Network News

X My Profile
View More Activity

What Does the Congressional Budget Office Do? An Interview With Doug Elmendorf.

PH2009090900309.jpgThe Congressional Budget Office is normally a sleepy place where budget wonks prepare little-read estimates on the cost of legislation. Right now, however, it's the center of the legislative universe: Its judgment of how much health-care reform will cost, save and require is dominating the discussion of the bills. Last month, I sat down with Director Doug Elmendorf to talk about what the CBO does, how it works, and whether it's in fact too conservative in its estimates. Also, he told me why he's stopped reading books.

Normally people don't hear much about the Congressional Budget Office. It's a quiet place where people talk about numbers. But every few years, when major legislation moves forward, it seems like the weight of the world is on your shoulders. So what is it you do?

CBO provides objective analysis of budget and economic policy for the Congress. It was founded in 1975. Congress wanted a source of information independent of what it was hearing from the administration. For the last 34 years, we have provided estimates of the cost of different proposals and of the effects of different proposals on the budget, the economy, and the well-being of individuals.

How exactly does that work? To paraphrase Schoolhouse Rock, how do a bunch of academic papers become a model?

On health care, the root of the model is that people work for firms and live in families and we think they are making purposeful choices about how to get health-care insurance that are sensitive to the cost and quality of health insurance. There are estimates that economists and health experts have developed over years of the ways that individuals and companies respond to different options. We build those estimates of how sensitive people and employers are into the model. If we can show to the model a set of insurance choices that are different than the set of insurance choices people are given today, we can model how they make decisions.

So you take existing evidence of how people react to changes and then test more changes against it.

That's right.

So you put it all into the model and then you get something called a score. What is a score?

A score can be different things. CBO does official cost estimates for legislation at certain points in the legislative process. Those cost estimates usually include the cost over 10 years of the legislation. That 10 years is called the budget window. Those numbers, and an explanation of where they come from, are the official cost estimate, But we also do a variety of preliminary analyses. The estimates we've released for the Senate HELP Committee health bill and House bills are preliminary estimates. They're complicated bills and we're still working through all the aspects of the calculations we need to do. But we provide the preliminary estimates so there's an understanding of the broad contours of the legislation.

People often say that the way you do your work is biased towards costs, because you can say firmly "we're going to spend this much money," and away from savings, because they're more speculative.

I don't think that complaint is well-founded. I understand many people have that concern. But we try very hard to offer numbers in the middle of the distribution on outcomes, both on the spending and cost side. It's true that if you want to spend $10 billion a year of preventive medicine, we can be pretty certain that it will cost $10 billion a year. So the uncertainty is low. But when we turn to the benefits of, say, preventive health benefits of health spending, we'll try to be in the middle of the range of possibilities. That's a wide distribution of possibilities. The uncertainty will be much wider. But we don't try to be conservative in the sense of being close to zero. We try to be in the middle.

I think people's frustration arises partly because one's intuition about the effects of certain policies on future federal spending is not always confirmed by the hard evidence. On the preventive side, most people's intuition is that discovering a disease earlier lets you treat them earlier and saves a lot of money. The difficulty with prevention is that you end up doing the test for a lot of people who wouldn't have the disease. Even if you're saving a lot of money on those who would have had the disease you don't save that much money ultimately. This isn't our research to start with. A paper in the New England Journal of Medicine last year found that only 20 percent of preventive spending saved money. A large chunk improved health at a low cost, so it may be a good deal. And then some was just extremely expensive and didn't do much to improve health. But that's the sort of evidence we're reading.

How do you deal with a new policy? One senator affectionately said that "the problem with the CBO is that they're like someone who looks in the rearview mirror to tell you where you're going." How do you deal with something that hasn't been studied because it hasn't been done? It's being built into policy because it doesn't exist.

It's a lot harder. Small changes in existing programs are easier for us to assess because we have a lot of experience watching those things change in the past. Things that are entirely new are more difficult for us. We try to find analogies where we can, from experience, but we also try to be very open about the uncertainties that exist. I have rarely given an interview or testimony or written a letter on health care where the word "uncertainty" didn't appear many times. But I think we still need to give Congress a ballpark of the various effects. If we just tell Congress that we don't know at all, that's not very helpful, and usually I think we can do a bit better. A recent example is a discussion about giving more authority to an executive branch agency to propose reforms for Medicare as we learn, over time, what does and doesn't work to save money in the Medicare system. There's no existing example of just that sort of agency. The analysis we did drew on experience with other sorts of agencies that have had some extra authority to propose changes in policy. We looked at how that's played out. The analogies aren't perfect. But they're the closest we can find.

You said that you've not given a testimony or released a paper that didn't use the word "uncertainty" many times. How often when that testimony is repurposed in a political speech or that paper is quoted in an article does the word "uncertainty" appear?

[Laughs.] I think the caveats that we apply to our analyses are often left behind. Even when we report ranges of numbers there's a tendency to report the midpoint of our range. I think that's too bad.

CBO has these two roles. One is to provide "the number" for the congressional budget process. But you also become a public agency in times of legislative controversy. You become the arbiter of the numbers people use. Is that a comfortable role for CBO?

We're not looking to be in the newspaper. Our job is to serve the Congress. We end up in the newspaper when there are particular issues where members of Congress disagree and where we've done some analysis that nobody else with our independence and analytic strengths has done. We're not looking for that. Most of our work we do with congressional staff day in and day out, with no particular spotlight shining on us, and we're happier that way.

Can you say, "I don't know?" Somebody comes to you and there's not a lot of data on their initiative. Can you pass?

In some cases. But when legislation is moving through the congressional process, we need to put numbers on that. I don't think we can just throw up our hands and say we can't provide an estimate.

Obviously your estimates are of great interest to many powerful people in town. What is your relationship with them like? Does Rahm Emanuel send you a dead fish? Does Max Baucus remember your birthday? Is there an effort to influence or browbeat you?

When we work with members of Congress we do that on a confidential basis. But as a general matter, we don't accept gifts. I've not been threatened about particular numbers. Mostly members of Congress complain that we haven't analyzed pieces of legislation quickly enough. We can't keep up with the pace of demands. Most of the angry calls I get are people legitimately frustrated that we can't address their issues as quickly as they'd like. I think people understand we're not susceptible to pressure about changing our numbers. We just aren't. We have a very strong culture of independence developed over 30 years that we will just report the answers we get. Browbeating doesn't work, it hasn't worked, and I think people have learned to stop trying.

You do occasionally see interesting exchanges in testimony. I remember one from a Finance Committee hearing. I was beginning to doze off a bit in the back. And then Chairman Baucus looks at you and says, "You know you're not God." Is that correct? Do you know you're not God? Do you keep a highlight reel of this stuff?

I'd heard that line before. Look, we're not all-powerful. We provide information. It's up to the Congress to decide what to do with that information. I said this to Chairman Baucus at the time. The hard decisions this year are his and his colleagues’. Our analyses only matter insofar as Congress takes them seriously. But it's up to them to decide what use to make of our analyses.

You can't agree with this question. But it's often seemed to me that some of the tensions come because CBO is a responsible agency within an irresponsible political culture. Do you ever sit in your office and think about how the work is going to be greeted?

We don't try to strategize about how or when to release information to influence the political process. That would be inappropriate. We meet our requests as quickly as we can. We publish findings when they're done. How that plays into the political dynamic in Capitol Hill or the country is not our concern.

A few weeks ago, one of your analyses was greeted by your predecessor Peter Orszag with a ... lack of favor. In particular, he said, "In providing a quantitative estimate of long-term effects without any analytical basis for doing so, CBO seems to have overstepped."

I don't want to respond to Peter's blog directly. On the general issue about what sorts of information we provide to Congress, we try to be as specific as we can consistent with the uncertainties in the world. Most of our quantitative analysis is over the 10-year budget window, but we do some analyses longer than that. We just released our long-term budget outlook. On an issue like health-care reform, that has long-lasting effects on the country and the budget, we'll try to provide as much real information as we can, and we have said that we will. When the crystal ball descends into fuzziness, we won't.

You're a blogger now.

It's a novel experience. The CBO blog, in contrast with many private blogs, is not the musings of a particular person on a particular day. We use it as a way of reporting the analysis we're doing in a more condensed or accessible form than our cost estimates, which can be rather involved. We use it as a way of making points that don't fall naturally into a report. For instance, we've done some entries on how we think about prevention and how it fits into our analyses. Ideally, we could've released a report on that, but we didn't have time, and this was a way of getting some information out for our staffs to read.

Do you write the posts?

Like many products at CBO, the blog postings reflect the work of a number of people, including me. But just like letters that go out over my signature are not all written by me, the blog posts aren't all written by me. I'm responsible for them and take them seriously. But Larry Summers was quoted years ago and said it was a cardinal sin to take credit for other people's work. As a treasury secretary, it was a mark of effective management. I think that's the role I play at CBO. I'm deeply engaged in the analysis we do. But there are 250 or so smart and knowledgeable people here besides me, and part of my role is to be a messenger for that analysis.

What about a MySpace page? Or a Twitter account?

We have talked about those things. At the moment, we're pretty preoccupied with the pressures of keeping up with Congress. But we're looking for ways to communicate more effectively.

In CBO's estimation, how do we save money on health care?

That can't be done in a short answer. The best I can do is refer you to the work we've written down. In particular, the letter we wrote in the middle of June to Sen. Conrad and Sen. Gregg summarized our analysis. Much of what we've said since then is out of that letter. There's a wide range of possible ways to reduce government spending on health or total spending on health. I think the main message is that there's a widespread consensus that a lot of the money being used is not being used to increase people's health. But extracting that money without inadvertently harming people's health is challenging.

Are any of these issues analytically harder than the others?

I think most issues we tackle are pretty difficult analytically. The things that are easy people don't ask us about.

What's the last good book you read?

Most of my reading is CBO reports and drafts of CBO reports. before I came, the reading was on the financial crisis. Mark Zandi's book, for instance. But one of the things I've given up for this job is the chance to read things that aren't directly related to our analyses. I've given up bigger things than that, like time with my family. But I've given up reading.

Photo credit: By Pablo Martinez Monsivais — Associated Press

By Ezra Klein  |  September 17, 2009; 3:28 PM ET
Categories:  Budget , Interviews  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Max Baucus's Legitimacy Problem
Next: Reporting!


Very nice interview.

Posted by: slantedview | September 17, 2009 5:16 PM | Report abuse

Ezra, that was an amazing interview.

Posted by: axlotl | September 18, 2009 1:35 AM | Report abuse

While its understandable the CBO must presume the worst possible case:

"The difficulty with prevention is that you end up doing the test for a lot of people who wouldn't have the disease."

In the real world, doctors have some ability and responsibility to weigh risk factors. This helps doctors decide whether to use an expensive screening (preventive) test, and how often.

So...if the more expensive types of preventive screening are used when indicated, and not otherwise....then the CBO conclusion that on the whole preventive care would raise total costs is just....misleading.

The real picture is more complex and interesting.

Posted by: jozzer | September 18, 2009 10:01 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company