This is admittedly hard, as economics has become more and more of a mathematical discipline in the past 20 years, to the point now where good schools won't accept students into the PhD program without strong math skills. Steve Levitt dealt with this question previously, deciding that he might get into Harvard or MIT, but not Chicago. (I think that's crazy, beause my understanding is that he had worked with top Harvard professors as an undergrad and would have had top-notch recommendations from them. And as much as math counts, fantastic recommendations from top economists are pretty valuable too.)
Mankiw suggests getting a Masters first (at somewhere like LSE) or following a closely related career track (such as law). These are both good ideas. Another idea is to spend some time working in policy (as a Fed/CEA/BLS/whatever researcher) or in the private sector as a researcher with an economic consulting firm. Either of these paths require smarts and economic intuition more than already honed math skills. (And, ok, an ability to learn Excel won't hurt....)
This can be a career, but most people do this only for a couple of years. The real benefit to this, though, is that it would allow you to:
- Get a sense of what an economist can do, other than teach/research;
- Develop analytical skills (and economic intuition) at work without explicitly doing math; and
- Take some math classes on the side to hone your skills.
I spent 1 1/2 years in the early 1980s as a student at Harvard Law School, and I think I could have forged a happy career with a law degree instead of a PhD. In the end, I decided that my comparative advantage was in economics rather than law, so I suspended my law studies. But I can always go back and finish the law degree if this economics thing doesn't work out for me.