The Washington Post has issued parts 2 and 3 of its series titled Top Secret America. Part two examines the role of private contractors. Jonathan Bernstein’s recent blog posting, cross-listed here at Salon, discussing the first part of the WaPo series is well worth reading, not least for the vitriol it elicited from some progressive bloggers because Bernstein had the temerity to suggest presidents do not control the national security apparatus. That pushback is indicative of a widespread misunderstanding, I think, of just what the federal government does, and how it does it.
Consider part two of the WaPo series, which deals with the role of private contractors in government. (This also gives me an opportunity to plug the work of two of my colleagues, Kateri Carmola and Allison Stanger, both of whom have written books about private contracting – see here and here)
There are two points to keep in mind when reading this portion of the WaPo series. First, the outsourcing of services to private contractors did not start with the War on Terror, and it isn’t restricted to the national security realm. Second, we should probably avoid blanket generalizations in assessing this practice. Any effort to do so needs to weigh the tradeoffs involved – the gain (or loss!) in efficiency versus loss of accountability, for instance – a calculus that I’ve no doubt differs across different policies areas and transactions.
The contracting of services to private firms, however, helps reinforce the larger point I’ve made regarding the misconceptions of presidents’ ability to control the bureacracy – misconceptions evident in some of the comments to Bernstein’s post. The following table, based on data gathered by Matt D’Auria for an article I just finished on this topic, shows the increase in private contracting just since 2000 (which is as far back as Matt could find data):
The trend toward greater reliance on private contractors, however, is simply a small slice of a larger movement dating back decades in which much of what government does is outsource functions to third parties. Consider the following table that shows the annual percentage increase in the government spending and regulations compared to the number of employees since 1940:
It shows that federal employment has grown incrementally since the end of World War II, whereas the growth rate in the size of the federal budget, and the number of federal regulations issued, has skyrocketed. What explains these different trends? It reflects what Hugh Heclo first identified in the 1970’s as a movement toward “government by remote control.” Simply put, most of what the federal government does is cut checks, either to pay for services provided elsewhere or for income maintenance programs (social security, Medicare, unemployment, etc.) In other words, the government is not in the widget-making business – increasingly it’s in the outsourcing business
Note that not all of this outsourcing goes to private entities; there’s been a steady increase in federal aid to state and local governments as well, as this chart indicates:
Now, this federal money comes with strings attached – that’s Congress’ means of holding recipients accountable. Look again at the chart above (Figure 2) showing the explosion in the number of pages of the federal register, which lists government rules and regulations. That’s the cost of doing business via third parties. Some might call this needless red tape, but that judgment can’t be made in the abstract – one needs to weigh any gain in efficiency with outsourcing via the cost of maintaining accountability through rules and regulations.
What does this mean for the President? Consider the Gulf oil spill. Clearly the MMS proved lax in its efforts to regulate the oil industry. As I’ve argued elsewhere, however, that cozy relationship was implicitly encouraged by members of Congress who were mostly focused on the revenues generated by off-shore drilling. When the Deepwater Horizon explosion occurred, however, the public didn’t hold Congress primarily accountable – they looked to the President for explanations. That includes not just understanding the faulty permitting process conducted by the MMS, but also responsibility for plugging the leak, and for the Gulf cleanup. And where does Obama turn to accomplish these last two goals? Toward a private company – BP – still another manifestation of government by remote control, with an emphasis on remote rather than control.
Many of my political science colleagues, as well as those who criticized Bernstein for his post on this topic, think that because the president is a single actor, he can exercise dominant influence over the executive branch. The reality, however, is that a system of government by third party does not lend itself to presidential control – but the unitary nature of the executive office does make it easier to blame the President when bureaucracies fail. In this respect, presidents get the worst of both worlds. The very diffusion of responsibility that accompanies a government exercised by remote control increases the tendency for people to want to hold someone accountable when things go wrong. That someone is usually the President.
Matt, do you think much of the public’s misconception about the President’s power comes from his “commander in chief” responsibilities. That power is very different from the bureaucratic powers he shares with Congress.
Jack – I’m not sure how much to blame on the commander in chief functions. Certainly we’ve seen instances, such as after 9-11, where the C-in-C functions loomed large in the public’s perception of presidential power. But even within this policy realm I think it is too easy to overstate the President’s real influence. Consider the McChrystal comments, and Obama’s choice to expand the war in Afghanistan. In one respect, of course, these (firing McChrystal, expanding the war) were Obama’s choices. But McChrystal’s comment reflect a military sentiment that suggests they had taken the measure of the president, and found him wanting. Obama’s expansion of the war, meanwhile, was based on information and options provided to him by his military commanders, and the success of the policies he has chosen will depend on their effectiveness in implementing them. More generally, I’m not sure there really is a huge difference in the president’s powers in national security versus other policy areas. There may be some difference in isolated cases, but on the whole, I tend to believe the difference is overstated. But this is an issue that deserves longer treatment than I can give in the comments section.