Sargsyan | 1 Gor Sargsyan Political Partisanship in the Tax Rates of Vermont Introduction: Why do certain societies adopt different policies than others? Much of the literature on political outcomes in the United States argues that partisanship is prominent among the numerous causes to the policies adopted in any given state. Each work encompasses its own view of the bigger picture, and though some argue that gender differences (Zipp and Plutzer 1985) and socioeconomic factors (Fritz 1982) are important, studies also emphasize the importance of a society’s aggregate partisan identity. I argue that political partisanship is the primary cause of tax rate changes. To study this hypothesis I will examine the role of partisanship on tax rate decisions made in town meetings in Vermont. In Vermont, partisanship has been born and nurtured within the town meetings, and the source of influence on voting outcome is from the groups of people in the meetings. In the sections that follow I outline a research design
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Sargsyan | 1
Gor Sargsyan
Political Partisanship in the Tax Rates of Vermont
Introduction:
Why do certain societies adopt different policies than others? Much of the literature on
political outcomes in the United States argues that partisanship is prominent among the
numerous causes to the policies adopted in any given state. Each work encompasses its own view
of the bigger picture, and though some argue that gender differences (Zipp and Plutzer 1985) and
socioeconomic factors (Fritz 1982) are important, studies also emphasize the importance of a
society’s aggregate partisan identity. I argue that political partisanship is the primary cause of tax
rate changes. To study this hypothesis I will examine the role of partisanship on tax rate
decisions made in town meetings in Vermont. In Vermont, partisanship has been born and
nurtured within the town meetings, and the source of influence on voting outcome is from the
groups of people in the meetings. In the sections that follow I outline a research design to
compare the evidence gathered in Vermont and assess which factors influence policy outcomes
in town meetings.
Voter turnout is based on countless factors that shape the general politics of our nation.
One main factor, most people will agree with, is population. In California, for example, the
percentage of people who vote in the presidential elections every year is less than half of the total
population. This means that the elected candidate, or chosen policy, is supported by the majority
vote of a minority group. This, in turn, creates doubt in the policy choices and sparks curiosity in
questioning the legitimacy of our governments. Yet, surprisingly, there is little known fact about
the nature of our government system and whether it should be changed. In this paper, I will
Sargsyan | 2
present a case study of the state of Vermont, emphasizing the presence of voter biases and how
the political partisanship within town meetings affects tax rates.
The significance of the subject of political partisanship as an influence in policymaking
lies in its providing of information that aids in the examination of our voting system, how clearly
our voting system represents the true voice of the people, and what exactly influences its
outcome in the results of policy decisions. This essentially affects how individuals, like you and
me, run our daily lives. As Gerber and Huber show in their article “Partisanship and Economic
Behavior - Do Partisan Differences in Economic Forecasts Predict Real Economic Behavior?”
the political party in control, which has a large influence on policy decisions, affects consumer
habits. Gerber and Huber explain that consumers change their habits of living and spending
based on the policies that they “predict” or “expect” that the new candidate will impose. The
expected outcomes are with respect to the winning candidate’s political views, and economic
plans for the duration of his term. Whether the candidate is a Democrat, a Republican, a liberal
or a conservative, consumers already have an idea of future changes to be implemented and
thereby adjust their lifestyles to future policies ahead of time. Examining partisanship as a cause,
in our government could also explain holes in our system and offer better, more effective, insight
on managing our government, not only in Vermont, but the rest of the US. The acting influence
in Vermont’s town meetings is applicable to larger populations as well, because population size
has little to no effect on the partisanship of a select community.
From the data collected on the 247 towns in Vermont from the 1970’s to the present, I
have discovered a strong pattern between tax rates and political affiliation of the towns. After
running tests on the relationship of partisanship and tax rates I concluded that tax rates, the
dependent variable, are directly correlated with partisanship, the independent variable. As
Sargsyan | 3
partisanship toward a Democratic party increases the tax rates increase, and in the same manner,
as Republican partisanship increases, tax rates decrease. This pattern is expected in most
Democratic systems of governments, and Vermont is a case study of this phenomenon.
A democracy is a system of government based on the consent of the governed. A
democratic government emphasizes the constitutional equality of each citizen in the decision
making process. In most of the country, elections provide citizens with an opportunity to
influence public policy. Usually, states tend to hold public election through the use of secret
ballots or by mail. In this way, anonymous public opinions are collected and voiced as a majority
decision and policy is created. However, not all states have the same voting procedures, and not
all towns hold annual town meetings to determine local policy. This makes us question the
accuracy and reliability of the final decision as it relates to public opinion, and the method of
collecting votes subject to possible variables influencing the voice of a majority.
In Vermont town meetings, locals decide these issues. The voting outcomes for elections
of state and federal officials appear to correlate with the amount of tax imposed in the town
meeting. Simply put, political partisanship at the voting booth should match the local policy
choices taken in town meetings. I expect that towns voting for Democrats will have higher taxes
and towns electing Republican parties will have lower tax rates. The focus of this paper will be
the change in tax rates specifically as it relates to a town’s election voting outcomes. Some
Deliberative Democracy scholars argue for example, that in a well-connected network of
discussion, like those of town meetings, partisan attitudes will harmonize, having provided
people with a unique setting of interaction. Stemming from the polarization of a network of
agreement within a town meeting, the gradual emotional attachment to specific candidates form
with disregard to logical decision-making based on factual information (Parsons 2010).
Sargsyan | 4
However, the tax rates in the towns of Vermont should still be relevant and proportional to the
voting partisanship of the individual. Do towns with higher tax rates tend to vote for democratic
candidates while towns with lower tax rates have Republican Party leaders? This question can
best be answered through data analysis of past presidential elections, used to determine political
orientation of Vermont towns, while keeping track of the tax rates established during the terms
of presidency. The collected election data from 1970 to 2012 will show us each town’s partisan
affiliation based on their vote for either the Democratic or Republican Party candidate. In
accordance, the data on tax rates will show us any patterns of proportionality to partisanship as it
changes, and to the extent that it changes, in each town.
Literature Review:
Group identity affects attitudes and behavior of individuals. This effect is equally and
actively present in political environments. The article “Party Affiliation, Partisanship, and
Political Beliefs” provides answers to the question of external influences acting upon voter
turnout in elections. Gerber and his coauthors discuss the connection that exists between voters’
political affiliations, and the decisions that the voters will make on Election Day. In particular,
Gerber suggests that there is a direct correlation between partisanship and voting turnout based
on a controlled experiment conducted by him for the first time. The subject in this case is the
state of Connecticut in comparison to Vermont. Unlike Vermont, Connecticut does not allow
citizens to vote unless they claim an affiliation toward a specified political party. There were two
groups of subjects in the experiment: the treatment group, in which the researchers intervened,
and a control group in which the researchers had no influence in pushing them to choose a
direction. The experiment, in regards to this, was conducted with a group of randomly selected
Sargsyan | 5
registered voters who claimed to be neutral. These registered voters were given a pre-survey
before the election without requiring voters to choose a political affiliation: the survey was
examined for tendencies toward a certain party. Before the election, each person was sent a letter
reminding them that they must choose a political party in order to be eligible to vote in the actual
election; most chose the party that they had shown tendencies toward on the pre-survey. When it
was time for the actual election, researchers took note of the changes in this same group of
people. When they chose a political party there was an increase in partisan identification. “Our
findings support the theoretical argument that partisanship is a group identity that induces
individuals to evaluate members of their group more favorably than members of opposing (party)
groups” (742). In this same way, I hypothesize that Vermont elections should show similar
partisan influence on the voters’ choices. Of course, partisanship does not imply citizen
participation in active politics, it simply points out the bias in voting behavior based on
individual preconceived notions and judgments.
Much of the existing literature emphasizes that partisanship is in fact a major component
in public policy. At least three conclusions, validating the strong presence of the influence of
partisanship in a community, can be drawn from the existing theories:
1: Partisanship has a strong effect on political and economic behaviors and attitudes.
2: Party affiliation is an underlying factor in candidate choice.
3: A citizen’s partisan identity toward a political party reflects their economic decisions.
Sargsyan | 6
Gerber has written a complementary article coauthored with others, called “Partisanship
and Economic Behavior.” He continues to support the idea of partisanship having a significant
role in elections, and extends those results to economic changes in consumer behavior. “There is
ample evidence linking partisanship to candidate choice and political participation” (Gerber and
Huber 2009, 423). Their studies have shown that before and during the election period, consumer
habits differ. The assumption is that consumers predict forthcoming economic policies based on
the likelihood of a candidate winning, and the actions they will take in reform of the system. This
study is less focused on a single state, however, and more focused on “mass politics.” “Citizens
appear to be adjusting their consumption based on expected income rather than today’s income”
(Gerber and Huber 2009, 424). When citizens expect a Democratic candidate to be elected and
thereby, raise taxes, they would cut down spending and save more in preparation, while the
opposite trends displayed dominance for a Republican candidate. In an expected similar case,
Vermont generates higher tax rates under Democratic leadership and lower tax rates in a
Republican government. Gerber points out some faults in his claim, clarifying that consumption
could take place by consumers from nearby counties that are of different political affiliations,
altering the survey results. That will be discussed in greater detail, however the conclusion
remains consistent that partisanship influences political outcomes. The partisanship in Gerber’s
argument lies in expected economic policy change that conducts political adjustments in the lives
of citizens, hence influencing their political decisions. This same type of partisanship leads me to
my research in its effects on decision-making in Vermont town meetings.
These town meetings have developed as a system of government, which Bryan
juxtaposes with the Democracy of the pre-modern world, over twenty-five hundred years ago, in
Athens, Greece. He calls this system a “real democracy” (Bryan 4, 2004), and in comparison,
Sargsyan | 7
argues that Vermont has kept its roots as a “real democracy” and most people are not even aware
of this. A “real democracy” is a jurisdiction in which all of its people are citizens and eligible to
participate in its politics. In Athens, those who were not citizens still abided by the laws. The
high taxes in place for the citizens of Athens also applied to non-citizens that were nonetheless
taxed equally, but did not receive the services and benefits as citizens. In the same way, Vermont
uses this tactic to its advantage. Fritz points out that this method of taxing non-citizens, or
tourists, was taken advantage of by Vermont as a monetary alternative to reduce the burden on
residential property tax (Fritz 382, 1982). Athens also manipulated agendas, as do towns in
Vermont. Items that may face opposition are placed at the end of the list so that minority groups,
such as farmers, would have to go home to milk, and, near the end, be underrepresented when
the agenda is called to a vote. In this sense, Bryan calls Vermont a “real democracy,” saying that
it comes with the bad and the good, but the bottom line is that a “real democracy” is not partial.
Every individual in a town makes the laws for themselves, and the officials don’t have much
power over them.
Theory and Hypotheses:
. Many questions arise when considering the development and strategy of a town’s tax
rate. Whether tax rates are based off of mutual feelings among policy-making groups or they are
based on strategic decision-making, there is reasoning behind every action in politics. Scholars
offer differing and, at times, controversial explanations for political outcomes in the state and its
towns. Based on collected evidence from historic records of over 240 towns in the state of
Vermont, I have gathered enough analytical data to offer a data-based argument on the cause of
state taxes and what external factors affect them. I also distinguish how these methods differ
Sargsyan | 8
from those of others. Decisions in politics are extremely complicated and require more than
pattern analysis because of the role of human nature in government. The biases that are born
from human environments may act as a valid explanation as to why policies are as they are. In
terms of tax rates in Vermont, I argue that partisanship plays a major role in the direction of
public policy, and its effects are strongly felt in Vermont. Local tax rate decisions are made in an
annual decision-making body consisting of eligible voters who choose to attend the annual town
There is a direct correlation between the levels of partisanship in any given democratic system as
it relates to the amount of tax rate imposed in that region.
This leads me to hypothesize the following:
H1: Partisanship has a direct effect on the imposed tax rates of a town in Vermont.
H2: If partisan identification favors the Republican Party, then tax rates are expected to be
lower.
H3: The higher the partisan majority in a town, the greater is the effect on the shift of tax rates
from the state average.
Even though partisanship is the main cause of differences in tax rates across Vermont
towns, there may also be other variables that affect partisanship and tax rates indirectly. There
are additional hypotheses that could also be factors in the tax rate decision process. These
alternative hypotheses give more insight as to how further influences may have an impact on
partisanship in Vermont.
Sargsyan | 9
H4: Partisanship may be directly or indirectly affected by external socioeconomic factors like
tourism, and urbanization in select Vermont towns, which in turn affect the tax rate.
H5: Variable income levels, and population size are the root causes to variance in partisan
identity in a town, which in turn affect the tax rate.
Each of the hypotheses can be tested with a common method. Since there is a mass of collected
data on town meetings, voting rules, tax rates and political affiliation in each of the towns in
Vermont, the simple method is to use regression analysis to compare the relative influence of
each of the independent variables on the dependent tax rate outcome. This method offers a
repeatable test of the various hypotheses. In order to test these hypotheses we must first explore
the relationship across towns between tax rates and partisanship. To observe this relationship, I
have collected data from the state archives of Vermont for each individual town. As I run
statistical analyses of the data, I anticipate finding a correlation to form between partisanship and
tax rates. A measure of higher Republican partisanship should correlate with lower tax rates, and
to the same extent a measure of higher Democratic partisanship with higher tax rates.
Comparable statistics would be a collection of ratios of democratic towns to republican towns
that would lead to aggregate tax rates over time. To run the tests and prove the hypotheses I
coded the data from the 1970s to the present, on Democrat or Republican votes in the
presidential elections, total tax rates, school tax rates, population, population under 18, and
annual wage. I first tested the relationship between tax rates and partisanship. From this we can
determine what exactly occurs as average partisanship in Vermont increases, so to say
Sargsyan | 10
partisanship in Vermont goes from least democratic, or highest republican, to highest democratic
level. I then evaluated the strength of this relationship by incorporating additional variables in
the statistical test. To do this I tested the same dependent variable, total tax rate, with different
independent variables such as population. From that, I examined the degree of partisan
orientation in each town by measuring the percent Democrat and percent Republican as it relates
to the total tax. Also, I measured the total tax rate in the same time period as the towns
experienced an increase in tourism to compare with the partisanship during its neutral economic
state.
Evidence:
The following visuals are a collection of statistical data to show the relationship between
partisanship and tax rates as well as to disprove possible alternate hypotheses relevant to the
data. As we examine the figures, it is important to note that the main figures in these tables are
the coefficient, the t-test and the p-value of the test. These figures will clarify the relationship
and its strength between the tested independent and dependent variables.
In the summary analysis of Figure 1, we can see information on all of the variables that
are involved in this project. Each section of the independent variables will be discussed
individually to isolate results for clarity and accuracy.
Total 11023.2328 3398 3.24403555 Root MSE = 1.4298 Adj R-squared = 0.3698 Residual 6942.39285 3396 2.04428529 R-squared = 0.3702 Model 4080.83995 2 2040.41998 Prob > F = 0.0000 F( 2, 3396) = 998.11 Source SS df MS Number of obs = 3399
. regress totaltax democrat year if population>1000
To go into detail on population, this regression analysis is a comparison of towns with
more and less than 1000 people. To classify small towns and large towns, I use population size
of anything below 1000 to be a small town, and anything greater than 1000, to be a large town.
Sargsyan | 19
These size classifications are in consistency with Fritz’s testing of small versus large towns in
terms of tax and vacation homes (Fritz 383, 1982). Furthermore, as run the regression we can see
that population continues to be so insignificant in the differentiation of total tax, which leads us
to conclude that population alone does not make a difference, and that small towns are not
different than large towns in terms of influence on total tax based solely on population size.
Richard G. Fritz provides an alternative hypothesis of having socioeconomic influences
on total tax rate of a town. He conducted a test to see how tourism and vacation homes affect the
tax ratesHis analysis was based on tourism in select towns, and the development of vacation
homes that are affecting the total tax rate of the town. He argues that the government provides a
monetary alternative in terms of paying taxes for hotel owners that attract tourists. In a town with
high tax rates, for example, these tourists would incur the same amounts of taxes without
receiving any of the governmental benefits that come from having high taxes. However, unlike
the tourist attractions, the impact of a vacation home, acts as a tax burden because it is an extra
portion of land to be responsible for. However, it is also possible, that the effects of vacation
homes, in small towns especially, would increase overall property value. Based on his evidence
and testing, Fritz’s results do not fully take into account the source of tax divergence between
urban and developed rural towns in Vermont, and need to determine if there are other equally
plausible explanations.
Conclusion:
Vermont is a state with its own unique set of procedures and rules. That is what makes
Vermont an interesting state to study; to examine their methods of conducting and developing
policy in the towns, which in turn would give us insight on our own methods of decision-making.
Sargsyan | 20
As we have discussed, partisanship has a strong correlation with the increase or decrease of total
taxes in the state of Vermont. Democratic towns are usually tended toward higher tax rates, and
Republican towns toward lower tax rates.
It is not without the consideration of alternatives that the reading of this paper comes to
an ultimate end. Undeniably, one cannot avoid considering the possibility that there are other
relationships, which lead to the growth or decline of tax rates. To take this idea into perspective,
the tax rates could be the independent variable affecting the partisanship as the dependent
variable instead. As is yet to be proven, the increase of tax rates could be the magnet force
drawing in surrounding Democrats, or if the tax rates are decreasing, then Republicans as the
population of the town shifts. It is true that we disqualified population as an acting influence in
the change in tax rates, but those tests, admittedly, are very limited to strict terms of numbers and
not demographics. The shift in population in terms of demographics, would involve the shift in
partisanship based on the shift in taxes. As the taxes increase, more Democrats choose to live in
the select town, and percentage of Republicans in the population declines in relation to the
growth of the number of Democrats.
Another explanation could involve, instead of a shift in population demographics, a shift
in ideals. The increase in tax rates could incentivize the Democrats who are not for such high
taxes to change their partisanship in order to redirect the flow of taxes in the Vermont towns and
vote Republican. This is to say that partisanship changes because of tax rates acting as the
independent variable influencing partisanship. Even though our tests show that Democrats tend
toward higher taxes, and that a higher percentage of Democrats leads to a higher percentage of
tax rates respectively, there are still many outliers present as is common in any situation
involving human nature. Our regression analyses are not based on an individual level, but at a
Sargsyan | 21
town level showing the partisanship of the entire state of Vermont as a collection of towns. Of
course, the greater the sample size, the more accurate the results will be, but the results, no
matter how accurate, cannot explain 100% of the cause. Republicans as well may also have a
change of heart as has often occurred in the past, and vote for higher taxes. This could be due to
philanthropic reasons, or simply because they believe it is more fair to pay higher taxes.
However, these suggestions are, again, outliers, in the results we have gathered, that should not
be unaccounted.
If partisanship is not taken into account, for example, a cause of tax rates increasing
could be because of a large amount of an under 18 population, forcing the government to provide
more funding to schools and build larger ones, for which they would need more money and
would have to raise taxes.
All in all, though there is simply a possibility for these ideas to be implemented, there is
not sufficient correlation or an initial expectation for the cause to be in lieu of the evident
observations in this paper. As we can see, all of the tests that were performed to investigate the
reliability of my argument, proved to be in very strong correlation to the likelihood of
partisanship being a cause to tax rate change. To summarize my explanations, the P value being
less than .05 suggests a large amount of confidence in the relationship.
Bibliography
Sargsyan | 22
Bryan, Frank M. (2010). Real Democracy: The New England Town Meeting and How It Works. University of Chicago Press.
Gerber, Alan S. Gregory A Huber, Ebonya Washington. 2010, “Party Affiliation, Partisanship, and Political Beliefs- A Field Experiment.” American Political Science Review. 104. No.4. (November), pp. 720-744.
Gerber, Alan S., Gregory A. Huber. 2009, “Partisanship and Economic Behavior- Do Partisan Differences in Economic Forecasts Predict Real Economic Behavior?” The American Political Science Review. 103. No. 3. (August), pp. 407-426.
Fritz, Richard G. 1982, “Tourism, Vacation Home Development and Residential Tax Burden: A Case Study of the Local Finances of 240 Vermont Towns.” American Journal of Economics and Sociology. 41, No. 4. (October), pp. 375-385.
Parsons, B. M. (2010). Social Networks and the Affective Impact of Political Disagreement. Political Behavior. 32 No. 2. pp. 181–204.
Zipp, John F., Eric Plutzer. 2004, “Gender Differences in Voting for Female Candidates- Evidence From the 1982 Election.” Oxford University Press. 49, No. 2. (Summer 1985), pp. 179-197