Private Sector Fairy Dust

Skip to: Prison privatization | School Choice | Paying not once, but twice | More Examples | Yes, we do need bureaucrats | Conclusion

It is common knowledge that the private sector is more efficient than the public sector because private businesses have to worry about going out of business, while government agencies do not. The private sector will work harder and smarter, since the need to compete creates incentives for innovation and hard work. Having private companies perform the jobs of government will make everything more efficient.

This common sense is actually entirely wrong.

At the federal level, there has been a push to contract out government work to private firms in order to do more work at a lower cost. This has been a spectacular failure:

The U.S. government is wasting billions of dollars each year paying contractors to do work that could be done for nearly half the price by federal employees...Unlike other studies that compared the salaries of federal employees to their private sector counterparts, POGO's analysis compared those salaries and benefits to what contractors actually billed the federal government for comparable services.

Of the 35 job classifications that POGO studied, contractor billing rates were on average 83 percent higher than what the government pays federal employees. In one instance, contractor billing rates were nearly 5 times higher than the full compensation paid to federal employees performing comparable services. Contractors were more expensive in all but two of the occupations—groundskeeping and medical records technician.

The executive summary is available here.

States have had a similarly disastrous history with privatization of government services. Texas hired the private firm Accenture to process applications for antipoverty programs, replacing a government agency that was not broken. Accenture's processing of Children's Health Insurance Plan and Medicaid applications was so incompetent that tens of thousands of poor children lost or were denied health insurance they were eligible for. Not only was the private firm incompetent, it was more expensive. The entire misadventure cost $100 million more than budgeted. Despite our common sense, public agency was more efficient and less expensive.

Accenture was similarly incompetent in its processing of food stamp applications. Accenture received $899 million to process food stamp eligibility, yet denied tens of thousands of people who were eligible. After three years, Texas is still catching up from Accenture's incompetence. Texan grocery stores have lost out on $1 billion in commerce from foodstamp use, to say nothing of the people (including children) who went hungry. When Texas finally moved to sever its contracts with Accenture, they were forced to pay an additional $246 million in severance fees. Once Texas returned eligibility determination to public workers, it took them years to clean up Accenture's mess and figure out who had been improperly denied benefits. Once again, there were no serious problems with the public agency Accenture was contracted to replace.

Indiana had a similarly disastrous experiment in privatizing their welfare program eligibility services. Indiana contracted out their eligibility services with IBM and other companies, replacing a public agency that was, again, without serious problem. There was no reason for Indiana to replace its public employees; before being privatized, error rates Indiana's eligibility system was below the national average and dropped calls and long waits were unheard of. After privatization, Indiana's error rates skyrocketed to 50% above the national average, and the system was riddled with dropped calls and long lines. The contract was for ten years, but the private firms were so incompetent that Indiana canceled the contract in just two years. At that time, the bloated, inefficient program was already 15% over budget. Though a disaster for people needing services, it was extraordinarily profitable for the private companies involved. IBM is currently suing Indiana for breaking the contract and Indiana may lose yet more money on this misadventure.

Prison privatization

Private prisons have seen incredible growth in recent years as politicians and the public have bought into the lie that the private sector is always better and cheaper than the public sector.

The reality is that any cost savings that come from privatizing prisons is not a result of greater efficiency. Rather, private prisons simply pay their employees less and cut corners on safety. Guards and other staff receive lower pay, fewer benefits, and less training, but are expected to do the same amount of work. The result is a truly disastrous safety record: Inmate on guard assaults are a shocking 49% higher in private prisons compared to public prisons. Inmate on inmate assaults are an incredible 54% higher in private prisons (though one study found that private prisons had more than double that of public prisons). Private prisons have higher rates of escapes and drug use among prisoners.

With more danger, less pay, and fewer benefits, private prison guard turnover is extremely high. Private per-year turnover rate of guards is an incredible 40.9%, more than double that of public prisons, at 15.4%. Texas' private prisons are apparently particularly incompetently managed; the guard turnover rate in Texas' private prisons is an astounding 90%. The cells of a private prison housing juveniles in Texas "were filthy" and "smelled of feces and urine." Inspectors "got so much fecal matter on their shoes they had to wipe their feet on the grass outside."

Texas is not alone in having bad experiences with prison privatization. Arizona's experiment in prison privatization has been a complete disaster. An audit by the Arizona government found that Arizona's private prisons are more expensive than public ones. This is an incredible statement for several reasons. First, private prisons do not adequately train their guards and other prison staff. While inspecting Arizona's private Kingman prison after an escape that led to the murder of two people, an Arizona Department of Corrections official who inspected the facility described the guards as being "not proficient". They simply lacked the proper training to be prison guards. But if the private prison is not spending the extra money on training, where is the money going? Second, Arizona's private prisons pay prison guards and other prison staff much less than do public prisons. They are also offered far fewer benefits. If prison guards and staff are being compensated so poorly, how can the private prisons possibly be more expensive?

This is a recurring theme in privatization schemes. Any cost savings from privatization comes from slashing wages, benefits, and training. Now, you may argue that prison guards are overpaid, over-compensated, or over-trained. But you are not some free market rocket scientist if your "innovation" or "increased efficiency" is simply slashing wages, benefits, and training.

As well as being more expensive, private prisons were less safe and incompetently managed. Arizona's Kingman private prison was so poorly managed that it did nothing to fix its alarm system that issued 200-300 false alarms per shift until an inmate took advantage of the vulnerability and escaped. With prison guards ignoring all the false alarms going off, it is small wonder no prisoner tried to escape sooner. This was the prison where guards were later described as being "not proficient." After the escape, the Arizona Department of Corrections removed the high risk inmates from the prison, but Arizona's contract with the prison owners stipulated that Arizona had to pay for empty beds. Management and Training Corp, the owner of the prison, was more than happy to accept payments for empty beds and dragged their feet, taking a full 8 months to fix the security problems. By the time they fixed the broken alarm system, Arizona had paid $3 million for Management and Training Corp to not house inmates.

Finally, Arizona's private prisons only house the healthiest, least expensive prisoners. How can Arizona's private prisons be more expensive than Arizona's public prisons when the private prisons only house the healthiest, least expensive prisoners?

So, to sum up: Arizona's private prisons are more expensive than Arizona's public prisons, even tough Arizona's private prisons pay and compensate their employees poorly, house only the healthiest, least expensive prisoners, and cut corners on training and safety. Due to these many compromises in quality, Arizona's private prisons are far less safe, with guards "not proficient" in necessary job skills. More expensive and lower quality is the very definition of less efficient. Private prisons are less efficient than public ones.

It has been argued that private prisons have proven cheaper than public prisons in California. While cost per prisoner is indeed cheaper, that private prisons are the solution is pure fantasy. Private prisons in California only house the very cheapest prisoners. Private prisons house exactly zero maximum security prisoners. They do not house death row inmates. They do not house prisoners with serious mental illness. Private prisons do not house juveniles or female prisoners. Private prisons do not house prisoners with serious health problems. In other words, private prisons are only house the very cheapest of prisoners--adult, male, low- or medium-security--while California public prisons have to house the more expensive prisoners. Of course private prisons will cost less than public prisons if they are only housing the cheapest prisoners!

Not only do private prisons in California only house the cheapest prisoners, but they pass a substantial amount of their costs on to the state. Private prisons only provide medical treatment for prisoners up to $2,500. The state of California must pay for the rest. Private prisons do not pay for any treatment of prisoners who are HIV positive. The state must pay for this extremely expensive treatment.

More states have had problems with private prisons. Apalling conditions were found at a for-profit Mississippi juvenile facility:

Walnut Grove Youth Correctional Facility, a juvenile prison in Mississippi operated by the GEO Group, is currently the target of a lawsuit and a Department of Justice investigation regarding conditions alleged to be so horrific that a former resident reportedly calls the facility "the deepest depths of hell." Another former prisoner indicates that violence is so pervasive that it has become "entertainment" for guards. The facility has averaged as many as three injuries per day due to violence. Oversight at the facility is highly questionable, as the GEO Group provides reimbursement for the salary of the individual appointed by the state to monitor conditions.

A lawsuit filed by the ACLU and the Southern Poverty Law Center in 2010 alleges a pattern of horrendous physical and sexual abuse by security staff, use of prolonged solitary confinement, abuse and neglect of mentally ill youth, and failure to provide basic mental health care. While juveniles allegedly suffer in atrocious conditions, private companies including the GEO Group have, according to one report, extracted more than $100 million in revenue from the facility's operation.

Conditions at Walnut Grove were so dangerous that one prisoner was assaulted and raped for 24 hours without help from guards; another has permanent brain damage from an attack. The management of this prison must have been extremely incompetent. If the prison housed unusually difficult prisoners, this level of violence might be understandable in a third world country, but 67% of prisoners at this facility were nonviolent offenders.

A New Jersey commission rejected a proposal to build private prisons, arguing:

Most objective cost studies show little or no cost savings to taxpayers coupled with an increased safety risk ... [P]rivatization does not appear to be a viable option for Monmouth County's maximum security facility due to the potential increased risk of liability and safety risks without proof of cost savings.

A meta study found no evidence that private prisons are cheaper than public prisons (here is an older meta study, published in 1999, demonstrating how politicians have ignored at least a decade of evidence). If there is no cost benefit for private prisons, why go with private prisons, which are less safe, offer less training, worse pay, and fewer benefits for prison staff?

Speaking of ignoring evidence, the Justice Department under President George W. Bush reached some very interesting conclusions about the incompetence of private prisons in 2001. It had been estimated 20% savings could be had through privatizing prisons, yet the Bush Justice Department found that privatizations had actually resulted in only 1% savings. Yet, prison guards and staff were being compensated far less. Where was all this money going?

The Justice Department found that private prisons were designed in essentially the same way as public prisons.* Private prisons were expected to save money in part because they would not function in the same way. It was expected that the private prisons would "innovate" and find new, more efficient ways to house prisoners. Obviously, the private sector has not been able to do so, though it is difficult to see how they could have. A prison needs guards, cooks, janitors, people to work the cameras and security systems (etc). How can this possibly be improved? Once again, you are not some free market rocket scientist if your "innovations" are simply slashing pay, benefits, and training of employees. We should not be surprised that the Justice Department found that private prisons worked essentially the same as public prisons. So much for private sector innovation.

*The United States Conference of Catholic Bishops issued the following resolution in 2000: "We bishops question whether private, for-profit corporations can effectively run prisons. The profit motive may lead to reduced efforts to change behaviors, treat substance abuse, and offer skills necessary for reintegration into the community." The bishops were not the first to point out the perverted incentives of private prisons. The goal of a prison should be rehabilitation--getting criminals to the point where they can be contributing, law-abiding citizens. But private prisons are driven by profit, and thus have no reason to rehabilitate their prisoners. Their profits will be higher if prisoners, once they leave prison, commit a crime and return to prison. This issue is was not taken up in this essay.

School Choice

Privatization has not worked for education, either. Conservatives have long championed the idea of school choice. Instead of public schools, parents are given vouchers to send their children to the private school of their choice. The idea is that schools must compete with each other for the vouchers, thereby unleashing the power of the free market. Poor student performance will result in parents pulling their children from the school and taking their voucher money with them. Bad schools will thus "go out of business." While this is in principle a good idea, it simply does not work in practice. The biggest experiment in school choice occurred over two decades with 20,000 students in Milwaukee, Wisconsin. The result in Milwaukee, as with everywhere else a school choice program was implemented, was complete failure. Tests comparing students in Milwaukee public schools and students in Milwaukee's choice schools showed no significant difference in student achievement. In other words, the education was just as good in the public schools as the choice schools. (If you are interested, here is the complete study)

Remember, the goal of school privatization was higher quality education, but this clearly did not occur. Nevertheless, proponents are still calling school choice a victory because it can provide equal education at a lower price. However, this is nothing more than a replay of the strategy in prison privatization. Private prisons (1) cut labor costs by cutting prison staff compensation, (2) only accept the cheapest prisoners (those without health problems, for example), and (3) shift as many costs back to the government as possible (for example, by capping per inmate medical costs). Choice schools follow the exact same strategy. First, Milwaukee's choice school teachers have far lower compensation than Milwaukee's public school teachers. You may argue that Milwaukee's public school teachers are overcompensated, but you may not argue that simply cutting teacher compensation represents a private sector "innovation" or "greater efficiency." Second, 19.5% of students in Milwaukee's public schools have a disability, while only 1.6% of choice students do. Obviously, students with disabilities are far more difficult and expensive to teach than students without. Choice schools would clearly not be so cheap if the proportion of students with disabilities increased more than ten times. Third, choice schools can expel students with behavioral problems, sending them back to the public schools. The Milwaukee public school district obviously does not have this luxury. They cannot simply expel a troubled student and leave their education up to somebody else. Obviously, having students with at-times dangerous behavioral problems is more expensive than not.

Finally, choice students have a potentially large advantage over others. Choice vouchers were allocated by lottery; some of the parents of public school students were invested enough in their children's education to enter the lottery, but all of the parents of choice students have this investment. Parents of choice students are, on the whole, more involved with their children's education. Despite this advantage, choice schools made no difference.

In the United States, we are obsessed with privatization and market-based strategies. The education system in Finland is consistently rated best in the world (or close), easily outperforming many other countries, including the United States. Yet Finland has no private schools. Although it would be foolish to assume that what works for a small country like Finland will work for a large country like the United States, Finland's success gives us more than ample reason to question our assumption that making education more like the free market will ultimately yield smarter students. (Finland and the United States may be more familiar than you think--follow the link for more information).

Overpay once, then pay again for antipoverty programs

At the federal level, private firms contracted to perform government responsibilities typically pay their employees so poorly that one in five workers earn sub-poverty wages and do not have health insurance. This is another incredible statement. As we saw above, federal contracts awarded to private companies, on average, cost almost double what public employees cost to do the same job. Yet private contractors pay their employees so little that a significant number of them are literally impoverished. And when people are impoverished, they become eligible for Medicaid, food stamps, the Earned Income Tax Credit, and other costly government antipoverty programs. In essence, the government winds up paying twice for privatization--once to overpay a private firm to do the work a public servant could do for much less, and again to cover the antipoverty programs for these private firms' employees.

More examples

The city of Chicago's experiment in privatizing parking meter enforcement has been a complete disaster, with huge hikes in meter rates, broken meters, and loss in business revenue due to customers' difficulties parking.

Other examples can be found here. Some highlights:

  • An audit report from the Wisconsin Legislative Audit Bureau revealed that the state's department of transportation wasted more than $1 million by outsourcing almost half its engineering work to private contractors over the past five years. The audit found that about 60% of these outsourced jobs could have been done at a lower cost by state workers, which would have saved the state $1.2 million.
  • In 2010, Gary, Indiana canceled its 10 year contract with United Water. In May 2008, a state inspection found that the district, under United Water's management, had violated discharge limits 84 times between 2005 to 2007; had at least 25 pieces of broken equipment; filed inadequate monitoring reports; and failed to meet mandated deadlines. By cancelling the contract and bringing water service back in-house, the city expects costs to decrease from $16 million to $8 million a year.
  • In a Cincinnati Enquirer investigation, the newspaper concluded that the State of Ohio and many local governments engaged in "casually administered" contracts with "lax controls." From 2000-2003, 116 state audits found that contactors misspent $97.7 million tax dollars.
  • In September 2008, Indiana was required to reimburse the private Indiana Toll Road operator $447,000 for tolls that were waived for people being evacuated during a severe flood. This requirement in the contract forced the state to pay money to a private contractor in order to ensure the public's safety.

(follow the link for more examples)

Yes, we do need bureaucrats

Here is a fascinating article about how cutting federal employees has generally resulted in increasing costs. How can this be?

The average voter may imagine federal bureaucracies as overstaffed, full of people leaning on their rakes and sharpening their pencils. But the truth is, most agencies are, if anything, understaffed. The government has grown tremendously in its spending and scope since the 1960s, and the population of the nation has grown by a margin of 100 million people, but the size of the federal workforce has remained remarkably static at about 2 million...

This is not to say that there aren't big bureaucratic reforms that need to be made that could lead to people losing their jobs. Many agencies, for instance, exhibit excessive layering in their management ranks (think job titles that start with "deputy-" or "under-"). And if Congress and the administration could agree to lift some of the outdated procedural requirements and redundant reporting demands that are the bane of the average civil servant's life, it might be possible for agencies to fulfill their mission as well or better with fewer people.

The result of our understaffed bureaucracy is that agencies are not able to do their jobs fully. This is a cynical, vicious cycle. Politicians cut the agencies' budgets to the point where they cannot possibly do their jobs fully. Politicians then attack these same agencies for not fully doing their job, using the failure for justification to cut their budgets further.

In any case, when government agencies do not do their jobs fully, the cost of the resulting mishap is greater than the cost of having a properly staffed and managed bureaucracy. The article highlights several examples:

Chronic manpower problems at the Minerals Management Service—the office within the Interior Department charged with regulating offshore oil wells—stretched all the way back to the 1990s, when a long boom in deepwater drilling coincided with a bust in the agency's funding. In December of 1996, the year when the agency's budget bottomed out, the Houston Chronicle reported that offshore fires, explosions, and blowouts had increased by 81 percent...since 1992.

Obviously, dealing with blowouts is more expensive than having a bureaucracy that can properly do its job. This was dramatically illustrated by BP's Deepwater Horizon oil spill in the Gulf of Mexico. Here is more about how competent regulation (led, of course, by federal bureaucrats) could have prevented the disastrous oil spill that spewed five million barrels of oil over a five month period into the Gulf of Mexico. Obviously, the economic and environmental damage this caused more than exceeded the costs of competent regulation. According to a federal investigation, "The root causes [of the spill] are systemic and, absent significant reform in both industry practices and government policies [regulating offshore drilling], might well recur."

The Securities and Exchange Commission:

"Preoccupied with its own staff reorganization," the FCIC says, the supervisory program went more than a year without conducting a major examination...In an op-ed headlined "Muzzling the Watchdog," three former heads of the commission wrote that the SEC "lacks the money, manpower, and tools it needs to do its job." [...] If there were any place where the federal government might have had a fighting chance to fend off or at least ameliorate the worst financial crisis since the Great Depression, it was at the SEC in the mid-2000s. Instead, the SEC divested itself of personnel and initiative. Consider the breathtaking consequences: not just the $700 billion bailout (most of which has been or will be paid back), but $400 billion in lost federal revenue as a result of the recession (that's in 2009 alone) and the $800 billion stimulus to get us out of it. Suddenly, shaving a few million dollars from the overhead costs of the SEC doesn't sound like much of a bargain.

Weapons acquisitions:

Every year, the Government Accountability Office analyzes a portfolio of major weapons contracts to see how the Pentagon is handling its acquisitions process, and the past decade has seen a staggering trend of increased cost overruns. In 2000, the average weapons system contract ultimately cost 6 percent more than originally projected. In 2009, the average weapons program ended up costing 25 percent more. Cost overruns for that year alone amounted to $296 billion. Among the causes of the problem, the GAO's 2009 report cited "shortages of acquisition professionals" as well as "degradation in oversight, delays in certain management and contracting activities, increased workloads for existing staff, and a reliance on support contractors to fill some voids."

Actually, the Pentagon is not the only branch of the government to have problems with weapons acquisitions. This article details how the Coast Guard spent a whopping $24 billion on ships that are so poorly built that they can not even be used. This should come as no surprise, since the Coast Guard had no requisition experts (bureaucrats). Requisition experts are trained engineers and businessmen who have the technical and business expertise to understand the weapons systems and ships being purchased. Because the Coast Guard did not have such a bureaucracy--expert businessmen and engineers--the systems they purchased turned out to be entirely useless. Paying bureaucrats to ensure the job was done right would have ensured that $24 billion was well spent. Instead, the Coast Guard is stuck with $24 billion worth of ships and systems that do not work. This is a danger to crews and people in American waters; a search and rescue mission was aborted because of a hull crack on a brand new ship.

Now, when this sort of nonsense is occurring, the executive branch of the federal government should be able to track down such waste. This is the responsibility of the Office of the Management of the Budget. But the entire staff of this office is 500 people. Obviously, 500 people cannot possibly monitor everything that the federal government does. But, as we have seen, this sort of oversight is clearly needed.

The paradox, then, is this: if the aim is to reform the civil service in order to put a lid on federal spending, what we really need are targeted increases in the federal workforce. A wise first move would be to double the size of the OMB.


Private firms are not always more efficient than the government. There is no "private sector fairy dust". In the private sector, good firms will grow and expand; bad firms will lose money and go out of business.

Government agencies do not have this feature. Good agencies will survive and thrive, but so, too, will the bad ones. While this is certainly a problem in the public sector, by no means does it make a private firm automatically better than a public one. If a public firm is well-run, it will be as good or better than any firm in the private market. But a poorly-run public firm will not, unfortunately, go out of business.

If you are a conservative reading this, you probably thought that Accenture--the application processing firm hired by Texas--was a poorly run company. Obviously, the plural of anecdote is not evidence, and not all private firms hired by a state will be so incompetent.

But this is precisely the point. When Accenture got a contract from the Texas government, they were no longer at risk of going out of business. Governments--privatized or not--do not work this way. If a poor family is trying to apply enroll their children in the Children's Health Insurance Program or Medicaid and get incompetent service from Accenture, their children still need health insurance. Competent service or not, they are forced to participate in the market. Public programs simply do not work like a private marketplace.

So, if the advantage of the private market is that poorly managed firms lose business and eventually go under, this advantage is taken away when a state contracts out to private firms. With a fat government contract, Accenture had no danger of losing money for incompetent service. Accenture should have been better managed--this is true. But if the advantage of the private market is eliminated, why should we expect a private firm to outperform a government agency? Is it because we believe in private sector fairy dust?

So, think of a dichotomy of firms not as public vs. private, but as well-run vs. poorly run. The Veterans Health Administration in the United States* is illustrative. Through the early 1990's the VA health care system was a national disgrace. But today it is--without exaggeration--the best health care system in the entire world:

Indeed, the VHA's lead in care quality isn't disputed. A New England Journal of Medicine study from 2003 compared the VHA with fee-for-service Medicare on 11 measures of quality. The VHA came out "significantly better" on every single one. The Annals of Internal Medicine pitted the VHA against an array of managed-care systems to see which offered the best treatment for diabetics. The VHA triumphed in all seven of the tested metrics. The National Committee for Quality Assurance, meanwhile, ranks health plans on 17 different care metrics, from hypertension treatment to adherence to evidence-based treatments. As Phillip Longman, the author of Best Care Anywhere, a book chronicling the VHA's remarkable transformation, explains: "Winning NCQA's seal of approval is the gold standard in the health-care industry. And who do you suppose is the highest ranking health care system? Johns Hopkins? Mayo Clinic? Massachusetts General? Nope. In every single category, the veterans health care system outperforms the highest-rated non-VHA hospitals."

Remember, the VA is "socialized medicine." Unlike Canada, where physicians, hospitals, and clinics are private, even VA doctors, clinics, and hospitals are government owned. Thanks to the hard work, leadership, and innovation of government employees, the VA went from being a national disgrace to the very best health care in the entire world. They are also more efficient; cost per patient and per procedure is vastly lower than the American private health care sector.*

For another example, we briefly looked at Finland's schools, which are entirely public (there are no private schools whatsoever) and the best in the world. While Finland's schools have always been entirely public, they have not always been the best in the world. In the 1980's, Finland's education was among the worst in Europe. But with hard work, skill, innovation, and perseverance, Finland's government bureaucrats turned Finland's education system from among the worst to the very best.

I lived for a summer in Guatemala. Guatemalans talk about their "public" buses, colorfully painted and extremely uncomfortable school buses, and the "private" buses, modern and comfortable coach-style buses akin to American Greyhound buses. The vast majority of transportation in Guatemala--both within a city and between cities--occurs via the "public" buses.

But make no mistake--the "public" buses are in no way public. The drivers are not paid by the government. The government does not design routes or service. The "public" buses are about as pure a free market as exists in the entire world. If a Guatemalan thinks he can make money operating a bus, he buys a bus, designs a route he thinks needs service, and starts driving. Despite the fact that there is no centralized planning, the buses are quick, efficient, and frequent. Everyone quickly gets to where they need to go, and the bus operators make a decent living. It truly is amazing what a free market can accomplish.

When a free market works this well, the government should keep their hands off. But--as we have seen--there are times where the free market does not work very well. It is the responsibility of the government to make sure that these unusual markets operate smoothly, just as it is the responsibility of the government to leave alone the private markets that are working without problem.

*This will be explored in more detail in a different essay, but it is not done yet. When it is done, it will be posted to the blog, the atom feed, and will be linked to here and on the homepages.