Relaxing electoral constraints in local education funding

If voters do not like federal tax policy, they can elect new representatives. At the local level, though, voters can directly deny the tax increases their elected officials propose. All but three states have a limit on either the taxing or spending abilities of local governments and, most commonly, state laws require that increases in local taxes receive a public referendum before they are enacted (Mullins, 2010). In the November 2020 election, voters nationwide decided on the fate of $52.7 billion of proposed funding, down from $70 billion 4 years prior (Pierog, 2020). Requirements that budgeting questions be put to a vote reduce the level of spending (Feld & Matsusaka, 2003; Funk & Gathmann, 2011). However, the rules that govern these votes vary in myriad ways. States differ in the types of taxes or spending the rules cover: Some only allow proposals for capital spending, while others also include current expenditure spending. States also often limit the total amount of tax revenue or the tax rate governments can propose, and vary in whether they adjust for inflation, for changes in population, or for growth in the property tax base. Certain election rules limit the timing of when governments can put proposals on the ballot, since local governments may use this timing flexibility strategically (Anzia, 2011; Kogan et al., 2018; Meredith, 2009). States also differ in the vote share required to approve the proposals, with many proposals requiring more than a simple majority. There is little empirical research, however, documenting how these differences in voting requirements may affect government budgets, local spending, and the provision of public goods and services. This is a significant gap in the literature given the millions of dollars in funding at stake in each local referendum and the billions at stake nationwide. In this paper, we study a proposition in California that weakened the constraints on some local governments by lowering the vote share required to approve capital funding for schools and community colleges. We use a difference-in-differences design around this policy change and data for over 4,000 local elections across the state over 2 decades. We estimate the effects of this policy change on the proposals made by affected districts, their outcomes at the ballot box, and on the eventual funding outcomes. We develop a theoretical model of the interaction between a school board and voters, building on the literature in local & & & In the school board a tax that the can or the school board to to than of the However, in how vote the of this We use this model to how a change in the required vote share the of the proposals and the vote and to how these effects on the in between voters and elected We study how the policy change affected the funding outcomes of local We that to in local government a more than This is a and that governments There effects on funding outcomes in more and we effects in with of and on the of funding is most by referendum model that the of a to a change in the vote share on the of between voters and their elected the level of in outcomes that the elected officials the empirical types of have or more election outcomes. There are these effects the of elected officials when funding and of voters when voting on the We that the policy change made more to a to However, the of the proposals by or a the proposals may in vote on a we that the policy change in proposals a of votes in However, this than the in the vote share In school and community in for their proposals that than the amount of the policy we change in the that affected of new funding, and increases in the that they new these proposals that from the they in increases in the of on and a on vote with the spending than the voters, when the of to of we to empirical on the and policy effects of tax This literature on the of and the on government in government officials have et the of these to proposals of and voting on school a model of school and voters and that school a level of spending than voters but are in their We this of by a policy and that the voting in proposals is with their We also to empirical literature on the effects of tax on the of public & & & 2010). study the of the of local tax in this we study a change in a of the limit et that and the effects of and requirements in the of their However, their data not in the vote We that the vote share required to public spending effects on the amount of spending proposed and this is the to to study the effects of the change in the voting have voting on the that in this policy and a model of for rules and that for the rules is to the in their school et that voters in with a more of school more to vote for the in the literature voting rules in the of voting & & to is for requirements with the of state and that requirements to This these by the of a change in the of a in the of public of voting on school and the of the funding significant and the of school funding in we study the of the school votes et the effects of the capital on and their effects on are of a literature that the effects of school the on on outcomes is school may to outcomes & 2011; & & However, to study school capital like we have of effects on outcomes et al., et al., is and effects on in et effects of capital on and and effects of spending on have that on of spending like et al., 2020). have effects on that a school and school we to a literature on the for public goods in We how the effects of the differ by the of the literature that is to for government et al., et al., However, some of these are not to et al., to is and that more than but of proposals they the to the policy change we that local elected officials in spending on public goods & that in more elected officials a level of spending than the of the is In the we local government funding and the policy change we in we in we the data we in we the empirical in we and in we study on funding for capital in local governments in governments with the to tax include school districts, community districts, and that a and In this we the for these governments in revenue for California local governments are from property taxes but can revenue by taxes or vary on the of property tax increases local governments can adjust and All these tax increases require However, and taxes are only to and not school taxes can be for can only be to capital and can be either revenue or do not require in are only for local governments with to In require are by the and of the and are with property taxes over many California school on than taxes are than state also a funding in school capital from require a At the of the vote on the state of the of new school and of the of local the amount of in the state for and funding school and funding for the This the of state for school In in the state the to and for with schools that to We the of these on in the school are not the only local that can There are community in the and the public in the have the to their capital the school community can and in they to on for the school state of community when the the but the can use local to the of In a of and the of the to the state in the of school to local on the state for local school to school this spending on capital In the of school to to school school a and put on the ballot in a required a to the school also required a to that school not this California voters in have made for proposed to the of in California voters have the voting from a to a simple majority. in November voters the to a to for and community by that requirements on school to for the required that a election or local election than a In required a of the board approve the than a and a amount for the tax rate to the proposals that not these requirements not for the vote not the vote requirements to the of the the new voting and rules in a of changes to the and are to be most the change in vote share that we on and the election timing of the of the vote in of proposed before and the change in vote the required vote share before and the of is that the of vote over this this to have at of the with the at of votes in in the years prior to the change and of votes in the This of is than the in the required vote by the in the share of proposed that the in the of the their vote share the of vote by and a the share and of proposed and to by school and community colleges. the for in the at the vote for the in to the at the vote for the to the of by school and community from data in the of the of local proposed and these each There is a change in the new policy of proposed and increases and they more in This is with the that when the referendum most and local elections in most are proposed in years to for the proposed in years but not are in local elections and are for the elections have than and the to elections may have the of requirements only to and not the effects we in this include the effects of the change in the and the change in change in the of voters have affected the level of spending and the vote share that given but is not in the to a election a on votes for officials have to put on election before the policy However, we many in and elections prior to the policy the Kogan et are the effects of elections on for public spending. elected officials may have that in elections in California more to than in data this when to the prior to when the vote not vary with the the to elections may have voting outcomes in either We to this when we in the of the of the of we to a voting model to the of the policy model how a change in the vote share required to a funding referendum can change the funding that local propose, the that and the In this we the model and the and of the model are in model on the by and the interaction between a with and voters approve their in We this model by the of the voting that are required to In model is to and voting on model the interaction between voters and a in a local expenditure on a public There is a level of funding, the to a for spending on the public If the or the to receive the required voters, the level of funding is voters a level of funding, their on the of the public and the of the taxes that be to We the funding level from the with policy given by We the funding is at the voters level is the between the and the We that is is with the of referendum that the spending. is to is with the voters, the that are in the of the funding level from their level of of in the model is the whether to a referendum to a public level, than the level, they a the level is they a the voters vote or If at share of voters vote in is is the vote share the of this the is We that is at we that the level of the public In to in the outcomes of proposals, and allow the model to proposals, we that are to the that the not at the they that given the vote share required is are whether they they their to their over the outcomes. policy change we study is to a change in the vote We are in how the proposed level of of and the public we the level of the of a in the vote on each when is and is not between voters and the on the level of the and voters the level of spending, a in the not change the or the of votes in the of the and, a the level of also the spending than the a in the in proposals and, vote in However, since the is the that the to in In the model a for the of a change in the vote on the of local and how the policy change that we study the proposals made by and how voters vote on for the of between voters and elected We of and data on public schools in California over the 2 decades. of data is the of election for local in California between and to et and others have from the California a of the for California at California We include elections from community districts, and school of that have or are to only but we include funding in each in the we the of the ballot the proposed for We also whether the and the of votes for and from we the share of voters for the we study There are types of of the and the for are the we on in the We the with of We use and on and on from the of We use from for and of each local from the are for and school are not for community districts, of the of in a community we their current from the for California with a of We the of with the In we is a This of from to districts, community districts, and each we the of elections and the of proposed and and the amount of funding proposed and is with the of elections between and we the and vote this the types of 2 of the by prior to the of and of the school proposed a of community in the and to this of This given that these have a of than school and community the and more to put types of funding proposals, changes in on the of and elections not vary across to be than the proposed by and In this we to the of the change in vote requirements on the spending proposals made by the the the spending proposals from voters, and the capital spending of local We use a difference-in-differences to However, they require types of We the we use to the questions government proposals and funding outcomes. We the we use to questions on local governments that proposed a We this the in whether or not they made a that We use this to estimate the of the policy change on the of and the policy the total amount of proposed and funding estimate these we use data on from to We use this to the of the policy change on the vote share in of the and the that a However, officials may to the policy change by in that they not have prior to the change this is of we in the be in be to this the of the policy change on the level of for spending. is the of the change on the level of for the proposals that the officials to We the three we whether the of school and community in of their of a and the of the This the in on and we whether the of from a of This the in on we study the effects of on funding the on government and the on voting outcomes. This the in on and the we to across and We the of the policy change on the and of proposals made by the from of the in the in the are the the interaction between for and the the of on the that a a in effects and a of while 2 the effects with effects for each to the to differences between and In the estimate is and not not more than to of however, that the is more There are changes in the timing of when are of a the in the across and in each and for of the of is that to for the vote governments not votes in elections are in the of the is with affected school and community the timing of their proposals in to for the on government of of the vote share to but to differ by while to in in is for whether the government proposed a that in is the amount of the government proposed that by the in when is allow for at the government and 4 of the of on the amount of funding proposed by estimate is significant and a in and a in of 2 of a the in the across and in each and for the of and elections the policy on proposed funding is not significant in most In a in we years since the on proposed funding not effects on the of the the policy change more to and more to propose. the when governments often new in are more to may proposals before is that the of these effects in a of total capital spending while more or to a we the of the policy change on the of the elections that are this we on of We use the election of election for a funding across the types of elections are proposed by school and community districts, and elections are at whether the of the of for to funding proposed by and 2 of 4 a in vote of the not include a of the amount that the vote when the from to in the the policy change governments to more spending, may the of voters that are to However, the change in the of of the that may have also not to the vote share down by the amount of the policy of the of vote the difference-in-differences each in study There is a in the years prior to the of by a in that in from a in a in the on the vote is than the in the vote In the of the is the on vote share and of proposed of of the vote share to 2 but to differ by while to in allow for at the government of 4 the of on the that a We that more to to in This is with the in the election vote than the in the vote share by of the of is a in the years prior to the of by in that in most these with from the we that the of proposed in the while the of their This that the between is on the new are proposed but they are more to This is with the a level of spending than the voters of the effects of on outcomes. the change in vote share required to affect this differ across we to the and are in outcomes at the of the policy affected are more to than However, of 4 that this and effects by the of proposals to however, the of the timing of their proposals is a change in on government level outcomes. of of the vote share to but to differ by while to in in is for whether the government proposed and or more that in is the amount of proposed and by the in when allow for at the government and 4 of that a on school funding outcomes. We that funding in by to a more than This is than estimate of the amount of proposed funding of the of the of of 4 that this on funding is in elections in In the we that the policy affected outcomes. There on whether a but in the total amount of funding that proposals, the vote share but the of effects to in in and in total funding are with the vote share the the in the the vote share but only the not a election, change to the timing of the affected the of voters in This have either or the spending of the voters voting on the If the level of spending voters to approve vote proposed funding and funding and is with the effects we are not in the In we include for the timing of the election and of the the effects in the with for whether a a or a with the with of the for election timing in the in by the effects with and and by the effects and a for whether the election in the are However, the in the are with the effects of a change in the vote share when school spending than that vote the funding these we in proposed a in vote in the of and in We to in the effects across with We do not have a to estimate the of on the of to we this theoretical model that the of a to a in the vote share by the of the between voters and their with the of the elected over how the the of effects types of have and that the between voters and is not the or we are in the to is the from a policy this in types of the change in funding to the policy This in on the of elected officials spending the We on the and of the the of and the level of funding the prior policy We are in whether the effects of differ by the of the We the of on and allow the on the level of funding to differ with of the in the of the for each with the the in that that with the most the the of and funding affected by the of the in of on capital funding of the of the vote share on the amount of by the in when to but is to differ by the to that In is to differ by of the in while in is to differ by of to funding, or of the each are on the allow for at the government the the most the in the of increases in funding to the only by whether the or the level in the significant between and in the on but the policy change a on the amount of funding by that the effects in the are in in with but not between of or may have funding of by of the share of the in the have of but in the have or of on whether the of the that the effects of are in in the of the of the these have these of are of of at of the the on with of between the and is for outcomes. of we is in the of the and have the of voters to be more and since they are to have in school also be to vote for taxes at school the this be by the that school can be of how of vary by of the share of the that is over There is a with the between the and is In the in we whether effects with of capital spending prior to the policy many not approve funding in the years of data we we with of funding and the funding the of these There is between funding and the effects of of this is to whether the funding gap between the by of the by of funding in the prior to the policy In that is between the amount of funding and while to a in funding in these are not in with to the by the most to and approve the amount of spending prior to the policy the policy effects the to reduce the gap between the and in funding and However, the gap between the and the more than a gap of We do not that the between the or prior funding of and the is is with of population, and and and are and also with more and to be for and the between these these types of the policy change affected the in that are more or have of and and for governments to proposals in the increases in funding in In to the we have we a we community from the a share of the and may be than school the for the and community colleges. are we limit the to years before for only years the of this is to at the effects of the is in some of the study that the effects are more in these are with the of the policy on the amount of funding proposed is and only significant at the This is with that the effects on funding in and on vote share is also However, the increases in the of a and in the amount of funding this on in differences the of to & In the may with of the but not with a of In we the of or a in a and funding outcomes in dollars of the effects the are or the There are with of these in like are many since many do not or funding in some years & & the to in the of that school have to is by and from types of funding, in is a in taxes on the of new and on the of new or when we include taxes the we a in the vote share and a not significant on This is but we in that school their taxes with proposals are than prior to only school taxes each is that the on taxes the on that we this change in proposals, of how can be the in taxes and the in that school their funding for current with funding for capital also have their revenue for capital funding to like and current total and current from the of We in the that we have that are of by the in each are the however, is a not significant estimate of the on This to a estimate on the we on a and also the dollars estimate of the of on dollars can be the effects of the policy in state funding for in around the of have for in the policy change we the state the in some school school in that schools that a level of the of is to whether this since schools often in of We a of schools from the California of and the school with schools from the are from the that the effects are not by this are in and and state also for funding new and school in made state for school to and may have school to to these also have is of schools or districts, is not to directly for in some effects of these increases in the of state to the are However, the timing and of are with a We or in either the proposed or between and the years in the state funding in but not In we total on whether are proposed over This is not with school to of state funding by when they This not that the state funding on school funding, state the local funding that we study in more funding than in the of state In this paper, we how a change to the voting required for a school in California affected the share of voters We that schools more than their funding to the In we that governments more to a but are more to a to the policy a of in a in the voting by from to We these in the of a model of the interaction between a and elected between voters and elected a lowering of the vote share not affect the proposals however, the that they we do in the proposals made and the that they This that is between the voters and the elected elected officials use their flexibility to more spending and that with a these in a in the amount of funding We that be with some to share a property tax with the that not to the policy changes in the tax rate by school affect voting in these capital funding in a school that voters tax for the or in they If that the the effects in the are the of these not the on the school that limit the of local elected officials and require tax to be referendum are the We study change in this and that the vote and more than funding from However, the policy change not the of school that proposed a In that funding that have proposed the This proposals may have and proposed policy change not have effects across effects in that not and This that or constraints on local elected increases spending more in with more to for this may officials in may have with referendum voters over their level of spending or have over referendum outcomes. have for policy that changes to local tax Some states the of these States change the with local officials can policy a in some to States also change the rules for in the that the of how these are can affect the amount of funding that and taxes that with is at the of is at of of data that the of this study are in the at is not for the or of by the than be to the for the