ASSESSMENT OF THE ELECTORAL GEOGRAPHY OF UKRAINE: CASE OF PRESIDENTIAL ELECTIONS OF 2019

Full text

International Scientific Conference GEOBALCANICA 2020
571
ASSESSMENT OF THE ELECTORAL GEOGRAPHY
OF UKRAINE: CASE OF PRESIDENTIAL ELECTIONS OF 2019
DOI: http://dx.doi.org/10.18509/GBP.2020.63
UDC: 328.13:342.843.1(477)"2019"
Kanat Makhanov
Eurasian Research Institute, Kazakhstan
ABSTRACT
The study attempts to explain the geographic distribution of votes on the presidential
elections of 2019 in Ukraine based on economic and location data of the districts
constituting the region of the country. Taking the traditional East-West electoral division
of the Ukrainian regions as a baseline the, analysis shows the persistence of the regional
cleavage in the distribution of votes. Based on available data, the study attempts to
incorporate the idea of the economic vote to explain the variations of vote casts across
the regions of Ukraine. The results of the estimates show that the size of the average wage
rate in Ukrainian districts was a relevant factor explaining the voting patterns. The
locational factor also proved to be a relevant variable affecting the vote casts. Being
located in the westernmost parts of the country, especially in the regions of Galicia, was
a significant factor explaining the vote cast differences between candidates.
Keywords: Ukraine, presidential elections, regions, vote casts, economic vote.
INTRODUCTION
Ukraine held two-round regular presidential election on 31 March of 2019. Since the
candidates failed to get an overall majority in the first round of the elections (a simple
majority), the incumbent president Petro Poroshenko and Volodymyr Zelensky, who got
the highest number of votes in the first round of the elections, faced each other in a run-
off on April 21 of 2019. According to the Central Election Commission (CEC), 2,344
international observers from 17 countries and 19 organizations officially registered to
monitor the elections admitted the compliance of the elections with democratic principles
and basic international standards [1]. These elections were regarded by observation
missions as generally reflecting the will of the Ukrainian people [2].
As it was expected by many [3], on April 23 of 2019 the Front-runner Volodymyr
Zelensky won a landslide victory gaining 73.22% of the vote and was declared President-
Elect by the CEC. The incumbent president Petro Poroshenko took 24.45% of the vote.
The election campaign was carried out in a generally professional and smooth manner
across the country [4], except for Separate Districts of Donetsk and Luhansk Regions, the
Autonomous Republic of Crimea and the city of Sevastopol with voter turnout of 62.86%
in the first and 61.37% in the second round of the election [1]. Figure 1 depicts the map
of the ultimate results of the second round of the elections.
The geographic distribution of the votes of the second round of the elections were rather
unusual with overwhelming absolute victory of Volodymyr Zelensky all across the
country except for Lviv region (electoral districts 115-126) and some parts of Tenopil
(electoral districts 163 and 165) and Ivano-Frankivsk regions (electoral district 83) [1].

Socio-economic geography
572
Figure 1. Results of the 2nd round of the elections
The abovementioned electoral districts were the only ones that registered majority of the
votes for Petro Poroshenko. Thus, Volodymyr Zelensky got majority of the votes in all
but one of 25 regions of Ukraine where presidential elections were held. However, this
kind of geographic distribution of the votes is rather similar to the previous elections held
on May 25 of 2014 when Petro Poroshenko got elected with 55% of the vote as a one
round winner [5]. At that time, Petro Poroshenko won absolute majority of the votes in
all across the country except for some parts of the Kharkiv region [1].
LITERATURE
Despite having democratic electoral traditions and having curious geographic electoral
patterns, Ukrainian election cases receive relatively little attention from scholars in
geography and most of the existing works are very recent. However, the among all studies
encompassing the electoral geography of Ukraine the issue of regional polarization and
East-West division is one of the most frequently discussed ones. Studies sets a
comprehensive baseline in the sequence of posterior studies identifying that historical and
cultural legacy of Austro-Hungarian, Polish, Romanian and Czechoslovak rule on the
modern territory of Ukraine is positively related to the pro-nationalist and pro-
independence electoral preferences in the western parts of the country. Historical legacy
and linkages of with Russian and USSR, on the other hand, is correlated with pro-
Communist and pro-Russian support in the eastern part of Ukraine [14].
There are studies reveals a continuous homogenization of the regions of Ukraine in terms
of electoral preferences throughout 2002-2014 except for the westernmost region of
Galicia and the easternmost region of Donbas during 2006-2014 [13]. [6] is one of the
studies done on the electoral geography of Ukraine studying the geography of the turnout

International Scientific Conference GEOBALCANICA 2020
573
in Ukraine throughout 2002-2014. Discussing the decreasing gap in turnout between
urban and rural areas in Ukraine [6] mentions the East-West divide of the country in most
of the parliamentary elections. He assumes it to be a matter of political culture and
differences in historical experiences of certain parts of the country. In another of his
studies, [6] identifies Russian social networks as an important factors in propagating pro-
Russian subscription and fomenting anti-Ukrainian sentiments in eastern regions of
Ukraine [7].
[8] focuses on the East-West dichotomy in the context of Ukrainian conflict resolution
attempting to map the imaginary “East-West” divide of the country outside of its borders.
She argues that the East-West dichotomy in Ukraine is neither innate nor fully
internalized. Her arguments suggest that the East-West polarization of the Ukraine
partially derives from external factors and global trends in political affairs. There is a
comprehensive work done by [16], studying the mechanism of the electoral behavior of
Ukraine. Her analysis detects the relevance of ethno-cultural variables in determining the
voting patterns in the regions of Ukraine. However, one of the interesting aspects of her
work is that she applies the “correct voting” measure used in [17] and [18]. Her finding
suggest that income level of electors are positively related with the degree of correctness
of their vote casts. In general, studies covering parliamentary elections of Ukraine are
more frequent than those focusing on parliamentary elections are [9], [10], [11], [12]
rather than presidential elections [14], [19].
EXPLAINING THE EAST-WEST ELECTORAL CLEAVAGE OF UKRAINE
Ukrainian Presidential candidates with pro-Russian electoral agenda and advocating
closer ties with the eastern neighbor have greater chances to gain much more vote cast in
eastern regions than in other parts of Ukraine [12]. On the other hand, pro-European
candidates with strong appeals to traditional Ukrainian culture and values usually get
great majority of the votes in western regions of the country. A clear regional division of
Ukraine in its electoral patterns were visible yet during the presidential elections of 1991
[19], which gives more relevance to the argument of a historical origin of the regional
cleavages in Ukraine [13], [20], [21], [22]. Most often for purposes of electoral research,
the regions of Ukraine are divided into East and West along the riverbed of the Dnipro
River and southern regions along the coast of the Black Sea usually have the same
electoral patterns as the eastern regions do [16] (Figure 3). Thus, regions like Zaporizhia,
Kherson, Mykolaiv, Odessa and Crimea present electoral patterns similar to the
traditional East of Ukraine. The borderline between the East and West is not fixed,
however, and was shifting eastward and westward in different periods of the electoral
records of the independent Ukraine. In periods of the maximum extension of the western
electoral camp like in 2004 and 2010, the division between the two parts of the country
looks as South-East and North-West. The maximum eastward extension of the line
between East and West of the country reached as far east as the Uman–Kharkiv line [23]
(Figures 5 and 6). The election of 2014 stands by as an exception because it was a snap
election during a conflict with Russia (Figure 7).
When it comes to voting patterns of the East and West of Ukraine, two historical regions
stand out in most of the cases. The most representative electoral behavior of the East is
persistent in the region referred as Donbas, the borders of which fit into the area of the
regions of Donetsk and Lugansk. The typical western electoral region of Ukraine is
represented by the historical region of Galicia, the area of which almost perfectly fits into
the borders of the regions of Lviv, Ivano-Frankivsk and Ternopil (Figure 2). In this regard,

Socio-economic geography
574
electoral preferences of some regions and territorial polarization of Ukraine often follow
modern administrative borders as well as historical borders, which came into existence as
mostly as a result of historical struggles between USSR, Poland in more recent historical
periods and standoff clashes between Austrian and Russian empires in earlier periods
[13].
Figure 2. Results of the
presidential elections of 1991.
Figure 3. Results of the second round
of the presidential elections of 1994.
Figure 4. Results of the second round
of the presidential elections of 1999.
Figure 5. Results of the second round
of the presidential elections of 2004.
Figure 6. Results of the second round
of the presidential elections of 2010.
Figure 7. Results of the snap
presidential elections of 2014.
One can note that higher degrees of polarization of the vote casts between East and West
sharpness of separation between them contributed to more fierce confrontation of the
general public opinion on the election results. The second round of the elections of 2004
is an illustrative case of clash of public opinion of the two sides when Viktor Yushchenko
got more than 93% of the votes in the western regions of Galicia and won with a very
high margin most parts of the western Ukraine. The opposite was true for his opponent

International Scientific Conference GEOBALCANICA 2020
575
Viktor Yanukovych, who received more than 90% of the votes in the easternmost region
of Donbass [14]. The turmoil around the electoral results rapidly developed into the so-
called “Orange Revolution”. The regional context was brightly manifested during the
turmoil caused by the outbreak of conflict in the eastern region of Donbas, which was
manifested in not only in parliamentary [12] but also in presidential election.
The East vs West cleavage of Ukraine during electoral events can be seen in multiple
dimensions and it extends beyond the electoral behavior of different parts of Ukraine.
Linguistic differences between the regions are often presented as explanatory factors in
shaping the cultural differences and attitudes towards politicians [24], [19], [26], [27],
[25], [13] (Figure 9). Based on linguistic factor, the electorate of Ukraine can be classified
as 1 - Ukrainian-speaking ethnic Ukrainians in the western regions of the country
supporting pro-western and pro-European candidates; 2 – Russian-speaking ethnic
Russians of east and southern regions of Ukraine, who prefer to vote for a pro-Russian
candidates; and 3 – bilingual population of the center and eastern center, who often
represent the swing vote [27], [30] (Figure 8).
Figure 8. Ethnic map of Ukraine.
Figure 9. Linguistic map of Ukraine.
Figure 10. Religious map of Ukraine.
Figure 11. GDP per capita in the regions of
Ukraine as of 2017.
Ethnicity of the population in the regions is also a one of the frequently mentioned and
studied explanations of electoral preferences [28], [29]. Some scholar propose the
relevance of historical, institutional, and cultural reasons in explaining regional
differences [21], [22]. Mapping of the population based on ethic and linguistic factors
backed by comprehensive data and control groups often coincide with traditional borders
between of the East, South, West, and Center of the country [31], [32]. Religion has also

Socio-economic geography
576
been tested as a relevant predictor of vote casts and political views of the population [33],
[34] (Figure 10).
Another important factor that makes the eastern regions of the country different from its
western parts is the degree of urbanization of the population, which also can be
understood as a predictor of political behavior of the population of Ukraine [23]. For
instance, all four of the existing cities of Ukraine with population more than one million
are located in eastern, southern and central parts of the country. Nine out of top ten largest
cities of Ukraine are located outside of the traditional West of the country. The city of
Lviv, which is considered to be the center of the historical region of Galicia, is the only
city with population above 500 thousand people located in the west of the country.
However, the urbanization of the population often correlates with other parameters such
as industrialization, employment and income level, making it extremely difficult to
identify the ultimate factor or to isolate the effect of every single predictor [14], [15]
(Figure 11). There are also studies suggesting that neither of the abovementioned factors
is good enough to explain the differences in electoral conduct between the East and West
of Ukraine. Instead, they suppose the predicting factor to be a complex amalgam of
economic and political legacies of the USSR [35], [36].
DISTRICT BASED ANALYSIS
Unlike many of the previous presidential elections in Ukraine, the elections of 2019 did
not present a stark East vs West geographic division of votes. Volodymyr Zelensky’s
landslide victory on the elections by getting majority of the votes in 24 out of 25 regions,
where the elections where held, was exalted by media as an electoral unification and the
end of East vs West dichotomy [37].
Figure 12. Percentage of distribution of votes for the incumbent candidate

International Scientific Conference GEOBALCANICA 2020
577
However, if we differentiate the regions by the percentage of vote casts for either of the
two candidates we will see that there are substantial differences in percentages of the
votes and that the border line between the East and the West is still there where it used to
be during previous electoral events. The map depicted in Figure 12 shows the distribution
of votes in the second round of the presidential elections of 2019 in Ukraine highlighting
the electoral districts with vote casts of at least 24% for incumbent President Petro
Poroshenko. 24% is the approximation for 24.45% of nationwide vote cast for Petro
Poroshenko. Thus, Petro Poroshenko got more percentage of vote than on average in ten
out of 25 regions that held the elections and of them correspond to traditional western
electorate.
Following the idea of the competence model of retrospective voting [38] suggesting that
the decision of voters are based on economic outcomes during the incumbent president’s
term [39] we apply it to the regional distribution of votes between the two candidates on
the election of 2019. The origins of the idea can be traced back as far as to 1960s to the
theory of economic voting [42], [43], [44]. The basic premise that is followed here is that
voters evaluate the performance of the incumbent head of the administration based on
economic performance in their respective areas comparing their wellbeing at the moment
of election to a certain reference point in the past. The concept was applied in many
studies in different forms and variables for a wide range of countries [40], [41].
Following a simple logic of the economic vote theory and applying it to the regions of
Ukraine we estimate the following model:
       
 - average wage in a particular district
 - percentage change of the nominal wage in a particular district
- dummy=1 for districts inside the Galicia region and dummy=0 otherwise
 – dummy=1 for districts inside the average West and dummy=0 otherwise
 - dummy=1 for districts inside the bigger West and dummy=0 otherwise
In order to estimate the model we refer to data on vote casts by each of the 199 electoral
districts inside of the country that were involved in the electoral process. The data is taken
from the CEC [1]. We also use data on average wage in each of the 623 districts that were
under control of the Ukrainian government on the moment of elections. However, due to
incompleteness of all necessary data on all variables we are interested in, the number of
districts was reduced to 555.
Table 1. Estimation results
Coefficient
Standard deviation
t-statistics
p-value
Const.
7.44710
1.30748
5.696
2.00e-08***

0.000639709
0.000159094
4.021
6.61e-05***

−0.0525071
0.0331761
−1.583
0.1141

23.9907
0.837539
28.64
4.67e-111***

11.5938
0.917114
12.64
2.48e-032***

4.00894
0.952446
4.209
3.00e-05***

0.802501

0.800702
Observations
555

Socio-economic geography
578
The estimation results shown in Table 1 indicate the relevance of four out of five variables
introduced in explaining the percentage of vote casts for the incumbent candidate on the
second round of the presidential elections of 2019 in Ukraine. In particular, the average
wage rates, and locations of districts inside all three types of the electorally shaped West
of Ukraine are statistically significant in explaining the vote casts for pro-western
incumbent candidate. Thus, higher wage rates in districts of Ukraine during the period of
elections contribute to higher votes casts for the incumbent candidate. The sensitiveness
it of votes to average wage rate is very low though. According to the model, the average
wage in districts of Ukraine should be increased to 1563 local currency units (nearly
$58.2) to get one additional percent of vote cast. Surprisingly, the change of the average
wage rate during 2017-2018 was not found to be a relevant predictor of votes the
incumbent candidate.
The coefficients obtained for dummy variables of location go in line with the idea that
western location of districts is strong factor that affects the percentage of votes for pro-
western candidate. Location of a district within the border of the historical region of
Galicia gave almost additional percentages of vote casts for a candidate that is perceived
as a pro-western. As it was expected, this coefficient is smaller for wider understanding
of the Ukrainian West. Location of a particular district just west of the Dnipro River
ensured 11.59% additional votes for the pro-western candidate. The fact of being located
in a wider understanding of the electoral West of Ukraine, which cover geographical
West, Center and North of Ukraine, gave additional 4% of votes for the incumbent
candidate.
CONCLUSION
A simple analysis and mapping of the results of the second round of the presidential
elections of 2019 in Ukraine show that the traditional East-West electoral cleavage of
Ukraine persists. The districts of Ukraine located west of the Dnipro River provided
significantly more votes for President Poroshenko, who was perceived as pro-western
candidate. Although it did not affect the ultimate outcomes of the election, it did change
the distribution of votes across the regions of Ukraine. By incorporating the idea of the
economic voting [42], [43], [44] along with the regional components to an OLS model
we found statistically significant estimators for some of the variables. In particular, the
model suggests that districts with higher wage rate relative to the rest of the country
provided higher vote casts for the incumbent candidate. On the other hand, the change of
the average wage rate was not significant enough to explain the variations of vote casts
across the districts of Ukraine. This fact questions the validity of the relevance of the
general idea of the economic vote. However, the major problem with testing the economic
vote is the availability of relevant data and the limitedness of this analysis does not allow
neither to confirm nor to reject it.
The model also showed that location of the districts did play a significant role in the
distribution of votes. The fact of being located in the regions of Galicia meant nearly 24%
more of vote cast for the perceived pro-western candidate. The tendency to vote for the
incumbent pro-western candidate were reduced as for districts located east of the core
West of Ukraine. Hence, geographic location of the Ukrainian districts was a relevant
predictor of vote casts.

International Scientific Conference GEOBALCANICA 2020
579
REFERENCES
[1] Central Election Commission of Ukraine, Results of voting in Ukraine (
https://www.cvk.gov.ua/pls/vp2019/wp300pt001f01=720.html).
[2] Preliminary Statement of the IRI International Observation Mission to Ukraine July
(https://www.iri.org/sites/default/files/2019.7.22_preliminary_statement_of_the_iri_internationa
l_observation_mission_to_ukraine.pdf).
[3] Grajewski, M. Presidential elections in Ukraine. EPRS | European Parliamentary Research
Service, Ukraine, p 1, 2019.
[4] International Foundation for Electoral Systems. Presidential Election in Ukraine: Post-
Election Report
(https://www.ifes.org/sites/default/files/ifes_2019_presidential_election_in_ukraine_post-
election_report_may_2019.pdf).
[5] Yakymenko, Y. Ukraine’s Party System Evolution: 1990-2017, The publication is supported
by the Ukrainian Office of Konrad Adenauer Foundation, 2017.
[6] Dobysh, M.P. Territorial Patterns of Turnout and Representation in Parliamentary Elections
2002–2014 in Ukraine, Vestnik: Geographic and Geological Sciences, Ukraine, vol. 23, pp 69-
81, 2018.
[7] Dobysh, M.P. Euromaidan and conflict in Eastern Ukraine in social networking sites:
Territorial differences of pro-Russian subscriptions in Ukraine, Hungarian Geographic Bulletin,
Ukraine, vol. 68, pp 51-64, 2019.
[8] Hobova, Y. East-West dichotomy in the context of Ukrainian conflict resolution, Cognition,
Communication, Discourse, vol. 17, pp 92–102, 2018.
[9] Clem, R. S. & Craumer, P. R. Orange, Blue and White, and Blonde: The Electoral Geography
of Ukraine’s 2006 and 2007 Rada Elections, Eurasian Geography and Economics, vol. 49, pp
127– 151, 2008.
[10] Craumer, P. R., & Clem, J. I. Ukraine’s Emerging Electoral Geography: A Regional Analysis
of the 1998 Parliamentary Elections, Post-Soviet Geography and Economics, vol. 40, pp 1–26,
1999.
[11] Shevel, O. The parliamentary elections in Ukraine, October 2014. Electoral Studies, 39
(October 2014), pp 159–163, 2015.
[12] Shevel, O. (2015), The parliamentary elections in Ukraine, October 2014, Electoral Studies,
vol. 39, pp 159–163, 2014.
[13] Dobysh, M. Territory, Place, Scale, and Polarization of Electoral Preferences in Ukraine,
2002-2014, Vestnik: Geographic and Geological Sciences, Ukraine, vol. 73, pp 63-68, 2018.
[14] Katchanovski, I. Regional Political Cleavages and Electoral Behavior in Ukraine in 1991-
2004, Conference: Annual Meeting of the American Political Science Association, Washington,
DC, 2005.
[15] Craumer, P. R., & J. I. Clem, Ukraine’s Emerging Electoral Geography: A Regional Analysis
of the 1998 Parliamentary Elections, Post-Soviet Geography and Economics, vol. 40, pp 1–26,
1999.
[16] Surzhko-Harned L.M. The Ukrainian Voter: Electoral Behavior in a New Democracy, PhD.
Dissertation, University of Pittsburgh.
[17] Zhukov, Y. M., & Talibova, R. Stalin’s terror and the long-term political effects of mass
repression. Journal of Peace Research, vol. 55, pp 267–283, 2018.
[18] Lau, R.R., David P. Redlawsk, and David J. Andersen. An Exploration of Correct Voting in
Resent US Presidential Elections, American Journal of Political Science, vol. 52, pp 395-411,
2008.
[19] Clem, R.S. P.R. Craumer, Shades of Orange: The Electoral Geography of Ukraine's 2004
Presidential Elections Eurasian Geography and Economics, vol. 46, pp 364-385, 2005.
[20] Zhukov, Y. M., & R. Talibova. Stalin’s terror and the long-term political effects of mass
repression. Journal of Peace Research, vol. 55, pp 267–283, 2018.

Socio-economic geography
580
[21] Abdelal, R. Memories of Nations and States: Institutional History and National Identity in
Post-Soviet Eurasia, Nationalities Papers, vol. 30, pp 459-484, 2002.
[22] Roper, S. & Flesnic F. Historical Legacies and Their Impact on Post-Communist Voting
Behaviour, Europe-Asia Studies, vol. 55, pp 119-131, 2003.
[23] Karácsonyi D., K. Kocsis, K. Kovály, J. Molnár & L. Póti. East–West Dichotomy and
Political Conflicting Ukraine – Was Huntington Right?, Hungarian Geographical Bulletin, vol.
63, pp 99-134, 2014.
[24] Barrington, L.W & Herron E.S. One Ukraine or Many? Regionalism in Ukraine and Its
Political Consequences, Nationalities Papers, vol. 32, 2004.
[25] Pirie, P. National Identity and Politics in Southern and Eastern Ukraine, Europe-Asia Studies,
vol. 48, pp 1079-1104, 1996.
[26] Ryabchuk, M.A, Future Ukraine: One Nation, Two Languages, Three Cultures?, Ukrainian
Weekly, vol. 23, 1999.
[27] Shevel, O. Nationality in Ukraine: Some Rules of Engagement, East European Politics and
Societies, vol. 16, pp 386–413, 2002.
[28] Bremmer, I. The Politics of Ethnicity: Russians in the New Ukraine, Europe-Asia Studies,
vol. 46, pp 261-283, 1994.
[29] Gee, G.K., Geography, Nationality, and Religion in Ukraine: A Research Note, Journal of
Scientific Study of Religion, vol. 34, pp 383-390, 1995.
[30] Barrington, L.W. & Herron E.S. One Ukraine or Many? Regionalism in Ukraine and Its
Political Consequences, Nationalities Papers, vol. 32, 2004.
[31] Inglehart, R. & Norris P. The Developmental Theory of the Gender Gap: Womens' and Men's
voting Behavior in Global Perspective, International Political Science Review, vol. 21, pp 441-
463, 2000.
[32] Inglehart, R. & P. Norris. Rising tide: Gender Equality and Cultural Change Around the
World, Cambridge, UK; New York: Cambridge University Press. pp 84-87, 2003.
[33] Layman, G. Religion and Political Behavior in the United States: The Impact of Beliefs,
Affiliations, and Commitment from 1980 to 1994, Public Opinion Quarterly, vol. 61, pp 288-316.
1997
[34] Lijphart, A. Religious vs. Linguistic vs. Class Voting : The "Crucial Experiment" of
Comparing Belgium, Canada, South Africa, and Switzerland, The American Political Science
Review, vol. 73, pp 442-458, 1979.
[35] Kubicek, P. Regional Polarization in Ukraine: Public Opinion, Voting and Legislative
Behavior, Europe-Asia Studies, vol. 52, pp 273–294, 2000.
[36] O’Loughlin, J.O. Political Geography of Civic Engagement in Ukraine, 1994-1998. Post-
Soviet Geography and Economics, vol. 39, pp 233-266, 1999.
[37] How Ukraine’s new president broke down a historic divide, washingtonpost.com,
(https://www.washingtonpost.com/politics/2019/05/01/is-this-new-era-ukrainian-politics/).
[38] Persson, T. & G. Tabellini. Macroeconomic Policy, Credibility and Politics, New York:
Harwood Academic, 1990.
[39] Alesina, A. & H. Rosenthal. Partisan Politics, Divided Government and the Economy, New
York: Cambridge University Press, 1995.
[40] Campello, D. & C. Zucco. Presidential Success and the World Economy, Journal of Politics,
vol. 78, pp 589–602, 2016.
[41] Virmani, A. Economic Growth, Governance and Voting Behavior: an Application to Indian
Elections, Indian Council for Research on International Economic Relations Working Paper No.
138, 2004.
[42] Downs, A. An Economic Theory of Democracy, 1st edition, New York, NY: Harper, 1957.
[43] Tufte, E. R. Political Control of the Economy, Princeton, NJ: Princeton University Press,
1978.
[44] Kramer, G.H. Short-Term Fluctuations in U.S. Voting Behavior, 1896–1964, American
Political Science Review, vol. 65(1), pp 131–143, 1971.