.

My instinct is that if you can save 10,000 more refugees this year, and the program works well, you can also lay the groundwork for saving more next year. But I can’t prove that.


Matt 02.12.23 at 11:26 pm

Harry mentioned the politics of the second-best in comments to his post on higher education the other day. I guess it falls into something of the same space as non-ideal theory or realism, as opposed to moralism. The basic idea is that we shouldn’t hold out for purity if doing so gets in the way of making the lives of many people, some of them with urgent needs, better.

There are lots of different ways of thinking about the idea of “2nd best”, of course, and I don’t want to claim that one is the “real” or “best” way, but there’s an important way of thinking about it, derived from a result in economics by Lancaster and Lipsey, that suggestions something deep and important, and worth keeping in mind. Lancaster and Lipsey’s result showed that even if “perfectly competative” markets were perfectly efficient, this was (almost always) irrelevant. The reason was that, if those conditions couldn’t be acheived, you cannot assume that approximations to them would get you closer to the best outcome. So, there’s no reason to think that minimizing “distortions” would bring us closer to efficiency. Rather, there are regularly cases where introducing further departures from “perfect competition” will bring us closer to efficiency.

This seems to generalize in lots of areas. We might think that the first-best world is one where there are no distinctions made on the basis of race, sex, etc. in hiring, promotion, and so on. But, given the reality of racism, sexism, etc., it can be the case that trying to approximate that will actually take us further away from equality than would using affirmative action policies. Of course, one of the morals of the “principel of the 2nd best” is that you can’t know this stuff a priori – you have to work it out emperically, and to do that well requires being open to trying different policies and revising them. (There’s also a lot of risk of being caught in a “local maximum” – one reason why recent attacks on ideal theory seem misguided to me – but that’s getting into another big set of issues – too much for a blog comment.)

To my mind Joseph Heath is one of the best philosophers taking this stuff seriously and making use of it, so I’d recommend his work to those who are interested. (More people should be!)


Ray Vinmad 02.13.23 at 6:17 am

It’s a little hard for me to figure about what counts as ‘second best’ in many different policy choices. Is it second best of what is possible, second best of what is optimal…and whose second best will count when there is conflict? We will fight over what is second best perhaps as readily as what is best. Thinking in terms of health care policy I’d wonder if ‘second best’ was the one that would preserve profits for healthcare companies even if that led to something that was below second best for patients and the public.

But of course, when it comes to most policies, I would be happy if we got second best since optimal or first best is so out of reach.

I also wonder when the ‘accept second best’ kicks in. It seems like it doesn’t really kick in until there is a disputed political option on the table. So perhaps this means we should continue to argue for what is best and not worry so much about second best until there are some options available. Again, it always feels like is presented as second best in politics (thought nobody quite presents it that way) is such a far, far cry from what is best.

I wonder if the rejection of second best by the left is often a reaction to the scarcity of honesty and the distance of the left from how the decisions are being made. People howl about all the compromises perhaps because they are so far from being able to influence the compromises and perhaps because the compromises are often presented in a deceptive-seeming manner as ‘we are absolutely boxed in, there is no choice at all, be happy for what you can get….’ Sometimes this is true, sometimes it is because people are unwilling to fight for something better because their loyalties are engaged elsewhere and they honestly don’t care that much. Often it’s hard to tell the difference! It’s easy to foment discontent later because a story can always be told about what might have been.

I think the caution of ‘what kind of nasty things does this allow’ is the right sort of caution. Some things can’t be compromised over because they do involve mistreatment and severe harms. So there’s clearly a distinction in the question ‘what kind of world do we want’ between ‘everyone’s desires being met’ and ‘everyone’s basic needs met’ and ‘letting some people suffer great indignities….’ I.e., the distinction between negative and positive rights for lack of better terms. Even if we don’t get to full equality we cannot and should not accept degradation and humiliation for some.


Frank Wilhoit 02.13.23 at 1:35 pm

THIS. Nothing is self-explanatory. There is no such thing as too much context. Shout it from the rooftops. Politics is the art of the possible only because it is the art of the necessary: without clear markers of what is necessary, it does not matter what is possible.


Thomas P 02.13.23 at 6:46 pm

There is also the trap where you are offered to choose between worse and worse choices every time, expected to pick the lesser evil. You need at least the threat of saying, “I don’t care what is the lesser evil, none of these choices are good enough for me to support!”


Mike Huben 02.14.23 at 2:05 pm

Theory of the second best is a two edged sword. I think the best use is denying that first best solutions are possible (almost always) or should be approached, which is very handy for refuting foolish Economics 101 arguments.

And as some others have pointed out, no unique second best solution can be found without experimentation, which allows greatly conflicting claims. I’m sure slavery can both be defended and opposed as a second best solution, for example.


Sashas 02.14.23 at 7:44 pm

I think it’s a mistake to use “politics of the second best” as a label. The granularity is too large, and covers several wildly different kinds of logic. Here’s my attempt to break it down a little more finely and some initial thoughts on the ethical implications of each:

Lesser of Two Evils arguments are generally (in my experience) used to refer to situations where I am being asked to make a choice, but someone else has constrained my options. (e.g. to two options both of which are Evil). This came up in the comments section and as it relates to “second best” politics it is to note that one way to address being in a Lesser of Two Evils situation is to attempt to refuse the choice entirely. Since I am generally assumed to not have the power to actually steer us away from one of the two choices given, we’re basically asking whether I should exert a little influence toward the lesser evil, or whether I should exert no influence while stating that both are evil.
Examples in comments: Ray Vinmad (5)?, Thomas P (7)

Strategic Voting arguments (sorry, I don’t have a better name for this one) are slightly different in that there is some hope of exerting influence toward the primary desired outcome. At that same time, it remains either risky or difficult to exert that influence effectively. Someone might make an honest determination that a high probability of success with lower reward is simply a better deal than a low probability of success for a high reward outcome. This is made thornier when, as in the OP’s open borders example, the lower reward outcome comes with morally wrong elements. I view this as a pretty straightforward type of optimization/balancing, and I think any attempt to set out a general rule for whether one should vote strategically is basically doomed from the start. We can talk about how to actually measure political risk and the potential for collective action, etc, but if we want to put together a “rule” about this type of second-best politics that’s what we’d have to do.
Examples in comments: Jake Gibson (2), CP Norris (3), Mike Huben (8)?

Harm Reduction arguments are a type of qualitative second-best politics. Rather than addressing a “problem” head-on, we note that in the absence of the desired first-best policy there is something that we should do to mitigate the harm caused by the lack of that policy. The example of this thinking from which I drew the name is the movement for harm reduction w/r/t drug abuse. There are live debates around tools such as needle exchanges. The needle exchange does not directly address the problem of drug abuse, but it does mitigate some of the harm that results from the problem being unsolved. I think there’s a very strong case for harm reduction, as long as efforts toward it do not seek to block the advancement of a first-best type solution. At the same time, I have seen proponents of first-best solutions attack harm reduction generally on the grounds that it “reduces demand” for a first-best type solution. I’m not convinced.


Moz in Oz 02.14.23 at 9:46 pm

Thomas P: There is also the trap where you are offered to choose between worse and worse choices every time, expected to pick the lesser evil.

This has been my experience. You compromise and compromise until you’re so thoroughly compromised that you have no idea what you were even trying to achieve in the first place. I see this often with anti-climate-change people, or environmentalists more generally, who travel a path from protest to greenwashing as they “progress their career”.

This is generally the case when the “right position” is completely opposed to common practice, as with environmentalism vs capitalism. It’s very easy to get caught up in “capitalism is going to destroy/monetise everything, the goal of environmentalists is to delay that process and mitigate the worst effects slightly where possible”. As brutally demonstrated by the proposed use of “ecological offsets” to replace Great Nicobar Island in India… what price making extinct a keystone species that’s also charismatic megafauna? Apparently a few hectares of monoculture tree farm. That’s what some environmentalists might be tempted to call a win, and it’s definitely second best.

Comments on this entry are closed.