Economists think about what things cost, and philosophers think about what things are worth. It might sound like those are two completely different tasks, but actually they’re not. The cost of spending an hour, or a dollar, on one worthwhile thing is the forgone opportunity to spend that hour or dollar on any other worthwhile thing. Therefore, caring about worthwhile things means thinking about what things cost.

Ironically, philosophers are not usually taught to think like that—in fact, sometimes they’re taught not to think like that. For example, imagine philosophers choosing between two policies to reduce automobile emissions, one more ambitious and one less ambitious. We usually ask, even if the more ambitious policy is more expensive, isn’t cleaner air worth paying for? Or, even if the more ambitious policy is more expensive, isn’t the point to put that burden where it belongs (on makers and buyers of cars)? Or, isn’t it part of civic virtue not just to try hard to reduce emissions, but to try as hard as we can? And so on.

Now, those are good questions, but meanwhile the economists are asking: if the more ambitious policy would make new cars a lot more expensive, wouldn’t a lot more people drive their old cars longer, creating more pollution in the end than if we had adopted the less ambitious policy? If so, then the more ambitious policy might show that our hearts were in the right place, but at the cost of more pollution than there had to be.

In the past year, I have come to appreciate how much I still need to learn about asking that second sort of question. We have to think about what really matters but we also have to think about what really works. If we’re not doing the latter, it’s hard to take us seriously when we say we’re doing the former. It’s not just love and marriage that go together like a horse and carriage. Justice and feasibility go together that way too, though Frank Sinatra never sang about it. But Adam Smith had a lot to say about it. For the founder of modern economics, the point was to understand what pulleys and levers actually have the effect of turning poverty into prosperity.

Unfortunately, this marriage has been on the rocks for some time. I am often asked to speak about ‘the good society’. I won’t say that the good society is a just society—that’s too obvious. What I like to say, is that there can be no justice without feasibility. I don’t mean merely that after we figure out which ends are just we then have to look for effective means. I mean that there is no such thing as a just end that stands apart from what we can feasibly do.

Think of it this way. There are two ways of approaching any sort of planning: we determine our priorities and choose which ends to pursue, then we rank the available means in order of feasibility. Or alternatively, we rank the available means in order of feasibility, then we prioritize among the ends we can pursue by those means.

Coaches know to take the second approach instead of the first. You don’t choose your favourite play and then try to figure out how the team you actually have is going to run it. You look at what your team can actually do, and choose the best play in their playbook. Neither approach guarantees victory, but the second does offer the chance to play your best game.

Good coaches put feasibility first. Unfortunately, policy makers—and the public at large—often do not: we frame policy by declaring what matters to us, and we leave it to the pulley-and-lever folks to make it all work. When we look at it that way, it’s hardly surprising that so many public policies should turn out to be counterproductive.

So, why is it so difficult to balance what matters with what works?

Barrier 1 : Know a hard case when you see one

One major barrier is the surprising difficulty of knowing a difficult case when we see one.

For example, when asked whether a new law should require more ambitious or less ambitious reductions in automobile emissions, it looks like an easy question: do we want a bigger improvement or a smaller one? The problem, though, is that we weren’t given a choice between two outcomes. We were given a choice between two policies, and just knowing we want the ‘greener’ outcome does not tell us which policy to choose. The road to unintended consequences is paved with easy answers to hard questions.

Psychologist Daniel Kahneman says that it’s actually very common for people to respond to a hard question by unwittingly replacing it with an easier question and answering that one instead. Economist Bryan Caplan suggests that this is exactly what happens in a lot of everyday thinking about public policy.

The questions we are asked are whether to have tougher pollution standards, or a higher minimum wage, or tighter rent controls, or bans on price-gouging. And yet the questions we hear are whether to save the environment, or to help low-wage workers, or to increase affordable housing, or to protect the desperate. The questions we hear are no-brainers, but they are not the questions we were asked. Of course we want to save the environment, but intentions aren’t results, and it’s hard to know what will actually have that result and therefore, what course of action to take.

To generate change, we need to develop the trained and informed habit of thinking about policy in terms of the levers and pulleys. That might sound obvious, but as political scientist Steven E. Rhoads points out, policy-makers are more likely to be trained in law than in economics, and so—like we philosophers—they tend to think of policy as shaping society mainly through the assignment of rights and duties, rather than in terms of incentives, coordination, and planning.

Barrier 2: Ends don’t always justify means, but often means do justify ends

We know that ends don’t always justify means, but we should also understand how means can justify ends. Suppose that we have a chunk of money to spend on reducing the death rate among young people. The obvious thing to do is to identify the number one cause of those deaths and spend most of the money on that. Paradoxically, though, that might actually save fewer lives. For instance, perhaps we don’t know very much about how to prevent the number one cause of early deaths, while a recent medical discovery has provided a low-cost cure of a disease that is only the number six cause.

The moral of this story is that we are more likely to succeed if we look first at what is really feasible in the world as we find it, and then set our priorities. In other words, ends do not always justify means: reducing the number one cause of early deaths is a noble end, but it may not justify the means of diverting resources away from preventing a lot more early deaths from other, more feasibly treated causes. As Rhoads says about this case, “We can know how high to set any one objective only if we know what we give up in progress towards other objectives.” Because ends don’t always justify means, sometimes means have to justify ends. Precisely because some things are precious, priorities cannot come first. Feasibility has to come first.

We are more likely to succeed if we look first at what is really feasible in the world as we find it, and then set our priorities.

Even so, doing things the other way around is still the norm in debate about public policy. For example, Martha Nussbaum, one of the best known public intellectuals on the planet, has since the 1980s been advocating a political structure that would involve world-wide, needs-based governmental distribution of resources, redistribution of income and property, environmental regulation, trade regulation, and labour standards, to name just a few highlights. These policies call for global political planning and, if you like, choreography at a level of ambition and subtlety that the world has never imagined, much less seen.

Now, Nussbaum is a philosopher, and she herself is careful to point out that her work has mainly taken the form of “announcing some ambitious goals,” in her phrase, and that a great deal more remains to be said about precisely how the approach can be used to generate political principles for today’s world. To some extent, this job is a practical job, a job for economists, political scientists, diplomats, and policy-makers. The message is as clear as it is simple: the big-idea brigade state the objectives and then the pulley-and-lever brigade make them happen. The rationale for that division of labour is clear—philosophers do just specialize in ends, and economists specialize in means. But what is the rationale for that order of labour—priorities first, feasibility second? When we leave questions of means for the bright sparks to work out after we’ve set our ends, we ignore the possibility that what the bright sparks have already worked out is that those sorts of ends just aren’t possible. Of course, Nussbaum herself says that her approach would be self-defeating if it ended up making people worse off. Exactly right, and that is precisely why questions about feasibility have to come first. Then, and only then, do big ideas have any point.

What happens in the world of public policy when feasibility is someone else’s problem? Usually, the result is exactly the opposite of what was intended.

For example, the No Child Left Behind Act of 2001 requires all students to perform at a certain threshold on standardized tests in subjects like maths and English, and any failures mean federal funds to the lagging school can be cut. The result is that teachers now have to shift more time to focusing on tests, and away from actually teaching maths and English.

My point is not that we shouldn’t have ambitious goals, but ‘ambition’ shouldn’t be a euphemism for making up policies and leaving feasibility to sort itself out.

Here’s another example. On this side of the US housing crisis, it is staggering to think that in 2002 President George W. Bush could speak into existence the goal of creating millions of new homeowners, and simply say, “I set an ambitious goal. Achieving our goal is going to require some good policies out of Washington.” Now, reaching that goal would require lowering the price of credit, for one thing, but by 2002 the Federal Reserve Bank had already been cutting interest rates to historic lows, eventually resulting in skyrocketing prices and then skyrocketing foreclosures. So the necessary means to Bush’s goal could have only hurt the very people he meant to help, but putting feasibility second is to decide in advance not even to hear that kind of warning, much less heed it.

My point is not that we shouldn’t have ambitious goals, but ‘ambition’ shouldn’t be a euphemism for making up policies and leaving feasibility to sort itself out. And the tell-tale sign of ‘feasibility second’ ambition is that acting on it usually would undermine the very things that make prosperity possible in the first place. What are those things?

Well, think about the sort of prosperity that’s represented by something as simple as an ordinary pencil. As economist Leonard Read observed in a classic essay, even making something as simple as a pencil requires the coordinated activities of a whole planet full of people.

For instance, it takes people who know how to fell trees, people who know how to transport and store them, how to cut them into logs, how to cut the logs into slats, and how to cut, cure, and stain the slats to make the half-shafts of the pencil. And it’s a similar story for the lead on the inside and the lacquer on the outside, the rubber on the end and the ferrule to hold it on, and that’s not to mention all the people in other industries that make possible industries like logging, transportation, and all the others that contribute something to every pencil. Now, even people who are evolutionists about biology tend to be creationists about economics, but let’s be clear: the knowledge of how to make a pencil resides with no one. The late economist and philosopher Friedrich Hayek got it right: knowledge like that resides only in a whole planet-ful of people coordinating with one another.

And that’s just pencils. Take a look at everything else we have. Apparently, prosperity happens not because of ambitious central planners but if anything in spite of them. What the world’s pencil-makers really need is an institutional structure that makes productivity both possible and attractive. They need secure ownership, so that they can both control access to scarce resources and know that they will reap what they sow. They need prices that actually tell them what consumers are prepared to buy and what suppliers can feasibly provide. And they need laws that are general and rule that is predictable, so that they can know how to plan their activities. When you spot an ambitious policy that would undermine one or more of these things, you can be sure that the word ‘ambitious’ in that case is nothing more than a euphemism.

Barrier 3: Work the problem, not the concept

Justice isn’t the name of an abstract concept; it’s the name of a solution to problems. Some will say that justice means “share and share alike,” others that it means “to each his own.” But justice isn’t identical to either of those slogans. Justice is what solves the problems we actually have, in peaceful and prosperous ways.

For example, Harvard philosopher Michael Sandel has been leading a charge against confusing justice with faith in the concept of market efficiency. However, Sandel’s alternative is to identify justice with a different concept: a shared, communal concept that should emerge from public debate, case by case, about the proper meaning of goods like “health, education, family life, nature, art, civic duties, and so on.” This, Sandel says, is the “great missing debate in contemporary politics.”

What happens when that great debate reveals that people just don’t agree—for instance, about what should be for sale on the market? Doesn’t the very fact that there is anything here for us to debate in the first place suggest that disagreement between reasonable people is a defining feature of the world we actually occupy? If so, then if justice is anything at all it must be a way of living together in peace in spite of disagreeing about how else to live. Identifying justice with a concept, any concept, therefore misses the whole point.

That way of thinking about justice also puts institutions like markets in a very different light. Sandel decries “the corrosive tendency of markets,” because markets allow people to engage in all sorts of transactions that may violate the meanings that you and I might attach to the things being bought and sold. But look at it this way. Michael Sandel’s book, What Money Can’t Buy, is one of the things that money can buy. Its publisher sells it along with even more titles in areas like marketing, finance and business. This for-profit venture presumably has no intrinsic desire to publish books on the corrupting power of profits. But the good news for Sandel is that in market society, the publisher can have a reason to publish that book even without an intrinsic desire for it. Likewise, the good news for a reader like me is that the only thing standing between me and a copy of that book is the price. I don’t also have to convince my neighbours that I should be allowed to buy it, any more than Sandel had to convince his neighbours that he should be allowed to sell it.

Thinking of justice as a concept is also perhaps the chief reason that feasibility gets pushed out of discussions of justice. Nussbaum identifies justice with a certain concept about allocation, and leaves questions of feasibility for later. Not to say that Nussbaum is wrong about what’s just; my point is that I don’t even know what it would mean to say that something is just in advance of asking whether it would leave us all destitute. If justice is about anything, it is about solving our problem of finding a way to prosper together in peace in the world as we find it.

A good society is a just society, but justice is an empty word without feasibility. Feasibility is about means to ends, but much more than that, it’s also about recognizing that our means are what set the boundaries within which ends can merit our consideration. That means feasibility has to come first. For one thing, we have to understand what’s feasible before we can know a hard problem when we see one. For another, we have to understand which means are feasible before we can justify any ends. And lastly, we have to understand that justice is about feasible solutions in order to be talking about justice in the first place. Feasibility is not the handmaiden of ambition. Feasibility is what determines whether ambition has any point at all.

If we mean well, we might set our priorities and then try to make our first choice feasible. When we actually do well, though, it’s because we first understand what’s feasible and then set our priorities. A good society therefore depends on more than knowing what works and knowing what matters. It depends on knowing those things in that order.

This is an edited and condensed version of Prof. Russell’s lecture at Cranlana 2012 in Australia, which he kindly shared with us.