CBO: Looking inside Congress’ most objective office


Written by

The 113th Congress may have been one of the least productive U.S. legislative bodies in history, but one congressional organization is always busy — the Congressional Budget Office.

Charged with providing independent economic analysis of legislation and budgets, CBO has earned respect and ire from Congress members for its by-the-book objective reports.

CBO is responsible for scoring a long list of political formulations including budget options, projections and baselines, sequestration projections and federal mandates.

But one duty is more frequent than all others: cost estimates on legislative bills. CBO compiles a formal report for virtually every bill reported out of committee, totaling between 500 and 700 a year, according to CBO’s website.

Each report must show how much a bill will cost and how it will affect spending or revenues over the next five or 10 years.

Take that, Nixon!

“CBO was created first and foremost to give Congress its own capacity to develop information on the economy and budget,” said Philip G. Joyce, former CBO employee, author of “The Congressional Budget Office: Honest Numbers, Power and Policy Making” and now professor at the University of Maryland, College Park.

CBO was born out of the Congressional Budget and Impoundment Control Act of 1974 in reaction to President Richard Nixon’s excess of executive power, according to Joyce.

Prior to the Impoundment Act, Congress only dealt with the budget in a piecemeal basis and the Office of Management and Budget supplied the cost estimates on bills and budgets.

Congress decided to assert itself in the budget process after Nixon impounded funds Congress had appropriated. The impoundment kept the money from going to its designated source.

“1974 not being the high watermark for trust between the legislative and executive branches, it didn’t make any sense for Congress to have its own budget and have to rely on numbers coming out of Richard Nixon’s budget office,” Joyce said.

And so, in that vein, CBO was created. But what is tantamount to its legacy, according to Joyce, was the first director of CBO, Alice Rivlin.

Rivlin decided early that CBO would not make recommendations to Congress. Though it would weigh the pros and cons, it would not try to influence Congress in anyway.

With that precedence, Rivlin set a tradition of hardcore objectivity that still reverberates into the present. Despite all the partisan bickering and lack of congressional comity in the chambers, CBO remains a stone-faced statue of nonpartisan judgment.

That doesn’t mean Congress members and staffers always like the results. According to Doug Holtz-Eakin, former director of CBO, gripes about how a bill is scored are pretty typical.

“If you said to someone even 35 years ago, ‘We are going to create this nonpartisan institution in the most partisan institution imaginable,’ I don’t think they would have said it would work,” Joyce said, “In a sense, it’s sort of amazing the organization has ended up being as well respected as it has been.”

The numbers game

How does CBO predict the possible cost of a bill, especially for something abstract or extremely complex?

Since its inception, CBO has grown to a staff of about 220. Each employee has a specialty in a budget account. Meaning, if a complex bill, such as the Affordable Care Act, will fall in many categories and 20 or 30 people may work on that one bill.

One of the biggest problems for CBO is figuring out which bills require a report. Once a bill is reported out of committee, the CBO report must be ready in time for the floor vote. But with about 5,000 bills introduced every year, it would be impossible to score them all.

To get a head start on reports, CBO works closely with committees and party leadership to find out what bills are likely to report out of committee, according to Joyce.

“It is a mistake to think the committee reports a bill and then it gets dropped in CBO’s lap,” Joyce said. “Often, there is informal dialogue between CBO and the committee.”

This dialogue extends beyond simply when a bill will be reported out. Lawmakers often call on CBO for advice on how much a bill they are writing will cost.

“People often express surprise that a committee sets a target on how much [a bill] will cost and then magically, CBO comes out with an estimate that is exactly that amount,” Joyce said. “That’s not because CBO has been rigged, but because of this dialogue.”

Once bills are assigned, CBO has to sit down and figure out what a bill actually does. According to Joyce, sometimes a Congress member will write a bill thinking it does one thing, but the language actually says another.

“The CBO is not in the business of trying to interpret what the sponsor of the bill is trying to do; they are only in the business of what [the bill] actually does,” he said.

From there, analysts will contact anyone who could provide information about what the bill will cost, including agency employees, interest groups, private corporations and experts within the field, or individuals who can give them raw, nonanalyzed data.

When it comes to possibly promulgating a law within an agency, the analyst may ask an agency employee what the likely steps are and how each piece of the bill may likely be promulgated.

CBO also relies heavily on laws put into action that are similar to the one it is working on. For example, if the government is building a website it might look at other websites of similar size and function to estimate the cost.

“[CBO] has a really hard job, typically projecting the cost of something that is new,” Holtz-Eakin said, explaining the challenge lies in CBO not having anything to compare the new bill with.

The analysts then draw conclusions based on the numbers and information to figure out how much a bill will cost.

But is it accurate?

Hitting the bull’s eye?

“Many of these estimates are not scientific; in many cases they are just the best guess,” Joyce said.

When CBO puts out a number, it should be interpreted more as a range. CBO tries to get to a number where there is as high a probability that the number will be higher as there is that it will be lower, according to Joyce. So, basically CBO shoots for the middle: if it says something will cost $10 million, think in the range of $5 to $15 million.

As far as accuracy goes, CBO can’t really check its cost estimates. Looking retroactively at a bill after it has become law is like comparing baby pictures to an adult; it’s very different.

According to CBO’s website, it is often difficult or impossible to determine the incremental impact on the budget of a particular piece of legislation.

“In many cases, the actual costs or savings resulting from enacting legislation cannot be identified; they may be a small part of a large budget account or revenue stream, and there may be no way to know for certain what would have happened if the legislation was not enacted,” the website said.

Furthermore, most of the bills CBO scores never even end up as law.

However, a few laws have been independent enough from these factors to be tracked. The American Recovery and Reinvestment Act was one of them. Through 2012, ARRA totaled about $555 billion; CBO’s estimate was $36 billion or 6.5 percent off the mark with its prediction.

Others were closer. ARRA spending on transportation spending was about $37.3 billion through 2012. CBO’s estimate was $0.5 billion, or slightly more than 1 percent off.

Regardless of how the final tally adds up, CBO takes its accuracy very seriously because it goes hand-in-hand with objectivity.

“It’s a culture that rewards getting the numbers right,” Holtz-Eakin said.

Past CBO directors have told Joyce they gauge their success in objectivity on whether Republicans yell at them as much as Democrats, according to Joyce.

It is a tough job predicting the future, crunching numbers and taking earfuls from both sides, but CBO is doing it.

-In this Story-

Alice Rivlin, Congress, Congressional Budget Office (CBO), data analytics, Doug Holtz-Eakin, Government IT News, House of Representatives, Management & Budget, open data, open government, Philip G. Joyce, Senate, Sequestration, Tech
TwitterFacebookLinkedInRedditGoogle Gmail