Complexity Is a Big Problem

Forget Big Versus Small — Our Government’s Needless Complexity Is a Big Problem

 

by Joshua Holland

Our ideological debates over the size of government and its role in the economy are often fierce, but what if we’re all missing a larger issue?

Steven Teles, a political scientist at Johns Hopkins, argues that the needless complexity with which our government operates costs us dearly – in dollars, transparency and trust in our institutions. And that’s true for both liberals and conservatives.

Teles borrowed the word “kludge” from the world of computer programming to describe the problem. “A kludge,” he writes, “is an inelegant patch put in place to solve an unexpected problem and designed to be backward-compatible with the rest of an existing system.” Things go wrong in a variety of ways “when you add up enough kludges” leading to “a very complicated program that has no clear organizing principle, is exceedingly difficult to understand, and is subject to crashes.” Teles adds: “any user of Microsoft Windows will immediately grasp the concept.”

Moyers & Company caught up with Teles to discuss how our “kludgeocracy” hurts our government’s ability to function as it should. Below is a transcript of our talk, lightly edited for length and clarity.

Joshua Holland: Tell us a little bit about some of the costs that we end up paying for all these kludges – you use the costs of navigating a ridiculously complex tax system as one example.

Teles: Well, with the tax system, part of the costs are simply what people pay to get people to do their taxes. That’s actually a fairly significant one. But also all the time people spend on their taxes adds up to well over $100 billion a year. There’s no human good that’s being produced by that. These are simply costs that are associated with us having a much more complex mechanism than we would have needed.

Holland: And of course, lobbyists advocate for this loophole and that loophole, and then in the end you have something that’s almost impossible for ordinary people to navigate.
Now, our kludgeocracy is a big problem for those of us who think an activist government can solve certain problems. Let me quote, just briefly, from your piece in National Affairs, and then ask you to flesh this out. You write, “Politicians may posture against corporate welfare, but kludgeocracy makes it hard for voters to see how much business profits from government, which makes it difficult to effectively target their anger. As a consequence, that anger diffuses into our system of government as a whole, leading to a loss of trust, and to skepticism of the possibility that the public sector could ever be an effective instrument of the public good.” Can you unpack that for us, Steven?

Teles: Yes. I think part of the issue is that some of these policies that have really complicated mechanisms were created precisely because Republicans, to be blunt, didn’t want fairly clean, simple policies that would make it easy for voters to recognize they were getting a government service and then reward the people who produced it for them.

Look at Social Security. For a very long time, it produced excellent results for Democrats because the recipients knew they were getting Social Security, they knew who produced it, and they rewarded them politically.

Increasingly, when more and more public policy is being done in what Chris Howard called, “the shadow welfare state,” people are not always aware that they’re getting a government service. So when you look at a lot of these areas – like the ways we help people go to college — a lot of them are done through the tax code in ways that people don’t really think of as government. When you ask them, ‘Are you a recipient of government services?’ people who are getting thousands and thousands of dollars a year don’t think of themselves in that way.

The consequence is that a huge amount of our middle class think of themselves exclusively as givers rather than takers in the welfare state, so they imagine there’s this whole other population who are the ones taking everything and they’re the ones giving, even though, in fact, they depend very substantially for their well-being on all these awkward, indirect ways of [delivering benefits]. Play that game out over and over again, and you have a welfare state that doesn’t make it clear to them how important government is for their own well-being.

Holland: That’s clearly a problem for liberals. In the piece, you say that a kludgy system is problematic for both liberals and conservatives. I’m not sure why it’s a problem for conservatives.

Teles: When government doesn’t work very well, it’s hard for people to punish it. So, in the case of these health care exchanges — is that the fault of the health care companies? Is it the fault of the government? Is it the fault of the contractors? It’s very hard to know exactly who to punish, and that’s actually one of our more direct programs where it’s a little easier to target. But when you think about the huge system of contractors doing huge amounts of government services, all of the arrows that would allow you to figure out who to hold responsible are hard to see.

The other part is that I would argue it’s better for conservatives for government either to be all in or all out. When government is sort of a little bit everywhere, it’s really hard for conservatives to say, “No, we don’t want our government here at all.” Now there are very low hurdles for government involvement so it would be better to have a world in which it was clear. As it stands, government can inch into areas in very small, often awkward ways, but once it’s done that, it’s hard to have the fundamental conversation about whether we want government intervention at all.

Holland: You say in the piece that Americans are “ideologically conservative but operationally liberal.” I put it a little bit differently. I often say that Americans tend to be suspicious of the idea of government in the abstract, but they actually like the things that government does in the specific.

And it seems to me that for conservatives, obscuring what government does is a benefit, in that they don’t necessarily have to come out and say exactly what services they would deprive people of in order to cut high-end taxes, which is — I guess I don’t really ascribe to conservatism a good-faith belief in smaller government. I think they’re mostly interested in lowering high-end tax rates and deregulating business, neither of which is obviously inconsistent with a kludgy system.
Teles: Right, there are conservatives and there are conservatives. I mean, certainly a lot of my Libertarian friends genuinely think that the intervention of government into all kinds of areas is pernicious. But there are obviously also substantial numbers of Republicans who mainly thought about who was on their team, and thought the government wasn’t rewarding the people on their team and hurting the guys on the other side. That certainly explains the structure of Medicare Part D, where eventually conservatives cut a deal where they would allow the government to get into this area, but only if it cut all their friends in on the deal.

I also think there are a lot of Libertarians who got sold on the idea of privatization. They thought that it was better to have some government service — if it had to be done at all – contracted out to private firms.

And I think they had really underestimated how much those private firms — once they were in this contractual relationship with government — would be a durable lobby for more government, and often more expensive government. It would be very hard to pull them out. In some ways it’s much harder to pull out a kludgy, complicated government that has all these links to the private sector, than it is to pull out a much cleaner public program. So I guess maybe I also attribute slightly more good faith, at least to some set of conservatives, than you do.

Holland: Suzanne Mettler wrote about what she called, “the submerged state.” She pointed out that virtually all Americans are, at various times in their lives, both “makers” and “takers,” but most people don’t realize that they receive federal benefits, even when they do. Can you just give us a little bit more of a detailed understanding of how this relates to a kludgy system?

Teles: The most obvious example, again, is the way we now pay for college education. If you think about the Pell Grant program, that’s a pretty good example of a non-kludgy program, right? People basically get income-related support for going to college. They know they’re getting it. They know it’s coming from government.

But we also have this other way, which is very much part of the shadow welfare state. That’s the tens of billions of dollars every year that goes through various kinds of tax-advantage savings — 529 plans, Coverdell IRAs — and they are generally not perceived of as part of our welfare state.
We made a quite conscious decision that we were going to build a lot of our welfare state around higher education into the tax system, but people generally view that as a service that’s coming from the private sector, especially given that it’s coming in the form of tax subsidies.

It means that for high-end tax payers, they’re only paying $2 out of every $3 of the money that they’re putting in, because the government is basically picking up the other dollar out of their taxes.
I think that’s a good example of a program liberals often supported, because they wanted more financial support for people who are going to college. But they did not want it to seem big-government-y, and this seems small-government-y. So in that sense, I think liberals were implicated in this needless complexity as well. This wasn’t just something that conservatives insisted on. In many ways, New Democrats liked this arrangement because it had the effect, they thought, of big government programs without looking like government.

Holland: The health care reforms are a perfect example of what you’re describing. Everything we’re seeing with the website right now is based on the inherent kludginess of the whole scheme, right? It has to transmit information securely between several state and federal agencies, verify your eligibility for subsidies, verify your citizenship status, and then it has to transmit a whole bundle of information to hundreds of private insurance companies.
And we can contrast that with a far simpler system, like opening up the existing Medicare program for all comers, and just using tax revenues to cover the health care expenses of the poor.

But let me ask you this: the Dutch system is ranked one of the best health care systems in the world and it follows the same basic structure. It has a mandate, there are subsidies, and then it relies on private insurers.

But the Dutch have a different view of the role of government and about the value of regulation. They have cost controls in place, and when they set up their system, they were going in a different direction – they were privatizing a single-payer system. Kevin Drum focused on this point. There was a big difference in terms of the paths taken to get there, and the Dutch had already accepted the idea of universal health care. They were arguing over how best to implement it. To what degree is our kludginess, then, a result of our political culture?

Teles: Well, again, I do think that Americans are operational liberals and ideological conservatives, in the sense that, at least right now, big, direct government seems somehow inappropriate. So that creates at least a finger on the scale toward things that help people while hiding the hand of the state.
But to go back to the complexity of the health care system, in some ways I think we can overstate this as a technological problem or a computer design problem. It is — and you were fairly clear about this — the underlying design of the program that’s really responsible. That is, it’s not just that, A) we’re contracting out the technology, and B) we have a bunch of private insurance companies involved. It’s also substantially means-tested, which is important. That means testing interacts with other kinds of means testing, it interacts with other parts in the system, and this is really the exact definition of a kludge. You’ve got one set of complexities that you’re already inheriting, but you can’t get rid of them, so you’re adding another complex fix on top of it, and every time you do that, you end up with all these complex interaction effects. Think about this now from the point of view of somebody who’s designing the website. It’s not just the website, it’s all the stuff behind the website — you’ve got an incredibly complicated task of simultaneously determining eligibility, merging all these different systems, and then having it all work in a single computer system that doesn’t have enough internal capacity for that process. So in a way, Obamacare is a perfect storm of all of these phenomenon and I do think it presents a really difficult issue for liberals.

I was, in the end, in favor of the Affordable Care Act. I was aware that this was an incredibly complicated, kludgy thing, but we were so desperate to get to something like universal care — or at least expanded care — that we all just swallowed it.

You know, I think people who were in favor of single-payer didn’t have much of a political answer about how they were going to pass it, but the nature of a kludgeocratic system is that it constantly throws up these problems and people have to keep accepting half-measures. I keep having to accept these complicated mechanisms in lieu of what I know would be a better operating system, and so in that way people who are liberals also get implicated in this degree of complexity and in a way contribute to it.

Holland: And on top of all those problems, we gave the states the ability to opt out of running their own exchanges, so you have the federal government running, I think it’s 35 or 36 exchanges that they didn’t expect to be running.

We have a system with a lot of veto points in it, that is to say, a lot of points in the legislative process where people and institutions can derail the process, and they get to make demands and get their little amendments and add-ons. To what degree does this add levels of kludges to the system?

Teles: Well, there are a couple of those choke points in Congress, and it’s not just the filibuster. We often have bills being worked out in multiple committees. And each of those committees can insist on some set of their own preferences. Often those preferences go to fundamental policy design that then have to get squeezed into something. And in the article, I say that you should think about all those veto points as toll takers, right? Those people can all insist on some role in the ultimate decision making, and when you add up all those extra roles, then you end up getting a policy that’s ultimately a combination of different visions that don’t necessarily mesh together.

The other part of this is that federalism itself is a kind of kludge. When you’ve got a system where people think that they’re supposed to be cutting the states in on every major public policy endeavor, then you’ve got to involve a whole other set of actors, their preferences, and the implementation chain gets even longer. In the paper I talk briefly about Hurricane Katrina. We think that both the states and the federal government should be involved in emergency response, and what you often get in cases like that are hold-ups, where both sides are waiting for the other one to do something. But that clearly is an institutional cause of kludgeocracy, and in the case of Katrina, it had terrible, devastating results.

Holland: You write, “kludgeocracy is now self-generating, as its growth has created a kludge industry that feeds off the system’s appetite for complexity.” Tell us a little bit about this kludge industry?

Teles: This has a couple of different dimensions. The most obvious is the whole system of contractors we have for doing almost everything the government does. That then becomes a lobby for continuing to do things through contractors, because contractors can recycle their profits from doing work for the government into lobbying. That’s, I think, the most obvious part.

The other way this is self-generating is through lobbying in general. There’s a great piece of work that’s being finished by Lee Drutman on business lobbying, and one of the things he shows is that now that we’ve got this very mature system of lobbyists, they have to convince their corporate clients that they’re producing value. Increasingly what these clients want is measurable, traceable results that can show some piece of government that the lobbyist got for them. And he argues that this is leading to ever more morselized government programs, where the lobbyists are pushing for little pieces separate from larger policy designs that they can attribute to their own work. So in that sense, the lobbyists are a part of that self-generating system of ever greater complexity.

Holland: Steven, I think my final question is a tough one. You write that “rather than the traditional left-right debate over the size of government, in the years to come, we’re going to have to debate its inherent complexity.” And if we understand that the kludgeocracy is born of our political culture and our basic system of governance — federalism, and having lots of veto points — then what exactly can we debate? Or I guess what I’m asking is simpler: Is the system just inherently kludgy as a result of conflicts that go back to the founding of the republic? Or is there something that we can do to reduce this complexity, short of radically changing our system of government?

Teles: Well, I’ve increasingly come to be more open to radically changing our system – certainly more open than I was just a few years ago. You know, there is some work that’s being done by my colleague Emily Zackin. She’s done great work on state constitutions, and one of the things she shows is that the state constitutional tradition is distinct from the federal constitutional tradition. States end up changing their constitutions relatively frequently, and they often have long constitutions, much more like European constitutions. In that sense, substantial institutional change is not actually outside of our constitutional traditions. It’s just outside of the constitutional traditions we associate with our federal government.

The other thing is that our Constitution doesn’t operate the way it’s supposed to. Once you get a system of highly polarized parties, the system operates radically different than it was designed to, and in that sense, we haven’t updated our Constitution to accommodate the system of parties that we’ve developed.

I’m not averse to some of the Supreme Court’s federalism jurisprudence which, in part, has the intention of forcing the government either to do things cleanly and through the national government or basically leave them to the states. That would be a generous interpretation of the Court’s position on the Medicaid expansion. And I actually think that if we knew in 2009 what we know now — that the Court was going to allow states to opt out of the Medicaid expansion –we would have ended up dealing with the expansion through the exchanges, which was problematic, but not nearly as complicated as letting states just opt out.

So I think the area I would be most eager to deal with is our system of cooperative federalism — start pushing policy in the direction of either having the federal government do it or leave it to the states. I’d move away from this complicated tapestry that weaves the federal and the state governments together.