IN AN IDEAL WORLD, the Legislature would overhaul our tax code, reform our budgeting process and restructure our government in one fell swoop.

The actions, after all, are closely related: While we can decide how we want to distribute the tax burden among various types of taxes and groups of people, it’s difficult to properly set the levels of those taxes if we don’t know how much money we need to operate government. We can’t know how much money we actually need until we reform a budgeting process that has little to do with identifying and funding our actual needs. The cost of funding our needs could be reduced if we structured our government in a way to avoid duplication and ensure cooperation from agency to agency.

Of course, in the real world, we’ll be happy with achieving those goals pretty much any way we can get them. The reality is that the very complexity and comprehensive nature that demands these tasks be interwoven also makes that politically difficult.

But whether done in concert or individually, these are the tasks before our General Assembly as it convenes for the 2004 session. Many other issues deserve action as well. But our lawmakers must not allow anything to stand between them and these tasks.

Our state and local tax code is characterized primarily by exemptions, handed out sometimes for good reason, sometimes to satisfy a favorite special interest or to quiet the loudest protesters. Changes have been layered atop each other so many times that it’s nearly impossible to answer such basic questions as whether the tax code is progressive or regressive, whether it is efficient, whether it encourages behaviors we want to encourage. Several proposals have been put forward to overhaul this system. Out of those needs to come a new code that is easy to evaluate and short on special exemptions, that the public perceives as fair, that doesn’t cost too much to administer or comply with, that brings in a predictable and growing amount of money each year, and that either has little effect on economic decisions or else encourages activities we want to encourage and discourages those we want to discourage.

Our budgeting process has been based on incrementalism, and on the assumption that we need to keep doing everything we’ve ever done. When the economy is good, legislators provide the extra money to keep doing the same things, and add on some new programs; when the economy is bad, they make everyone share the pain. Legislators need to routinely eliminate programs that are no longer needed and redesign programs that aren’t working. That means asking difficult questions about what our state government needs to do, how that can best be done, and how much that should cost. Then they need to fund those needed programs appropriately. The governor last week provided a good model to follow in accomplishing all of the above.

Our government is based on the 19th-century notion that power must be dispersed widely, because voters aren’t smart enough to hold anyone responsible for exercising it badly. The result is agencies that duplicate each others’ work and compete for attention and funding — a process that increases the cost of government, renders innovation extraordinarily difficult and makes it nearly impossible for the public will to be exercised. Governors should control nearly all executive agencies, from the prisons and the Commerce Department to the state Education Department and the colleges.

These are not new or radical ideas. These changes won’t guarantee that our government will better serve us, but not making them practically guarantees it won’t.