One day, I received a phone call out of the blue from the Chicago Federation of Labor asking if I could help it put together a study on how to cut that city’s budget. This is not the usual union request—nor is it the usual efficiency study I get asked to conduct. But Rahm Emanuel had just been elected mayor and had set about slashing city spending, privatizing services, and reducing the city payroll. And he issued a challenge to the city’s unions: If you don’t like my plans, come up with another way to save as much taxpayer money.  

And so they did. My favorite example: Construction workers repairing sidewalks were working the standard eight-hour shifts, five days per week. Pouring a load of cement, though, takes about five hours to complete, meaning they could only pour one load per day, or five per week. If the city went to a 10-hour, four-days-per-week schedule, however, as the unions suggested, the crews could pour two loads per day, eight per week—a 60 percent increase in productivity at no additional cost and with savings on fuel and equipment rental.

Under the leadership of Jorge Ramirez, the unions worked with my firm to develop a plan to save $242 million a year. It included many proposals like the new schedule for pouring sidewalks. But the whole plan started with $40 million from something called “managed competition,” in which public employees compete against private-sector firms for contracts to provide government services.  

Chicago is not alone in pursuing managed competition. It is ubiquitous in the United Kingdom, which has had at least 3,500 such competitions. In this country, cities including Phoenix, Charlotte, Indianapolis and Philadelphia have pursued managed competition as a strategy for cutting cost and improving service.  

The Phoenix trash-hauling operations have become legendary as the fountainhead of managed competition. In the late 1970s, the city decided to outsource trash collection, splitting the city into several zones for bidding purposes. But first, it compared the private bids received to what could be achieved in-house. City employees proposed business as usual, so the in-house benchmark wasn’t at all competitive and the first zone was outsourced. Stung by the resulting privatization, the city’s workers realized they needed to adopt new practices and new equipment, introduce greater flexibility on wages and working conditions, and seek professional guidance. The public employees succeeded in winning the competitions for the remaining four zones, and even eventually won back the contract for the zone they initially lost.

Former Indianapolis Mayor Stephen Goldsmith is similarly famous as an innovator and privatizer of government services. In 1992, Goldsmith targeted roughly 80 city services for privatization. Again, as public employees began to realize that their futures lay in competing, they began to win more and more of the contracts. Indianapolis’ fleet services department got so good at what it did, in fact, that it started expanding its “business” to service other quasi-governmental entities.

It may surprise some people to know that, in most instances where such “managed competition” has been tried, the in-house government workers have almost always beaten out the private sector. Thus, even the pro-privatization administration of George W. Bush found itself ending a long-standing policy that discouraged the government from competing with the private sector, revising guidelines to facilitate managed competition, and eliminating the practice of outsourcing without first considering whether government might perform the work more efficiently.

Of course, not all government services can readily be privatized. Many government activities are obvious candidates for private competition while others, such as components of the regulatory system or such “public goods” as basic research, generally belong in government. Where to draw those lines, though, is debatable. The Bush administration designated performance audits as inherently governmental (hey, my firm offers those!), while Massachusetts has long operated a privatized environmental “Rent-a-Regulator” program. It’s not really clear, despite the Bush administration’s assessment, what’s “inherently governmental”—at least in this country—about “heraldry services.” Maybe that’s just a Bush thing.

In sum, there’s no good reason why the private sector shouldn’t be allowed to compete to provide many, if not most, government services—but there’s no good reason why government shouldn’t, and can’t successfully, compete back. But as these instances of managed competition illustrate, realizing lower costs within government often requires exposing the government to competition in order to force government managers and workers to become more efficient. In short, the answer isn't privatization—it’s competition.

For instance, in Edmonton, Canada, a voucher system forced public schools to compete. Within two years, their quality had so improved that there were hardly any private or parochial schools left in the market. Similarly, a 2005 study of a federal prison privatization found no meaningful cost savings over traditional prison operations. But the report noted that “the threat of outsourcing” and the resulting “perception of competition forced all of the [public] facility managers to monitor their usual methods of ‘doing business.’”

This holds larger implications for what government can and should do in the 21st century. Take, for instance, one of the most long-criticized operations of the federal government: the Post Office.

Postal services were quasi-privatized in the US decades ago, but that hasn’t done much to reverse their financial problems. In many ways, that’s an industry- and world-wide problem. But Congress also has restricted the US Postal Service (USPS) from competing successfully, At one point, it prohibited the provision of services ancillary to mailing, like wrapping and copying. The USPS has done its best to make money despite the strictures placed on it, cleverly capitalizing on its reputation for slogging through rain or snow or sleet or hail by producing a line of branded all-weather clothing. Other countries allow their postal services more ambit to compete like a business—primarily in offering financial services, a not-illogical fit, with Japan Post owning the world’s largest bank.

In this country, though, the idea that the government might compete successfully against private enterprise tends to provoke opposition and incredulity. But the postal example illustrates what governments can potentially accomplish. In a world in which private companies increasingly offer government services, and governments increasingly must compete against the private sector, how do you draw a distinction between the two? What is a “government”? I’d suggest that, the more exclusively an entity is the provider of a service, and the more essential that service is, the more it is comes to resemble a government. For example, as Facebook came to function as a natural monopoly (and one central to people’s lives) it began to face growing demands from its customers for rights somewhat akin to people’s expectations of governments—and to attain popularity ratings akin to those of the IRS. As conservatives have rightly grumbled, governments are monopolies—but in many respects, monopolies are also, essentially, “governments.”

Natural monopolies are a declining phenomenon; they generally occur in public goods or public utilities—often the province of governments. Since their infrastructure reaches most citizens, they can also—like Facebook—form a “backbone” off which other, profitable “apps,” both public and private, can be constructed, as postal services worldwide have done.

In the health care reform debate at the beginning of the Obama Administration, University of Chicago law professor Richard A. Epstein expressed the opposition of many in declaring, “If … the government is a nimble competitor, we should welcome its entry into all markets, not just health. We should have government airlines and oil companies to keep the private sector in line.” Well, that’s exactly the point. As Jacob Hacker and Rahul Rajkumar have written, private and public programs often coexist:

Each side of the divide has strengths and weaknesses, but in every case the public sector is providing something the private sector cannot: A backup that's there if and when you need it; a benchmark for private providers; and a backstop to make sure costs don't spin out of control.

If governments can be nimble competitors, we should welcome their entry into markets and, where they can’t compete, they should yield to superior private-sector products or services. All of this could perhaps cut the Gordian Knot in today’s public-versus-private debate. Take the most contentious arena in which it arises: privatizing a portion of Social Security. There are, of course, complications to any such effort, like the massive cost of forward-funding such a system, which are well beyond the scope of this discussion. But, in theory, there isn’t any reason why Americans shouldn’t be able to choose to have their retirement savings managed by private fund managers—but there’s also no reason why they shouldn’t be allowed the option of keeping those savings with the government. A crazy choice? Not so. The U.S. Social Security Administration delivers better returns net-of-fees than private managers, over time—and virtually risk free. You won’t become the next Warren Buffett that way, but public Social Security accounts would be the best choice for most Americans.

And so would a competitive government.