The idea that there is one thing called "government"--and that you can measure it by looking at total spending--makes no sense
You hear all the time that the government must get smaller. John Boehner said it the day after the elections: "We're going to continue and renew our efforts for a smaller, less costly and more accountable government." Discussing negotiations over raising the debt ceiling, Barack Obama sounded a similar note: "We have agreed to a series of spending cuts that will make the government leaner, meaner, more effective, more efficient, and give taxpayers a greater bang for their buck." And a large majority of Americans agree in the abstract (while simultaneously opposing any significant spending cuts).
Conservatives like to point to high levels of federal spending--23.8 percent of GDP last year--as evidence that government is too big. Liberals are less likely to talk about the size of government, but generally prefer a "larger" government that does more things for more people. But the idea that there is one thing called "government"--and that you can measure it by looking at total spending--makes no sense.
First of all, the number of dollars collected and spent by the government doesn't tell you how big the government is in any meaningful sense. Most government policies can be accomplished at least three different ways: spending, tax credits, and regulation. For example, let's say we want to help low-income people afford rental housing. We can pay for housing vouchers; we can provide tax credits to developers to build affordable housing; or we can have a regulation saying that some percentage of new units must be affordably priced. The first increases the amount of cash flowing in and out of the government; the second decreases it; and the third leaves it the same. Yet all increase government's impact on society.
When people say government is too big, they often have in mind something like the Consumer Financial Protection Bureau--a regulatory agency that tells businesses what they can and can't do. I happen to think that having fewer screwed consumers is a worthwhile idea. You may not, and we can argue about it. But that's not where the dollars are. The CFPA's budget is about $300 million, or less than one-hundredth of one percent of federal government spending.
The money is in programs like Social Security ($740 billion), which, per dollar, has a relatively small impact on the economy. Social Security doesn't say what businesses can or can't do, and it doesn't say what people can do with their money. It mainly moves money from people's working years to their retirement years, which means that it's doing something that they might have done anyway.
If you want to reduce the impact of government on people's lives, that's fine, but it's absurd to think that shrinking the federal budget is the way to get there. Instead, an obvious strategy would be to go after tax loopholes. Eliminating a loophole reduces government distortion of the economy (hooray!) and reduces the deficit (double hooray!). But because the Republican Party is fixated on top-line taxes and spending, popular antipathy toward the regulatory state (deserved or not) has been translated into an attack on popular entitlement programs.
More generally, one-dimensional discussions of government size overlook the fact that the federal government does many different things. Some feel like traditional government activities, like national defense and the federal court system. In these cases, tax revenues are pooled to pay for general services, not individual benefits.
But almost half of all spending goes instead to Social Security and Medicare, which are more like (compulsory) individual transactions that people enter into with the government. Social Security is a combined pension and insurance plan; Medicare is a prepaid health insurance plan. In both programs, you make specific, dedicated contributions and you get back specific, individual benefits. So rationally speaking, your opinion about Social Security or about Medicare should be based on how much you put in and how much you get out--not on the gross size of the program, and not on how big the rest of the federal budget is. Yet instead the total size of the budget has become the driving force behind potential structural changes in Social Security and Medicare.
If we don't make budgetary decisions by assuming some maximum size of the budget, then, what should we do? The answer is simple: we should make decisions on a program-by-program basis, just like a business is supposed to do. A company that wants to expand into China doesn't automatically pull out of Europe because there's some arbitrary limit on the size of its business; it invests in every opportunity that it thinks will be worth it.
That's the attitude we should have today. If there's a program that the American people, through our democratic system, agree will provide benefits greater than its costs, we should do it, independently of the existing spending level. And if there's a program that isn't covering its costs, we should kill it. This is obvious, but instead Washington seems locked in a debate about the total spending level and the total tax bill. And that's a recipe for bad decisions.