Grants Expenditure and Capital Expenditure Pattern Analysis

The aim of this research to determine the Political Budget Cycle's (PBC) patterns in grants expenditure and capital expenditure before, during, and after the regional election in Indonesia in 2012. The method used in this research is a mixed-method, where the first testing used simple regression (quantitative approach) then followed by content analysis (qualitative approach) for getting more finding from the data. Sample method used in this research is based on a purposive sampling method. The results showed that there were Political Budget Cyle's patterns though there was no significant effect of Regional Election on Grants Expenditure and Capital Expenditure. The conclusion indicated that there were Political Budget Cycle's patterns in Indonesia on Grants Expenditure and Capital Expenditure.


Introduction
Regional autonomy policy in Indonesia has brought many fundamental changes to the relation between Local Government (executive) and Dewan Perwakilan Rakyat Daerah (DPRD) or legislative. Before the year 2005, local leaders and their deputies were elected by Dewan Perwakilan Rakyat Daerah (DPRD). Since the application of Undang-Undang Nomor 32 Tahun 2004 about Regional Government which was an alteration from UU No. 22 Tahun 1999, local leaders were elected directly by people through Regional Election (Pemilihan Kepala Daerah and Wakil Kepala Daerah or Pilkada for short).
The relation between the executive and legislative has been through some changes, they are no longer as agent (executive) and principal (legislative), but both act as agents of the voter's community. Using sample from the United States of America, (Benito, et al. 2012:342) showed that "the size of the electoral cycle, especially Political Budget Cycle (PBC) did not depend on whether the governor was lame ducks or not." "Lame ducks opportunistic behavior can also be explained by their desire to help their colleagues to enhance the chance to get a better job outside politics," (Rosenberg, 1992:2).
Based on Peraturan Pemerintah Republik Indonesia Nomor 58 Tahun 2005 about Pengelolaan Keuangan Daerah, that state-local leader as the regional government leaders have the authorities of the regional financial management and represent local government concerning the ownership of the separated regional wealth. With the power that they have, incumbent as well as lame ducks, they have a http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 considerable chance to misuse the expenditure in APBD for their benefits. This budget abuse is known as Political Budget Cycle (PBC), which concentrates more on the utilization of fiscal policy.
PBC (Political Business Cycle) is an effort to manipulate the economy by using fiscal and monetary policy (Sakurai and Filho, 2010:3). PBC (Political Business Cycle), " with a myopic voter, thenincumbent or lame ducks government can use macroeconomic policy for their benefit so that the voters will respond by supporting those incumbent or lame ducks," Nordhaus (1975:20). Ehrhart (2010:4), "manipulation for the benefit of the election will cause damage to economic condition, and there is the fact that economic condition recovery can last several years." It is crucial to understand what type of expenditures can be manipulated.
Political Budget Cycle (PBC) is a common phenomenon all around the world, according to Shia and Svensson (2006:2). However, most Political Budget Cycle (PBC) testing in those countries was more concentrated on the general election that determines central government, especially presidential election. Benito, et al. (2012:342). Local governments, as well as provincial government, are part of the government that is closest to the community, whose policy can be observed carefully by the community. As consequences, incumbent or lame ducks in local governments and also provincial government increase greater fiscal policy and budget manipulation to influence potential in local government as well as in provincial government than in the central government.
Political Budget Cycle (PBC) testing can be done by testing expenditure policy fluctuation taken by incumbents and also lame ducks as explained by Veiga and Veiga (2007:3) that "there was evidence of expenditure composition manipulation, which incurred far greater in election year on expense classification that is highly visible by the voters while other expenditures that are less visible by the voters will stay the same or decrease". This explanation was confirmed by Drazen and Eslava (2005:4). The phenomenon in the region which its expenditures were highly visible by the voters was West Java villages fund program organized by West Java province in the year 2013 where every village receive fund amounted Rp. 100.000.000,00 through grants expenditure scheme.
The increasing highly visible expenditure phenomenon was confirmed by the early study. This phenomenon can be seen in attachment 1. There were grants expenditure fluctuations in the year 2011-2013 with the conducting of the regional election in 2012. There were increases in expenditure in 20 out of a total of 30 regions a year before the regional election compared to year during the conducting of the regional election. Moreover, there were decreases in grants expenditure in 22 out of 30 regions a year after regional election compared to the year during the regional election was being held.
Aside from highly visible expenditures increasing phenomenon, there was a decreasing phenomenon from the less visible expenditures. This phenomenon can be seen in attachment 2. There were capital expenditure fluctuations in the year 2011-2013 when the regional election was held in 2012. 17 out of a total 30 regions experience a decrease in capital expenditure a year before the regional election compared to the year when the regional election was being held. Moreover, 17 out of 30 regions experienced capital expenditure increase a year after regional election compared to the year during the regional election was being held.
Based on that background, the identification of the problem in this research is: were there Political Budget Cycle

Literature Review
Business Cycle is a fluctuation in the economy. (Mankiw, 2008:15). PBC (Political Business Cycle) is an effort to manipulate the economy using policy and monetary policy (Sakurai and Filho, 2010:3). This theory assumes that the business cycle phenomenon can be connected with political activities, such as election to vote for president or in other word cycle is also connected with the election cycle.
Meanwhile, Nordhaus (1975:188) gave a conclusion about what is PBC (Political Business Cycle), "with a myopic condition that the voters had, the incumbent or lame ducks government can use macroeconomy for their own benefit so that the voters will respond by supporting those incumbent or lame ducks". So, it can be concluded that Political Business Cycle (PBC) is "efforts done by incumbent government to get reelected in the next election by manipulating economic policy for short term benefit or lame ducks that will give signal to enhance the opportunity to get a better job outside politics or their desire to help their colleagues", (Rosenberg, 1992:2). In Ehrhart (2010:4), "manipulation for the election benefit will cause damage to the economic condition, there is a fact that economy condition recovery can last several years." It is crucial to understand the types of expenditure that can be manipulated.
Theoretically, the opportunistic model and partisan had developed in two phases. In phase one, the model was based on the assumption that voters have adaptive expectations which are assuming that voters will do mostly the same things in the future even though they were systematically "tricked." In the second phase, the model was developed based on the assumption that voters are rational and they cannot systematically be "tricked" based on available observation of the available information. In this case, political cycles arise based on the asymmetry between the voters and incumbents or lame ducks (Benito et al., 2012:344) From the first writing about PBC, we can arrange a diagram about the development of PBC theory until today as follow: PBC with Opportunistic (Oportunis) model started with Nordhaus scientific work (1975:187) that concluded incumbent government will put their best effort to get reelected by making conducive macroeconomic condition approaching the election, does not matter with which political party the incumbent government were affiliated. Opportunistic PBC has an assumption that voters will give a reward for the incumbent government if the incumbent government (as well as the party that became incumbent government affiliation) able to provide the best macroeconomic condition during the upcoming election.
Government policy that was not neutral during upcoming election caused PBC opportunist to occur by using monetary or fiscal policy. From these useable tow policies, emerge a derivative of Opportunist PBC theory, which is Political Monetary Cycles (PMC) and Political Budget Cycles (PBC).
According to Krause and Mendez (2004:752), Other PBC models are "Partisan models developed by Hibbs (1977) and Alesina (1987) who basically mentioned that PBC can also happen because of ideological difference between alternating parties holding the power, one party (for example the left wing) concerned more about unemployment rate and support expansionary policy to reduce the output gap with the risk of certain inflation rate, and the other party was more sensitive to inflation rate and did not put much attention to unemployment". Hibbs (1977Hibbs ( :1467, "was also focused to Phillips curve (like in Opportunist theory) and view that political party has a different position along that Phillips Curve, depend to the policy of each political party". But in reality, Partisan Model became opposition of Opportunist Model because Partisan Model emphasize on politician behavior to maximize their own objective function, different with Opportunist theory who only conclude that opportunist politician only wants to get re-elected and to have their original post. Along with the introduction of Rational Expectation concept, the revision of Hibbs' Partisan Model was done by Alesina in 1987 and1988, Alesina and Rubini in the year 1992 and many other researchers.

Partisan model, developed by
One of the derivatives of Political Business Cycle (PBC) is Political Monetary Cycle (PMC). Political Monetary Cycle (PMC) can be defined as taking advantage of a short-run trade-off from the Phillips curve near regional election to gain benefit (Nordhaus, 1975:187). This trade-off happens to inflation and unemployment, where inflation inversely proportional to unemployment. Assumption underlying PMC is "non-independent central bank so that, incumbents, as well as lame ducks, use their power in money supply policy," Benito et al., (2012:342). "Incumbents, as well as lame ducks, increase the amount of money supply so that unemployment rate can be reduced near regional election making the voters reelect the incumbents or supporting their colleagues or increasing the chance to get a better job outside of politics" (Rosenberg, 1992:2).
Other derivatives of Political Business Cycle (PBC) is the Political Budget Cycle (PBC). Political Budget Cycle (PBC) can be defined as "periodic fluctuation in fiscal policy influenced by election cycle", (Benito, et al. 2012:342). Political Budget Cycle (PBC) is a common phenomenon across the world according to Shia and Svensson (2006:2), Political Budget Cycle (PBC) is a "universal phenomenon both in developed countries as well as in developing countries regardless of the form of the government in each country, such as, monarchy, republic, etc".
However most Political Budget Cycle (PBC) testing in those countries were concentrated to election that determines central government, especially presidential election. As expressed by Benito, et al. (2012:342) that "most Political Budget Cycle (PBC) literature focused into central government, only a few works of literature http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 put their attention Political Budget Cycle (PBC) in provinces or Regency/cities". Local governments, as well as provincial government, are part of government that closest to the community, whose policies can easily be monitored by the community as consequences, incumbent or lame ducks in local government or provincial government increase greater fiscal policy or budget manipulation to influence voters in local government or provincial government rather than central government.
Based on those theories, the hypothesis of this research are: pattern in grants expenditure before, during, and after a regional election. H2: There is no PBC (Political Budget Cycle) pattern in capital expenditure before, during and after a regional election.

Research Method
The method used in this research is a mixed-method, where the researchers first did testing using descriptive and verification quantitative approach, then followed by a descriptive qualitative approach. The population in this research are the local governments in Indonesia conducting the regional election in the year 2012. The total population for this research is 82 Local Governments (Pemerintah Daerah (Pemda), while the research samples are 30 Local Governments.
The sampling method used in this research is based on a purposive sampling method. The referred criteria are the availability of local government financial statements period 2011 till 2013 and the availability of all data needed during research method and availability of grants expenditure data submitted to Komisi Pemilihan Umum (KPU) in the year 2012. Using the selection criteria mentioned above, 30 Province/Regencies/Cities were chosen, consisting of 3 Province, 6 Regencies and 21 Cities. This research took 3 years time.
Governor, Regent, and Mayor Elections are the election of governors, regents, and mayors democratically in Negara Kesatuan Republik Indonesia based on Pancasila and Undang-Undang Dasar Negara Republik Indonesia Tahun 1945. (UU no. 15 Tahun 2011 Penyelenggaraan Pemilihan Umum). This regional election was measured by reviewing regional election execution report posted in the official KPU website. Capital expenditure is budget expenditure for fixed assets and other assets acquisition that give benefit for more than one accounting period (Halim and Kusufi, 2012:107). Capital expenditure is measured by looking at capital expenditure value presented in Local Government Financial Statements.
Grants expenditure is "expenditure used to budget grant in the form of money, goods, and service to government or other local government, and a group of community/individual whose allotment has been assigned specifically " (Permendagri No. 39 Tahun 2012). Grants expenditure is measured by looking at grants expenditure value presented in Local Government Financial Statements.
In this research, the researcher used data gathering techniques commonly used in research, which are field research and literature research. Data analysis methods used in this research consist of two approaches, which are a quantitative and qualitative approach. In a quantitative approach, the researcher used the event study analysis taking advantage of a regional election in the year 2012. This event was treated as an event that can give a chance for the incumbent as well as lame ducks to perform opportunistic behavior concerning fiscal policy or budget manipulation which Political Budget Cycle (PBC) hoping to be able to influence the voters.
While in a qualitative approach, grants expenditure and capital expenditure fluctuations analysis were done using http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 content analysis Content analysis is a "research technique used to determine the occurrence of words or concepts in writing or corpus" (Busch, et al. 2012:1). Generally, according to Busch et al. (2012:3), there were two types of content analysis which were conceptual analysis and relational analysis "But, traditionally, conceptual analysis is a content analysis most often used. Conceptual analysis can be used to determine the existence and frequency of the concept displayed in most occurrence word in a text, so we can determine how many times the word occurs" (Firdaus, et al. 2010:11). "While, content analysis can be used to observe trend" (Busch, et al. 2012:3).

The simple linear regression model with a dummy variable
The simple linear regression model with a dummy variable used defined as followed: 1) BH = + 2012+ 2) BM = + 2012+ where, BH = Grants expenditure in a certain year BM = Capital expenditure in an certain year = excluded variable = coefficient variable 2012 = dummy variable regional election (t worth 1 if it falls on the regional election and 0 in other year and so forth) = residual error.

Result and Discussion
According to descriptive statistical testing result, the grants expenditure was forming Political Budget Cycle (PBC) patterns, which can be seen from the mean value fluctuation and maximum value in 2011-2013. The mean value in 2011-2013. Grants expenditure mean value in the year 2012 is bigger than grants expenditure mean value in 2011 and 2013 and grants expenditure maximum value in 2012 was bigger than grants expenditure maximum value in 2011 and 2013. This means that, descriptive statistic result confirmed the researcher's hypothesis that there was an increase in grants expenditure during regional election where the increase was the result of opportunistic behavior misusing the available fiscal capacity for their own benefit or their colleagues or in other words they were doing PBC (Political Budget Cycle) by moving budget allocation used to support them or their colleagues.
Meanwhile, for capital expenditure, forming Political Budget Cycle (PBC) patterns can be seen from the mean value fluctuations, minimum and maximum value in 2011-2013. Capital expenditure means value in 2012 is smaller than capital expenditure means value in the year 2011 and 2013. These findings were confirmed by minimum and maximum value fluctuation in 2011-2013. This means that, descriptive statistical result confirmed the researcher's hypothesis that there was a decrease in capital expenditure during the regional election where the decrease was the result of opportunistic behavior misusing the available fiscal capacity for their own benefit or their colleagues or in other words they were doing PBC (Political Budget Cycle) by moving budget allocation used to support them or their colleagues. The allocation move from less visible expenditures to highly visible expenditures.
Analysis tool used to estimate the equation in this research is simple regression analysis.
Before doing a regression test, the gathered data is first analyzed using classical assumption test. Classical assumption test used in this research is normality and heteroskedasticity. In this research, all classical assumption test have been done and the result showed that equation model with grants expenditure as the dependent variable is not distributed normally, then root transformation was done and the result of this root transformation was free from heteroskedasticity and normal data distribution. While the equation model with http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 capital expenditure as a dependent variable was free from heteroskedasticity and normal data distribution.
According to the result of statistically data processing above, there were a Political Budget Cycle (PBC) patterns on grants expenditure. This show that there were grants expenditure patterns based on Sakurai and Filho (2010) and Ritonga and Alam (2010) that during regional election, local government tend to show opportunistic behavior by misusing available fiscal capacity for their own benefits or their colleagues benefit, or in other word, performing PBC (Political Budget Cycle) by moving budget allocation used to support them or their colleagues grants expenditure (highly visible). Even so, the conducting of a regional election in this research has less real influence (R) to grants expenditure and insignificant to grants expenditure.
The small influence was caused by the cautious attitude of the local leaders, both incumbent and lame ducks. This cautious behavior can be seen by BPK findings in grants expenditure if comparing between years 2007 to 2012. There were BPK findings in grants expenditure in LKPD Kab. Lampung Barat, Kota Cimahi, and Kabupaten Halmahera Tengah year 2007 but those findings were not found in the year 2012. Besides that, allocations were distributed to other costs such as the local leaders' inauguration through goods and service expenditure scheme.
The difference in sample research might influence that the result of this research was not significant, because out of 82 local government which was the population, researcher only took 30 to be the research sample, meaning 36,58% out of all local government conducting regional election in year 2012 due to data limitation concerning local government grant limitation to KPU regarding regional election.
From the available data, it can be seen that in five local governments, there is an increase consistently from the year 2011-2013, which mean that grants expenditure increase was considered proper considering APBD tendency to keep increasing from year to year and the increasing need for grants expenditure, causing the consistent increase. The same with other ten local governments, even though there was a decline from the year 2011-2012, from 2012-2013 the deficit value became smaller in four local governments, and even there were positive values in other six local governments.
The result of the regression analysis for hypothesis 2 testing, a simple linear regression equation as follow was obtained: Y2 = 0,208 -0,016 D1 + e. The explanation is as follow: α = 0,208 is constant meaning capital expenditure will be 0,208% when the regional election is not conducted, which has a value of 0. β1 = -0,016 is a regression coefficient for regional election variable of -0,016, meaning regional election was conducted, which has a value of 1, then capital expenditure while decline 0,016% assuming other variables are constant.
The result of statistical data processing was that there were Political Budget Cycle (PBC) patterns in capital expenditure. This shows that there were capital expenditure patterns according to Sakurai and Filho (2010) that during Regional election, Local government tend to show opportunistic behavior misusing fiscal capacity for their benefit or their colleagues, or in other word, doing http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 PBC(Political Budget Cycle) by moving budget allocation used to support them or their colleagues in capital expenditure (less visible). Even so, the conducting of a regional election in this research have little real influence (R) towards capital expenditure and not significant to capital expenditure.
The small influence was caused by the cautious attitude of the local leaders, both incumbent as well as lame ducks. This cautiousness can be seen in BPK findings in capital expenditure while comparing the year 2008 with 2013, where BPK findings in capital expenditure in LKPD Kab. Ciamis year 2008 but the findings were not found in the year 2013. Besides that, the allocations were distributed to other costs such as the cost of local leaders' inauguration through goods and service expenditure scheme.
The researcher also did qualitative data analysis to explain further the result of quantitative data analysis. Qualitative data analysis was done through content analysis during the early grants expenditure and capital expenditure study by the researcher. The conformity of PBC (Political Budget Cycle) pattern with the researcher's early study was shown in attached figure 1 From the data that the researcher gather, it can be concluded that as many as 4 Local governments out of 30 Local governments or 49,63% have PBC (Political Budget Cycle) patterns in grants expenditure allocation even though it is not the majority of the pattern. The result of the qualitative approach supports the result of quantitative approach done before that, where quantitative result showed that there were a PBC (Political Budget Cycle) patterns in grants expenditure allocation, but it was not significant.
Meanwhile, out of 14 local governments where Political Budget Cycle (PBC) patterns existed in grants expenditure, 10 of local governments were incumbents. It means that majority 71% of local governments who had Political Budget Cycle (PBC) patterns were local governments with incumbents status. Meanwhile, 30% from total local governments were incumbents local government which had Political Budget Cycle (PBC) patterns in grants expenditure. It means that there were higher Political Budget Cycle (PBC) patterns in grants expenditure local government with incumbents status. The result of the qualitative analysis was confirmed by the result of a quantitative calculation, where correlation coefficient and coefficient of determination in local government incumbents were higher than the local government with incumbents and lame ducks combination.
Meanwhile for capital expenditure patterns, from the data that the researcher gather, it can be concluded that 11 Local governments or 36,67% had PBC (Political Budget Cycle) patterns in capital expenditure allocation although they were not the majority. The qualitative approach supports the quantitative approach done before that, where the result of the quantitative approach that there were PBC (Political Budget Cycle) patterns in capital expenditure allocation but they were not significant. Out

Conclusion and Suggestion
In this research, researcher test how Grants expenditure and Capital expenditure patterns analysis before, during, and after Regional election in Regional Government in Indonesia Budget Year 2012. The first hypothesis testing showed that there were Political Budget Cycle (PBC) patterns in grants expenditure. The result of this research was in accordance with Sakurai and Filho (2010) and Ritonga and Alam (2010) that during Regional election, Local government tend to show opportunistic behavior misusing fiscal capacity for their own benefit or their colleagues, or in other word doing PBC (Political Budget Cycle) by moving budget allocation used to support them or their colleagues in grants expenditure (highly visible). Even so, the conduction of the Regional election in this research only has a small influence on grants expenditure and the influence was not significant.
The result of the qualitative approach supports the result of quantitative approach done before that, where the quantitative result showed that there were PBC (Political Budget Cycle) patterns in grants expenditure allocation, but it was not significant. There were PBC (Political Budget Cycle) in 14 Local governments out of 30 Local governments researched or 49,63% meaning that there were PBC (Political Budget Cycle) patterns in grants expenditure allocation even though they were not the majority. Moreover, there were higher Political Budget Cycle (PBC) patterns in grants expenditure in local government with incumbents status.
The result of the second hypothesis testing showed that there were Political Budget Cycle (PBC) patterns in capital expenditures. This result was in accordance with Sakurai and Filho (2010) that during Regional election, Local government tend to show opportunistic behavior by misusing available fiscal capacity for their own benefit or their colleagues', or in other words they were doing PBC(Political Budget Cycle) by moving budget allocation used to support them or their colleagues from less visible capital expenditure. Even so, the Regional election in this research has little real influence on capital expenditure, and it was insignificant.
The small influence in both hypothesis testing was caused by the cautious behavior of the local leaders, both incumbent as well as lame ducks. This cautiousness can be seen in decreasing BPK findings in Capital expenditure between the year 2008-2013. Besides that, Capital expenditure allocation was distributed to other costs such as local leaders inauguration cost through goods and service expenditure scheme. Sample research difference might influence the insignificance of this research's result because out of 82 local governments that became population, and the researcher only took 30 to research sample because of the limitation of local government grants data submitted to KPU for the benefit of the regional election.
The result of the qualitative approach analysis showed that there were PBC (Political Budget Cycle) in capital expenditure allocation, but they were not significant. The same thing is shown in the table above, where there were PBC (Political Budget Cycle) in 11 local government or 36,67%, meaning that there were PBC (Political Budget Cycle) in capital expenditure allocation although they were not the majority. Also, there were higher Political Budget Cycle (PBC) patterns in capital expenditure in local governments with incumbents status.
For the Local Government, to review the effectiveness of local regulation concerning grants expenditure and capital expenditure so that the damage to the economic condition due to Political Budget Cycle (PBC) can be minimized and can be allocated to other items that can have a http://jurnal.unpad.ac.id/jaab -ISSN: 2614-3844 multiplier effect towards the economic condition of each region. Even though the research showed that Political Budget Cycle (PBC) had little influence and insignificantly or has minority result in a qualitative approach, there still Political Budget Cycle (PBC) pattern in Local Government Financial Statement (Laporan Keuangan Pemerintah Daerah (LKPD)). For Badan Pemeriksa Keuangan (BPK) in every province to do more in-depth supervision for the two kinds of expenditure. For the next researcher, to enlarge the sample studied so that problems concerning correlation coefficient, determination coefficient, and insignificance of hypothesis testing can be resolved. Subsequent researches using other variables such as, goods and services expenditure, which can substantiate this research are also expected.