Unconvential Wisdom: Lessons from the Tortured History of LGA

“We believe a gradual reduction of aid to cities is possible and desirable. At a minimum, we recommend that the state not increase general purpose aid to cities or take on the job of preventing future city tax increases. In fact aid can be reduced gradually in real and absolute terms in the future, though not without opposition and pain….”

Many would consider this advice to be a disastrous and irresponsible policy recommendation and an unwarranted attack on city governments. What may be surprising is that this recommendation is lifted from the executive summary of a 1990 report from the Office of the Legislative Auditor (OLA) on local government spending. ¹

If it seems like we are caught in a time warp over arguments about property taxes and aid to local governments, it's because we are. The questions of whether we should give aid to cities at all, how much we should give, and how it should be distributed have been a staple of policy debate in Minnesota for well over four decades. With concerns over LGA cuts and property taxes reaching a crescendo yet again, it's a good time revisit the arguments for and against LGA and what we've learned about its practice.

Should LGA Even Exist?

The debate on state aid to local government begins on a philosophical level – is there a reasonable public policy purpose for having it at all? Counties have a strong claim on aid because they administer a large number of health and human service programs for the state as well as highway and law enforcement programs with statewide significance. Cities provide services of mainly local benefit. The "right" level of city services is what local residents decide they want and are willing to pay for.

The case for having an LGA program is based on three ideas:

• Cities vary greatly in property wealth and personal income, and those with low levels of wealth and income find it difficult to provide reasonable levels of city services without financial hardship.
• Some cities have regional impacts – positive and negative – and the "spillover effects" result in benefits and costs imposed on other cities.
• Since the property tax is regressive and unconnected from ability to pay, local services should be supported in part by state revenue sources.

All three arguments have some merit, but equally compelling counter arguments can also be made. Adequate services at reasonable prices are a laudable policy objective, but it's equally crucial to hold officials accountable for their spending decisions. When the state pays for a significant share of the cost of local services, it reduces the apparent cost of those services, increasing the possibility of higher levels of spending, wasteful or inefficient spending, and higher levels of public employment and compensation than would otherwise occur. Spillover concerns have to be placed in the context of other state spending priorities and their own cross-boundary effects. LGA is in competition with spending on education, human services, environmental protection and countless other state programs where the spillover issues are much larger and thus the rationale for state-level financing much stronger. With respect to regressivity, everything is relative. The property tax on homeowners after factoring in state refund programs is less regressive than state sales taxes. The recreation of the Voss database (matching household income with property tax bills) by the Department of Revenue later this year will provide tremendous insight into how much concern over the ability to pay issue is truly justified.

Lessons from the Practice of LGA

These arguments suggest that some aid to city governments may be appropriate in very carefully defined circumstances. The practice of LGA introduces politics into the equation. A number of important lessons can be learned from this forty-year marriage of theory and practice.

Lesson #1: Property tax relief from state aid has been temporary and comes at considerable taxpayer cost

In 1967 the state enacted a first-ever 3% sales tax with proceeds dedicated entirely to property tax relief. Per capita property tax levies declined for two years but by 1971 were 17% above 1967 levels.

In response, 1971 marked the launch of the "Minnesota Miracle" which was, in practice, a massive doubling-down on the idea that state involvement in local finance could buy lasting property tax relief. Details included an expanded homestead credit, passage of the LGA program, and local levy limits.

The Miracle legislation indisputably provided property tax relief but at substantial cost to state taxpayers. To pay for the Miracle, corporate and personal income taxes were increased significantly, the sales tax rate was increased to 4%, and sin and severance taxes were increased. From 1972-1982, a decade of high inflation, property taxes increased less than 5% per year. But during the same ten-year period, all other state and local taxes increased nearly 171 %. In the second year of the miracle, property taxes per capita were down by $15.08 from 1972 but all other state and local taxes had increased by nearly $133 dollars per person. ²

Miracle legislation could not provide permanent property tax relief because state subsidies to local governments provided incentives to spend more. Figure 1 is a telling graph appearing in the 1990 OLA report which tracked per capita city expenditures over 20 years (adjusted for the considerable inflation which existed during much of that time). Even after adjusting for inflation, state aid to Minnesota's cities grew four-fold during this period. From implementation of the Miracle in 1971 to 1987, inflation-adjusted city spending on several of the "high profile" city services which are the topics of aid cut testimony (fire, police, and parks) was essentially flat. Spending on streets increased modestly. In contrast, growth in the spending category labeled "other" far exceeded all other spending categories. The Legislative Auditor noted that "over this twenty year period inflation adjusted city expenditures grew by 65% and this period of fastest spending growth corresponds to the time of rapid growth of state and federal aid to cities." According to the OLA, the results suggested cities used 82% of the additional aid to finance increased spending and 18% to reduce property taxes. When the engine of the Miracle (large increases in personal income tax receipts due to unindexed tax brackets combined with inflation-fueled wage growth) sputtered, the spending continued and property taxes as a percent of personal income steadily increased through the mid 90s.

Lesson #2: LGA Cuts Trigger Very Diverse Responses by Cities

What happens when state budget deficits make it difficult or impossible to maintain current levels of local government aid? Unwinding local spending is politically and practically difficult. Some argue that LGA cuts guarantee a property tax increase - the only issue being whether the pain will be severe or unbearable. Yet history has shown city responses to local aid cuts are much more nuanced. A 2006 study examined city actions in response to $157 million in LGA cuts from 2002-2004 ³ . During this period, the median percentage of a city's lost aid replaced by property tax revenue was 61%. However, this fails to capture the extraordinary range of city actions in response to LGA cuts. About half of the cities either replaced less than 16% of their lost aid with property taxes or replaced more than 160% of their lost aid with property taxes. Nearly one in five cities that lost aid from 2002-2004 actually reduced their property tax levies. Coming on the heels of cuts earlier this decade, upcoming decisions will likely be even less pleasant. Nevertheless, this diversity of responses suggests matching local preferences with accurate tax pricing does work. The lockstep cause-and-effect relationships portrayed in aid debates are far from certain.

Lesson #3: A lot of LGA likely goes to the wrong places

The amount of LGA reform and "formula-fiddling" since its inception give testament to the notion that distributing LGA based on neediness is an exceptionally difficult political task. House Research staff identifies 7 "eras" of LGA design, and each era featured minor and major formula tweaks to correct for various consequences and problems that became apparent and make the aid system more politically acceptable.

The one major investigation to design an empirically robust aid system to target cities in need was a 1991 study commissioned by the Minnesota Legislative Commission on Planning and Fiscal Policy which came to be known as the "Ladd Study." ⁴ After an investment of thousands of labor hours by legislative and agency staff, stakeholder groups and three nationally-renowned public finance scholars, the report and its findings landed with a giant thud at the Legislature, The problem: the study's findings regarding optimal aid distribution bore no resemblance to the actual distribution of local government aid. The findings caused the late Dr. John Brandl, a state senator at the time, to note that it appeared that Minnesota could be just as efficient at distributing LGA by throwing money out of a helicopter across the state. The proposed formula reform was not adopted.

It is quite likely that the current LGA formula is in some cases preventing some cities with strong "need-based" claims from receiving any aid, A 2003 MTA study concluded that a number of large and growing cities have strong LGA claims but are being shortchanged by a formula only recognizes population decline as a "need factor." ⁵ The report concluded that many for cities over 2.500, LGA was likely making inequities among cities worse.

Lesson #4: LGA likely supports a significant amount of spending which has nothing to do with equalization goals (adequate services at reasonable tax prices)

A fundamental concept behind the notion of LGA is that state aid subsidies should be used only to help compensate cities for cost factors over which they have no control. A city with high poverty and crime rates requires more resources to achieve an adequate level of public safety and thus may merit some state aid. The problem comes in matching this idealistic objective with a general purpose aid program that allows cities to use the money for any purpose they see fit.

Our recently-released report from MTA's Center for Public Finance Research ⁶ flags one example of the dangers of general purpose local government aid. We found statistically significant evidence that LGA is strongly correlated with higher levels of expenditure on employee health benefits after controlling for other factors that may influence a city's tab for employee health care. Aid that enables greater subsidization of employee health care hardly passes the public purpose test. Among other recommendations, we argued that any LGA cuts should be prioritized based on deviations from state spending averages so as not to unfairly penalize those cities which have already invested in health care cost containment efforts. Critics argued that prioritizing LGA cuts in this way amounted to state meddling in the affairs of municipal contract agreements. The irony of this argument is seemingly lost on these critics – providing LGA at all is, by definition, the largest purposeful intervention of the state into local affairs and municipal contracts ever invented. As our study indicates, LGA already influences local government contract negotiations.

Echoes of the Past Still Apply

LGA is an idea that has some conceptual merit but faces enormous implementation hazards. Local government aid can ensure citizens around the state have access to adequate levels of services at reasonable tax prices. It also can stimulate spending levels beyond what local taxpayers would otherwise approve of paying for with their own property taxes. Distribution mechanisms have a huge influence on whether the result is a defensible, economically-efficient program or a significant waste of state taxpayer dollars that could be spent for other purposes. Given the debatable assumptions behind LGA and the historical lessons, a compelling case can be made that aid to cities should be subordinated to other claims on the state budget.