The Center on Budget and Policy Priorities has a very strong reputation in Washington for providing honest, careful assessments of important public policy questions. In May of 2008, Phil Oliff and Iris J. Lav produced a report suggesting that the Massachusetts property tax cap had a number of hidden consequences. I find some of the claims advanced in the report a bit to hard to understand, e.g.,

A tax cap won’t make government services cost less. A cap does not prevent employee health insurance costs, special education costs, or other costs beyond localities’ control from rising much faster than the cap allows. Nor does it hold down the cost of heating buildings, buying gas for police and fire vehicles, and operating schools buses when the world price of oil is skyrocketing. When these things occur, as they have in Massachusetts, other services have to be cut to fit total expenditures under the cap.

But as President Obama argued in making the case for the excise tax, and for proposing steep reduction in Medicare provider payments tied to various reforms, price signals can be useful for disciplining providers. And to the extent other services are cut, this presumably reflects the values and preferences of voters. Massachusetts voters are allowed to override the caps to pay for services. When confronted with a choice between raising property taxes and offering very generous deferred compensation to firefighters and police officers or holding the line on property taxes and offering somewhat less generous deferred compensation, some municipalities will choose the former and some will choose the latter.

Claims that caps will produce large savings through “efficiencies” are overblown. There are fewer efficiencies to realize from squeezing down revenues than cap proponents generally suggest. One person’s “efficiency savings,” such as the elimination of a police or fire station, may represent the loss of a critical service for another person. Ultimately, a property tax cap is highly likely to lead to reductions in basic community services and a deterioration in the quality of life in many communities — particularly in communities that cannot routinely override it.

Interestingly, this is the argument the Medicare chief actuary raised about the proposed cost savings in the Medicare program under PPACA. I’m encouraged to see CBPP express skepticism about this notion. It is certainly true that some efficiencies won’t materialize, and that voters must recognize that there are trade-offs: lower taxes will often, though not always, mean a lower level of public service provision. But as to the conclusion, we’d need more evidence to determine if the deterioration in the quality of life in Massachusetts communities has outweighed the gains in purchasing power that result from the lower tax burden. One assumes that there is at least some value in private consumption.

CBPP made a number of other observations that made a great deal of sense, e.g.,

Changes in school enrollment can have a big impact. The adoption of Proposition 2 ½ coincided with a decline in Massachusetts’ K-12 enrollment, allowing schools to operate with less revenue.If another state adopted a property tax cap during a period of steady or rising enrollment, it would be forced to impose much more extensive cutbacks in teachers, classes, and programs than those seen in Massachusetts.

The first observation seems fair. The latter observation, however, discounts the possibility that new compensation strategies for teachers, new strategies for achieving effective school discipline, and the embrace of student-centered technology could sharply reduce the costs of delivering a particular educational outcome.

Wealthier communities will override a tax cap more frequently than poorer ones. This has contributed to a growing spending gap between local governments in high-income communities and all other communities, despite Massachusetts’ progressive system of state aid. This is likely to occur in other states that implement a cap.

This is hardly surprising. The logic of Tiebout choice suggests that some communities will specialize in offering a generous and expensive basket of public services. The danger here is that this will entrench pernicious inequalities in educational provision, which is a strong case for statewide choice programs. Oliff and Lav go on to suggest that middle class communities might suffer the most from a cap. But one has to assume that middle class communities reap at least some of the benefits from tax restraint.

Josh Barro of the Manhattan Institute offers a very different perspective in a very smart, rigorous report on the lessons of Massachusetts for New Jersey. Here is his basic assessment:

Overall, Proposition 2.5 has succeeded in restraining growth of property-tax collections, total tax collections, and per-pupil education spending in Massachusetts. These fiscal successes have not come at the expense of the state’s educational outcomes, which are the nation’s best, consistently outperforming—or at least tying—New Jersey’s results on national school exams. Massachusetts’s advantage persists even within certain traditionally disadvantaged demographic groups.

A number of Josh’s findings are in tension with the findings of Oliff and Lav, including:

* Massachusetts’s lower spending levels cannot be explained by a lesser need to serve hard-to-teach students. Even school systems in that state with similar percentages of students who were not proficient in English, or who were from low-income families, spent thousands of dollars less than their counterparts in New Jersey.
* Despite their lower spending levels, Massachusetts’s public schools are the country’s clear top performers, as measured by National Assessment of Educational Progress (NAEP) exams administered by the U.S. Department of Education. In 2009, Massachusetts outperformed New Jersey in both reading and math in grades four and eight (though for grade eight, the gap in reading performance is within the margin of error).
* Massachusetts’s stronger NAEP performance is not explained by favorable demographics. Within most demographic groups, students in Massachusetts achieved higher average NAEP scores than their counterparts in New Jersey. Hispanics and Asians/Pacific Islanders were a notable exception. Massachusetts students eligible for subsidized lunch or who were English-language learners also matched or outperformed their counterparts in New Jersey on the NAEP exams, despite lower spending in districts with high concentrations of such students.

I hope and expect that everyone interested in state and local budget issues takes a look at both studies. I also hope that CBPP will give due consideration to Josh’s findings, and that Oliff and Lav will offer constructive criticism regarding where the MI analysis might have missed something important.

The Newark Star-Ledger spoke to Lav about New Jersey and Massachusetts:

Iris Lav, the author of the Center for Budget and Policy Priorities’ study, does not deny that New Jersey has a problem with rising property taxes. But she says the state should consolidate its mess of towns and school districts instead of lowering an artificial cap onto the current system, and emulate Massachusetts in other ways, such as spending money more wisely on education.
“You can become more efficient by becoming more efficient — but not by putting a cap on,” Lav said.

First, it’s not obvious that putting a cap on will preclude states and municipalities from becoming more efficient by becoming more efficient. Indeed, one assumes that it would provide an added incentive to do so, just as President Obama’s excise tax and payment reforms will presumably provide incentives for medical providers and private insurers to become more efficient.

And as I’m sure Lav understands, there are diseconomies of scale as well as economies of scale. In 2008, Jim Manzi wrote an excellent post on Jon Corzine’s call for consolidating towns and school districts, and he found the proposal wanting.

If one were to make the heroic assumptions that these costs differences are caused entirely by town size, and that every town below about 6,000 people could find a nearby town or towns to merge with to get to about 10,000 people (but not more than 15,000, at which point costs would start to rise), and that therefore all small towns could get to the same per capita local government costs as the lowest-cost towns, then you could get some costs out of the system. How much? Under the unrealistic assumptions provided above, about $0.3 billion per year. In comparison, New Jersey’s 2008 state level budget is about $33.3 billion. This is up about $2.2 billion over 2007. So, this – again, totally unrealistic – benefit could be achieved simply by increasing spending from $31.1 billion last year to $33.0 billion this year instead of increasing it to $33.3 billion. Of course when you add in local spending, total state plus local spending is much higher than this in New Jersey. New Jersey’s 2005 total state and local spending was the eight-highest in America, at about $8,900 per capita. Nearby Connecticut had the tenth-highest level of state and local spending, at about $8,550 per person. Simply spending at Connecticut’s level (which isn’t exactly like saying “become Alabama”) would reduce total state and local spending by about $3 billion per year. Finally, under the same set of assumptions that lead us to think we can get $0.3 billion by forcing the consolidation of hundreds of towns, we would presumably want to break up the large towns into smaller towns of 6,000 – 15,000 people each, since this is the lowest-cost town size. That would, under the same assumptions, save more like $1.5 billion per year. Don’t hold your breath.

So yes, Lav’s proposal might save $1.5 billion per year, or slightly more adjusted for inflation, once the extraordinary costs involved in consolidating towns have been fully paid. How does this compare to pension and wage reform?

In a recent issue of NR, Josh wrote the following:

Without a freeze, most New Jersey teachers can expect a raise on the order of 4 percent this year. If they did not receive raises, and if a just-passed reform requiring teachers to contribute toward health insurance were implemented immediately, the state would save $765 million, almost enough to offset the funding cut without raising property taxes, laying off teachers, or buying fewer textbooks and less classroom equipment.

This suggests that there is a great deal of low-hanging fruit in the New Jersey budget. Note that municipal consolidation is very difficult to achieve, and so, in a sense, it is the perfect dodge. And then you confront the fact that it would yield trivial savings relative to pension and wage reform.