Also, when the barriers to entering a new field are high, the government gets to choose from a whopping 2-3 potential vendors. Some competitive choice!
And right here is where the argument breaks down, because the customer the private companies are competing to serve isn't the public, it's the same manager of a government agency whose motivations we were suspicious of before
Nice.
You get funny effects when private managers try to move their motivation to please the customer directly onto the employees- it just pisses off the employees even more and they end up doing things like secretly spitting in customers' food.
It sounds like what you're describing needs something like objective accountability, but this is a bust too- in publicly funded education, if you try to set up objective standards, then the motivation becomes making those measurements look good (teaching to the test).
1 is important...gov't contracting can work fine, if the system is set up so that you actually get the job based on best performance & lowest price rather than political clout...anonymous bids and all that. There are also functions where the gov't doesn't currently employ people who know how to do this job, and it's a discrete job so that it makes more sense to hire contractors rather than build up and train a set of government employees. This isn't really a question of the gov't being inefficient while the private sector isn't--private companies often contract with other private companies for the same reason.
The real problem is when you hire contractors to do what really are gov't functions: industries that are natural monopolies or have higher barriers to entry so that you have very little recourse if your contractor is incompetent or corruptl; industries that affect people who aren't your paying customers (what financial incentives to private prisons have to see that inmates are treated with a basic level of decency? It's not like they can choose to be jailed in a competitor's prison); etc. etc.
There are also functions where the gov't doesn't currently employ people who know how to do this job, and it's a discrete job so that it makes more sense to hire contractors rather than build up and train a set of government employees.
This is certainly true.
It could be argued that, due to various endemic quirks of human nature, that managers can never consistently translate market incentives into employee performance, because after three or four (or more) levels of management, the attenuation of the signal from the top is just too great. In this case, it wouldn't matter what kind of incentives the top dogs have--organizations over a certain size (say, a few hundred employees) simply have endemic inefficiency and disconnect from market pressure.
This isn't to say that *all* big organizations are like this, only that there's no way to consistently overcome these problems. Any successes at overcoming them are non-replicable.
When a lame government worker is negotiating a contract with a sharp private-sector salesperson, who gets skinned?
When an underpaid government worker is negotiating a multi-million dollar contract with a company that can afford to hire them at double the pay, how long after the contract is signed do they stay on the job?
Forced to choose etween fiscal conservativism and shovelling out money to campaign supporters, how will the Republicans decide?
I know one government agency that has a large fraction of its staff as contractors, but they're pretty firmly entrenched. While there is nominally competition among the contracting vendors, the workers are oddly untouched by it - when the large bid is lost by one vendor, the winning vendor generally offers to hire all of the same people under the new name - this has happened twice in recent memeory. Currently, all of the vendors are Alaskan-native-women-owned corporations, though the relationship of that to the employees is negligible.
One motivation for this, from the point of view of the government agency, is that the federal hiring process is very, very slow. As a result, it's common for hiring decisions of the contractor to have substantial agency input, and for some of the better contractor employees to be poached by the Feds after a few years, where they can run the long process while the person is still working for them.
It may cut costs as well - in contrast to the one person mentioned in the NYT article the other day, the contractors aren't paid as well and have substantially worse benefits than the federal employees.
Hence the Ford Factor, the 3x that costs rose on the UK railways after privatisation in 1996 according to Roger Ford, editor of Modern Railways.
"Hey, thanks."
"Me, too" is the last resort of the incompetent blogger.
Or lazy one, at least. Thanks for spending the time.
The interesting distinction seems to be not between public and private, but between monopoly and competition. The problem with some government work isn't that it's lacking the magic private sector fairy dust, but that it's a de facto monopoly.
Also, can I just say that when we talk about troops in Iraq we should probably count the security mercenaries as troops, and that maybe it's just not a great thing to run a military by contract?
10: I have a friend who has worked at the same job for nearly seven years for three different firms. Not even the office furniture has changed.
My husband's written the same newsletter since the eighties -- he's now its fourth publisher. Or fifth, if you count the last time he was self-published as a separate event.
This seems like a partial summary of the reasons people (usually Republicans) favor forcing decisions down to the lowest level available--it allows for accountability. I've always found that argument attractive, and it has grown (for not good reasons) more attractive with time.
I think that oversimplifies -- devolution can make accountability easier in some areas, but you're going to end up with a lot of stuff that gets done badly that way too.
I agree that privatization is not a magic bullet.
You say:
"... the difference is that the self-interested motivation of the owner of the private company to serve its customers is presumably stronger than the public-spirited motivation of the managers of the agency to serve the public by carrying out democratically agreed upon policies, and so we expect the private owners to be better managers of their companies than public servants are of the agencies they run."
I think there is more to it than that. It is often easier for the manager of a private enterprise to tell when he is doing a good job. Consider being the manager of a chain of supermarkets, you have available a vast amount of data about how well the chain as a whole as well as individual stores is doing. Compare being head of a school district, it is much harder to tell how well you are doing. This makes managing efficiently harder independent of motivation.
Also the manager of a private enterprise generally has more authority over his subordinates than a public sector manager, making it easier for him to impose his goals on the organization.
Regarding incentives I think you understate the extent to which lower level employees can be subject to market incentives. Walmart is a giant enterprise but individual store managers are easily evaluated on the performance of their particular store giving them a strong incentive to manage it well. Sales people and the like are often paid on commission and have a strong incentive to keep their customers happy (sometimes too strong so that they make overly generous deals).
If the government wants to ensure contractors actually do save money, they should start using more creative, results-based contracts. I, in fact, am a government contractor and have seen these in action. One option is called benefits funding. Under a traditional contract, a company would come in and say something like "(process) currently costs you $X per year. If you pay us $20 million, we'll build you a computer system that will let you do it for $Y per year." If the savings aren't realized, the government is SOL. Under a benefits-funded contract, a company says "(process) currently costs you $X per year. We think you can do it for $Y per year if you had a new computer system. We'll build you that system for free and you give us half the difference between $X and $Y for the next 10 years." They have much bigger incentives to do it right that way.
I agree that there is a lot of waste in the way a lot of government work is currently outsourced to the private sector but some of that is based on how the work is contracted out and the contracts are written. Contracting work out can be cost-effective but it's hard work to set up the contracts so that there are the proper incentives.
15, 16: The problem being that the lowest level tends to have more morons than higher levels. Or that's the usual situation, the last six years to the contrary notwithstanding.
18: Attractive conceptually, but sophisticated contractors will quickly get very good at designing compensation formulas that end up working like licenses to steal.
Or, at least, there are a lot more people at the lower levels, and some of them are going to be morons.
If we had a nationwide standardized math curriculum, it'd be a reasonable one -- teaching math at the elementary and secondary levels is something that people have been working on for long enough that anything ridiculous in a nationwide curriculum would get fixed. With every school district in the country developing their own curriculum, on the other hand, probably 90% of them would be fine, and some of those would be great, but the last 10% would be worse than anything that could have happened at the federal level.
It's been my experience that people who think government is as efficient as the private sector have never worked in both arenas. Anyone who has knows that both are inefficient, but it's not even a contest. Government jobs are considered to be a property right from a constitutional perspective, and workers treat it precisely as that, working 9-5 with 1 hour lunches, and surfing the internet and chatting around the office for a good 75% of the 7 hours of "work" in between. The private sector has its own silly set of demons, but it's certainly not this.
18 -- That seems like a good sort of arrangement, and one that I have seen recommended as a way to budget for environmental efficiency savings as well.
Government jobs are considered to be a property right from a constitutional perspective, and workers treat it precisely as that, working 9-5 with 1 hour lunches, and surfing the internet and chatting around the office for a good 75% of the 7 hours of "work" in between. The private sector has its own silly set of demons, but it's certainly not this.
Raise your hands, everyone who comments here and works for the private sector.
I rarely have time to read during the day and have even less to comment. I'm employed in the private sector, but it is on a government contract.
I thought of this thread last night when I saw the play Little Dog Laughed. In it, a character asks a rentboy what it's like to have people pay him for sex and he says that people don't pay him for sex; they pay him to leave afterwards. What a lot of people don't understand is that's also what makes contractors attractive to government -- they don't pay us to build them something; they pay us to build it and then leave. It's the "leaving" part that saves (or should save) the government money -- the $104 an hour mentioned in the article is expensive compared to the hourly rate of a government worker but not so expensive in the long run compared to hiring someone who is then entitled to guaranteed employment, a generous retirement benefit, and health insurance in retirement. If they really only need the extra employees for a year or two or three, that $104/hour can be a good investment. It's when that year becomes forever that the government has a problem.
Everything worth saying has been said by Charlie Sheen.
I would raise my hand, but I don't want my cubemates to ask why.
22: I've worked in both, and Office Space contains much wisdom.
Bob Slydell: You see, what we're actually trying to do here is, we're trying to get a feel for how people spend their day at work... so, if you would, would you walk us through a typical day, for you?
Peter Gibbons: Yeah.
Bob Slydell: Great.
Peter Gibbons: Well, I generally come in at least fifteen minutes late, ah, I use the side door - that way Lumbergh can't see me, heh heh - and, uh, after that I just sorta space out for about an hour.
Bob Porter: Da-uh? Space out?
Peter Gibbons: Yeah, I just stare at my desk; but it looks like I'm working. I do that for probably another hour after lunch, too. I'd say in a given week I probably only do about fifteen minutes of real, actual, work.