Ever since I came into contact with government, both state and federal, and especially in the four decades since first going to work in it, I've been struck by the reality gap between what many Americans expect of it and what it's actually good at doing.

I've previously remarked that we tend to have the greatest faith in the prowess, efficiency, and effectiveness of whatever level of government we've had the least direct experience with. In other words, those who are deeply enmeshed in the toils of local or state government look to Washington to get things done, while those with long federal experience are apt to say, "We'd better make sure that the states are responsible for meeting this next challenge."

In general, American government is better than some, worse than others. It's rarely corrupt and eventually completes most of what it sets out to do. In this it differs from, say, India, Russia, and most of Africa. But it's usually slow, plodding, feckless, and not very responsive. In this, it differs from Singapore-indeed, from much of East Asia, including even China, at least now and then. (China is a mixed bag. It seems to be hopeless at public health matters, for example, even food safety, yet had drilling equipment in place within two weeks of deciding to build a big public-works project. And the modernization of Shanghai, Beijing, and other cities over the past two decades is staggering. Of course, government boosts its efficiency when it is known to shoot those who impede what it views as "progress.")

Why this rambling discourse on governmental competence? Because a leitmotif of the Obama era, particularly in conjunction with the economic downturn and major recent screw-ups by the private sector, is that "government should run things" or, at least, make many more of the big decisions about how things will run.

We hear this in connection with banks, for example, with AIG, with automobile companies, with health care, and much more. In education, we hear it-from the administration-in connection with student loans and now, from all sides, in connection with the use of "stimulus" dollars to effect a host of long-overdue reforms that schools, districts, and states have failed to accomplish on their own.

Look harder and you'll see three themes entangled here.

• The use of federal investigative and regulatory power to try to compel, coerce, or frighten people and institutions into doing something different. (Look at Chrysler's restructuring or the "stress testing" of major banks.)
• The use of federal dollars to "bribe" changed behavior (more charter schools, say) or underwrite activities that wouldn't otherwise attract enough money (e.g., modernized state education-data systems).
• Direct operation (or very close supervision and/or actual ownership) of enterprises that previously ran outside the governmental orbit. That's what we're seeing in the student-loan sphere, for example, and in a host of financial institutions.

Try to get even simple things done by government and you will see how arduous it is. Try to get your Social Security check deposited in a different bank account or to sign up for a different Medicare drug plan. Try-if you work there-to hire or fire someone, or promote or reprimand them. Try to get a grant or contract issued. Try to get a regulation revised or a publication distributed.

It's really hard. It takes a lot of effort and tons of time and the initiative often misfires or simply vanishes into the ether. Sometimes it even gets you into trouble for having tried-thereby reinforcing the attitude of don't rock any boats or try anything different.

Government, in short, has enormous difficulty fulfilling its current responsibilities, coordinating its various parts, and accomplishing its present objectives. You don't have to romanticize the private sector's competence to harbor serious doubts that giving government even more duties is a formula for disappointment. That's true in education and in much, much else.

Shall we, then, try once more to reinvent government so that it works better? Good luck. On a good day government does well to get the checks delivered to the right addresses. It can't run thousands of banks-much less tens of thousands of schools.

And it's naïve to expect otherwise.

Chester E. Finn, Jr. is a senior fellow at Stanford's Hoover Institution, president of the Thomas B. Fordham Institute, and senior editor of Education Next. From 1985 to 1988, he served as Assistant Secretary for Research and Improvement and Counselor to the Secretary at the U.S. Department of Education. Earlier positions include Staff Assistant to the President of the United States; Special Assistant to the Governor of Massachusetts; Counsel to the U.S. Ambassador to India; and Legislative Director for Senator Daniel Patrick Moynihan.