Earl Devaney, Chairman of the Federal Recovery Board, Talks About Building and Running Recovery.gov

Devaney leads effort that's reshaping how governments nationwide approach transparency.

by / July 11, 2010

On Feb. 17, 2009, Recovery.gov went live, and the concept of government transparency hasn't been the same since.

The site, along with FederalReporting.gov, was designed to give American taxpayers the tools needed to track how and where $787 billion in federal stimulus funds are being spent. Though initially derided for lacking data and organization, the sites have evolved with more robust data sets, maps and usability features. These efforts now set the standard for shedding light on how public agencies spend taxpayer dollars, and they have become the model for state and local government transparency sites throughout the nation.

Earl Devaney chairs the Recovery Accountability and Transparency Board, which built Recovery.gov and FederalReporting.gov and is charged with maintaining and improving the sites. The 12-member board also is responsible for finding and stopping waste, fraud or abuse of stimulus funds. The board, an independent government entity with 45 staff members, divides its time between website operations and fraud prevention.

Devaney, inspector general (IG) for the U.S. Department of the Interior and a former Secret Service agent, admits he's not an obvious choice to lead an effort that's using the Web to transform government transparency.

"I have been unable to find the member up on the Hill who thought it would be a great idea for a bunch of IGs to build a website," he said. "Having said that, I suspect they wanted a group of people who will probably play it straight down the middle and guard the integrity of the data, not spin it either way politically, and give the American public the best shot at seeing something unfiltered."

The hastily constructed initial version of Recovery.gov was replaced by a $9.5 million redesign in October 2009 that included a host of new functions. And the board is now working with Edward Tufte, who is professor emeritus of political science, statistics and computer science at Yale University. .

"I'd like to think we continue to add new features to the website and do things differently," Devaney said. "We're listening to our stakeholders and our users - they are not bashful about telling us about what they like and don't like."

In a rare interview, Devaney spoke to Public CIO about the creation of Recovery.gov, the future of government transparency, and how these efforts drive the fight against waste and abuse.


In your mind, what's the impact of Recovery.gov on government transparency?

There have been great groups in Washington, D.C., for many, many years - like OMB Watch, the Sunlight Foundation - that have been talking about transparency, and then along comes an administration that actually endorses that. The folks who are doing the IT in the White House are very bullish on it, and so this is a grand experiment in transparency and accountability. I suspect what we do here will create the platform for what comes in the future.

I don't think we're ever going to get back to the point in time when somebody says, "Let's do it the old, nontransparent way." I fully intend to retire in the not-too-distant future. I'm in my 40th year, so the folks who come after me hopefully will have a good foundation to build on. Even after the Recovery Board sunsets, and it's scheduled to do that in 2013, you take the word "recovery" off the door and you still have a viable organization that can monitor and be transparent about the way the government spends big chunks of money for the future, no matter what kind of money it is.

So you're hoping it will transition from a novelty to just the standard operating procedure?

That's a good way to put it. I do think it's transformational. I think that the government is going to do business like this in the future. This is probably the biggest experiment going on right now, and they'll take the good and build on it and we will probably learn a lot of lessons as we go along. For instance, I think FederalReporting.gov, the data collections site, has an enormous capacity to be used for other things like collecting any sort of data from states and other kinds of recipients.

The real difficulty is the fact that we have a variety of different users. We have average Americans who we hope will come on this site and may even return once or twice to see the money being spent in their own neighborhoods. Then there are very sophisticated folks who want to visit our site, take that data and do things with it that perhaps the federal government would never dream about doing. So that's what I hope transparency drives - I hope it drives innovation. I'd like to be the authoritative source for a lot of great websites that ought to come in the future.

When you look at those two different kinds of users and everyone in between, how do you go about designing a website for that broad of a user base?

There has to be some sort of compromise because we can't be so high tech and oriented toward sophisticated users that the layperson can't get on and get something out of it as well. And the sites can't be so basic that they don't deliver what high-end users want. So we threw out the rule book, which basically says you find your user niche and you build a site to address it. Because our niche is everybody, we did a lot of testing, we did a lot of focus groups around the country to see if we were hitting the mark, and the bottom line is we're probably satisfying most users.

Let's talk about the technology that makes it go. How similar is the technology to what runs Maryland's StateStat, Baltimore's CitiStat and those types of applications?

Well, certainly the mapping piece is a redo of those previous efforts. We wanted a site that was highly interactive, and the mapping was something we gravitated to early on. We had seen the Maryland efforts, and that's akin to what we wanted to do. We want to be able to have Mr. and Mrs. Smith go on the site and put in their ZIP code and see what's going on there. We hope as well from an accountability standpoint that when Mr. and Mrs. Smith see that a contract has gone to the mayor's brother who may have a different name, that Mr. and Mrs. Smith are going to click on the fraud, waste and abuse button and let us know. So there's a little method to our madness.

It turns out transparency is a little easier to talk about than it is to practice. For instance, the downside to transparency is often embarrassment. When we went live [with stimulus spending data] in October, there was a lot of outcry about the data. Some of the data was bad. There are 99 data elements that the recipients [of stimulus funds] had to send in to FederalReporting.gov and get right. One of the big snafus was that turns out not many of the recipients knew what congressional district they live in. So they just put in two numbers, they didn't care if they were the right numbers because the system allowed them to move on to the next question. Well, when we had a database full of incorrect congressional districts, that didn't make Congress very happy. There's a technical fix to that; we tied the ZIP codes to the congressional district, and if the ZIP code entered by the user doesn't match the congressional district, the user is told there is a problem. That corrected that problem in the next reporting period, so that data got a lot better.

But there was a lot of embarrassment, and I tend to think embarrassment drives self-correcting behavior. I published lists of people who haven't reported like they should have, and those lists are public. You can download those and reporters can go on there and look. Lo and behold, the next quarter you see those lists starting to shrink. So transparency drives accountability, and it provides a bunch of people who are good at looking at fraud indicators in an enormous database that we've never had before. We're looking at every single contract grant and loan for indicators of fraud. We've never been in a position to do that as an IG community.

With so many agencies getting something out of the American Recovery and Reinvestment Act, how closely can a 12-person board work with state and local governments to seek out that fraud and abuse?

We leverage the resources of the IG community. There are 28 federal agencies that got recovery money to give out, and there are 29 inspectors general that got money to provide oversight. The unfortunate piece is that state and local entities couldn't get any oversight money. We've done an awful lot of training for them. We're trying to leverage our resources so we can help states that don't have the ability to do the oversight that they would like to do.

The board has become a place where state and local folks come to get help. We'll facilitate getting them the right answer, and we sometimes end up holding their hands in reporting situations because states do a lot of bulk reporting and it's difficult. Our technical people work with their technical people right up to the last minute to make sure they get in all that they need to get in. So it's a very good story with respect to leveraging resources and partnerships. There haven't been any sort of issues that I know of, quite frankly, where the feds are saying, "Well no, this is federal business, stay out of our business."

Have there been cases of abuse or fraud that you've been able to nail down. If so, what did you do about it?

So far, we have about 250 cases nationwide being worked by various inspectors general and U.S. Attorneys' Offices. A typical fraud case unfortunately takes about two years from cradle to grave, so you're not seeing yet, although I suspect we're coming up on the time when we might start seeing arrests and indictments being made. This is white-collar crime - it's labor intensive, paper intensive. It's not the kind of thing you're going to see on television where we do a big sweep and load a bunch of people on a bus and duck walk them in front of the courthouse. These are going to come out one at a time, over a period of time. But the bottom line is, yes there has been fraud. Yes, we are working some, but quite frankly, not as much as I would have imagined for the amount of money that is on the table.

Why not?

Normally government money being doled out goes down to the state and ends up in a big pool of money, which is very difficult to track. The Recovery Act requires that money to be tracked cradle to grave. So when it does go down to the state and further down to contractors and subcontractors, it remains recovery money. It's clearly identified recovery money, and we have managed to track that money on the website.

Second, we have what I call a million more citizen IGs who are looking at these projects as they come on board and presumably making an assessment as to whether or not they want to report fraud and waste to us - and they are doing that, by the way. We have reporters who get up every morning and look around the recovery site for their next story. Some people are prolific at doing that, and I suspect that's a deterrent for bad guys as well. Politicians are watching this money like I've never seen in my 40-year career, from the vice president down to governors, mayors and local politicians. Everybody knows this money is very sensitive. If it's wasted, there is going to be a big story about it.

When the websites launched, you got some criticism for not having the kind of data people maybe were anticipating. As you look back on what you've been able to do, how pleased are you?

We're very pleased. After all, we were asking recipients to do something they have never done before. This was all recipient-loaded data. There were a lot of mistakes made during the first round in October that weren't being made in January when the next reporting period was done. And we just finished another reporting period; my initial look at that is that there are even fewer mistakes this time. I always anticipated as the quarters went on, recipients would get better at loading the data in a more accurate fashion. We also get better at doing quality control of that data. Ultimately the site will have probably the most quality-controlled and most accurate data of any government site.

We will never be to the point where all the data is accurate and there won't be any silly mistakes being made. But the agencies will have a much better process in place a year or two years after the recovery mission started than they did in the beginning.

What are the biggest misconceptions that you find about what the board does?

People seemingly don't understand this is very different than in the past, where we are talking raw recipient data and putting it up for the public to see. I always anticipated that the public might not like what they see. They may say, "Oh my God, I can't believe the money is being spent that way." The truth is, that's the way money has always been spent, but they've never seen it. I've come to believe that the public deserves to see how their money is spent. And I suspect over time, if they don't like what they see, they'll vote differently next time. We're not trying to pass any judgments, by the way.

This is another question we sometimes get asked: Why did you give out such and such money to so and so? Well the fact is, we're not in the business of giving money out, nor are we in the business of making judgments as to whether that was a good bridge or not, or whether it was a smart use of money to build a particular road. What I want to do is put the facts up on the website, provide some tools that users can use to slice and dice those facts and then they make their own judgments.


Chad Vander Veen

Chad Vander Veen previously served as the editor of FutureStructure, and the associate editor of Government Technology and Public CIO magazines.

Platforms & Programs