Regulatory capture happens when Government officials bias policies in favor of the firms supposedly being regulated. Usually there is no overt corruption involved in capture. The bias can exist even for the most scrupulous regulators because it often works at the subconscious level. Economists usually discuss the capture issue with respect to the finance, healthcare, and other industries, but what about defense?
Luigi Zingales has a nice framework for thinking about capture. He finds three main channels through which the regulators become captured by business: (1) career concerns; (2) information; and (3) environmental pressure.
(1) Career Concerns.
The prospect of a high-paying job in the regulated industry is the most discussed aspect of capture. This channel is strongest when the wages of the government official are far below that of equivalent positions in industry.
It is no secret that top defense officials often make their way into lucrative positions at contractors. This has been going on for some time. Back in 1969, Senator William Proxmire found that the top 100 defense contractors employed 2,072 retired military officers of the rank of colonel or Navy captain and above. Lockheed alone had 210 of them. Proxmire complained:
… former officers may even negotiate contracts with their former fellow officers…
In addition , there is subtle or unconscious temptation to the officer still on active duty. After all, he can see that over 2,000 of his fellow officers work for the big companies. How hard a bargain does he drive with them when he is one or two years away from retirement?”
In 2006, a GAO survey of Internal Revenue Service data found that 52 contractors employed 2,435 former Department of Defense senior military and acquisition officials. Today, we see the Project on Government Oversight picking up the reins and creating a new database on the revolving door. (I wonder whether it counts General Petraeus going to a $73 billion private equity firm.)
One way to mitigate the revolving door problem is to increase compensation for government work. Singapore paid its Prime Minister about $3.8 million in 2011. That compensation has come down and Singapore is relatively authoritarian, but even the British paid their bureaucrats much better in WWII than the US. The wage differential between Government and industry was pointed to as a source of wartime inefficiency for the US.
There are, perhaps, more politically palatable ways about it than overhauling compensation (which still needs to be done). For example, shaming or other cultural norms could be effective. Maybe promoting the right people with the right skills, regardless of age or rank, could be effective at keeping the best in Government.
At any rate, “The best cultures are aspirational.” That means giving your best people the freedom to make innovative decisions and be held into account for those decisions. When employees feel like they’re truly contributing to new technologies and the nation’s interest — rather than executing the standing orders of a bumbling bureaucracy — then concerns from a revolving door can be mitigated.
(2) Information
It is very often the case that regulators need information from the industry they are regulating in order to make good decisions. In one version of this channel, you may see Government officials bargaining with industry to obtain the information, providing a way for the regulated to “trade” information for favorable treatment.
A stronger version of the information channel is that even if industry provided raw data because it was required by law, the narrative coming out of the data biases the analysis toward the interests of industry.
Defense officials might know their own military requirements, but they generally don’t know about technology solutions. They relies on the contractors for information on engineering specifications and pricing.
The Department of Defense has long ago stripped itself of production knowledge. The Army and Navy used to have strong in-house technical capabilities. Over the course of the 1960s and 70s they were reduced to relatively small technology labs which no longer participated in major systems development. Technical knowledge and knowledge of pricing are very often found together, particularly when we move out of the realm of commodity goods.
A clear example of the kind of information capture I’m talking about comes from contractor business base analyses. For each major contractor, the Government receives a ton of data on current and expected contract orders, fringe costs, overhead allocations, and so forth. This justifies pricing rates for contract proposals, as well as reimbursement rates for incurred costs.
The faulty assumption is that a growing amount of direct contract work at a plant always reduces overhead rates, making the next unit of output cheaper than it would be otherwise.
Therefore, Government officials find themselves planning the business base of contractors. Officials want a steady or increasing amount of work at each business unit to prevent spiraling overhead costs. And this is a major barrier to entry for new firms.
Generally the way it happens is that the Government provides an orderly competition between major contractors which ensures a “fair” return on capital for the destruction of excess capacity.
For example, if a contractor is lacking future orders, they will signal that to the Government by justifying higher rates in the forward pricing rate proposal. Usually this would make the contractor less competitive and eat into profits. But, to keep rates from going out of control — possibly ending in market exit — defense officials funnel contracts back to the supplier.
In this way, what’s best for the contractor is also seen by the defense official to be what’s best for national security. A steady share of the work keeps production predictable and rates low — minimizing costs, maximizing complacency, and destroying competition.
This was but one half-explained example of many in which information bias plays a role.
What’s interesting is that defense officials are not just regulating market activities, as though they can sit and observe from the outside. Its not like regulating health insurance or fuel emission standards. For defense, officials are working in concert with contractors to create or operate weapon systems. It is a joint venture in the “production” of national security. The Government relies heavily on contractor information to do its own business.
(3) Environmental Pressures
Because of the tight relationship that arises when the regulated is doing business directly with Government officials, it is also true that nearly the entire audience concerned with regulatory policy is concentrated in the contractors. Bias seeps in from the environment.
This is Zingales’ third channel. The regulators gather formal training in the business area and become impressed with its importance. They also gather information from the regulated which often leads to social activities. The Government regulators start adopting the views of the regulated industry.
When Government officials make policy, most of the feedback comes from industry. After all, it is their operations and bottom-line at stake. If the official can err to one side or another on a big issue, he’d rather err in favor of industry because the public isn’t going to notice. If he erred toward the public’s interest — in favor of competition — then he’d certainly hear about it from industry with knock on effects on his career.
What Zingales found is that the process can create a situation where the regulators are biased — even if just unconsciously — toward policies that are pro-business and anti-market.
The Federal Acquisition Regulation, for example, is a monstrosity by any measure. Costs of FAR compliance are such that new entry is almost prohibited, even with inducements for small business and others factors. Scaling to a major prime would entail many obstacles in MIL-SPEC, auditing, program management systems, wage negotiations, documentation and reporting, inspections, oversight reviews, and so forth. That’s why it has basically never happened. We’ve only seen consolidation of the defense industry within WWII era firms (SpaceX aside, which nevertheless has received a huge, if temporary, pass on a number of regulations).
You never see Lockheed or Boeing complaining bitterly about the FAR. When a new rule is under consideration, contractors don’t fight against it. They fight to influence the language. Whenever a new policy goes through coordination at the Washington Headquarters Service, it has a comment period. You might not be surprised that nearly all comments come from industry. From observing a few such policy changes, I would estimate that the responses followed a 10/1/0 ratio: 10 comments from industry, 1 from the bureaucracy, and 0 from the rest of the public.
An even bigger hurdle to new firm entry than the FAR is the entire requirements, acquisition, and budgeting process. It takes a very long time, requires a huge amount overhead costs, and necessitates being “plugged in” at the start. The big primes would never criticize this process — they are the only ones that can play in it. Now of course it’s complicated because Congress has made much of the process by law, but whether the policy comes from within or above the bureaucracy, industry’s voice is the only one giving feedback.
Officials don’t get lobbied from the firms outside of defense, who might happily enter defense under some alternative regime. The problem is that they couldn’t even imagine such a welcoming defense regime. And even if they could, there’s a slim chance that the logic behind it could conquer all the existing environmental pressures toward the status quo.
(I’ve argued in various places that organizational budgeting could benefit competition, for example, here, here, here, here, here, and here.)
Conclusion:
Capture is a complex problem that manifests itself in many different forms, including career concerns, information, and environmental pressure. The way it all plays out depends on the particular industry and context.
For defense, the capture process is especially problematic because it biases the military force structure toward legacy platforms and higher costs. Big bureaucracies in industry and the DOD need competition from outsiders who are not attached to the equipment and methods of the past. Usually the bureaucracy rejects disruptive innovation before its feasibility can be proven because of honest disbelief.
The one true obstacle to competition — and the innovation it excites — is barriers to entry. However, barriers don’t only reside on the supply side. Ultimately, the suppliers must cater to the demands of the consumer. Competition in industry therefore requires pluralism on the Government’s side, a kind of tolerance that accepts competing developments of new system concepts.
And yet despite the arguments above, the capture process also provides for a conversation between military requirements and technology development. Its ends are not only legitimate, but often productive. If requirements reside in the Government, and technology in the contractors, then there must be some efficiency to the mixing of personnel, information, and culture. Indeed, this “cozy” relationship was thought to be optimal for high-technology sectors because of the long lead times associated with complex systems. I will pick this point up in a later post.
Leave a Reply