'Should I stay or should I go?" That's not just a rock lyric. Businesses are always asking whether it makes sense to stay where they currently operate or look for a more profitable location.

Unfortunately, rather than concentrating on the retention side of that question by creating incentives for current firms to expand, states typically focus their efforts on attracting new firms or retaining companies that threaten to leave. This failure to devote resources to growing the broader business base has real economic implications.

Politicians love to announce that they have brought a new company into the state or saved hundreds of jobs in a firm that was discussing moving out. But when it comes down to it, the tax breaks usually just move firms around a region and, on net, rarely yield much.

It is estimated that about 80 percent of new jobs are created by employers already operating in a state, while almost all the rest come from new business formations. Less than 1 percent comes from relocated or retained companies. And if the tax breaks create competitive advantages for the tax recipients compared with firms already located in the state, some jobs could be lost. Not smart.

A recent report by the Tax Foundation and KPMG takes on the issue of how firms not eligible for tax breaks compare with firms that receive them. The study created seven different types of companies and then determined their tax burdens as if they actually did business in each of the 50 states. This is a comparison that real-world firms might undertake when deciding where to locate.

How did New Jersey, Pennsylvania and Delaware fare in this analysis? When it comes to helping the new but doing little for the current business base, New Jersey may be the national champion. The state's corporate tax burden is one of the 10 highest for five of the seven business types.

Only capital-intensive manufacturers do OK. That's nice, but manufacturing makes up only about 6 percent of total payrolls. Corporate headquarters, call or distribution centers, retailers, or firms involved with research and development face nearly the highest taxes in the nation.

Yet New Jersey opens its wallet wide when it comes to new companies. After factoring in all the tax breaks available, five of the seven company types wound up with tax burdens that were lower than in 46 other states. If a business comes to New Jersey, its taxes are low. But if you are already here, well, good luck.

Pennsylvania and Delaware, however, tend to treat new companies only modestly differently from current operators. Pennsylvania is a fairly high-tax state, with five of the seven types ranking 29th or worse in tax burden. New firms don't do any better. In six of the seven categories, the tax rates put them at 30th or worse. Whether you are a newcomer or a long-timer, you still get taxed fairly heavily.

As for Delaware, it tends to be in the middle of the pack for mature firms and a little better, in the top 20, for new firms.

So, what does this tell us? First, let's be clear: Tax rates alone don't make a good location. Good public services, a skilled workforce, a good-quality transportation system, markets for the products, and/or a desirable lifestyle are also important. These results just compare the corporate tax burdens.

But the implications are also clear. Pennsylvania and Delaware don't make new companies the winners and current companies the losers in the tax game. By contrast, New Jersey's approach seems almost irrational. In a state already having major growth problems, it makes no sense to put so many resources into attracting or retaining a few special companies when that activity doesn't create many jobs. Meanwhile, the state socks it to companies it should be helping: the resident firms that are the real employment machines.

This disparity needs to change, or a lot more firms may decide that they should go rather than stay.

jnaroff@phillynews.com