A good chunk of those savings comes in the form of reduced labor costs. Today, 30% of public-sector workers are unionized, typically receiving pensions, good health benefits and better-than-average pay. Just 7% of the private sector is unionized. (Public-sector unions also tend to support Democrats -- another reason privatization is a big issue for the largely Republican Tea Party movement.)

"Oftentimes it's about breaking unions," says Dean Baker, a co-director of the Center for Economic and Policy Research, a progressive Washington, D.C., think tank. "Insofar as you can get a lower-cost work force, you can get savings."

Cutting costs by driving down wages

Like many public agencies, Jackson County, Ore., considered the bottom line when it tried to reopen its 15 library branches, shuttered after a loss of federal funds in 2007. While negotiating with the county workers union, it put management of the libraries out for bid.

Library Systems & Services, the only private company to bid, won the contract with a bid that cut costs by about 40% over what the union had proposed. With library hours reduced, the company rehired 70 of the 110 laid-off library members, at their same salaries but with reduced benefits, says Amy Blossom, the manager of the Ashland branch.

An analysis by the Oregon State Library found that while overall library staffing in Jackson was reduced by 36%, the number of librarians was reduced by 52% and the number of those with advanced degrees -- common for the position -- by 57%.

This net result -- lowered compensation overall and fewer benefits -- is typical with privatization, according to the American Federation of State, County and Municipal Employees, a union.

"It's certainly been one of the factors in the growing inequality" between rich and poor that led to the recession, Baker says. "You get rid of those jobs, you put more downward pressure on the wages of other jobs."

Critics also say there's no guarantee that work will be done less expensively, or better, with privatization. In fact, the accounting books at private companies often remain closed. Jackson County, while happy with Library Systems & Services, has no idea exactly how the company is spending tax dollars. (The company declined to comment for this story.)

That veil of secrecy can be costly. Consider the case of New York City's hiring of computer consultant Science Applications International for a project called CityTime. After more than a decade, there's no end in sight, and the New York Daily News reported recently that more than 400 consultants on the project have billed the city an average of $400,000 a year. The city's controller is looking into why a project that was supposed to cost $68 million has cost more than $700 million, the newspaper reports.

Asked to comment for this story, Science Applications provided written statement saying: "CityTime is working now for 73,000 employees with a 99.9 percent accuracy rate and SAIC is eager to have the system up and running in the remaining agencies once those agencies give us the green light. We believe CityTime brings a great value and savings to New York City taxpayers and we are pleased to help ensure its completion."

Ironically, the project was for equipment to prevent time-clock abuses by city employees.

Privatization "is a very mixed bag. Just because something is done in the private sector doesn't mean it's done more cheaply," says Baker, particularly once you factor in "highly paid executives."

"A government that's incompetent to deliver a service is not going to be any more competent to monitor a contractor," Sclar adds.

Get rid of government?

Even when privatizing services does save money, the question remains: How far do we want to go?

Privatization purists say taxpayer dollars should never help pay for programs that would not turn a profit in the private sector or where there's a private alternative. Some Tea Party candidates have gone as far as calling for privatization of Social Security and Medicare, the nation's biggest social programs.

Take Amtrak. Experts think half of its routes would be profitable if run by private companies, says Chris Edwards, an economist and the editor of Downsizing the Federal Government, an online guide of the libertarian Cato Institute. The routes that wouldn't be profitable -- largely in rural areas and outside the Northeast -- "don't make any sense and shouldn't be run," he says.

"I don't think anything should be off limits," he says. He'd like to see the Federal Aviation Administration -- essentially the police of the skies -- privatized, along with airports. The media and federal accounting office could monitor operations, but ultimately customers would drive performance.

If Dulles International Airport in the Washington, D.C., area hired a lousy security company, passengers would be more likely to use nearby Reagan National Airport instead, Edwards says.

Another area ripe for privatization: highways and toll lanes, something that's already being done in some areas. Let private companies raise capital and charge fees, and if you can't pay, slum it on the public pavement.

Life in second class

But how much of this would the public stand for?

Sure, we'll accept first class in the skies. In fact, bizarre pricing gaps among travelers make flying affordable for the average Joes in the sardine cabin.

But should there be two classes of airport security, two classes of fire protection, two classes of public safety? And what happens when those paying to drive on private roads get tired of paying a gas tax to support public roads? It can be annoying to pay both, Edwards concedes.

When Sclar, whose work takes him around the globe, lectures on privatization, he shows slides of what he calls "the transportation of the rich in São Paulo, Brazil, and the United States."

The first slide shows a helicopter, the safe mode of transport in Brazil, where public services have been heavily privatized and public areas are often unsafe. The second shows a New York City taxi -- a very democratic mode of transportation.

"When you don't have good municipal services, societies break down," Sclar says. "The social costs of that can be enormous over time."

So when do you let it burn?

The Tennessee fire has helped put the high-level debate into perspective.

When city officials ordered the firefighters of South Fulton, Tenn., not to put out Gene Cranick's fire, their reasons made good business sense: If you provide a service for free, no one will pay.

But was it the right decision for a community?

As Cranick mourned his losses, strangers across the country called him a freeloader, an ingrate and a jerk. Firefighters from other departments said that the South Fulton officers should be ashamed of themselves. After the fire, one of Cranick's relatives went to the fire station and reportedly assaulted the chief.

Even privatization proponents hedged. "The Obion County fire seems a clear example of government failure, not market failure," wrote Thomas Firey for the Cato Institute. "It's not difficult to imagine what a private fire service would do in an event like the Obion fire: It likely would extinguish the blaze and then send the homeowner a bill."

On the National Review Online, writer David Foster suggested this right business decision may not have been the moral one. Others said the fire was not a good example of privatization anyway: When the subsidized city department entered the market, it unfairly blocked out competitors.

But even in an entirely free market, the question would remain: When your neighbor's home is burning, do you have a responsibility to help, or do you let it burn for a lack of a fee?

Or perhaps, if you're not paid up, we just assume you're not really a neighbor. That's what one poster at The Agonist seems to suggest, blaming Cranick for not having just paid the fee and adding, "Jeebus people, step up and be part of the community already."