Letter to the Editor

(from Maine Townsman, November 2004)
By Evan Richert

By Jeff Austin’s article (Maine Townsman, July 2004) on my paper, “Regionalism, New England-Style,” sent me scurrying for my notes. It’s been one-and-a-half years since I wrote the piece, so I couldn’t be sure: had I really “smeared” municipalities, as the article claimed? Called for replacing town government with “large county-style government?” Advocated for just 25 to 50 municipal districts? And no footnotes?

To my relief, the answers were: no, no, and no. And nothing in the review changes my mind that we can save tens of millions of dollars through local alliances along the lines of “municipal districts.”

I didn’t “smear” New England’s towns; I called them frugal and icons of democracy. I also said that, confined to their 30 or 40 square mile boundaries, they are losing control of their destinies, and, so far, they haven’t been willing or able to adapt. That is a diagnosis, not a smear. Having made the diagnosis, I offered a proposal by which towns might transform themselves to retain their pre-eminence in our system of governance.

Nowhere did I say that the answer was to replace towns with county government. Peer states outside of New England tend to have small but strong county governments. But there was no proposal – not even a hint of one – for “large-style county government.”

Rather, there are things to learn from the peer states that could be used by towns willing to reinvent themselves – not as counties, but as small groups of towns joined together with enough critical mass to gain control of costs, become forces in economic development, and to rein in sprawl. What I said was: “The challenge is to find a form of regional governance that resembles the New England town in its accessibility, frugality, and volunteerism, and to allow that form to take shape through the same principles of self-organization that allowed the original town form to evolve.” I even put that in bold face. Jeff couldn’t have missed it.

The proposed municipal districts would have budgeting, tax-rate setting, land use planning, and K-12 authority, with popular governments based on their own charters. Jeff misrepresented the proposal by failing to say that these would be voluntary alliances, and by not pointing out that the proposed criteria for sparsely populated rural areas would be different than in the more settled areas. This is important because the rural areas couldn’t meet the urban requirements, so he made the idea seem impossible to your readers in rural towns.

He then threw around the numbers “25 to 50” such districts. Those weren’t my numbers. He made them up. Nor did he mention that, once joined together in municipal districts, the individual towns would remain as bodies politic and continue to deliver services unless they decided through the budget process that they should be delivered jointly.

Much of the article was aimed at disputing that municipal districts could save money and reduce property taxes. But the writer took several wrong turns.

The first was failing to understand that the correlation between fragmented, general-purpose local governments and the number of employees required to deliver local services came from analysis of nationwide statistics, not just from a few peer states. That correlation is notable and statistically significant—there is almost no chance that it is a random occurrence. The greater the number of general-purpose units of local government with taxing authority, the stronger the likelihood that there will be higher proportions of full-time equivalent employees (FTEs).

The next was misunderstanding the basis for choosing the peer states. Why not Alaska, or states like Minnesota and Wisconsin (where FTE rates are similar to Maine’s, and thus improve Maine’s comparison with the peers)?

I looked for states with similarities to Maine in population density and size, demographics, resource-based economies, and northern latitudes. But they could not have local governments similar to Maine. That was the point: find peers that are similar in key ways, except for the variable being tested.

This eliminated extremes like Alaska (extreme size) and Wyoming (extremely low density). It also eliminated Wisconsin and Minnesota, where townships are numerous and have similar functions to towns in New England (see U.S. Census, “Government Organization, 2002 Census of Governments,” Dec. 2002). You’d expect higher FTEs in these states, and that’s what you get.

As to the peer states – Oregon, Idaho, North Dakota, South Dakota, and West Virginia — Jeff did his own math to downplay the significance of these states’ lower FTEs. But he punched a couple of wrong numbers.

First, he claimed that it must be because local governments in Maine are responsible for more services than local (county) governments in the peer states. Not so: it is the other way around. And when you add up local and state government totals, and do so on an apples-to-apples basis, the result for 2001 was: Maine—519; average of peer states—454, or 12% less. Jeff’s mistake was that he did not account for variations in the states’ higher education systems.

Jeff figured his home run would be in the large role that K-12 employment plays in Maine versus the peer states, thus leading any conclusions about high FTE rates away from the municipal side of local government. Here, he again ran into an apples-to-apples problem, this time by not factoring out the higher education services provided by local (county) governments in the peer states. A majority of the peer states’ local governments run community college systems. When this is accounted for, and considering only the municipal side of local government, Maine has 12 FTEs per 10,000 more than the peer states.

That translates into an extra 1,500 FTEs statewide at a cost of about $44 million per year, based on average 2001 payroll figures from the Census. In terms of property tax relief, reducing costs by that amount would be like doubling the state’s original $7,000 homestead exemption.

The extra non-instructional K-12 FTEs in Maine compared with the peer states cost $28 million per year. So now we are up to potential savings of $72 million.

That leaves the instructional K-12 staff. Maine in 2001 had 47 more FTEs per 10,000 than the average of the peer states, accounting for about $190 million of additional cost. Jeff offered some self-serving explanations (Mainers like their schools, the shift to S.A.D.s failed to reduce costs, it’s a political conspiracy).

The simpler possibility is that public school enrollment simultaneously is falling in Maine and shifting outward to suburban communities. This has led to student-teacher ratios that, in the declining municipalities, are well below levels considered acceptable for “personal and supportive teacher-student interaction,” as Cape Elizabeth puts it.

Cape Elizabeth, one of the highest performing districts in the state, sets these levels at 18:1 for K, 20:1 for Grades 1-3, and 22:1 for Grade 4 and up. In the declining municipalities, it is not unusual to see ratios as low as 12:1. Meanwhile, recipients of the shifting population have to add teachers to stay within acceptable ratios.

The fragmented school district geographies not only drive up administrative costs; they also make it impossible to easily move teachers from places with falling enrollments to places with rising enrollments. (See “Final Report: Task Force on Efficiency and Equity in the Use of K-12 Educational Resources,” January 2004, which documents the costs of having one school district for every 720 students and one administrator for every 200 students in Maine).

If just a fraction of the extra K-12 instructional cost were eliminated, this, combined with the earlier savings, could increase total savings to $125 or $150 million per year and still leave teacher-student ratios well below the recommended maximum.

Numbers aside, Jeff completely missed the premise of the proposal for municipal districts. If Maine moved toward municipal districts or something like them, the engine for the savings would be embedded in their very structure. A rational decision by local decision-makers in a municipal jurisdiction of 100 to 250 square miles will be different than a rational decision in a jurisdiction of 30 or 40 square miles.

Faced with a question of whether to close a school in one town and build a brand new one in the next town over, or to build new public safety buildings a few miles apart, or to have separate dispatch and assessment services, the rational answer will change. There are simply more opportunities to avoid duplication in staffing and facilities and to take advantage of scale.

No one knows for sure what the savings would be. On the municipal side, they likely would be greatest in regions where more and more towns are passing the 2,500 or 3,500 population marks – the point at which suburban services and paid staffs really start to kick in, and redundancy takes off. Twenty-seven percent of all Maine towns now exceed 2,500, including 20% with more than 3,500.

But there is a way to find out. Let’s ask the state to conduct a pilot program. Backed by generous financial incentives, the state would ask for competitive proposals from interested groups of towns, select the best two and track the results. In my speaking around the state, I think some towns are ready. What do you say?

Meanwhile, thanks for keeping the issue alive. A year-and-a-half later, I thought it might be dead. And those footnotes? There were 30 of them. How in the world did Jeff miss 30 footnotes?

JEFF AUSTIN’S REPLY: Evan Richert just doesn’t seem to get it. The reason for my article was to shed additional light on the FTE “gap” that Richert claimed existed between Maine and our peers. I believe that Richert misused Census data to justify a radical restructuring of local government in Maine and that the Portland Press Herald picked up on his conclusions without properly analyzing his research and the data upon which it was based.

Furthermore, I believe the FTE gap and Richert’s “bigger is better” proposal are totally unrelated. Whatever its merits or flaws, Richert’s approach is in no way justified by the Census data.

Unfortunately, in the above “Letter to the Editor”, Richert spends most of his time defending his regionalism idea rather than addressing my contention that he misused the data. His reference to higher education services being delivered differently in the “peer” states is weak because it was never mentioned in his original research. It’s ironic that he now wants to factor out higher education provided by local governments in the peer states, but says it’s inappropriate to make comparisons between municipal services provided by Maine communities (and to what degree they are provided) and those provided by local governments in his peer states. [I am aware that in one of his peer states, West Virginia, municipalities have virtually no road maintenance responsibilities… they are handled almost entirely by the State.]

Richert’s response has several inaccuracies, but rather than having a back and forth, ongoing debate, let’s look at some quotes from a column on this very topic written by Economist Charles Lawton in the Portland Press Herald last July.

“For local government in Maine, education is the 800 pound gorilla . . . fully 74 percent [of local FTEs] were in education.”

“The overwhelming importance of education in Maine’s local government is seen even more clearly in comparing [Maine local FTEs] to the U.S. average. For every category of employment except education, Maine had a lower share than the national average.”

This last comment was my exact point. The Census data regarding local government FTEs doesn’t point a finger at municipal service delivery costs in Maine. For anyone willing to take an unbiased look at the data, it clearly shows that Maine is more committed to its schools than our peers. Using the Census data to justify a radical restructuring of municipal service delivery in Maine is not substantiated by the facts.