Regional Heads' Discretionary Spending in Local Election
FITRA DHARMA
Department of Accounting,
University of Lampung,
Bandar Lampung City,
INDONESIA
Abstract: - Various research in several nations showed that incumbent regional leader who ran for a second
term exploited their budgetary authority to win re-election. The budget components used are primarily
expenditures on people and capital. The purpose of this study was to examine whether discretionary spending
on social assistance and capital expenditure occurred during Indonesia's simultaneous local elections in which
the incumbent was re-run for a second term. We used panel data regression analysis to examine district-city
leaders' 2017 and 2018 simultaneous local elections. The sample of 504 municipal districts showed empirical
evidence that the incumbent increased spending on social assistance, equipment and machinery, and road
irrigation and network maintenance. This finding is consistent with prior research indicating the incidence of
discretionary spending in incumbent led local elections.
Key-Words: - Determination CEO, CEO Attributes, Coal Based Companies, Firm Performance
Received: December 4, 2022. Revised: May 5, 2023. Accepted: May 20, 2023. Published: June 1, 2023.
1 Introduction
Local elections in Indonesia allow voters to assess
regional leaders' performance over five years.
However, because the regional head shares
budgetary authority with the local legislature, the
incumbent can craft the Regional Revenue and
Expenditure Budget (APBD) to suit their political
interests. Regional leaders who run for re-election in
local elections for the second term have the
opportunity of using budgetary authority to win re-
election, [1], [2], [3], [4], [5].
Regional heads increase budget components
directly relevant to the voters and increase
infrastructure spending in the year preceding the
local election to demonstrate their ability to
administer their regions, [1], [6], [7]. The political
budget cycle, discretionary spending, or pork-barrel
spending are all terms used to describe the
phenomenon of using spending for political
purposes. Discretionary spending refers to various
expenditures other than mandatory government
spending given or allocated to specific constituents
without reference to an objective formula, [6].
Previous studies in Canada, Russia, Colombia,
and France, budget allocations for social services
and local government infrastructure increased in
each country in the run-up to local elections. In
Canada, provincial governments increased spending
on social services and roads in the run-up to the
general election, [2]. As the regional head elections
approach, Rusian's local governments increase
public spending directly related to the people, [1].
Colombia's local governments have also increased
direct public and infrastructure spending,
particularly on roads and energy and water
infrastructure, to sway voters to re-elect the
incumbents, [7]. When facing mayoral elections in
France, mayors reduce tax revenues and increase
capital expenditures, [8].
Additionally, the study discovered that spending
to advance the political interests of incumbent
regional heads occurs in Indonesian regencies/cities
that hold direct local elections, [9]. The expenditure
category "other" increased in Indonesia's direct
election years, but the total administrative
expenditures do not, [9]. In Indonesia, political
budget cycles occur solely during direct elections;
indirect elections did not occur. Discretionary
spending or the political budget cycle has not been
widely carried out in Indonesia. Study in Indonesia
regarding the political budget cycle was carried out
when the Indonesian government had not
implemented accrual-based accounting standards, so
there were differences in spending components [9].
This study continues, [9], [5], and examine a more
detailed budget components, namely social
assistance spending and capital spending.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1178
Volume 20, 2023
2 Hypothesis Development
Politicians, including politicians in local
government, are agents who represent and act on
behalf of the people as principals, [10. The
regent/deputy regent and mayor/deputy mayor as
regional heads in regency/municipal government are
political positions. They are elected through
regional head elections by the voters of each region.
Because regional heads are rational individuals, they
tend to maximize their satisfaction and interests,
[11], [10]. One of the motives of politicians in
making government policies is the increase in the
number of votes that voted for them in the next
general election, [12].
People expect the performance of regional heads
after the general election to be influenced by the
budget issued by the regional head before the
regional head election, [13]. Regional head
incumbents with high competence who participate
in the contestation of regional head elections for the
second period increase regional spending and reduce
regional revenues better than incompetent regional
heads, [14], [13]. The regional head's budget policy
in the year leading up to the regional head election
is to increase social and infrastructure spending, [1],
[2], [3], [14], [15].
Social Assistance spending is a program in the
local government’s budget that is not directly
related to implementing programs and activities.
The provision of social assistance to the community
depends on the policies of local government
officials so that they could be used for the regional
heads' political interests. Previous research by, [16],
[17]. found that regional heads in Bali and North
Sumatra provided "material benefits" to the
community to win their hearts. Beneficiaries receive
social assistance expenditures in cash or goods.
This study argues that in the fiscal year leading
up to the local election, regional heads who will run
for a second term will increase social assistance to
the electorate. The increase in electoral related
budgets demonstrates the regional leaders' ability to
lead their regions. Abundant social assistance
spending will signal the community that the regional
head cares for them, [4], [7], [18], [13].
Based on the description above, hypothesis (1) is
then constructed as:
H1: The local election that regional heads
participate in has a positive effect on
increasing social assistance spending.
The Regional Head will increase infrastructure
expenditure in the local election year.
Regents/majors want to show that they can lead the
region. Infrastructure budgets, especially roads and
energy and water construction, increased in
Colombia’s municipal, [7]. Provincial governments
in Canada also increase their budget allocations
ahead of the general election, especially on road
spending, [2]. Irrigation and agriculture expenditure
also increase in India's subnational government at
the timing of the election, [19].
This study argues that to demonstrate the
competence of regional heads to the electorate, in
the year leading up to the general election, regional
heads will increase capital expenditures for both
equipment and machinery expenditures, building
and building capital expenditures, as well as road,
irrigation, and network capital expenditures.
Based on the description above, the proposed
hypothesis (2) to (4) is as follows.
H2: Local election that regional heads participate
in has a positive effect on increasing
equipment and local government machinery
spending
H3: Local election that regional heads participate
in has a positive effect on increasing
buildings and infrastructure spending.
H4: Local election that regional heads participate
in has a positive effect on increasing
irrigation road capital and network spending
3 Materials and Methods
We apply regression model as follows:
BVit = β0+ β1BV-1 + β2LEit + β3Incit +
β4Elect*Inc + β5Spendingit + β6LGit +
β7Islit + ε
Where:
BVi: The budget variables are Social Assistance
Spending (SAS), Equipment and Machinery
Spending (EMS), Building and Construction
Spending (BCS), Road, Irrigation and
Network Spending (RINS) in the local
election year, measured by the percentage
increase in spending in year y and year y-1.
BVit-1: The budget variables are Social Assistance
Spending (SAS), Equipment and Machinery
Spending (EMS), Building and Construction
Spending (BCS), Road, Irrigation and
Network Spending (RINS) in the local
election year, measured by the percentage
increase in spending in year y-1 and year y-2.
LEit: Local elections in year y, measured with
(dummy variable), 1 if there is a local election
in district/city and 0 if not, there is an
election.
Incit: The local election that the incumbent regional
head's participation. It is measured by
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1179
Volume 20, 2023
(dummy variable), 1 if the incumbent regional
head participated in the district/city election
and 0 if not.
Spendingit: Local government spending, to proxy of
government size area, measured by the
logarithm of total local government spending.
LGsit: Different types of local government,
measured by (dummy variable), 1 for district
government and 0 for city government
Islit: Differences in geographic areas, areas outside
Java and Bali in Indonesia score 1 and local
governments in the Java and Bali islands are
given a score of 0.
Ha is accepted if the test result is significant <
0.05 level of significance. If probability > 0.05 then
Ho is accepted. If probability < 0.05 then Ho is
rejected.
4 Results
4.1 Sample Selection
This study examines the utilization of local
government spending during local elections and
emphasizes the election of the incumbent regional
head's successor. The regional elections under
consideration are the 2017-2018 simultaneous
regional elections. For the same level of study,
district and municipal governments throughout
Indonesia were employed as the sample in 2017-
2018. The data on local government spending comes
(LKPD), which the BPK has audited. In contrast,
the data on local elections comes from the General
Election Commission (KPU).
According to BPK data, 542 local governments
were audited in 2017-2018. There are 508 district
and city governments after 34 provinces are
eliminated. Three district and city governments had
insufficient data and thus must be eliminated from
the sample, leaving a total of 505 district and city
governments. During the two-year observation
period of 2017-2018, 1010 observations were made.
The following Table 1 details the sample selection
technique.
4.2 Descriptive Statistics
Table 2 summarizes descriptive statistics; in total,
504 city districts were sampled in 2017 and 2018,
resulting in 1008 observations. Seven hundred sixty-
seven were drawn from city districts that did not
hold simultaneous elections in 2017-2018. One
hundred two regencies/cities held local elections
without the incumbent regional leaders present,
while 138 regencies/cities held local elections with
the incumbent regional heads present.
The mean value of total sample social assistance
spending in the year of the local election (SAS-y) is
1.449. it is a 144.9 percent increase in social
assistance spending over the previous year. For local
governments that hold non-incumbent local
elections, the average rise in SAS-y is 1,407, or a
140 percent increase over the previous year. Local
governments that conduct local elections followed
by an incumbent have the most significant gain,
who average 1.690, a 169 percent increase. This
average confirms that social assistance spending
will grow in local elections won by incumbents.
On average, equipment, and machinery spending
(EMS) is 0.088 for the total sample in the local
election year. The mean value of EMS-y for
incumbent local governments running in local
elections is 0.167, up 16 percent from the previous
year. More than the local government, which held
local elections but did not have any incumbents,
which increased by only 0.11 percent.
The mean value for roads, irrigation, and
networks spending (RINS) fell in election years, to -
0.032 for the overall sample and -0.134 for
incumbent local governments that did not participate
in local elections. Meanwhile, the average figure for
local governments conducting elections and the
incumbent voting is -0.1. The average RINS for
local governments that have an incumbent running
in the general election is higher than the average
RINS for local governments that do not have an
incumbent running. Prior to the local elections, road,
network, and infrastructure spending grew by about
0.1291 percent (12%) for local governments that
held local elections with the incumbent as a
participant.
For all samples, the mean value of Building and
Constructions spending (BCS-y) in the local
election year is 0.158. It indicates that during 2017
and 2018, district and municipal governments in
Indonesia raised building and construction spending
by an average of 15%. However, the mean value of
districts and cities where the incumbent was rerun
for the election was -0.046, a decline of 4%.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1180
Volume 20, 2023
Table 1. Sample Selection Procedure
Table 2. Descriptive Statistics
SAS-
y-1
SAS-y
EMSy
-1
EMSy
RINS-
y-1
RINSy
LGSy-1
LGSy
BCS-y-1
BCSy
N stat
Mean
1,140
1,449
0,146
0,088
0,172
-0,032
12,063
12,054
0,127
0,158
1008
Mean A
1,113
1,410
0,162
0,070
0,204
-0,005
12,061
12,052
0,139
0,223
768
Mean B
1,102
1,407
0,127
0,110
-0,006
-0,134
12,040
12,029
0,005
-0,052
102
Mean C
1,317
1,690
0,071
0,167
0,1291
-0,100
12,092
12,083
0,151
-0,046
138
Min
-1
-1
-1
-0,998
-0,991
-1
11,603
9,346
-0,969
-1
1008
Min A
-1,00
-1,000
-0,923
-1,000
-0,962
-1,000
11,630
9,346
-0,969
-1,000
768
Min B
-1,00
-1,000
-0,705
-0,736
-0,852
-0,752
11,654
11,662
-0,771
-0,774
102
Min C
-1,00
-1,000
-0,998
-0,711
-0,991
-0,772
11,603
11,656
-0,878
-0,889
138
Max
29,28
27,754
9,826
18,624
10,176
20,491
12,898
12,912
17,401
17,996
1008
Max A
29,28
27,754
9,826
7,558
10,176
20,492
12,898
12,912
17,401
17,996
768
Max B
24,42
22,714
4,115
3,250
1,609
3,952
12,828
12,806
2,510
1,106
102
Max C
24,78
2
26,376
1,811
18,624
4,577
4,707
12,743
12,786
4,726
2,495
138
Std. Dev.
3,833
4,090
0,669
0,832
0,758
1,044
0,212
0,241
1,023
0,927
1008
Std. Dev A
3,807
4,005
0,698
0,600
0,809
1,158
0,212
0,248
1,100
1,024
768
Std. Dev B
3,988
3,851
0,675
0,541
0,414
0,508
0,198
0,202
0,541
0,451
102
Std. Dev C
3,882
4,703
0,463
1,683
0,634
0,558
0,219
0,228
0,832
0,446
138
Skewness
4,113
3,505
4,875
12,304
5,791
14,917
0,734
-1
10,310
9,979
1008
Skewness A
4,220
3,570
5,228
4,097
5,932
14,200
0,717
-1,757
10,610
9,558
768
Skewness B
4,231
3,380
2,815
2,470
1,266
5,352
0,868
0,922
2,042
0,747
102
Skewness C
3,596
3,283
0,896
9,796
2,845
5,414
0,739
0,843
3,300
1,805
138
Description:
A: local government without local election
B: local government conduct local election without incumbent
C: local government conduct local election with incumbent
SAS = Social Assistance Spending EMS = Equipment and machinery spending
BCS = Building and Constructions spending RINS = Roads, irrigation, and networks spending
LGS = Local Government spending
4.3 Discussion
Table 3 shows the results of panel data regression.
The dependent variables are social assistance
spending (SAS-y), equipment and machinery
spending (EMS-y), Building and Constructions
spending (BCS-y), Roads, irrigation, and networks
spending (RINS-y). Discretionary spending in the
year of the local election occurs if the interaction
between the local general election and the
participation of the incumbent (LE*inc) has a
significant effect. The panel data regression results
show that the election followed by the incumbent
regional head affects increasing local government
social assistance spending. So H1 is supported.
Description
Year
Total
2017
2018
Local government in Indonesia
542
542
1.084
Minus
Provincial government
34
34
68
District & cities
508
508
1016
Minus
Incomplete data of District & cities
4
4
8
Total samples processed
504
504
1.008
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1181
Volume 20, 2023
The results in Table 3 indicates that when an
incumbent re-run an election, equipment and
machinery spending (EMS), Roads, irrigation, and
networks spending (RINS-y) increases. However,
building and construction spending negatively
affects during an election year. As a result, H2 and
H4 are supported, but H3 is not.
Social assistance spending increases during
municipal election years when the incumbent
participates. In Indonesia, social assistance spending
refers to the gratuitous distribution of goods or
money to the community. Social assistance
spending is one of the most straightforward budget
instruments that regional leaders can utilize to win
the election.
Equipment and machinery expenditure has also
increased. This expense is mainly utilized by local
government personnel to enhance community
services. This surge in spending during the election
season demonstrates the regional chiefs' alignment
with government servants. Because they are
intimately connected to the people during the
government process, bureaucracy becomes a
powerful political machine for regional chiefs
during elections besides the political party.
Roads, irrigation, and networks spending also
increase during municipal election years. This
capital investment is inextricably linked to the
public's needs. People in cities and villages alike
require roads and supporting infrastructure to be in
good shape to conduct their different activities. The
rural community also requires irrigation facilities for
agricultural uses. Regional chiefs running for a
second term boost this expenditure to demonstrate
to the public that they can lead and develop the
region effectively. In the view of the community,
damaged roads will produce a negative perception
of regional leaders. In cities and regencies with
inadequate road infrastructure, non-incumbent
politicians can easily accuse the incumbent of being
incapable of leading and showing a lack of concern
for the people.
Local election years typically see a fall in
building and construction spending. It is because
buildings and construction are inextricably linked to
the community's interests, as are the advantages that
accrue in the year after completion. Hence, regional
leaders do not exploit this expenditure instrument to
advance their electoral political objectives.
Table 3. Analysis of Regression
Variable
Hi
SAS y.
EMSy
BSC y
RINS
Coeff
Prob.
Coeff
Prob.
coeff
Prob.
coeff
Prob.
C
-2,2838
0.5750
-0.8927
0.0426
0.33709
0.1241
-1.316
0.0000
SAS y-1
-0.0959
0.0000
EMS y-1
-0.2881
0.0000
BCS y-1
-0.1094
0.0000
RINS y-1
-0.0237
0.0000
LE
-0.0026
0.9961
0.0514
0.0022
-0.29321
0.0000
-0.1133
0.0000
INC
-1,4787
0.0000
-0.2762
0.0000
0.52868
0.0000
-0.06930
0.0000
LE*INC
H
1,7738
0.0000
0.2692
0.0000
-0.50846
0.0000
0.08623
0.0000
LGS
0.2718
0.4386
0.0912
0.0135
-0.01229
0.5106
0.10865
0.0000
LG
-0.2155
0.0000
-0.0759
0.0000
0.00836
0.0002
0.01225
0.1647
Is
0.9250
0.0000
-0.0778
0.0000
0.01772
0.0000
-0.06438
0.0014
R-squared
0.0161
0.4156
0.33886
0.15916
Adjusted r-squared
0.0092
0.4115
0.33423
0.15327
S.E. of regression
4,071
0,793
0,9032
1,003
F-statistic
2.341
1.016
7.322
2.704
Prob(f-statistic)
0.02250
0.0000
0.00000
0.000000
Description:
SAS= Social Assistance Spending
EMS = Equipment and machinery spending
BCS= Building and Constructions spending
RINS= Roads, irrigation and networks spending
LE = Local Election
Inc = Incumbent
LGS= Local Government spending
LG =Local Government
Is: Islands
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1182
Volume 20, 2023
These findings demonstrate that local Indonesian
administrations' discretionary expenditure on local
elections. The research's findings corroborate, [1],
[8], [2], [7], [9]. During an election year, the
incumbent regional head demonstrates to voters that
he can lead and deserve re-election by raising
capital and social spending.
The research's findings corroborate the agency
theory, indicating that the regional head, acting as
an agent, will maximize his utility to win re-election
by raising regional spending that elicits public
sympathy. During an election year, the rise in
spending demonstrates the regional heads' political
motivations for re-election. In Indonesia, regional
leaders employ short-term tactics such as increased
expenditure to win the election.
To mitigate regional heads' opportunistic
behaviour during election years, the central
government must control local government
budget planning to ensure that it remains
focused on the community's long-term
objectives. The local government's annual
budget must be based on medium and long-term
planning; no annual budget should be devoid of
a medium-term plan. Regional medium-term
plans must be based on regional needs and
compatible with the central governments. Thus,
it is hoped that social and capital expenditure
will not increase in specific years, particularly
election years.
5 Conclusion
The study found that the phenomenon of
discretionary budget for regional head elections
occurred in Indonesia. These results are consistent
with the research of, [1], [8], [2], [20]. Regional
leaders exercise fiscal authority to advance their
electoral political interests. These indicate that the
regional head performs opportunistic actions that
benefit him as an agent.
References:
[1] Akhmedov, A., & Zhuravskaya, E. (2013).
Opportunistic Political Cycles: Test in a
Young Democracy Setting. 119(4), 1301–
1338.
[2] Blais, A., & Nadeau, R. (1992). The electoral
budget cycles. Public Choice, 74(4), 389–403.
https://doi.org/10.1007/BF00137686
[3] Brender, A., & Drazen, A. (2013). Elections,
leaders, and the composition of government
spending. Journal of Public Economics, 97(1),
18–31.
https://doi.org/10.1016/j.jpubeco.2012.08.011
[4] Chortareas, G., Logothetis, V., & Papandreou,
A. A. (2016). Political Budget Cycles and
Reelection Prospects in Greece’s
Municipalities. European Journal of Political
Economy.
https://doi.org/10.1016/j.ejpoleco.2016.02.001
[5] Darmastuti, D., & Setyaningrum, D. (2019).
The effect of discretionary spending on
incumbent victories in elections. Humanities
and Social Sciences Reviews, 7(4), 685–693.
https://doi.org/10.18510/hssr.2019.7488
[6] Curto-Grau, M., & Zudenkova, G. (2018).
Party discipline and government spending:
Theory and evidence. Journal of Public
Economics, 164, 139–152.
https://doi.org/10.1016/j.jpubeco.2018.05.008
[7] Drazen, A., & Eslava, M. (2010). Electoral
manipulation via voter-friendly spending:
Theory and evidence. Journal of Development
Economics, 92(1), 39–52.
https://doi.org/10.1016/j.jdeveco.2009.01.001\
[8] Binet, M. E., & Pentecôte, J. S. (2004). Tax
degression and the political budget cycle in
French municipalities. Applied Economics
Letters, 11(14), 905–908.
https://doi.org/10.1080/135048504200026864
2
[9] Sjahrir, B., Kis-Katos, K., & Schulze, G.
(2013). Political Budget Cycles in Local
Indonesia. Southeast Asian Studies. Uni
Freiburg. …, 17(April), 541–550.
[10] Zimmerman, J. L. (1977). The Municipal
Accounting Maze: An Analysis of Political
Incentives. Journal of Accounting Research,
15, 107. https://doi.org/10.2307/2490636.
[11] Jensen, M. C., & Meckling, W. H. (1976).
Theory of the Firm: Managerial Behavior,
Agency Costs and Ownership Structure.
Journal of Financial Economics, 3, 305–560.
https://doi.org/10.1177/0018726718812602
[12] Downs, A. (1957). An Economic Theory of
Political Action in a Democracy. Journal of
Polilitical Economy Vol 65(2) April 1957, Pp
135-150, 65(2), 135–150.
[13] Rogoff, K. (1990). Equilibrium Political
Budget Cycles. The American Economic
Review, 80(1), 21–36.
https://doi.org/10.1057/9780230226203.0037
[14] Candel-Sánchez, F. (2007). Incentives for
budget discipline in the presence of elections.
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1183
Volume 20, 2023
European Journal of Political Economy,
23(4), 863–884.
https://doi.org/10.1016/j.ejpoleco.2006.09.006
[15] Galli, E., & Rossi, S. P. S. (2002). Political
budget cycles: The case of the Western
German Länder. Public Choice, 110(3–4),
283–303.
https://doi.org/10.1023/A:1013089504557
[16] Sedana, I. N. B. S. (Universitas U. (2014).
Konstruksi Berita Kampanye Pilkada
Kabupaten Gianyar tahun 2008 pada Surat
Kabar Lokal Bali.
[17] Simandjuntak, D. (2012). Gifts and promises:
Patronage democracy in a decentralised
Indonesia. European Journal of East Asian
Studies, 11(1), 99–126.
https://doi.org/10.1163/15700615-20120008
[18] Priyono, S., & Wagner, N. (2018). Political
Budget Cycles At Subnational Level:
Evidence From The Indonesian Simultaneous
Regional Election In 2015. Jurnal
Perencanaan Pembangunan: The Indonesian
Journal of Development Planning, 2(3), 203–
218. https://doi.org/10.36574/jpp.v2i3.38
[19] Sáez, L., & Sinha, A. (2010). Political cycles,
political institutions and public expenditure in
India, 1980-2000. British Journal of Political
Science, 40(1), 91–113.
https://doi.org/10.1017/S0007123409990226
[20] Bohn, F., & Veiga, F. J. (2019). Elections,
recession expectations and excessive debt: an
unholy trinity. Public Choice, 180(3–4), 429–
449. https://doi.org/10.1007/s11127-019-
00647-x
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
The author contributed in the present research, at all
stages from the formulation of the problem to the
final findings and solution.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
No funding was received for conducting this study.
Conflict of Interest
The authors have no conflict of interest to declare.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on BUSINESS and ECONOMICS
DOI: 10.37394/23207.2023.20.105
Fitra Dharma
E-ISSN: 2224-2899
1184
Volume 20, 2023
