The Department of Veterans Affairs scandal roiling the Obama administration — leading to Friday’s resignation of VA Secretary Eric Shinseki — effectively springs from a classic tenet of public administration. Simply put: what gets measured, gets done.
This maxim has been attributed to various people (for Peter Drucker’s version see here), but the late, great James Q. Wilson put it this way, in his book “Bureaucracy”: “work that produces measurable outcomes tends to drive out work that produces unmeasurable outcomes.”
We’re familiar with this basic dynamic from the standards and accountability movement in education. If you test reading and math, you will teach less history and music. And if one’s pay is linked to test results, you will see “teaching to the test” and even cheating on the test.
But with the VA came another twist. The department had promised in 2012 to see people within 14 days of their preferred appointment date, and even linked administrative performance reviews to that outcome. However, despite five years of substantial budget growth at the VA (from $50.6 billion in FY2009 to $66.3 billion for FY14), there aren’t enough VA doctors to make good on that promise. So, veterans seeking appointments (some 1,700 of them at the Phoenix facility alone, apparently) were not put on official waitlists and thus did not count against the 14-day window. They wound up instead listed on legal pads or in separate files – anywhere, really, they couldn’t be “measured.” And since they weren’t measured, they didn’t have to be “done.”
Is this a reason not to measure? No. The problem with denouncing “teaching to the test,” after all, is that we want students (or the VA) to attain certain substantive outcomes. Covering the ground necessary to get there is, by definition, teaching to the test – but that is only problematic if it is a bad test. Keeping waitlists helped the VA know how far behind its goals it was – but not how to fix the problem. Thus the problem is in the fix — in the management, not the measurement.
While there there has been a wave of complaints about the Obama administration on the management front more generally (for a sampling, see here, here, here), it is true that most presidents find the putative title of “chief executive” hard to live up to. After all, they head an executive branch filled with nearly 3 million employees in myriad, quasi-autonomous agencies answering not only to the White House but to Congress, interest groups, and bureaucratic reality. As Yale’s Peter Schuck recently observed on this blog, the problem of “failure” in government is often systemic. Wilson’s above-cited book is a primer on that point.
Still, presidents can surely help themselves on this front – at the least, to avoid the “friendly fire” of unknown unknowns, as Donald Rumsfeld might have put it. One commonality in the various scandals and pseudo-scandals that have beset the Obama administration, from the VA to the IRS to Healthcare.gov and even to Benghazi, has been that they have been reactive, that the White House has argued it didn’t know what was going on until alerted by the din of the media. This necessitates crisis management to replace the administrative management that should have taken place earlier. As former National Security Agency head Michael Hayden observed, White House attention to the agencies is often “not unlike one of those electronic pet fences: you don’t know you’ve gone beyond your limits until you feel the electric shock.”
Which brings us to another story last week which was rather more upbeat but got a lot less press. The headline was that President Obama selected San Antonio Mayor Julian Castro to head the Department of Housing and Urban Development. But the more intriguing footnote was that current HUD secretary Shaun Donovan will be nominated to direct the Office of Management and Budget (OMB).
Donovan’s remarks that day cited “OMB’s unique role as one of the most critical in government.” He’s right about that. OMB is often referred to as part of the White House staff: it is not, and this matters. Historically and culturally, OMB differs from the White House staff in that it serves the president of the time as part of the institution of the presidency, while the latter centers on a specific incumbent’s personal and political needs. This has been the case since Franklin Roosevelt created the Executive Office of the President in 1939. OMB’s staff are mainly careerists, not political appointees. Their job is to give the president institutional memory; a factual basis for decision-making; and also, crucially, an early warning about what is going on in the bureaucracy, through long-standing processes of central clearance covering not just budgeting but legislation, testimony, regulation, and managerial improvement.
While the agency has had its own difficulties over time living up to the “M” part of its name — not least for some of the structural reasons noted above – it remains the single best tool at the President’s disposal for this purpose. This makes the White House’s seemingly passive-aggressive attitude towards maximizing that asset rather odd. On the one hand, President Obama has appointed very talented people to run the agency. On the other hand, he hasn’t kept them there. Peter Orszag served 18 months; Jack Lew 15 months; and Sylvia Mathews Burwell barely a year. The top position has been left vacant for more time than any single director has filled it. And while Orzsag left the administration, the other transitions are down to the president: Lew left to become White House chief of staff, then Treasury Secretary, and Burwell of course to be nominated as Secretary of Health and Human Services. Likewise Jeffrey Zients, the deputy (and twice acting) director, was recently tasked with the emergency fix of Affordable Care Act implementation and now has become director of the National Economic Council in the White House.
In each case, the president clearly thought he needed OMB’s good people somewhere else. That’s his prerogative – and presidents have notably short time horizons. But that is exactly why they need a stable and empowered OMB — to help them be more pro-active managers, and measurers.