Ezra Klein offers a response to my previous post, and there’s a lot that’s good it in. I appreciate Klein’s characteristic effort to provide a nuanced and balanced take. Nevertheless, I can’t say that I am persuaded.

Before going into the substance of Klein’s piece, I want to clarify that my prior post was not a response to the controversy that has arisen following a letter from four former CEA chairs about Gerald Friedman’s projections of the effects of Sanders’ proposals. My piece was written last Monday, the CEA letter was published Wednesday, I promise you I have no moles on Mount Olympus. I do have an opinion about the “kerfuffle” provoked by that letter, and maybe I’ll express it in a future post. (Or maybe not.) But my post on theories of politics was not addressed to that controversy.

Anyway. Klein and I agree, I think, that “in a democratic polity, wonks are the help”. Elections are where voters set the interests and values in service of which wonks’ technical expertise will later be deployed. But Klein is quite correct to point out that we don’t elect disembodied interests and values, we elect people, and the competence of those people along myriad dimensions will determine whether they are capable of translating the interests and values they represent into meaningful social outcomes. I agree with Klein that voters are called upon to evaluate not candidates’ competence, but their competences, to trade off weaknesses against strengths, and sometimes to trade off evaluations of competence against their preferences with respect to interests and values.

As Klein very aptly puts it, “debating the details of campaign proposals is, on some level, fantasy football for wonks.” But, he argues

Watching a candidate run his campaign’s policy processes is one of our best ways of predicting how he would run his White House.

The key word there, by the way, is run. Some of the most important decisions the president makes are about how to run the processes that translate vision into policy. Those decisions include whom to hire, which advisers to listen to, which ideas make sense, which strategies are likely to work. The presidency is one damn decision like that after another. Obama, famously, is so exhausted by the decision fatigue of the job that he wears the same color suit every day so he has one less thing to decide in the morning.

This is one way in which campaigns give us insight into presidencies. Presidential candidates also have to decide whom to hire, which advisers to listen to, which ideas are truly good ones, which strategies are likely to work. To make those decisions well, they need a sound philosophy, yes, but they also need to want to hear good advice, they need to want advisers who will tell them when they’re wrong, they need to have good instincts for when something they want to believe is true simply isn’t, and they need to be realistic about the strategies that are likely to work and the ones that aren’t.

A White House has to be run, for sure, and Klein is eloquent and correct on the work and care that entails. But I don’t think a campaign’s policy processes tells us anything much about that. Presidential campaigns are not presidencies. The goals, incentives, and constraints are entirely different. The “policy process” of a campaign begins first and foremost with the work of a campaign, which is to signal the interests and values of the candidate. Policy details are, for the most part, elaborated reactively, as competing candidates try to work out inconsistencies between opponents’ broad policy visions and the interests and values of the electorate they are vying to win over. Policy details are also elaborated to signal competence and to concretize interests. Detailed white papers are more than most voters are willing to work through, but their existence, especially if certified by trusted experts, may persuade voters of a candidate’s competence. One way to appeal to voters’ interests is to be able to make claims like “the typical American family will save $5200 per year if my policy is enacted”, but making that sort of claim without some pretense at a projection opens candidates to attacks on character and competence. Candidates face tradeoffs between the benefits of making speculative or ambitious claims and potential costs in perceived credibility. They also face tradeoffs between perceived benefits of making promises and the constraints those promises may impose on their future choices or credibility should they be elected. These are the factors shape the policy details they come to offer.

Certainly presidents have to sell policies to legislators, so you might argue that the sales job that is an election may not be not entirely alien to the process of governing. Mostly that’s wrong. What gets legislated is a function of constraints imposed by the legislature, and the interests that shape those constraints are very different from those of the open electorate. President Obama signaled in a State of the Union address his strong support for Universal Pre-K, as Hillary Clinton or Bernie Sanders probably both would in the context of this election. Universal Pre-K is good policy on the merits, and is popular with the public, but President Obama has made no effort whatsoever to transform his call into legislation. The skill that a President requires to get policy enacted is the ability, somehow, to get the constraints of the legislature and her own policy agenda aligned. That can be done Lyndon-Johnson-style, by backroom armtwisting; it can be done per Coase’s Theorem, with various kinds of side payments; it can be done as the prior Clinton administration did with welfare reform, by letting the administration’s agenda shift to become more hospitable to the constraints of the legislature; it can be done by creating outside threats, in terms of adverse public opinion and potential electoral risk, as Bernie Sanders proposes to do with his “political revolution”.

One certainly can draw connections between skills on display during an election and skills that will eventually be required to get policy enacted. It is not unreasonable to suggest that Hillary Clinton’s locking up of endorsements by political insiders at all levels might indicate real skill in Johnsonian arm-twisting and Coasean bargaining that will help her “get things done”. It is also not unreasonable to argue that Bernie Sanders’ remarkable success at organizing a grassroots donor and volunteer base and his clear intent (unlike President Obama) to ask those grassroots to engage publicly in conflict with political opponents, suggests a competence at creating electoral risk. Voters can and should consider these competing competencies, along with their adjudications of values and interests. But these competencies have little to do with the quality of the campaigns’ policy white papers. For the most part, wonk-quality is not a useful differentiator of presidential candidates.

Let’s look at an example. As Klein notes

There are plenty of criticisms to be made of Obama’s presidency, but I think the baseline competence of his administration has begun to dim memories of how important presidential management really is.

My own view of the Obama administration is mixed, but I am happy to second Klein here. I think the administration takes its policy process seriously, and generally does fine work along dimensions of technical competence. That doesn’t mean I endorse all of its work, or that the administration’s technical work is apolitical. It is not. All policy is ideological. Technical work carries ideology with it in how things are framed, what is assumed, and what is considered. Still, within the boundaries the Obama administration has chosen to define, its technical work is high quality.

But as Klein reminds us, eight years ago, on the core wonk controversy of 2008, Barack Obama took a position of astonishing technical incompetence. You can accuse Bernie Sanders’ campaign of touting optimistically shaded estimates or underplaying some costs, but his proposals are fundamentally sound in wonkish terms. Single-payer healthcare does work, and would eventually reduce costs, but the politics of getting there may be impossible, or we might have to tolerate a long transition period during which costs remain high to appease incumbents. What Barack Obama proposed in 2008 — universally accessible health care through individually purchased insurance without an individual mandate — was sheer absurdity. As Paul Krugman has reminded us many times, the Obamacare approach to universal insurance is a three-legged stool: guaranteed-issue leads to a death spiral without an individual mandate, and an individual mandate is unaffordable without subsidies, community rating + mandate + subsidies. (See also Ezra Klein.) If you think policy details during a campaign are the best predictor of the policy process of an administration, you would have expected very little of Barack Obama’s tenure. And you would have been wrong. I suspect that even at the time, Klein was not terrified that Obama would become his caricature of Jimmy Carter as a man whose good values were eclipsed by bad management. Obama signaled competence in ways that Klein could recognize, despite the fact that he took a ridiculous position on the core technocratic issue of the campaign in order to be more appealing to voters.

What I really think is going on is that humans, proles and elites both, have a wide variety of largely unconscious ways of reading candidates. Wonkishly inclined people don’t see themselves reflected in Bernie Sanders, as they did in Barack Obama and do in Hillary Clinton. For them like everyone else, that leads to skepticism and mistrust. But wonkishly inclined people tend not to leave it there. They are verbal, they regard themselves as rational, and so they rationalize. But Klein’s rationalization is, I think, not ultimately supportable. Neither the wonkishness of the candidate nor the quality of campaign white papers is a predictor of the success of a presidential administration. As Klein points out, George H. W. Bush was no wonk, but he was a good manager and in retrospect a decent president. From the perspective his supporters, Ronald Reagan was a phenomenally successful president, but he was no wonk, and I suspect his campaign white papers weren’t anything special either. Johnson, JFK, FDR, admittedly, the farther back we go, the harder it is to “measure” analogues of the contemporary wonk. But I don’t think you’ll find sustainable a theory of campaign policy details as a reliable predictor of actual policy competence. I think wonks, like other humans, have a tendency to see what they want to see, and to reach for theories that flatter their own views and that elevate their own role in the democratic process.

I support Bernie Sanders in the Democratic primary. I don’t support Sanders because I think he is brilliant in some academic way. I don’t support Sanders because I am particularly impressed with the details of his policy proposals, although they are not nearly as hopeless as some self-proclaimed technocrats make them out to be. A democracy is not a graduate seminar.

It is not that I am for Bernie Sanders, but that Bernie Sanders is for me. Bernie Sanders, more than any politician who has ever had a serious shot at the office of United States President, represents my interests and values. By that I don’t mean my interests in a narrow, self-interested sense, but in his vision for what kind of country my country can and should be.

A democratic polity does not elect a technocrat-in-chief, but politicians whose role is to define priorities that must later be translated into well-crafted policy details. Paul Ryan’s various budgets haven’t been wrong because they require giant magic asterices to make the numbers add up. They have been wrong because the interests and values Paul Ryan represents are wrong. The magic asterices don’t reflect dumb mistakes, but smart politics. The problems of our polity do not arise because one faction or another is too stupid to do high quality science. If your interests are the interests of the fossil fuel industry, and you are unwilling or unable to transcend the narrowness of those interests, then confusing the public about the science of climate change is a mark of intelligence, not stupidity. Being smart is great. You may be proud of your GRE scores, your PhD, your Nobel Prize even. And deservedly! But raw intellect is not scarce, and no faction holds anywhere near a monopoly.

In a democratic polity, wonks are the help. The role of the democratic process is to adjudicate interests and values. Wonks get a vote just like everyone else, but expertise on technocratic matters ought not translate to any deference on interests and values. If your theory of democracy is that informed citizens ought to cast votes based on the best social science, you have no theory of democracy at all. If you are honest, you will follow your own theory where it leads, as Bryan Caplan has, and work to limit democracy. But Caplan, whom I love, is mistaken, because he begins with a mistaken theory of politics. If you want to see how that theory of politics works in the real world, look no farther than the European Union, which is a real-time experiment in demoting democratic adjudication of values in favor of technocratic adjudication of facts. I know, you don’t agree with their science. Their economists haven’t died quickly enough to realize that a decades-old consensus has been discredited. Technocracy, like communism, like capitalism, has never been tried. Elevating technocracy above democracy is similar to, and as insidious as, letting military power escape civilian control. The problem with life under military rule is not that the army lacks patriotism, or that it doesn’t mean well. But the interests of the military are not the interests of the polity, and we invented democracy because human beings have a tendency to confuse their own interests with the public’s. The interests of the class of humans who might reasonably qualify as technocrats are also not the interests of the polity.

So, I am for Bernie. I am not against Hillary. But just as it’s foolish to say that Democrats and Republicans are “all the same” because they are both corporatist parties, it is foolish to claim that Bernie and Hillary do not represent meaningfully different interests and values. I’ll enthusiastically support either Bernie or Hillary over a Jeb Bush, Marco Rubio, Ted Cruz, or Donald Trump. But it is Bernie Sanders who is for me, and I’m supporting him without apology. If your interests and values are my interests and values, I hope that you do too.

The word “progressive” is used in all kinds of ways. I’m not interested in the term as a label of people. I won’t even self-identify as “a progressive” (or almost any other thing), even though I often find myself making common cause with people who do. I’ll leave arguments about who is a progressive to others.

One frequent use of the term “progressive” is to refer to the distributional effects of an arrangement or policy. The income tax is “progressive” because it takes more (in absolute terms and in percentage terms) from high earners, thus reducing after-tax income inequality. A carbon tax, it is often claimed, would be regressive, because the poor spend a high fraction of their incomes on heating and transportation than the well-off, who therefore pay less (in percentage terms), accentuating after-tax income inequality.

It is easy to forget, in these kinds of conversations, that “progressivity” is not scalar. Policies cannot be ordered by their progressivity, because most nontrivial policies make different groups of people better off and worse off in ways that don’t monotonically map across the income distribution.

Let’s look at a few charts to make this clear. In all of our hypersimplified drawings, the x-axis will define three groups, which we’ll call “poor”, “middle”, and “rich”. For our purposes, it matters not at all whether we are talking about income, wealth, health, or ponies. There is some good thing and two alternative arrangements. We can think of one arrangement as the status quo and another one as a proposed new policy. Or we can think of both arrangements as hypotheticals we are considering and comparing.

Let’s first look at an example under which one alternative is plainly more progressive than the other:

Arrangement B is plainly more progressive than Arrangement A. The distribution of, um, ponies is more equal across the distribution. Both the rich and the poor are brought towards the middle.

However, suppose we are evaluating the following two policies for their relative “progressivity”:

Very few people would, I think, describe Arrangement B as “more progressive” than Arrangement A, because Arrangement B clearly screws the poor. The term “progressive” is garlanded with moral connotations about helping the poor, so this would not qualify. However, Arrangement B does include a transfer from the rich to the middle, and it reduces the inequality between the rich and the middle substantially. If your definition of progressive is “redistributes from richer to poorer”, you cannot definitively order one arrangement as more or less progressive than the other.

One might describe a proposal to go from Arrangement A to Arrangement B as “insider-outsider egalitarian”. It is not an irrelevant case. Although historical experience suggests they are mistaken, people who oppose unions frequently argue that they have precisely this effect. A better-off, unionized portion of the workforce, in this story, sees wages increases, at the expense of wealthy owners and managers (who have to pay unionized workers more), but also at the expense of the not-unionized (or not-employed) portion of the workforce, who see wages stagnate and prices rise.

While “insider-outsider egalitarian” is not an irrelevant case, it is not an ethically confusing one either. Holding constant the identity and the ordering of the humans, very few people explicitly argue that screwing the poor for the sake of egalitarianism among insiders is a good thing. Arguing that this will be the effect of a policy is a way of opposing it. (When identities are not held constant, the moral intuitions become less clear. Suppose under Arrangement B the poor are new immigrants whose poverty masks a great improvement relative to circumstances in their home country?)

Let’s consider a third case:

This is the case I want to call quasiprogressive. Again, if our definition of “progressivity” is redistributing from richer to poorer, we can’t rank Arrangement A against Arrangement B. Going from Arrangement A to Arrangement B, ponies flow in both directions, to the poor, sure, but also to the rich from the middle. Nevertheless, people are tempted to call a shift to Arrangement B “progressive”, because it is good for the poor. Our ethical intuitions about this change may be positive. Or we may describe it negatively, call it a “hollowing of the middle class”. Is it a vase or a love affair? Suppose that a trade agreement will put middle-class workers in developed countries in competition with low wage workers elsewhere, leading to both a decline of middle-class wages and a proliferation of cheap consumer goods. The poor will benefit from cheap stuff, the rich will benefit from cheap stuff and cheap help, but the middle class will get screwed. Should we call this agreement “progressive” or “regressive”? Taking into account only domestic welfare, would it be a good thing or a bad thing?

If our only choices are Arrangement A and Arrangement B, we might be tempted to dispel once and for all with this fiction that Barack Obama doesn’t know what he’s doing. Short circuit. It doesn’t compute. Do we screw the poor or gut the middle class?

Fortunately, those are our choices only if the space of available policies is constrained artificially. In reality, our decision space usually looks more like this:

The black bars in this chart represent the status quo. We are decided to enact a reform. Shall we jump to Arrangement A or Arrangement B? Absent political considerations, most of us would go for A. But suppose, not implausibly, that an ideological and institutional apparatus exists whose function is to defend the material interest of the wealthy? Then perhaps only Arrangement B will seem “politically possible”. Since Arrangement B does help the poor, it may seem unethical not to take it. If (as is usually the case) the harms to the middle are indirect — no one will “take” anything from anyone, market forces will do the dirty work — wealthier people of good will be tempted to support the reform. They stand to lose nothing, and maybe to gain something. The poor will be made better off! Harms to the middle will seem hypothetical, and in any case will amount to just desserts adjudicated by an impersonal market. Didn’t Rawls tell us that the welfare of the poorest should be our touchstone? Less wealthy people of goodwill may not be so sanguine about the reform. They may be torn. When a union member opposes a trade agreement that she understands will lower prices on many goods, but that also might put her out of a job, is she fighting the good fight for the middle class, or is she just a special interest screwing others to preserve her place at the trough? So long as Arrangement A is off the table, it will be very hard to mount an effective opposition to accepting Arrangement B, which after all benefits the poor so greatly.

Policy changes all the time, and the circumstances that favor reforms that look like Arrangement B over those that look like Arrangement A are permanent. Since wealth and influence correlate, each accommodation tilts the political economy towards the next capitulation. A weaker middle is less and less capable of opposing quasiprogressive alternatives that look both desirable and virtuous to the rich. In a world where the wealthy are genuinely interested in helping the poorest off but also vigilant caretakers of their own interests, quasiprogressive solutions are a political-economic sweet spot. The only way to prevent this dynamic without actively screwing the poor is to fight for solutions that work for the poor without sacrificing the middle. Those solutions will always be politically challenging. Even where a policy change is “positive sum”, as trade agreements are argued to be, policymakers almost unconsciously gravitate towards arrangements that segregate benefits towards the powerful rich and the sympathetic poor but shift burdens towards the less salient, less powerful middle.

This sounds very abstract, so let me name some examples where I think the American polity has chosen quasiprogressive solutions. We’ve talked about trade. The choice of quantitative easing rather than fiscal policy to stimulate the economy at the zero-lower-bound is another example. QE-assisted asset-price rises undoubtedly helped people at the margins get or keep their jobs while they also helped wealthy assetholders. Fiscal stimulus would have helped the marginally employed at least as well and might have put pressure on middle class wages too. But bonds might not have performed so well. The bank bailouts of 2008-2009 are frequently justified on quasiprogressive terms. Surely, had there been a catastrophic financial collapse, the poor would have suffered most of all! Of course they would have. They always do, that is the essence of poverty. But we might have averted a financial collapse with transfers to underwater homeowners, for example, or to be more fair, with helicopter money for everyone. Instead, we let middle-class homeowners collapse into negative equity and, too frequently, foreclosure, even as we provided generous support and regulatory forbearance to ensure that financial institutions and their creditors would be made whole. Most of the fiscal stimulus came not from Obama’s headline $800B program, but via automatic stabilizers that kick-in for the poor and unemployed. We rescue the rich. We congratulate ourselves for paying so much to help the stigmatized, traumatized poor. We tell the middle to suck it. That’s the quasiprogressive way, and it’s at the heart of American political economy.

Let’s do one more chart.

Which one, Arrangement A or Arrangement B is more progressive? If our definition of “progressivity” is redistributing from richer to poorer, we absolutely can rank a shift between these two arrangements. Going from Arrangement A to Arrangement B is clearly progressive. Ponies are shifted from both the rich and the middle to the poor.

So should our moral intuitions applaud this change? I don’t think so. I call this one “token progressivity”. While helpful to the poor, a transition from Arrangement A to Arrangement B extracts a small concession from the wealthy but burdens the middle disproportionately.

A lot of conversations bog down into arguments over whether the effects of some event were quasiprogressive or token-progressive. It just doesn’t matter very much. Uniformity of the direction of altered outcomes is too simplistic a criterion to rest moral intuitions on. A society is more than a summation of hypothetical utilities. The shape of the distribution of ponies matters as well. If we want to live in a middle class society, we have to fight for policies that protect the middle class. If we content ourselves with a progressivism of the politically achievable, we may find ourselves torn between doing nothing at all or celebrating accomplishments that may genuinely help the poor, but at the expense of an erstwhile middle class on a path of downward convergence. The work of good policy lies more in expanding the menu of choices than in optimizing over what is straightforwardly achievable.

At this particular moment, I want to emphasize “expanding the menu” needn’t mean preferring single-payer healthcare to reforming ObamaCare, or preferring universal free tuition over means-tested financial aid. But it is an implication of this analysis that the centrist acceptance of net budgetary impact as a criterion for valuing programs — where “net” means net of user fees like insurance premiums or tuition contributions — deserves special scrutiny. The first question should always be “value for the money”. Regardless of how a program will be financed, are the goods to be purchased worth the opportunity cost of the resources that will be expended? If we decide that a program is worthwhile, we then face a political trade-off. The smaller its net budgetary impact, the easier our program will be to pass. However, every shift in financing from general revenue towards means-tested user fees moves us precisely along a path from progressive, to token-progressive, to quasiprogressive, as defined in the discussion above. There are sometimes colorable reasons for user fees, besides political expedience. User fees can address fairness concerns about programs whose benefits aren’t universally available and yield few positive spillovers. Importantly, program value may not be independent of fee-structure. For example, copays may discourage frivolous doctor visits, and so enhance “bang-for-the-buck” of expenditures overall. These are sometimes valid concerns, but they often serve as fig leaves for imposing fees just to reduce net budget impact. That is, they serve as politically motivated excuses to reduce the progressivity of programs in order to increase their burden on the middle class relative to counterfactuals under which the programs would be financed from general revenue.