Disunited States.

Call it the Aladdin principle: Be careful what you wish for -- you just might get it. For years, the nation's governors (especially the Republicans) have insisted that if the federal government would just give them responsibility for big programs like welfare, they could do a much better job. Their genie came in the form of the 1994 Republican Congress, elected on a surge of animus toward the federal government. And with that, the rhetoric of devolution -- sending authority back to the states -- has been turning into reality. So the irony was rich several months ago, when the National Governors Association, including top Republicans like New York's George Pataki, complained that the states were being left holding the bag on welfare.

The difference between the theory of devolution -- beloved mantra of the GOP, slogan of New Democrats, and widely popular with the public -- and its application in reality is the topic of Disunited States. Written by John Donahue, an associate professor at Harvard's Kennedy School who served as a top lieutenant in the Reich Labor Department, Disunited States is a timely and evenhanded reminder that devolution has a downside.

Disunited States begins with a simple,if daunting, question: How will American government change as authority is moved away from Washington, D.C., to the 50 states? The answer, quite simply, is that states will have more leeway to lower their taxes and decrease their spending on social services in the already heated competition to attract well off residents (who consume few services) and large corporations (which bring jobs).

Already the bidding war has developed into a race to the bottom. Take the case of the incentive packages states offer corporations. In 1980, for example, Ohio offered Honda a package worth about $20 million to attract a new plant. That worked out to around $3,800 per job. By 1992, South Carolina was offering BMW a package worth upwards of $65,000 per job. As the war between the states intensifies, more and more money flows from state treasuries (where it could be used to build schools, maintain parks, or lift families out of poverty) to the profit lines of the companies.

Similarly, in order to lure high-income residents, states have made their tax policies increasingly regressive. States tend to rely more on consumption taxes, which disproportionately hurt those on modest incomes. Only nine states have top income tax brackets above $50,000, and state tax revenue raised from corporations has fallen b a third since 1980. Under such systems, the more work states are asked to do, the heavier the burden will become on those least able to pay.

While there are no easy measures to determine whether the bidding war between states also encourages regressive spending policies, it's not hard to imagine that the same forces that skew tax policy in favor of the rich apply here as well. And with more and more money going to tax cuts, state governments caret afford to be as activist, no matter what their intentions.

At its extreme then, national devolution threatens to become a kind of national dissolution. As the federal government steps back, states fight increasingly damaging bidding wars with each other. Regressive policies are the natural winners. The rich get richer, the poor more hopeless; the middle class falters. This insight is the book's greatest contribution.

But Disunited States is not without its flaws. The writing is at times dry and academic and, more seriously, the overall emphasis is muddled -- a product, I imagine, of the book's origins in the author's research on state competition for business investment. Important themes are missed.

One example: Donahue fails to explain clearly the political rhetoric which surrounds the devolution debate. And to ignore the rhetoric is to lose the argument. indeed, one of the driving forces behind "sending power to the states" is not a judgment that states can serve the people better -- but a cynical calculation that talk of empowering the states is good cover for downsizing government, especially social services. When, under the banner of "new federalism," Reagan consolidated scores of programs into state block grants, the end result was less funding. Or take welfare reform, the most important example thus far of devolution: It's clear that a good number of the people who talked the talk of "helping people help themselves" were actually more interested in helping themselves balance the budget on the backs of the poor. That's why Pataki and friends, the governors who will get stuck paying more -- or explaining to their constituents why legal immigrants shouldn't be eligible -- are having second thoughts.

Donahue does provide some rhetorical correctives to devolution chic. He reminds us that states are not necessarily more efficient, or even "closer to the people," than the federal government. And it is state and local government, not the federal government, that has been getting truly big. But these points, too, are made only in passing; a coherent discussion of the ways conservatives argue their case -- and why liberals are sometimes afraid to disagree -- would have served the reader better.

Disunited States would also have been a stronger (and more useful) book if Donahue had tried to address the question it naturally raises: How can government do a better job as states behave more and more like a pack of dogs with not enough bones to go around?

One important answer to this question is a bit of common sense familiar to Monthly readers: Government needs to create an environment in which good ideas flourish and bad ones die quickly. States are often given the "laboratories of democracy" moniker, but, as Donahue mentions, states are slow to learn from the successes and failures of one another. There is not much use in experimenting if the lessons are not applied. So the federal government should take improving state government as a core mission: gathering useful information about what is working well, and making sure word gets around.

That is a topic in which politicians should be much more interested. Then again, if politicians were to spend as much time talking about how to fix government as they do talking about how to break it down, we wouldn't need to be waiting for genies.