March 21, 2006

A FAULTY FIX:
Repairing the "Ratchet" Will Not Repair TABOR

Coloradoans recently voted to suspend the Taxpayer’s Bill of Rights (TABOR), the nation’s most restrictive limit on state taxes and expenditures, because it has caused a sharp deterioration in services ranging from education to public health to transportation.

Those pushing for the adoption of TABOR-style measures in other states are seeking to dissociate their proposals from Colorado’s experience under TABOR by arguing that Colorado’s problems stemmed entirely from one feature of its TABOR, the so-called “ratchet.”  (The ratchet, part of the formula for determining the state’s annual spending limit, makes it extremely difficult for state services ever to recover from an economic downturn.  See box on next page.)  TABOR proponents further argue that they have fixed the ratchet problem in the TABORs being proposed in other states, so these other states would not experience the problems found in Colorado.[i]

These claims are inaccurate.  Fixing the ratchet would not prevent TABOR from causing public services to deteriorate.  It is the basic TABOR formula, not the ratchet, that undermines states’ ability to fund services.  This is evident from the fact that public services in Colorado declined significantly before the 2001 recession began, and thus before the ratchet could have had any effect.  For example, between 1992 (when TABOR took effect) and 2001, Colorado fell from 35^th to 49^th in the nation in K-12 education spending as a percentage of personal income and from 23^rd to 45^th in access to prenatal care, a sign of funding shortages in local health clinics.

Moreover, the changes proposed to deal with the ratchet in other states do not necessarily eliminate the problem; they only moderate it.  In most cases, states would still be unable to restore services in the aftermath of an economic downturn.

TABOR Formula Squeezed Services Even Before Ratchet Took Effect

Colorado’s TABOR is the most restrictive type of tax and expenditure limit (TEL) in the nation.  It is a constitutional amendment that restricts annual growth in state and local revenues to the rate of growth of population plus the inflation rate for consumer goods.  It also requires voter approval to override the revenue limits and requires that revenues in excess of those limits be refunded to taxpayers.[ii]  

TABOR’s “population-plus-inflation” formula prevents a state from collecting the amount of revenue needed each year to maintain existing public programs and services; as a result, it progressively shrinks state budgets and public services over time.  This is because the formula is an inaccurate measure of the costs facing state governments.  The cost of health care, education, corrections, and other areas of government generally grows more rapidly than the cost of consumer goods.  Moreover, certain segments of the population — such as schoolchildren and the elderly — require more public services than others.  When these segments grow more rapidly than the population as a whole, the cost of government grows more rapidly than the TABOR limits allow. [iii]

Unlike the TABOR formula, which each year reduces a state’s ability to provide adequate public services, a ratchet component of a TABOR formula comes into play only in the aftermath of recession, as explained in the box above.

The extent to which TABOR shrinks services even without the operation of a ratchet can be seen by the deterioration in services in Colorado between 1992 (when TABOR took effect) and 2001 (before the recession began to depress state revenues, and thus before the ratchet had any effect): 

• Between 1992 and 2001, Colorado declined from 35th to 49th in the nation in K-12 education spending as a percentage of personal income.[iv]  In 1992, Colorado’s average per-pupil K-12 funding was $379 below the national average; by 2001, it was $809 below the national average.  Thus, even as Colorado was becoming more prosperous during the economic boom of the 1990s, TABOR was forcing it to weaken its commitment to K-12 education.

• Between 1992 and 2001, Colorado declined from 35^th to 46^th in the nation in spending on higher education as a share of personal income.

• Between 1992 and 2001, Colorado declined from 23^rd to 45^th in the nation in access to prenatal care, a sign of funding shortages in local health clinics.

• Between 1992 and 2001, Colorado declined from 18^th to 44^th in the nation in health coverage for low-income children.  During this period, the share of low-income children who lack health insurance rose from 16 percent to 24 percent in Colorado, even as it remained virtually unchanged in the nation as a whole.

Thus, even before the ratchet had any effect on Colorado, the TABOR formula had seriously undermined the state’s ability to provide adequate public services.[v] 

Further evidence that TABOR’s fundamental problem is its basic formula rather than the ratchet can be seen in the graph [on the next page], which shows what would have happened if a TABOR without a ratchet had been in effect in all states since 1990.  Between 1990 and 2004, state spending averaged 6.9 percent of the gross domestic product (GDP).  Had a TABOR with no ratchet been in place in all states starting in 1990, state spending would have declined steadily as a share of GDP, from 6.5 percent in 1990 to 5.4 percent in 2004.[vi]  In 2004, states would have had $158 billion (20 percent) less ability to support services than they actually had.  Put another way, if all states had had a TABOR without a ratchet, they would have experienced a deterioration in public services that is similar to what Colorado experienced prior to the recession. 

Ratchet Exacerbates TABOR’s Problems

The preceding discussion shows the damage that TABOR can cause even without a ratchet.  A ratchet makes the damage even worse by preventing public services from fully recovering from economic downturns.  (The existence of a ratchet typically is hidden within the wording of a TABOR proposal.  Any TABOR proposal must be carefully scrutinized to determine whether it contains a ratchet.)

State revenues often decline during a downturn, and states often cut expenditures to balance their budgets within those lower revenues.  Normally, states can restore these expenditures when revenues rebound during the recovery.  Colorado, however, was unable to do this because the low point of revenues during the recession became the new base from which the subsequent year’s TABOR revenue limit was calculated.  This ratchet effect prevented Colorado from returning to the level of services the TABOR limit would have permitted had there not been a recession.

Colorado Eliminates the Ratchet, While Other States’ Proposed Fixes Do Not

As noted in the box above, a ratchet occurs when revenues stagnate or decline during a recession and TABOR’s inflation–plus-population-growth formula is applied to that depressed level of revenues (or expenditures supported by those revenues) to determine the subsequent year’s TABOR limit.  This is the way Colorado’s TABOR operated prior to the passage of Referendum C in November 2005.  Referendum C suspends TABOR for five years and modifies TABOR so that when the suspension expires, the population-plus-inflation adjustment will be applied to the amount of allowable revenues in the previous year, not the amount of actual revenues in the previous year.  This means that if Colorado is forced to cut services during a future recession, it will be able to restore some of those services when the economy improves.[vii]

Now proponents of TABOR in other states claim they, too, have fixed the ratchet in their new proposals.  But they have not; the new proposals do not do what Colorado has done to eliminate the ratchet.

There is a new crop of TABOR proposals referred to as Stop OverSpending Initiatives (or “SOS”) being put forward in at least five state— Michigan, Missouri, Montana, Nevada and Oklahoma.  These proposals often include two provisions aimed at fixing the ratchet.  The provisions are an alternative limit calculation and a budget stabilization fund that is to be used when revenues fall short of the TABOR spending limit.  The SOS limits are expenditure limits, so the following discussion will focus on a spending limit rather than a Colorado-type revenue limit.  The effects of revenue and spending limits are similar.

Alternative Calculation

Under the alternative limit calculation, the TABOR limit for a given year would be set as the greater of previous year’s spending adjusted for inflation plus population growth or the previous year’s limit.  (Note that the second formula is the nominal level of the previous year’s limit, not the new Colorado formula, which is the previous year’s limit adjusted for inflation and population growth.)

Using the previous year’s limit freezes the TABOR spending limit during an economic downturn and keeps it frozen until revenues recover.  

It would work as follows.  If expenditures in any year are lower than the TABOR limit — because revenues are down and the budget stabilization fund is not available or not used (see discussion below) — then under the alternative calculation, that year’s limit would continue into the subsequent year without adjustment for inflation and population growth.  The frozen limit would continue to apply until there is a year in which the previous year’s actual expenditures adjusted for inflation and population growth are higher than the frozen limit.  The frozen limit might remain in place for two or three years, since fiscal crises have lasted for a number of years in many states in recent downturns.  If the limit were frozen, public services would fall substantially behind even the standard of need recognized in the TABOR formula; the population would continue to grow and inflation would continue to push up the cost of services over those years, but the limit would not be adjusted to reflect this growth.

In a state without a TABOR limit, there is an opportunity to restore services that may have been cut during a downturn.  Coming out of a fiscal crisis, revenue growth is often unusually strong, and that “extra” revenue can be used to reassess the cuts that were made during the fiscal crisis.  Under the alternative calculation, however, that opportunity would not exist.  At the end of the fiscal crisis, expenditure would be allowed to grow from the frozen level by only inflation and population growth; no make-up growth would be allowed.  Thus, the SOS proposals moderate the ratchet effect but do not eliminate it.[viii]

Budget Stabilization Fund

In the best of all circumstances, the budget stabilization fund could eliminate the ratchet effect.  Here is how it could work.  Suppose actual revenue falls from one year to the next due to a severe economic downturn (as occurred in Colorado between FY 2001 and FY 2002) and that revenue is now less than the spending limit.  (In a balanced budget context, a state cannot spend more than its revenues.)  An amount equal to the difference between actual revenues and the allowable spending limit would then be transferred from the budget stabilization fund to the general fund, so that the state could spend up to the limit. The next year’s limit would be calculated using this higher level of spending and not the level that would have occurred if the state had only been able to spend the amount of actual revenue it collected. If this worked, there would be no ratchet effect.

In order for this full ratchet fix to work, however, four significant conditions must be met.   There is evidence that these conditions will not or cannot be met in some states under their TABOR proposals.

• Design of the Budget Stabilization Fund (BSF).  Budget stabilization funds typically have caps — a limit on the amount of monies that may be held in the fund.  In order for the ratchet fix to work to the full extent, the BSF cap must be high enough to allow the state to spend up to the limit (assuming enough funds actually are deposited in the BSF).  In some states, this clearly is not the case.  For instance, in Nevada the BSF is capped at 5 percent of the spending limit, while in Oklahoma it is capped at 10 percent of the spending limit.
  
  In recent years, state fiscal crises have been deep and have lasted for two, three, or even four years in some states.  For example, the states together entered the most recent downturn with reserves equivalent to 10.4 percent of a year’s expenditures.  While this is much more than was on hand in either of the previous two downturns, it turned out to be far from sufficient.  Indeed, it constituted only one-fifth of states’ cumulative budget gap over the multi-year fiscal crisis.  This history suggestions that a cap at 5 percent or 10 percent of annual spending is likely to be insufficient.

• Adequate Amount in the Budget Stabilization Fund. Even if the cap is high enough, the BSF must actually contain enough funds to allow a transfer to compensate for revenue shortfalls. The level of funds in the BSF will depend on how mature it is (i.e. how many years it has had to accumulate funds before a fiscal crisis hits), the strength of the state’s revenues (i.e. if revenues are not growing faster than population plus inflation, then no money will be deposited in the BSF), and whether filling the BSF is a high priority use for revenues above the limit (i.e. many proposals require that any excess revenues first be deposited into an emergency fund before going into the BSF and/or they limit the percent of excess revenues that go into the BSF. As Figure 2 shows, at least four states — Michigan, Montana, Nevada and Oklahoma — limit the amount that can be transferred into the BSF to 50 percent of any excess revenues. The remaining excess is returned as tax rebates.)  If the BSF does not have enough funds to fully make up for budget shortfalls during a fiscal crisis, then the ratchet effect will not be fully eliminated.

• Use of the Budget Stabilization Fund.  The BSF has to be used to fill the gap between revenues and the spending limit.  This requires either political will of the legislature or explicit language in the constitutional amendment requiring a transfer in the amount of the budget shortfall. The Oklahoma proposal, for example, lets the legislature decide whether or not to use the BSF to cover shortfalls.  Even if the legislature decides to use BSF funds, it limits the amount of money that can be transferred in a way that could prevent fully eliminating the ratchet.[ix] The Missouri proposal also makes the transfer optional, and the Montana proposal is unclear on this point.

• Budget Stabilization Fund Transfers.  The BSF transfers must count toward the spending limit.  In other words, the amount transferred from the BSF must be counted as expenditures so that the subsequent year’s limit would be the amount of expenditures including expenditures from the BSF, adjusted for inflation and population growth.  If the transfers do not count toward the limit but only allow the state to spend up to the limit in the year of the transfer, then there is no effect on calculating future years’ limits and the ratchet would not be fixed.  As Figure 2 shows, there are at least three states — Michigan, Montana, and Oklahoma — where the language of the initiative is not clear on this point.

Conclusion

TABOR’s main problem is its “population-plus-inflation” formula, which starves a state over time of the funds it needs to maintain current services for residents and severely limits the state’s ability to meet new challenges as they emerge.  The ratchet effect exacerbates the funding shortfall caused by TABOR, but it comes into play only in the aftermath of a recession, so fixing the ratchet would not fix the problems inherent in the TABOR formula.  

The SOS TABOR proposals discussed in this report purport to eliminate the ratchet by providing an alternative formula for the annual TABOR adjustment and establishing a budget stabilization fund.  The alternative formula only moderates the ratchet; it does not cure it.  In the best of all circumstances, a properly designed and used budget stabilization fund could eliminate the ratchet in a TABOR.  But for a variety of reasons, those circumstances are unlikely to all be met.

No matter what bells and whistles are added, it remains problematic to use a rigid, constitutional formula to determine revenue or expenditure levels.

End Notes:

[i] Referendum C, which suspends TABOR refunds for five years and permanently fixes the ratchet effect, passed by referendum on November 1, 2005.  Not all TABOR advocates think the ratchet needs to be fixed.  A recent Heritage Foundation report argues that, “a better alternative [to eliminating the ratchet effect] is to take advantage of the ratchet effect to force state government to seek the same types of innovation that the private sector relies on to achieve a competitive edge.” Alison Acosta Fraser, Colorado’s Taxpayer’s Bill of Rights Should Not Be Breached, The Heritage Foundation, No. 1873, July 28, 2005.

[ii] The TABOR amendment also locked in the existing statutory spending limit of 6 percent annual increases, making it in effect a constitutional spending limit as well.  At the state level, inflation is measured by the Consumer Price Index (CPI-U) for the Denver-Boulder-Greeley metropolitan statistical area. The CPI-U measures the price changes for a basket of goods purchased by the typical urban consumer, such as housing, food, clothing, and entertainment.

[iii] For additional information on the problems with the population-plus-inflation formula, see David H. Bradley, Nicholas Johnson, and Iris J. Lav, The Flawed “Population Plus Inflation” Formula: Why TABOR’s Growth Formula Doesn’t Work, Center on Budget and Policy Priorities, January 2005.

[iv] Center on Budget and Policy Priorities calculation of National Education Association and Bureau of Economic Analysis data.  Current expenditures provide funding for the operating costs of K-12 schools.  Expenditures include items such as salaries, fixed charges, transportation costs, school books, materials, and energy costs but do not include capital expenditures and interest payments on debt.

[v] The recession caused a further deterioration of services.  Had Colorado voters not voted to override TABOR for five years and permanently eliminate the ratchet, the further deterioration would have become permanent.  Colorado would not have been able to restore its pre-recession level of public services — as minimal as those were in many areas.

[vi] This paper shows maximum permitted expenditures.  Since states have balanced budget requirements, a revenue limit results in limited expenditures.  Some proposed TABOR limits directly limit expenditures; the ultimate effect is similar.

[vii] Colorado may not be able to restore all of the lost services, since the operation of the basic TABOR formula would still prevent full funding of services at their pre-recession levels.

[viii] In addition to this SOS proposal, there also is model legislation from the American Legislative Exchange Council (ALEC) that has proposed freezing the TABOR revenue limit when actual revenues decline and keep it frozen until revenues equal or exceed the frozen limit.

[ix] Oklahoma’s proposal (State Question 726) stipulates that if a revenue failure has occurred, an amount equal to no greater than 35 percent of the total amount in the budget stabilization fund may be transferred to the General Fund and any other revenue funds.

KEY FINDINGS IN THIS REPORT:

 The main problem with tax and expenditure limits like Colorado’s TABOR is the arbitrary formula used to limit revenues each year. Fixing the “ratchet” (a feature of TABOR that worsens revenue shortages after a recession) would not avert the sharp decline in public services Colorado experienced under TABOR.

 Services in Colorado deteriorated severely even before the ratchet took effect — evidence that the ratchet is not the major problem with TABOR.

 The current set of TABOR proposals — known as “SOS” — claim they fix the ratchet through an alternative formula for annual adjustments; the change moderates the ratchet but does not eliminate it.

 The SOS proposals also include Budget Stabilization Funds. A BSF could fix the ratchet in the context of these limits — but only if it is designed and used properly. The evidence suggests that would be unlikely in most states.