Printer Friendly

Do legislators vote their constituents' wallets?

I. Introduction

Do the economic consequences of policies affect the political behavior of elected officials? In particular, do the costs and benefits that flow to households in legislative districts affect representatives' support of policy proposals? In recent years, scholars have investigated this question by regressing roll-call vote decisions of legislators against measures of the economic interests of constituents and the ideology of the legislators |3; 5; 8; 9; 10; 11; 13; 14~. With some exceptions, such as Peltzman |13~, most scholars who perform such regressions conclude that constituents' economic interests are not the sole determinant of legislators' behavior. Representatives' own policy preferences explain some of the cross-sectional variance in their support of proposals. In the words of economists, representatives shirk from their constituents' interests.

Many authors who regress roll-call vote behavior against measures of constituents' economic interests and representatives' ideology are aware that their measure of the latter (summary roll-call vote ratings by groups such as the Americans for Democratic Action (ADA)) is not a pure measure of representatives' policy beliefs because it is merely a summary of previous roll-call behavior. To eliminate the contamination, scholars regress the ADA rating against numerous economic, demographic, and regional variables. The unexplained residual variance is then used to measure representatives' ideology.

Those who use this technique obviously hope that these residuals represent personal ideology and not other possible causes of vote decisions. The reasonableness of this assumption depends on whether the variables regressed against the ADA rating capture all the other causes of congressional behavior. Kalt and Zupan |9, 293; 10, 112~, Carson and Oppenheimer |5, 172~ and Kau and Rubin |11, 64~ exclude three important classes of variables: the positions of interest groups and executive-branch actors, and the likelihood of proposal passage, all of which affect the decisions of legislators. This exclusion increases the probability that the residuals in the ADA equation represent not only the personal ideology of representatives, but also the ability of executive-branch and interest-group actors to affect reelection.(1)

Even though critics of this line of research focus on the nature of the ADA residuals, the measures of constituents' economic interests are equally problematic. In Peltzman's |13, 184~ words, "The usual procedure in modeling legislators as agents has been to treat all the residents of a legislator's district as his principals. That is, the vector (of exogenous economic characteristics) usually consists of a set of average resident characteristics, that is, per capita income, education, and so on." Individual-level economic data are not used to measure the economic affects of policy proposals or to estimate the political salience of the economic effects.

In this paper, we do not attempt to ascertain the effects of constituents' economic interests and legislators' ideology on the policy choices of the latter. Instead, we estimate the economic impact of a policy change at the household level and then ask whether the behavior of legislators toward the policy change is consistent with any one of several plausible aggregations of households' economic interests into a political referendum under majority rule. Following Coates and Munger |6~, we also test whether the cross-sectional variation in legislators' tendency to make policy decisions that are in accord with their constituents' economic interests varies with the margin of victory received by the legislator in the previous election. In addition, to the extent that legislative behavior does not correspond to any plausible aggregation of household economic effects, we examine whether voters react in the subsequent election by punishing those legislators who did not vote in their constituents' economic interests.

II. Economic Factors and Legislative Behavior: A New Approach

In this paper, we examine policy choices in which it is likely that the effect of economic benefits will be salient and we use data that allow us to measure those benefits at the household level. The fiscal reforms enacted by New Jersey in 1990 and 1991 were (and are) extremely salient, and legislators did not use parliamentary tactics to disassociate themselves from their decisions to enact the changes in taxes and aid |2~. If legislators ever respond to the economic effects of policies on constituents, we surely should observe such behavior in circumstances like those experienced in New Jersey in 1990 and 1991.

We use a unique data set of matched income and property tax data for all New Jersey homeowners to estimate the net costs and benefits at the household level of the system of income taxes, property taxes, property tax rebates, and school aid during 1987. We then estimate for every household the net benefits of the massive fiscal changes enacted by the New Jersey legislature in 1990: a large increase in income taxes, school aid to poorer jurisdictions, and the redistributive character of property tax rebates. We do the same for additional reforms enacted in 1991: a decrease in school aid to the poorest jurisdictions and an increase in general municipal aid. Finally, for both these policy changes, the household data are grouped into legislative districts to determine whether state senators' decisions to support or oppose the reforms enacted in 1990 and their partial repeal in 1991 reflect the economic interests of households calculated either cardinally (the mean voter, dollar-weighted) or ordinally (the median voter).

These data allow us to directly observe the economic effects of various tax and expenditure policies at the household level and eliminate the need to use the imperfect proxies so often observed in roll-call studies. In addition, these data make clear that the determination of whether the economic interests of constituents have been adequately represented depends on whether the costs and benefits are aggregated across individuals (1) cardinally or ordinally and (2) district by district or statewide.

III. The Incidence of School Aid and Property Tax Rebates

The incidence of the state and local government tax and spending system is usually approximated by the sum of state and local government taxes. Intergovernmental grants, a form of state government spending, are usually considered in isolation, because the incidence of spending is usually treated separately from the incidence of taxes. However, even though grants are an expenditure item for the state government, they represent an important source of revenue for local governments. Hence, any analysis of a change in grants distribution that does not account for the resulting change in local taxes will be incomplete. It is important, therefore, to consider the role of grants as a local revenue source and their impact on local voters as an increase in local government resources. Aside from this change in the treatment of grants, we follow the rest of the literature on tax incidence and do not explicitly consider the incidence of state or local government spending.

The government receiving an intergovernmental grant may direct the money either to reducing local taxes, increasing local government spending, or both. In order to identify the effect of the grant on individual taxpayers, one must make some assumptions about the allocation of the grant between these two uses.(2) We assume that reductions in local taxes and increases in local government spending are valued equally and identically distributed among local taxpayers. In other words, we will model the incidence of a change in grants as being the same as the incidence of a change in local taxes. We now turn to the question of determining the incidence of local taxes.

The standard theory of tax incidence suggests that in an open economy (such as New Jersey and its constituent municipalities)(3), the burden of taxes will be borne by relatively immobile factors of production. The most immobile factor is real estate (sites). For the purposes of our analysis, we treat sites as the only immobile factor, and assume that all gains and losses resulting from the shift in fiscal policy are reflected in site values.(4)

It is straightforward to implement our incidence assumption using our data on New Jersey homeowners. Every property has an value assigned to it by the local government for tax purposes. Clearly, one can total these assessments over all properties within a jurisdiction. Define an individual's tax share to equal his assessed value divided by the total assessed value in a municipality. Note that the property tax levy and the total assessed value are both determined by the local government. (The tax rate is uniquely defined by this process, as it equals the tax levy divided by the total assessed value.) An individual's property tax payment is simply his tax share multiplied by the total property tax levy. If the total levy falls as a result of intergovernmental grants, then an individual's property tax falls by the amount of his tax share multiplied by the grant. This is the way in which grants are allocated among individuals in the empirical work that follows.(5)

The incidence considered in this work is the "impact" incidence of intergovernmental grants. The long run incidence of the grants will depend on the reactions of individuals and on whether individuals are able to transfer tax increases to others. The long run incidence of a reform is not necessarily identical to the impact incidence. This may be a shortcoming in the analysis as pure economics, but it may not be a flaw when modeling the political choice process. It depends on the calculations of tax and expenditure incidence that voters and their representatives are making when choosing among competing policies. One cannot dismiss the possibility that the impact incidence is the measure of costs and benefits used by the participants in the political process. The use of impact incidence may be only an approximation to the way that the economy works, but may be an exact model of the public sector decision making process.

IV. The Incidence of Taxes and Grants in New Jersey

Table I displays the average income and property taxes, school and municipal aid, and property tax rebates paid (or received) by owner-occupied households in New Jersey in 1987 and after the reforms of 1990 and 1991.(6) Our calculations imply a tax and aid system in 1987 that is regressive for those households with less than $30,000 gross income and roughly proportional thereafter.(7) For expositional simplicity, our results do not include the change in federal tax liability for those who itemize, although doing so would not affect any of the results in this paper (details available upon request).

In June 1990, the New Jersey legislature, at the request of Governor James Florio, substantially increased income taxes, especially on higher income households, and used the proceeds to increase aid to poorer school districts. Simultaneously, the progressivity of property tax rebates to households was substantially increased.(8)

As Table I shows, the net result of these initiatives at the household level was to increase taxes for owner-occupied households with incomes above $70,000. Also the school aid system was altered from one that gave essentially the same aid to all income classes ($904 aid per household in the $0-$5000 class and $908 aid per household in the $150,000 and up class) to one that gives $1120 per household to the poorest class and $1026 per household to the richest. The aid system, which is progressive at the jurisdictional level, is much less so at the household level because the variance in incomes within jurisdictions is almost as large as the variance in incomes between jurisdictions.

The changes in the household property tax rebate had a more progressive impact on the fiscal system than the school aid changes. The benefits to the lowest income households were nearly doubled from approximately $216 per household to approximately $430.

The 1990 reforms created numerous protest movements among taxpayers. As the data in Table I show, the reason is not so much the redistributive character of the policy changes, which is quite modest at the household level, but the net increase in government revenue, most of which was not returned in the form of school aid and property tax rebate as advertised. The average household's income taxes rose by $674 while the school and property tax aid received only rose by $195--a net income tax increase of $479 per household.

With elections looming for all state senators and members of the assembly in November 1991, the legislature enacted changes in aid to schools and municipalities in March 1991. School aid to the poorest jurisdictions was reduced and redistributed to non-poor school districts. In addition $360 million of the reduced school aid was distributed to all municipalities to lower property taxes.

As Table II shows, the revisions did not affect the statewide average taxes paid or aid received very much. The purpose of the 1991 changes was strictly distributional: to create benefits for the non-poor and reduce the redistributive character of 1990 reforms in an effort to appease voters given that the 1990 income tax changes generated more revenue than was distributed in aid. In general, poor jurisdictions lost revenue from the 1991 changes, but the results are so varied that they are not easily described. For example, Camden, a large poor city, lost $19.2 million in school aid but received $12.7 million in increased municipal aid for a net loss of $6.5 million. Newark, another large poor city, lost $15.6 million in school aid but gained $18.7 million in municipal aid for a net gain of $3.1 million.

V. The Politics of Policy Change

Tables III, IV, V, and VI convert the economic effects described in Table I into simulated political decisions using several assumptions.(9)

(1) Households favor policies that lower their net taxes.(10)

(2) Married households receive two votes; all others receive one.

(3) Households' tendency to vote varies with income according to the following probabilities |16~:
Gross Income (in $1,000) Voting Probability

 0-5 .347
 5-10 .413
 10-15 .477
 15-20 .535
 20-25 .578
 25-35 .640
 35-50 .703
|is greater than~ 50 .756

Table II. Statewide Average Tax, School Aid, and Rebate
Payments for Owner-Occupied Households under the Status Quo,
the Quality Education Act of 1990, and 1991 Revisions

 Status Quo QEA Revised QEA

Income Tax 1630 2305 2305
Property Tax 2205 2205 2205
Homestead Rebate 192 184 181
School and Municipal Aid 944 1147 1201
Net Tax 2699 3179 3128

(4) Senior citizens (over age 65) are assigned a probability of voting of .688 regardless of income.(11)

Table III displays the economic effects of the 1990 reforms for each income class as well as the number of votes the reforms and the status quo would receive in a referendum. The average household under $70,000 income benefits from the 1990 reforms, but a majority of voters favor the changes only up to $55,000 in income. Between $55,000 and $70,000, the number of voters whose taxes increase under the reforms exceeds the number whose taxes decrease, but the average tax decrease is larger than the average tax increase. These results demonstrate the sensitivity of predicted electoral outcomes to the way in which voters' preferences are assumed to be aggregated. This finding, coupled with our results at the legislative district level, casts doubt on the standard approach of looking only at some "representative" individual. In our simulated statewide referendum, 54% (128,205) of the voters opposed the 1990 reforms while 46% (107,839) were in favor.

Table IV displays the economic effects of the 1991 reforms for each income class and the results of our simulated statewide referendum. The break-even point after the 1991 changes is also $70,000. The average household below $70,000 would enjoy reduced taxes and the average household above $70,000 would pay higher taxes relative to the status quo. Under the 1991 reforms, however, the majority of voters in each income class also favor the change for incomes below $70,000. The 1991 reforms reduce the gains of some of the winners between $55,000 and $70,000 under the 1990 reforms and give the proceeds to enough losers to create a majority of winners. The 1991 changes also increase the majorities in favor of change among the households with incomes between $25,000 and $50,000 (a large fraction of the population). As a result, our simulation suggests that the 1991 revisions to the Quality Education Act would command a statewide majority of voters (53.5%).

At the aggregate statewide level, the behavior of the legislature is consistent with economic rationality. The original Quality Education Act was passed in 1990 in great haste at the request of Governor Florio. As negative voter sentiment became apparent and the 1991 elections drew closer, the Democratic legislative leadership redistributed school and municipal aid to create a majority of middle-class households that were net beneficiaries.

The Behavior of Individual State Senators

At the level of the legislative district, political behavior seems less responsive to economic effects, at least as these effects are measured by our methodology. Tables V and VI display the results of TABULAR DATA OMITTED the voter simulations conducted at the state senatorial district level. The actual vote in the New Jersey Senate on the Quality Education Act of 1990 was 21-17 in favor with two abstentions. The vote was largely along party lines--21 Democrats favored and 16 Republicans and 1 Democrat opposed. Our simulation, however, predicted a vote of 16 in favor and 24 against the Quality Education Act that was less split along party lines than the actual vote--12 Democrats and 4 Republicans in favor and 11 Democrats and 13 Republicans opposed. In a simulation with cardinal voting (votes weighted by the dollar amounts the voters gained or lost) the results were even worse for the QEA--11 in favor (9 Democrats and 2 Republicans) and 29 opposed (14 Democrats and 15 Republicans)--because the losses of those whose taxes went up exceeded the gains of those whose taxes went down (the mean voter was worse off then the median voter).


Table VI shows the results of our simulation of the behavior of the New Jersey State Senate toward the revisions of the Quality Education Act in March 1991. The actual vote was 22-18 in favor of the revisions. Democrats voted 22-1 in favor and all 17 Republicans voted against the changes. Our predicted vote was 24-16 in favor of the changes. Seven Democrats were predicted to vote no; only one did (Senator Zane of the third district), and he was not predicted to do so. Eight Republicans were predicted to vote yes and none did so. Only 23 of 40 senators were predicted correctly, not much better than one could do by flipping a coin.

Table VII displays the estimated relationship between the incidence of economic costs and benefits in legislative districts and the roll call behavior of legislators. Senators whose districts have larger majorities that benefitted from the QEA or the revised QEA according to our simulations TABULAR DATA OMITTED TABULAR DATA OMITTED were more likely to vote in favor of both. The relationship is slightly larger and more likely to be different from zero in the vote on the revised QEA, but exists in all six simulations. The economic benefits do not predict roll call behavior much better than flipping a coin (the percentage of senators whose behavior is correctly classified in the probit models ranges from 55 to 62.5), but the difference is statistically significant. If the size of the majority in a senate district that benefits from the revised QEA changed by twenty percentage points from 40% to 60%, our simulation estimates that the probability of a favorable vote by a state senator changes from .358 to .597, an increase of 23 percentage points.
Table VII. The Effect of Household Economic Benefits on Roll
Call Behavior of State Senators under Various Participation

Simulation QEA Revised QEA

Turnout Varies by Income 2.1 3.0
Married Households Two Votes (1.9) (2.2)
Turnout Varies by Income 2.3 3.3
Married Households One Vote (2.0) (2.3)
Universal Turnout 2.3 3.4
 (2.0) (2.4)

Entries are probit coefficients. T statistics are in
parentheses. Dependent variable is roll call vote (yes or no)
of 40 state senators on Quality Education Act of 1990 (June 20,
1990) as reported in the Newark Star Ledger (June 21, 1990) p.
46 and the revisions to the QEA (March 7, 1991) as reported in
Star Ledger (March 8, 1991) p. A1. Independent variable is the
percentage of voters who favored the QEA or revised QEA
according to our political simulations as described in text.

Do Safe Seats Create Shirking?

Coates and Munger |6~ argue that the tendency for legislators to support policies that are not in the interests of the median voter in their districts is endogenous to previous political support. They argue that incumbents with "safe" seats (large previous electoral majorities) are more likely to challenge their constituents' interests than legislators in competitive districts with small previous electoral majorities. To test this proposition we regressed our measure of shirking, the size of the majority of voters in every district benefitted by the QEA interacted with the actual vote of the legislator, against the percentage vote won in the 1987 election. The coefficients are positive but not significant, the opposite of the Coates and Munger prediction. The larger a senator's 1987 electoral majority, the larger the majority benefitted by the senator's behavior on the QEA. We think the positive sign arises because many suburban Republicans and urban Democrats had large 1987 electoral majorities and voted in accord with their districts' interests. Suburban Democrats and rural Republicans voted against their districts' interests and had varied electoral safety.

Were the 1991 Revisions Directed at Vulnerable Democrats?

Did the Democratic leadership of the legislature, as a part of the 1991 revisions, direct benefits towards those districts represented by Democrats that were net losers under the original Quality Education Act? Ten Senatorial districts were represented by Democrats who voted in favor of the QEA in 1990 even though both the median and the mean voters in their districts paid more taxes as a result. According to our calculations, the 1991 revisions redistributed benefits so that a majority of voters became net beneficiaries in only four of the ten districts. Of the ten districts, TABULAR DATA OMITTED five were competitive districts, defined as those in which the incumbent won the last election with less than 60% of the vote. Of the five competitive districts, only two became net beneficiaries under the revisions. Economic benefits were not systematically directed toward those districts represented by marginal Democrats.(12)

How Did Voters React to Their Legislators' Behavior on the QEA?

Studies of the relationship between the economic effects of policy choices and legislative behavior usually end at this point. In this study we ask one further question. How do voters react to legislators who vote against their constituents' economic interests? In particular, do voters electorally punish those legislators who do not vote in accord with citizens' economic interests? To answer this question, we regressed the percentage vote obtained in the November 1991 election on a variable that captures whether legislators voted in their constituents' economic interest, along with standard political variables (the party of the candidate, whether or not the incumbent was running, and whether or not the election was unopposed).

Table VIII displays the results of six regressions. Controlling for whether the incumbent ran for reelection, was a Democrat, or had opposition, the percentage vote received by a senator in the November 1991 election was larger if his position toward the Quality Education Act of 1990 was in the economic interests of his constituents.(13) A unit change in the value of this variable would cause a 21 to 24 percentage point change in the vote received by a senator. Constituents' behavior is consistent with the assumptions in all of the three simulations of the benefits flowing from the QEA but in none of the simulations of the benefits flowing from the revisions to the QEA. The legislators' attempt to curry favor with their constituents through revisions to the QEA that diluted its redistributive character was discounted by voters whose reactions toward incumbents were governed mainly by the senators' behavior toward the original QEA in June 1990.

Also of interest in table VIII is the effect of party allegiance on electoral success. Just being a Democrat in November 1991 lowered one's vote total by 11 to 13 percentage points. One interpretation of this effect is a protest against the increase in the sales tax (from 6 percent to 7 percent) enacted at the same time as the QEA.(14) Though voters took out their anger on Democrats, in general, they nevertheless differentially supported senators who voted in the interests of their constituents more than senators who did not.

What Are the Lessons from New Jersey?

Many commentators have argued that the large defeat suffered by the Democrats in New Jersey in the November 1991 election suggests that redistributive programs such as the QEA are not compatible with the electoral relationships of a representative democracy, even if such programs are appropriate from a normative viewpoint. How much did the QEA alter the outcomes of the November 1991 elections?

We used the results of table VIII (column 1) to simulate elections in which the votes of senators on the QEA were switched.(15) Eight of the forty Senate races would have been won by different candidates. Seven Democratic incumbents lost in November 1991. Six of them voted against their districts' economic interests on the QEA. Of the six, three would have been reelected had they voted differently on the QEA. Four Democratic incumbents did not run for reelection and all those districts were won by Republicans. According to our calculations, only one of these seats would have remained Democratic if the incumbent had switched his QEA vote.

The Democratic majority in the state Senate was 21-17 before the election. After the election, the Republicans held a 27-13 majority. According to our estimates, of the 11 Democratic seats that became Republican (1 Republican seat became Democratic), only four would have remained Democratic if the incumbents had altered their QEA vote, too few to have retained a Democratic majority. The -11.5 coefficient on the party variable, which we interpret as a measure of voter sentiment against the sales tax increase, was a large factor in the Democrats' defeat.

VI. Concluding Remarks

In this paper we have considered a new approach to modeling voter decisions regarding government policy. We have shown that using the household as the fundamental unit of analysis makes possible a consideration of vital questions impossible to address under previous methodologies. In particular, we have shown that previous research focused on municipal or legislative district aggregates is likely misleading about the extent of underlying support for a given proposal. Hence, earlier attempts to determine whether voting behavior was primarily driven by "economic" or "ideological" considerations were not able to distinguish between the two because of aggregation biases. We have also provided the groundwork for a theory of economically motivated voter turnout integrated with a theory of voter behavior, by associating with each household a dollar amount to be gained or lost from various tax and grant regimes.

There are two questions posed in the title of this paper. We have provided the groundwork for answering the second, by specifying and implementing a model of economically motivated voting behavior at the household level. The first question remains: Do legislators vote their constituents' wallets?

Our answer is yes in a situation in which the economic effects of policy choices are salient like the New Jersey reforms of 1990 and 1991. At the statewide level, after initially voting against the economic interests of a majority of voters as the result of pressure from Governor James Florio, the legislature redirected aid to the middle class and created benefits for a statewide majority of voters. At the district level our results are weakly supportive of an economic explanation of legislator behavior. The votes of twenty-four legislators were predicted correctly in the 1990 reforms (60%) and twenty three in the 1991 changes (57%), statistically significantly better than flipping a coin, but not much better. A stronger finding is that in the next election voters gave fewer votes to those senators who had not voted in their economic interests, everything else being equal, thus restoring to some extent the relationship between constituents and their legislators through negative signaling. However, if all senators had voted in the economic interests of their district's median voter on the QEA but did not change their behavior on the sales tax, the Democrats still would have lost their overall majority in the New Jersey state senate as a result of the November 1991 elections.

Appendix: New Jersey Institutions and Dataset Description

The dataset used in this paper consists of all homeowners in New Jersey who filed both a 1987 New Jersey Income Tax return and a valid Homestead Rebate (a property tax relief program) claim. Of 3.3 million income tax returns and 1.5 million Homestead Rebate claims, 1.4 million observations were matched. All individual identification was deleted prior to any analysis being carried out. All of the analyses reported in this paper were performed on a randomly selected sample of 198,918 households representing 563 of New Jersey's 567 municipalities. The Homestead Rebate claim provides information on the assessed value of a person's house, the property taxes due, the amount of rebate, the town of residence, and whether or not the homeowner is over age 65 (self-reported). By combining these data with the figures from the income tax return, a comprehensive picture of the impact of both state and local taxes can be drawn.

Income Tax Liability

A measure of household annual income is used to represent ability to pay for tax purposes. This number, Gross Income (to be defined shortly), is also the point of departure for calculating state income tax liability. Here, the mechanics of the income tax calculation are presented.

In order to determine income tax liability, the following steps are necessary. First, determine Gross Income, equal to the sum of wages, dividends, interest, and other income received during 1987. The major omissions from Gross Income are Social Security benefits, unemployment insurance benefits, and lottery winnings. The dividend, interest, and other income figures also have some exemptions and deductions built into their calculation.

After obtaining Gross Income, one subtracts "other retirement income exclusions" (a supplemental income exclusion for taxpayers over age 62 who claim less than the maximum pension exclusion) to obtain Adjusted Gross Income. If Adjusted Gross Income is less then $3,000, then the tax liability is zero. Otherwise, subtract exemptions (one per dependent, one for being over age 65, and one for being blind) and deductions for allowable alimony and medical expenses to obtain Taxable Income. The tax liability for the 1987 status quo is then calculated according to the following schedule:
Taxable Income Tax Liability

less than $20,000 2% * Taxable Income
$20,000-$50,000 $400 + 2.5% * (Taxable Income - $20,000)
more than $50,000 $1,150 + 3.5% * (Taxable Income - $50,000).

After the reforms the tax liability is calculated as follows:
Singles and Married Filing Separately

Taxable Income Tax Liability

Less than $20,000 2% * Taxable Income
$20,000-$35,000 $400 + 2.5% * (Taxable Income - $20,000)
$35,000-$40,000 $775 + 5.0% * (Taxable Income - $35,000)
$40,000-$75,000 $1000 + 6.5% * (Taxable Income - $40,000)
More than $75,000 $3275 + 7.0% * (Taxable Income - $75,000)

Married Filing Jointly/Surviving Spouses and Heads of Households
Taxable Income Tax Liability

Less than $20,000 2.0% * Taxable Income
$20,000-$50,000 $400 + 2.5% * (Taxable Income - $20,000)
$50,000-$70,000 $1,710 + 3.5% * (Taxable Income - $50,000)
$70,000-$80,000 $2,410 + 5.0% * (Taxable Income - $70,000)
$80,000-$150,000 $2,910 + 6.5% * (Taxable Income - $80,000)
More than $150,000 $7,460 + 7.0% * (Taxable Income-$150,000).

Property Taxes and Aid to School Districts

Calculating the property tax payments and the pattern of reduced property taxes assumed to result from grants is straightforward. The property tax payment is that reported on the Homestead Rebate claim. This figure includes taxes for use by municipalities, county governments, school districts, and special districts.

School districts in New Jersey are either coterminous with municipalities or incorporate several municipalities into a regional school district. Regional school districts apportion their costs on the basis of the relative "equalized value of property" in their constituent municipalities. About half of the municipalities in the state (239 out of 567) participate in regional school districts. Most of these municipalities also operate local school districts. For example, a town will operate a K-8 school district while participating in a regional high school district.

Assessments in New Jersey are supposed to represent the full market value of property annually updated. In practice, they are updated infrequently. As a result, the Division of Taxation "equalizes" property assessments across municipalities to the market value standard on the basis of evidence from the ratio of assessed value to sales price for properties that sell during the year. This equalization process does not affect the distribution of property taxes within a municipality, only the allocation of regional school costs (and hence property taxes) among municipalities.

Grants are allocated among municipalities constituting a regional school district in the same manner that costs are divided. The reduction in property tax implied by a given set of grants is then a household's tax share multiplied by the sum of the grant to the local school district and his municipality's share of the grant to the regional school district.

Homestead Rebate

The Homestead Rebate program was enacted as a form of property tax relief concurrently with the adoption of an income tax in New Jersey. Owner-occupiers are eligible for a rebate on the property taxes paid on their principal residence. In practice, everyone who is eligible applies for and receives a rebate. There are approximately 1.7 million properties classified as residential in the state. Over 1.5 million households, or about 90 percent of the potential recipients, received Homestead Rebate payments in 1987. When one considers that the remaining properties include rental housing, out of state owners, and some types of small businesses, the response rate as a fraction of eligible homeowners is surely close to 100 percent.

The amount of rebate is calculated in the following manner. Let EV represent the equalized value of a house (assessed value divided by a town specific assessment-sales ratio), EPTR the equalized property tax rate (tax levy divided by the total equalized value in a town), and W the smaller of $10,000 and 2/3 * EV. The Homestead Rebate is the smaller of

H|R.sub.a~ = (0.015 + (0.125 * EPTR)) * W

H|R.sub.b~ = 0.5 * EPTR * EV.

Since almost every homeowner's equalized value (EV) is greater than $15,000, W is in fact $10,000 for the vast majority of the population. In addition, H|R.sub.a~ is smaller for almost all realistic levels of local taxes (the mean value for EPTR is about 0.026). If this is the case, then the only variation in the Homestead Rebate payment will be among homeowners in different municipalities, that is the program is like a town specific lump sum grant. There are additional rebates of $50 for people that are either age 65 or older, permanently and totally disabled, or qualified surviving spouses. These payments are included when calculating an individual's overall tax burden.

After the 1990 reforms the Homestead Rebate is calculated as follows:

Single Taxpayers

Gross Income |is less than~ $35,000

Rebate = Property Taxes - (.05 * Gross Income)

Minimum = $150; Maximum = $500

$35,000 |is less than~ Gross Income |is less than~ $70,000

Rebate = $150

$70,000 |is less than~ Gross Income |is less than~ $100,000

Rebate = $100

$100,000 |is less than~ Gross Income

Rebate = $0

Married Taxpayers and Heads of Households

Gross Income |is less than~ $70,000

Rebate = Property Taxes - (.05 * Gross Income)

Minimum = $150; Maximum = $500

$70,000 |is less than~ Gross Income |is less than~ $100,000

Rebate = $100

$100,000 |is less than~ Gross Income

Rebate = $0.

1. See VanDoren |17~ for a more complete discussion.

2. See Fisher |7~ for a summary and extension of the economics literature regarding the allocation of grants between reduced taxes and increased spending.

3. An open economy is one in which factors of production (labor and capital) are free to enter and exit.

4. See Bogart, Bradford and Williams |4~ for a detailed discussion of the incidence of state and local taxes.

5. Data on total school and municipal aid before and after the reforms were obtained from a press release dated March 5, 1991 and distributed by the office of then New Jersey State Senate President John Lynch. We thank Richard McGrath for these data.

6. Those households that rent housing are not analyzed in these data. The effect of renters on our analysis would depend on whether property tax reductions reduce rental payments by tenants or create capital gains for owners |12~. If the former view is correct and renters are mostly lower income, the economic gains to low income people would be higher than we have estimated. However, these economic gains might not affect our results very much because renters comprise only 38% of the population |15, Table 2~ and more importantly vote at a very low rate (39.8%) |16, Table C~. If property tax reductions create capital gains for owners, our results might be affected to a larger degree. To the extent owners of property are higher income individuals, the capital gains could offset higher income taxes, and because the affluent vote at a higher rate, increased support among the affluent for the reforms would alter our predictions.

7. Statements about the overall progressivity of tax systems must include measures of central tendency (mean taxes) as well as their dispersion (standard errors). Our calculation of the standard errors for Table I (not shown in the table) suggests that the variation in tax liability within income classes is large enough that statements about the progressivity of the system either before or after the reforms are very problematic.

8. The Quality Education Act of 1990 (QEA) and the Homestead Property Tax Rebate Act of 1990, New Jersey Laws, 204th Legislature, 1st Session, Chapters 52, 61.

9. We also conducted simulations using two other sets of assumptions. In the first, all households (including married) receive only one vote. All other assumptions are unchanged. In the second, turnout is universal (does not vary by income) and the number of voters per household is the number of adults in each household (imputed from the filing status). All three scenarios are used in the regression analyses discussed later in section V.

10. In this paper we do not modify our estimate of household net benefits to include the willingness of some household to pay increased taxes to support increased educational spending in poorer districts, even though we know that such households exist. Instead we ask whether the behavior of legislators is consistent with a narrow conservative estimate of net economic benefits to voters.

11. We did not assign a separate probability of voting for each income class for households headed by a person over age 65 because of severe problems in measuring the income of elderly households accurately.

12. Detailed regression results available upon request.

13. These equations were estimated with ordinary least squares even though the dependent variable was bounded at zero and one. See Achen |1, 40-41~ for a discussion of the differences between the linear probability model and a probit specification. Inspection of the residuals did not indicate the presence of heteroskedasticity.

14. In specifications omitting the party variable, the value of the size-of-majority variable did not change. In specifications omitting the size-of-majority variable, the party coefficient also did not change in value. The two variables' effects are separate and distinct.

15. For each of the 40 districts, we calculated twice the absolute value of the size of the majority for the QEA and multiplied it by the coefficient from table VIII column 1 (21.4638). This is the change in the general election vote that our equation predicts would have occurred if a legislator changed his vote on the QEA. We then compare this to the actual majority won by the legislator in the general election. If the predicted vote change is larger than the actual majority, then the outcome of the race would have been reversed by a change in the legislator's vote on the QEA.


1. Achen, Christopher H. The Statistical Analysis of Quasi-Experiments. Berkeley: University of California Press, 1986.

2. Arnold, R. Douglas. The Logic of Congressional Action. New Haven: Yale University Press, 1990.

3. Bernstein, Robert A. and Stephen R. Horn, "Explaining House Voting on Energy Policy: Ideology and the Conditional Effects of Party and District Economic Interests." Western Political Quarterly, June 1981, 235-45.

4. Bogart, William T., David F. Bradford, and Michael G. Williams, "Incidence Effects of a State Fiscal Policy Shift: The Florio Initiatives in New Jersey." National Tax Journal, December 1992, 371-87.

5. Carson, Richard T. and Joe A. Oppenheimer, "A Method of Estimating the Personal Ideology of Political Representatives." American Political Science Review, March 1984, 163-78.

6. Coates, Dennis and Michael Munger. "Nonideological 'Shirking' and the Economic Model of Politics: North Carolina General Assembly Voting on the Southeast Compact." University of North Carolina Political Science Department, Manuscript, 1991.

7. Fisher, Ronald C., "Income and Grant Effects on Local Expenditures: The Flypaper Effect and Other Difficulties." Journal Of Urban Economics, November 1982, 324-45.

8. Kalt, Joseph P. The Economics and Politics of Oil Price Regulation. Cambridge, Mass.: M.I.T. Press, 1981.

9. Kalt, Joseph P. and Mark A. Zupan, "Capture and Ideology in the Economic Theory of Politics." American Economic Review, June 1984, 279-300.

10. ----- and -----, "The Apparent Ideological Behavior of Legislators: Testing for Principal-Agent Slack in Political Institutions." Journal of Law and Economics, April 1990, 103-31.

11. Kau, James B. and Paul H. Rubin. Congressmen, Constituents, and Contributors. Boston: Martinus Nijhoff, 1982.

12. Mieszkowski, Peter and George Zodrow, "Taxation and the Tiebout Model: The Differential Effects of Head Taxes, Taxes on Land Rents, and Property Taxes." Journal of Economic Literature, September 1989, 1098-1146.

13. Peltzman, Sam, "Constituent Interest and Congressional Voting." Journal of Law and Economics, April 1984, 181-210.

14. -----, "An Economic Interpretation of the History of Congressional Voting In the Twentieth Century." American Economic Review, September 1985, 656-75.

15. U.S. Bureau of the Census. 1980 Census of Housing: Characteristics of Housing Units: General Housing Characteristics: New Jersey. Washington: U.S. Government Printing Office, 1982.

16. U.S. Bureau of the Census, "Voting and Registration in the Election of November, 1988." Current Population Reports series P-20, no. 440. Washington: U.S. Government Printing Office, 1989.

17. VanDoren, Peter M., "Can We Learn the Causes of Congressional Decisions From Roll-Call Data?" Legislative Studies Quarterly, August 1990, 311-40.
COPYRIGHT 1993 Southern Economic Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:And how would we know if they did?
Author:Vandoren, Peter M
Publication:Southern Economic Journal
Date:Oct 1, 1993
Previous Article:Reassessing the male-female wage differential: a fixed effects approach.
Next Article:Productive and financial performance in U.S. manufacturing industries: an integrated structural approach.

Related Articles
The personal vote: constituency service and electoral independence.
Tips for new legislators.
Legislative voting and the economic theory of politics.
Activating a grass-roots lobbying network.
15 tips for being an effective legislator.
Advocating effectively: why it's important to meet with public policy officials. (Board Primer).
Constituent communication in cyberland: using the internet to communicate with constituents can be very effective.
Physician executives and the political process.
Proactive Americans: get involved: concerned Americans are becoming personally involved in the political process and finding success as they defend...

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters