Mayors have three jobs, but most enjoy only one or two of them. The ones they like are creating public policy or building political support for those policies. Almost universally, the one they don’t is managing the city government.

This should come as no surprise. Most people who run for elected office are drawn to public policy (How do we reduce homelessness? How do we keep our young people? How do we connect land use and transportation?) or politics (How do we create a coalition of interests? How do we persuade the city council? How do we get our friends elected?). They don’t run because they want to manage employees. If they wanted to do that, they’d probably be a city manager.

And yet, local governments are complex and challenging organizations and, as we were reminded in the 2008-09 recession, sometimes poorly managed ones. And despite how a mayor feels about the grind of organizational management, it’s a job that cannot be ignored. To do so is to risk every smart policy initiative a good mayor can create and every shrewd political move she can make.

So what does a mayor–or, for that matter, any civic leader–need to know about managing a local government? Here are four of the basics:

During good times, prepare for the worst. The problem with a good economy is that it creates a false sense of security. If a city depends on development-related fees (such as impact fees) or sales taxes (which tend to spike in good times), beware: They can make leaders complacent. The best things to do with boom-time revenues are pay down debt and make capital investments. The worst things to do are hire more employees and hand out generous pay raises and pension increases.

When you make capital investments, ask how they’ll increase productivity. Productivity, of course, is doing more work with the same or fewer people, and corporations have made great gains in productivity in recent decades. Governments? Not so much. This has to change, and the place to do that is with capital investments.

Think back to the Econ 101 class you took in college. Remember the three “factors of production,” the resources that create goods and services? As Adam Smith explained two centuries ago, the factors are land, labor and equipment (which Smith called “capital”). Labor and equipment have a special relationship in that you can substitute one for the other; that is, you can buy labor-saving equipment to. . . well, save labor. And that’s exactly what governments should do. Every time a local government wants to build a building, buy a new computer system or purchase a sanitation truck, elected officials should ask how many hours of labor it will save. If the answer is zero, don’t buy it. If the answer is that it’ll require more employees, run away as fast as possible.

Don’t just ask about productivity–measure it, set goals for it and hold managers responsible for achieving the goals. Let’s not beat around the bush here: The objective is to reduce the city’s workforce while delivering better services to the citizens. It’s simple math really: 70 to 90 percent of local governments’ operating budgets are salaries and benefits. If you can improve the productivity of workers through labor-saving equipment, good training, better work flow and smarter management, you can reduce your headcount, please the citizens, pay higher salaries and hold taxes in check. It is, as economists like to say, “the only free lunch in town.” It’s also the surest route to re-election.

But change never comes easily even when it promises great benefits. Employees don’t like being told to work differently–say, moving from paper forms to electronic–and neither do their managers. So offering better ways of doing things is just the start; you have to ensure the better ways are implemented. And here is where it is critical to have reliable measurements of productivity (number of potholes filled per worker, number of business licenses processed per hour, etc.). It’s the only way to be sure you’re actually doing more work with fewer people.

The good news is that there has been a revolution in measuring productivity in city governments in the last 15 years. For a glimpse of what’s possible, visit New York’s elaborate NYCStat performance-reporting system. You’ll be amazed at the detail with which New York’s services are tracked. They should be tracked just as diligently in your city too.

Stop feeding long-term liabilities. You know what I’m talking about here: pensions and other benefits. We didn’t create our country’s $1 trillion public pension problems overnight. Oakland, Calif.’s city government hasn’t made a full payment on its pension obligations in a decade and a half, which explains why Oakland owes more in unfunded pension liabilities than its entire $400 million annual city budget.

Neither Oakland nor any other local government is going to solve its pension problems overnight, but we can make a start. This means, at a minimum, moving new public employees to defined-contribution retirement plans (like the 401(k) accounts private workers have), ending health insurance benefits for employees retiring early–and never, ever adding to these crippling liabilities again.

You’ve probably figured out by now that these are related. Productivity gains will help governments dig out of their financial holes. Budget discipline, particularly in good times, will keep us from digging future holes. Capital investments are a key strategy for boosting productivity–especially when married to measurements, goals and accountability. And all of it depends on mayors facing up to their least-favorite job: managing the city government as if it were a serious organization. Which, come to think of it, it is.