COLCHESTER -- It's time to re-evaluate Vermont's education financing system and test its fairness to both students and taxpayers. That was the consensus at an education symposium convened Tuesday by Gov. Peter Shumlin and the Legislature.

The event kicked off what's intended to be a session-long discussion of the state's school funding formula, with the aim of understanding how it's working in the current economic and educational environment -- and whether it could be improved.

The current funding formula was created in 1997 with the passage of Act 60, and amended in 2003 with Act 68. Now all of the laws' complex parts and funding mechanisms, everything from per-pupil costs and outcomes to property tax structures and income sensitivity, are on the table.

Held at St. Michael's College in Colchester, the symposium included a six-member panel discussion and three breakout sessions for wide-ranging conversations. It was organized as a response to growing complaints about high property tax rates and persistent concerns about educational outcomes -- particularly for schoolchildren from the state's poorest districts.

Panel moderator Lawrence Picus, author of a 2012 report on the state's education finance system, will prepare a briefing paper for the administration and Legislature by the end of January. The follow-up will summarize and expand upon Tuesday's discussion.

The six-member panel of experts from Vermont and around the country weighed in against a backdrop of divergent assumptions: that rising property taxes are unpopular compared to an income-based system, for example, or that the housing market collapse and lowered property valuations that resulted from the recession spurred only a temporary spike in property tax rates that could very well come back down.

Along with those issues, panelists looked at a separate issue at play in Vermont: ever-rising rates of growth in school budgets.

Vermont's per-pupil cost of education alternately takes first or second place as highest in the nation, according to Daphne Kenyon, public policy consultant from New Hampshire and a visiting fellow at the Lincoln Institute of Land Policy in Cambridge, Mass.

The state's achievement scores don't illustrate a return on that investment, Kenyon said, when compared to neighboring states like New Hampshire and Massachusetts, which spend less on average per student.

Patrick Walsh, an economics professor at St. Michael's College, also challenged the conventional wisdom that more spending results in better outcomes -- though he acknowledged that correlation can vary depending on which population of students is studied. Extra money directed to disadvantaged kids tends to make a bigger difference than a boost for better-off schools, for example.

But the impact of school spending is just one assumption that should be tested, according to both Shumlin and several panelists.

"Do we have a challenge with income sensitivity driving school spending beyond sustainable rates?" Shumlin asked in his opening remarks. (Under Vermont's education finance formula, the state uses income sensitivity to reduce property tax burdens for households that earn below a certain threshold.) Shumlin said some people believe passionately that tax reductions from income sensitivity shield too many people from the true cost of their votes on school budgets. But he wonders if that theory is just a myth.

About two-thirds of Vermont households qualify for income sensitivity reductions that apply to their property tax bills. Several panelists suggested that school budgets likely would be lower if more taxpayers had more "skin in the game."

The need for more and better data to answer such questions was a running theme in the breakout sessions. A desire for "data-driven decisions" was a mantra that applied to both the base education cost per pupil -- is it set correctly? -- as well as the quest for what the ideal class size and school district size is. That has long been a key issue in this small rural state with small schools.

If one takeaway could be gleaned from the discussion, it came from those outside Vermont: The system here works well, they said. Yes, it's 17 years old and likely needs to be updated. But it's got "good bones," as Picus put it. Adjustments may be called for, but not an outright replacement.

Aside from adjusting income sensitivity, one component of the formula that may be tweaked is the "high-spending threshold" that kicks in as a method of moderating school budget increases. Designed to curb school spending, the function spikes tax rates if a school budget exceeds a certain level of annual increase.

"It seems like that threshold is not really doing its job (of preventing) runaway spending," Walsh said.

Tom Downes, associate professor of economics at Tufts University, suggested reclassifying taxpayers beyond just residential and non-residential groups as is done under current law. Vermont could consider different tax rates for residential, commercial, industrial and "everybody else," he said, the latter category being a potentially profitable way of raising taxes on out-of-state homeowners.

Introducing multiple thresholds for income sensitivity, which would taper tax implications could also help, several panelists said. Stepping up the tax burden more gradually may dissuade people from manipulating their household income from year to year to avoid a big bump in their tax bills.

Assessing a property owner's ability to pay taxes by a more accurate measure of income may also be a good strategy, suggested Michael Wolkoff, deputy chair of the economics department at the University of Rochester.

Shumlin, Picus and several panelists cautioned that, whatever aspect of the school financing system may be altered, some people will be unhappy.

But the fundamental goal, House Speaker Shap Smith, D-Morrisville, underscored, is to achieve both equity in financial obligations and equal access to education for all schoolchildren in the state. Those two principles cannot be separated, he said.