TEL It Like It Is

Do State Tax and Expenditure Limits Actually Limit Spending?

Many states have attempted to slow state and local government spending by adopting tax or expenditure limitations (TELs). This study looks at the impact of these TELs on government spending.

In 1976, New Jersey became the first state in the Union to enact a tax or expenditure limitation. It was a statutory limit on state spending that forbade legislators from growing expenditures faster than state income growth. Though legislators let it expire just six years later, the New Jersey statute kicked off a new experiment in constitutionally limited government. In the next decade, nearly two-dozen states would enact TELs of their own. Today, 27 states operate under TELs, while a 28th state—Colorado—has temporarily suspended its (particularly restrictive) TEL until 2011. (Other states limit local spending by cities and/or counties, but this is not the focus of my research.)

Do TELs limit budget growth? Early tests of this question concluded that they do not. As time has permitted more data and more sophisticated means of testing it, however, some subsequent research has concluded that certain varieties of TELs can limit spending in certain circumstances. In recent years, studies of TELs have tended to follow one of two tracks. They have either looked at the circumstances in which TELs are applied, or they have looked at the properties that make some TELs effective and others less so.

Studies examining the circumstances in which TELs have been applied have tended to focus on whether TELs have a different impact in high-income states relative to low-income states. Since many TELs (like New Jersey’s 1976 TEL) tie state budget growth to state income growth, scholars have hypothesized that TELs in low-income states will be more limiting than TELs in high-income states. Indeed, that is what the data suggest: TELs seem to be associated with lower levels of government spending in low-income states and higher levels of spending in high-income states. The latter finding is worth emphasizing: these studies have not simply found TELs to be ineffective limits on state budgets in high-income states; they have actually found that TELs are associated with greater than average levels of spending in high income states. It may be that in high-income states, TELs increase spending by acting as an excuse for elected officials to spend up to the limit.

A second (and less-developed) class of studies has focused on the variety of forms that TELs can take and has concluded that TELs can effectively limit budget growth, but only when they take certain forms. For example, Michael New (2001 and 2003) has argued that TELs limit spending so long as they: a) are based on the relatively restrictive “inflation plus population” formula, b) are passed by citizen initiative, c) immediately refund surpluses to taxpayers, and d) mandate reductions in the limit when the state devolves a function of government to the localities.

This study combines the two approaches described above to evaluate TELs based on where they are applied (high- vs. low-income states) and based on how they are structured. A more detailed and comprehensive dataset permits me to explore the various structures of TELs in greater detail than previous work.