There's a lot of bureaucracy in there that needs changing. Photographer: Andrew Harrer/Bloomberg
There's a lot of bureaucracy in there that needs changing. Photographer: Andrew Harrer/Bloomberg

The scandal at the Department of Veterans Affairs is not going to bring down the administration of President Barack Obama. Nonetheless, it does seem to have legs. Even the administration’s supporters are resorting to half-hearted attempts to blame Congress, or George W. Bush, rather than dismissing the charges as a nonscandal being hyped by ideologues.

As it happens, I believe the administration’s supporters when they say that this scandal was a long time in the making, and that no one president can be entirely to blame. Fundamentally, this points to the difficulty of reforming any institution, and especially a government agency. Yet it also points to one of the cardinal weaknesses of Obama’s presidency: his prolonged hubris about how much a really smart, caring president could change the way government operates.

Obama's experience with the VA is a testament to the danger of big promises -- and high expectations. As Jordain Carney and Stacy Kaper write in their two-part series on the VA for National Journal:

He pledged to end the claims backlog while simultaneously making a string of moves that summoned a flood of new claims to the department.

The administration made it easier for veterans to get compensation for both post-traumatic stress disorder and exposure to Agent Orange -- a Vietnam War-era defoliant now tied to a long list of neurological disorders. Those moves extended help to long-suffering veterans, but they weren't matched by the VA reforms needed to adequately address the new claims. Agent Orange alone took up 37 percent of the Veterans Benefits Administration's claims-processing resources nationally from October 2010 to March 2012, according to a Government Accountability Office report.
And as claims soared during Obama's first years in office, so did wait times. In 2009, there were about 423,000 claims at the VA, with 150,000 claims pending for more than four months (the official wait time it takes a claim to be considered "backlogged"). By 2012, claims had exploded to more than 883,000 -- and 586,540 of those sat on the VA's backlog list.

The administration did request -- and get from Congress -- additional funding for the department. The VA's budget totaled $100 billion in 2009. In 2014, it was up to $154 billion. But that money doesn't instantly transfer into an expanded capacity to meet veterans' needs: It takes approximately two years to fully train a claims worker; the blame for the staff crunch doesn't rest on Obama's shoulders alone.

The influx of claims has since fallen, and the backlog is greatly diminished -- though there is controversy over how the administration has dealt with the claims. (For more on the administration's struggle to fix the VA, see Part One.)

"As a candidate, Barack Obama promised veterans the moon, but in many cases he hasn't delivered as president," said GOP Rep. Jeff Miller of Florida, the chairman of the House Veterans Affairs Committee. "VA's disability claims backlog grew to historic levels on his watch. … Candidate Obama promised the most transparent administration in history, but his VA is a case study in how to stonewall the press, the public, and Congress."

This is the sort of turnaround that a lot of corporate chief executive officers promise: We’ll handle more customers, but faster! Most of them fail, too. And corporate CEOs have a weapon that the president doesn’t: They can fire most of the staff. When looking at corporate turnarounds for my book on failure, I came across a lot of stories of successful turnarounds, and a lot of them started with just that step.

I know, that sounds cruel. Capital against labor! And actually, it is pretty terrible for workers who get the sack. On the other hand, it may be necessary to save the company.

Over time, institutions develop a strong culture, a set of institutional practices, customs and norms that control what the organization is capable of doing. To see what I mean, imagine the staff of the New York Times producing Gawker -- or the staff of Gawker Media producing the New York Times. This is functionally what companies are often trying to do in a turnaround situation: transform a company that had a profitable niche in one part of the industry into the very different sort of company that competes in a different niche.

Consider what would happen if Times publisher Arthur Sulzberger told his reporters they needed to turn themselves into Gawker. If they worked hard enough at it, for a long enough time, they might manage to produce a reasonable facsimile, at twice the cost and half the speed of the folks who do this every day. But almost everything about the group would be wrong for the task at hand. The people who have been selected (and self-selected) into Times-style journalism would mostly lack the interest or the, um, verve to write Gawker pieces. All the rules drilled into them about what constitutes good journalism would be in conflict with what they were now expected to write; their consciences would rebel, and their colleagues would egg them on in feeling aggrieved.* The writers would be unhappy, and constantly fighting to change back. The best ones would leave. So would a lot of the audience.

The best way for the New York Times to turn itself into Gawker would be to fire almost everyone. If you leave too many in place, the old newsroom culture will reassert itself, because the newcomers have no common culture, and so they will tend to assimilate into the existing group. (This is why Americans speak English even though most of our ancestors didn’t.) Firing much of your staff doesn’t necessarily work, but it can. When you can’t fire everyone, it’s very hard to make huge changes.

Note that I’m not claiming that this is a problem of lazy bureaucrats versus committed reformers, or some other tired cliché of the political beat. This is a problem that every organization in need of deep change faces -- and the difficulty of solving it is what keeps bankruptcy courts in business.

But the inability to make this kind of radical change does hamper would-be government reformers. So does the way that our government now functions after decade upon decade of prior reform: which is to say, it prioritizes processes, which can be measured, over outcomes, which often can’t be; rules over discretion; and rights over trade-offs.

What that means in plain English is that when you put reforms in place, you can’t just rip out the stuff that’s not working and do something different. What you’re actually reforming is the process, and because many of the current elements of the process are functionally mandated by other government rules, or court rulings, or bits of legislation that your reform effort didn’t amend, you have to layer your reform on top of the system you wanted to reform, rather than in place of it. Many of your reforms simply stack another layer of bureaucracy on top of the bureaucracy that was already causing problems. This is a problem that CEOs don’t face, unless they’re in some heavily regulated business such as banking or oil refining.

Most important, it is easier to change some parts of the system than others, and much easier to give something than to take something away. So it was relatively easy for Barack Obama to tell the VA that they had to do more to treat post-traumatic stress disorder. It seems to have been very hard to change the claims process to make it move faster or to hire more staff to help things move more quickly. The result was an even bigger backlog -- and, since the reforms commanded the staff to move more patients, more quickly, the temptation to “juke the stats” to make the waiting lists appear shorter than they were.

This is not something that could have been averted by caring harder or being really smart and committed to good government -- which is what Obama seems to have assumed. That’s not his fault. A better understanding of how hard crucial parts were likely to be might have resulted in a more modest reform with better results.

* Not a slam on Gawker, incidentally; writers going in the other direction would feel much the same way.

To contact the writer of this article: Megan McArdle at mmcardle3@bloomberg.net.

To contact the editor responsible for this article: James Gibney at jgibney5@bloomberg.net.