Paying bribes in :

A survey of business

1

202

January RESEARCH BRIEF BRIEF RESEARCH

AUTHORS: Paul Kenny

and Eve Warburton

ACKNOWLEDGEMENTS:

The authors would like to thank Paul Hutchcroft and Edward Aspinall for comments on earlier versions of this article. We also thank Burhanuddin Muhtadi and colleagues from Lembaga Survei Indonesia (LSI) for their important assistance in the design and Policy Brief prepared for Australia’s implementation of the survey. Finally, this Department of Foreign Affairs and Trade for the report was made possible by funding from ‘Supporting the Rules-Based Order’ project at Australia’s Department of Foreign Affairs and the Department of Political and Social Change, Trade for the ‘Supporting the Rules-Based ANU. For further details, see, Paul D. Kenny and Order’ project of PSC. We thank them for their Eve Warburton, “Hidden Profits: Sector-specific support. corruption in Indonesia”, APSA Pre-Prints, Sept. 2020, https://doi.org/10.33774/apsa-2020- nssz5 THE AUTHORS: Paul Kenny is Professor in the Institute for Humanities and Social Sciences at the Australian Catholic University. He specializes in political economy and comparative politics. DISCLAIMER: He received his PhD in political science from This article is part of a New Mandala series related Yale University and has degrees in economics to the Supporting the Rules-Based Order in and political economy from Trinity College Southeast Asia (SEARBO) project. This project is Dublin and the London School of Economics. run by the Department of Political and Social Prior to joining the ACU, Professor Kenny Change at the Australian National University and worked at the Australian National University, funded by the Australian Department of Foreign where he was Associate Professor and Head Affairs and Trade. The opinions expressed here of the Department of Political and Social are the authors' own and are not meant to Change. represent those of the ANU or DFAT.

Eve Warburton received her PhD from the Australian National University’s Coral Bell School of Asia and Pacific Affairs. Her doctoral dissertation examined the political economy of resource nationalism, and in particular the role that domestic firms play in shaping COVER IMAGE: nationalist policy trajectories. She is now "Kantor Pusat KPK - 2015" by Ya, saya finalising a book manuscript based on her inBaliTimur is licensed under CC BY-ND 2.0 dissertation, Reclaiming What’s Ours: A Political Economy of Resource Nationalism in Indonesia. Eve’s broader research interests include identity, representation, and the political economy of policymaking in young . democracies. Her work focuses primarily on Asia, and especially Indonesia.

1

Leveraging an original survey of 672

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying orruption is a pervasive Indonesian managers and business owners, C we set out to explore how different sorts of feature of Indonesia’s contemporary political companies experience corruption. We economy. In-depth studies investigate patterns of corruption at different and cross-national surveys administrative levels and across each of alike emphasise the ubiquity Indonesia’s major economic sectors — of illicit fees and agriculture, mining, construction, across the public and private manufacturing, and financial services. Like sectors, and in the everyday any survey on corruption, we expected some lives of Indonesians. of our participants to be ambivalent about Transparency admitting to engaging in or observing illicit

International’s Corruption activity (social desirability bias). However, Introduction business corruption business Perceptions Index, which is while we cannot be sure of the absolute based upon the views of experts and number of firms truly exposed to corruption, business people, gave Indonesia a score of 40 the variation we uncover allows us to draw out of a possible 100 in 2019, where 100 is reliable and important inferences about very clean. The score is an improvement on differential patterns of corruption across recent years, but reflects Indonesia’s firms and sectors.

continued struggle to change deeply Overall we find that about one third of entrenched practices of rent-seeking, graft companies have been asked to make illicit and bribery. payments to state officials — but that figure

The term ‘corruption’ refers broadly to the varies substantially across sectors. For misuse of public office for private gain. Such example, a majority of companies in the terminology suggests public sector actors to mining and construction sectors have be the agents of corruption. However, the received these sorts of requests, compared to intensity and character of corruption a small minority in sectors such as finance depends to a substantial degree on the and manufacturing. Across a range of similar incentives and constraints on those on the measures, sectoral differences matter for other end of corrupt exchanges: people in the how companies engage in and view informal private sector. In this report, we examine flows of money. Mining and construction corruption from the perspective of stand out as especially afflicted by corrupt Indonesia’s business people. practices. Firms in these sectors experience the most requests for bribes. But they are

2

also more likely to alter their financial the terms of market access, regulations, and

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying reports, which we suggest is because hiding effective tax rates. Corruption excludes costs and profits is easier for firms in actions, however illegal, that do not involve extractive sectors and construction than at least one party acting in an official capacity most other sectors, allowing them to make (Holmes, 2015, pp. 14-16). back the losses incurred due to state Corrupt exchanges have characterised state- . This argument implies that business relations throughout Indonesian attempts by Indonesian policymakers and history, but became especially endemic international organisations to diagnose, during the . Under , measure, and reduce corruption could more bureaucrats were encouraged to treat their efficiently focus their programs on specific office like a “franchise”, building networks of problem sectors. private sector clients and personal associates

for whom they facilitated lucrative business corruption business 1. Background: government contracts, import licenses, or firms and opportunities to partner with foreign investors (McLeod, 2011). The rents from corruption in such transactions made senior state officials immensely wealthy; but rents were also Indonesia channelled upwards through the Corruption is a major global challenge. bureaucratic hierarchy. For the business Recent estimates put the cost of corruption community, opportunity and success often at more than 5 percent of global GDP (US$ required close and personal relationships 2.6 trillion) with over US$ 1 trillion paid in with bureaucrats, ministers, or the president bribes annually. Corruption reduces himself. economic productivity, deepens inequality, The pro-democracy movement that forced and may even harm human health (Fisman & President Suharto’s resignation in 1998 Golden, 2017, pp. 83-120; Holmes, 2015, pp. demanded an end to corruption and 18-35). Corruption potentially includes a , and the reforms that followed vast set of actions, ranging from the illegal to brought greater transparency to state- the immoral. As different forms of corruption business relations. New anti-corruption are likely to have different causes, for the NGOs proliferated in the early years of purposes of our investigation, we focus on Indonesia’s democracy, new laws were corruption in the business sector, which passed, and independent watchdogs were set primarily includes bribery and extortion over

3

up to investigate illicit economic activities oil permits, quickening the pace of

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying and the misuse of public funds. The deforestation and land degradation, and Corruption Eradication Commission (Komisi prompting new and sometimes violent Pemberantasan Korupsi, KPK) in particular conflicts in resource-rich regions (Burgess, has been a major actor in efforts to reduce Hansen, Olken, Potapov, & Sieber, 2012). corrupt behaviour among political and In response, activists, NGOs and the KPK economic elites. The Commission’s sting have launched a wide range of programs to operations led to 1,152 arrests between combat corruption between the state and 2004 and 2019. Of those arrested, 297 came private sector. In the natural resource from the private sector. sectors, for example, the KPK launched a Yet corruption remains endemic to series of investigations into corrupt practices Indonesia’s post-authoritarian political in the mining industry, and worked together

economy. Democracy and decentralisation with NGOs and local activists across the business corruption business did not reduce corruption to the extent that country to make data public on irregular reformists had hoped; instead, the political mining licenses and environmental permits. transition reshaped prevailing patterns of The KPK also turned its attention to rent seeking and extortion. In post-Suharto corruption at the subnational level. Under Indonesia, companies interact with multiple the Jokowi administration, the number of levels of government, and the delegation of local leaders arrested by the KPK has spiked: authority for many business licenses and during President Yudhoyono’s decade in land-use permits to local executives opened power, the KPK arrested 45 district leaders up new opportunities for corrupt exchanges or their deputies; between 2015 and 2019 between the private sector, district-level alone, the KPK sent 74 district leaders to bureaucrats, local security apparatus, and prison. criminal entities. Competitive elections In general, however, the Jokowi government created new incentives for politicians to has emphasised preventative rather than extract rents from the private sector too, as punitive measures in the fight against electoral campaigns became increasingly corruption. The most powerful illustration of expensive (Muhtadi, 2019). The effects of the president’s aversion to the punitive and corruption in democratic Indonesia have often disruptive approach of the KPK came in been especially pernicious in the natural late 2019, when he approved a set of legal resource sectors, where collusion between revisions that weakened the watchdog’s companies and local state officials led to an investigative powers. Instead, Jokowi’s explosion in the number of mining and palm

4

reform program has focused primarily on constraint on their firm’s operations. Such

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying reducing the layers of red tape and wide-ranging assessments not only bureaucratic obstacles for investors, which underscore the difficulties inherent in create delays and incentives for payoffs. As a measuring corruption, but also suggest that result, Indonesia rose 19 positions in the companies in Indonesia experience corrupt World Bank’s Ease of Doing Business survey exchanges in very different ways. between 2017 and 2018. In the survey we This report contributes to efforts to better conducted for this study (described below), understand the nature of corruption in we also found that 65 percent of business contemporary Indonesia, and to develop representatives felt the current regulatory appropriate remedies. It looks specifically at climate helped rather than hindered their how businesses experience bribery and firm’s operations. extortion, and how those experiences differ

But when it comes to reducing corruption, depending on the sector, firm type, and on business corruption business the results of Jokowi’s approach are mixed. the different actors with whom firms Indonesia has improved steadily on interact. To achieve these objectives, we Transparency International’s Corruption conducted a survey of Indonesian business Perceptions index, moving up from 137th elites. place in the world in 2005 to 85th in 2019. On other sorts of measures, however, there

2. About the appears to be have been little change. For example, the volume of reports that the KPK Survey receives from the public each year regarding In collaboration with the Indonesian Survey potentially corrupt behaviour has, overall, Institute (Lembaga Survei Indonesia, LSI), remained relatively stable since 2004. we administered a face-to-face survey with Businesses have mixed opinions on the 672 business representatives between July trajectory of corruption too. In the 2017 2019 and February 2020. The sample frame World Economic Forum’s Executive Opinion was designed to reflect the structure of the Survey, companies reported corruption to be Indonesian economy in terms of sectoral the most significant challenge for their contributions, the geographic spread of operations in Indonesia. Yet, in the World economic activity, and the size of the firms Bank’s 2015 Enterprise survey (which draws operating in each sector. The sample, on a larger sample but with a more limited therefore, is representative of companies sectoral reach) just 13 percent of business operating in nine provinces (DKI Jakarta, people stated that corruption was a major East Java, West Java, Central Java, Riau, East

5

Kalimantan, North Sumatra, Banten and their sector. Specifically, we ask respondents

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying South Sumatra) and across the largest (1) how frequently their firm has been asked economic sectors that together contribute to to pay illicit fees or bribes, (2) how approximately 70 percent of Indonesia’s GDP frequently they have paid illicit fees or (agriculture, mining, construction, bribes; and (3) how common it is for firms in manufacturing, financial services, and retail their sector to pay illicit fees or bribes. trade). This level of sectoral representation Overall, some 33.2 percent of firms report is rare. Other business surveys, such as the that they have been asked to pay fees outside World Bank’s Enterprise Survey, exclude the official requirements (i.e., extortion, agriculture and mining, and in doing so leave facilitation, or security money), 30.6 percent out two major arenas of economic activity in report having paid such fees, 35.7 percent Indonesia. believe such illicit fees are commonly paid by businesses in their sector. These figures are The survey combines original business corruption business marginally higher than results reported in questions with some that have been asked the World Bank’s Enterprise Survey from before in previous studies, allowing us to 2015, where 30 percent of Indonesian firms make comparisons over time. In formulating (in the manufacturing, service and retail the questions, we use indirect phrasing sectors) stated they had experienced at least techniques, such that respondents could one bribe payment request. avoid admitting to participating in illegal

activities, which is a common approach when Notably, responses vary significantly by conducting surveys about such sensitive industry. Table 1 breaks down the results topics. For example, our questions often ask according to sector. The highest proportions respondents to offer their observations on of firms reporting being extorted, paying the behavior of companies in their sector bribes, and believing the practice is common generally. These efforts produced a relatively in their sector (columns 1 to 3) are found in strong response rate, and the data offer the extractive industries (47.9, 42.7, 53.1 unique insight into the nature of corrupt percent) and in construction (49.5, 44.2, and exchanges in contemporary Indonesia. 51.6 percent), while the lowest proportion of firms is in the financial sector (17.0, 16.0, 3. Findings and 22.3 percent). It is worth noting that in most cases (trade and logistics being the We first examine firms’ responses regarding exceptions), the perceived incidence of their experience of corruption, and their corruption is higher than the reported perception of the prevalence of corruption in experience of corruption.

6

Table 1 Percentages of Firms Experiencing

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying Corruption by Sector:

Asked for bribe Paid bribe Bribery Pay over Alter present in 2.5 % in financial sector bribes reports (1) (2) (3) (4) (5) Agriculture 28.9 26.7 30.0 7.8 8.9 Extraction 47.9 42.7 53.1 5.2 15.6 Manufacturing 32.7 27.6 36.7 5.1 8.2 Construction 49.5 44.2 51.6 11.6 16.8 Trade 25.0 23.0 25.0 2.0 5.0 Logistics 31.3 30.3 31.3 7.1 12.1 Finance 17.0 16.0 22.3 1.1 2.1 Total 33.2 30.1 35.7 5.7 9.8

To estimate the magnitude of the corruption that such practices were commonplace (9.8 problem, we also asked firms to estimate the percent overall, with a non-response rate of

percentage of their revenues that is taken up 4.0 percent). Again, however, variation business corruption business by illicit fees and costs (column 4). The across sectors is revealing. Responses ranged proportion of missing responses was higher from highs of 16.8 percent in the (7.4 percent) than for the question that construction sector and 15.6 percent in the simply asked for the frequency of firms extractive sector to a low of 2.1 percent in making such payments (3.1 percent).1 The the financial sector.

responses are highly skewed with most firms Next we set out to get a sense of the

responding that they paid nothing in such regulatory burdens and other operational costs. The majority of firms in all sectors challenges experienced by firms in Indonesia. reported paying no unofficial fees or costs, We asked respondents whether they thought but 5.7 percent reported paying more than that the rules and regulations related to their 2.5 percent of their revenue in illicit costs. line of business make their firm’s operations Firms in the construction are again much difficult, whether they thought that the more likely to report paying over 2.5 percent implementation of rules in their sector is of their revenue in bribes (11.6 percent) than consistent, and whether the security firms in other sectors. conditions in their sector are safe (from We next asked respondents whether firms in theft, for example).

their sector ever manipulate financial reports (column 5). Very few companies responded

1 We recoded N/A responses as 0 if respondents answered “never” to the question of how frequently they had made illicit payments (column 2 in Table 1).

1 Table 2. Percentages of Firms Reporting

Regulatory/Security Difficulties by Sector Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying

Regulations Rule implementation Insecure difficult inconsistent

(1) (2) (3)

Agriculture 35.6 30.0 11.1

Extraction 31.3 36.5 9.4

Processing 35.7 32.7 6.1

Construction 43.2 32.6 23.2

Trade 29.0 20.0 7.0

Logistics 30.3 21.2 6.1

Finance 20.2 8.5 2.1

business corruption business Total 32.1 25.9 9.2

We found that 32.1 percent of firms report The results reported here are raw that regulations make their operations percentages. It is possible that additional difficult, 25.9 percent that rules are applied factors such as firms’ size or profitability

inconsistently, and 9.2 percent that security confound these relationships. That is, if firms is a problem for their business (only 3.6, 3.4, in the mining sector are simply more and 0.3 percent of responses were missing). profitable, then it would make sense for Again, however, responses differ corrupt officials to target their activities in substantially across sectors (Table 2). Firms this sector. It is also possible that some local in the construction sector are much more governments are more poorly run than likely than average to report difficulties due others. In the academic paper that to regulation (43.2 percent) and insecurity accompanies this report we run multivariate (23.2 percent). There is less variation in regression analyses that control, not only for perceptions of rule implementation, although additional firm characteristics, but also hold firms in the financial sector are less likely to constant a firm’s region (Kenny & report issues than those in other sectors (8.5 Warburton, 2020). The sectoral differences percent). persist even with these more robust modelling choices. We find that firms in extractive and construction industries are

2

substantially more likely to report being What our survey suggests, however, is that

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying asked for and to admit paying bribes to extractives and construction share another government officials. feature that has so far been overlooked in studies of corruption. The fact that firms in These sectoral differences reflect, in many these two sectors are much more likely to ways, global patterns of corruption. Mining alter their financial reports indicates that, and other resource extraction industries, potentially, hiding firm profits is easier and especially in middle-income and developing more common. Specifically, both the countries, are particularly vulnerable to construction sector and natural resource corrupt practices on the part of firms, industries are characterised by bespoke (as politicians and bureaucrats. Analysts have opposed to standardised) and uncertain long argued that natural monopolies, such as inputs and outputs. This in turns make in petroleum oil extraction or timber logging, manipulating reports and hiding revenue provide opportunities for rent extraction on business corruption business from auditors much more viable. Firms in the part of government agents (Ross, 2001). these sectors can thus ‘make back’ the losses Additionally, all around the world, from incurred through bribery and extortion. developed to developing countries, construction sectors are notoriously corrupt For example, Indonesian construction firms and present state officials with opportunities typically pay “arranger fees” to state officials for kickbacks and bribes (Reuter, 1987). In on a project tendering committee, and to Indonesia, previous research has shown that other individuals that can influence the construction and public infrastructure is tendering process, like local government especially prone to corruption (Aspinall & leaders, parliamentarians and senior Van Klinken, 2011; Olken, 2007). The state members of political parties (Aspinall & Van issues crucial licenses and plays a substantial Klinken, 2011). Then, in order to “menutup regulatory role in these sectors, making them biaya” or “make back” the losses associated more prone to rent-seeking on the part of with those illicit fees, firm managers will state officials as well. Firms also benefit collude to inflate the costs of construction directly from engaging in corruption (access material, services from subcontractors, to licenses and permits, for example) and equipment and the like. Also common is the thus have strong incentives to invest in reporting of “pekerjaan fiktif” or “fictitious corrupt exchanges, especially in developing work”, where a construction company pays countries where autonomous judicial subcontractors for incomplete or oversight is weak (Olken & Pande, 2012). substandard work, or for the procurement of goods that were never used or were used in a

3

smaller volume. An investigator with the KPK conceal their true profits, and recover the

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying noted in an interview with the authors that costs associated with state extortion. monitoring and uncovering cost inflation for

substandard work or incorrect volumes of concrete and other inputs is immensely 4. Investing in difficult. Detecting fictitious work is more feasible, but the practice is so common it is reform almost impossible to police effectively across Firms engage in corruption because paying the sector.2 bribes ‘gets things done’ in the short term.

Similarly, mining inputs are often bespoke. But research shows that the firms which pay Each mineral, oil or coal deposit is different, bribes lose time, suffer greater inefficiencies, such that estimating and monitoring the real grow more slowly, invest less, and their

costs of constructing a mine in a given profits suffer too (Svensson, 2005). And, business corruption business location is immensely difficult, making this especially for large global firms, bribery phase of an extractive project ripe for constitutes a serious legal risk. It follows that financial manipulation (McPherson & firms might be willing to invest in MacSearraigh, 2007). Unreported production institutional remedies that can ease the can then be diverted onto the black market financial burden associated with bribery. beyond the purview of tax collectors. Such Like most Indonesians, business people in illicit production is difficult to uncover. But our survey expressed immense trust toward the Indonesian government estimates, for the KPK as the (once) independent body example, that some 30 to 40 million tons of tasked with investigating and prosecuting unreported coal gets exported from the corruption cases. In our survey, over 85 country’s shores each year (Indonesia percent stated they trust or very much trust Investments, 2014). For these reasons, we this institution. So, we asked respondents suggest extractives and construction are whether they would be willing to increase more systematically targeted by state actors their tax contribution if it meant that for extortion compared to other sectors of demands for informal payments would be the Indonesian economy, in part because eradicated via a program instigated by the firms in these sectors can more easily KPK. Specifically, we gave respondents the following scenario:

2 Interview with former KPK commissioner, 16 August 2020.

4

Imagine that the World Bank and the KPK are

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying implementing a new program that aims to Table 3: Percentages of Firms Willing to stop bribery and extortion between the Increase Tax Burden for Corruption Reform private sector and government. Experts believe the program will succeed. However, to finance the program, the government needs to increase the corporate tax rate. How Sector Willing to pay much additional corporate tax is your for reform company willing to pay if bribery and Agriculture 27.8 extortion can be eliminated?

Extraction 41.7 The non-response rate to this question was 14.9 percent. 32.3 percent of firms reported Processing 24.5

a willingness to pay some additional taxes to business corruption business reduce corruption, while 52.8 were unwilling Construction 39.0 to pay anything. 14.3 percent were willing to pay less than 1 percent more, 7.4 percent Trade 24.0 were willing to pay between 1 and 2.5 Logistics 33.3 percent more, and 10.6 percent were willing

to pay more than 2.5 percent. There is some Finance 36.2 variation by sector. In Table 3, we convert the willingness to pay for reform into a Total 32.3 binary variable, with any value above 0 reclassified as 1. Firms in the extractive (41.7 percent) and construction sectors (39.0 percent) are the most willing to pay some additional taxes towards reform. Notably, however, firms in the financial sector are also more likely to respond positively to this question (36.2 percent) even though they are the least likely to experience extortion in the first place.

5

— in particular extractives — are especially

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying 5. Conclusion prone to predation by state actors because

This report makes two principal firms in these sectors can potentially hide observations about the nature of corruption their true revenues with greater ease. and state-business relations in Indonesia. What do these findings mean for measuring First, our survey shows that in Indonesia the corruption generally, and for anti-corruption intensity of corruption varies from sector to policy in contemporary Indonesia sector. As in many parts of the world, specifically? First, measuring corruption at Indonesia’s construction and extractive the national scale, and making cross-country sectors are especially prone to bribery comparisons based on those state-level requests from state officials. These two observations, might produce misleading sectors alone accounted for almost 20 characterisations of the nature of corruption percent of Indonesia’s GDP, and so the in a country such as Indonesia. Transparency business corruption business leakage of corporate profits has potentially International’s Corruption Perceptions Index broader economic implications for the and the Global Corruption Index by Global Indonesian economy. Risk Profile, for example, allocate states an

Importantly, firms in these sectors were also overall score when it comes to the intensity more likely to manipulate their financial of illicit economic activity. Empirical

reports and hide their true costs and profits. differences between economic sectors are

Qualitative evidence of how corrupt usually lost in these aggregated indexes; yet transactions operate in these sectors of such differences are potentially critical for Indonesia’s economy substantiates our designing appropriate anti-corruption theorized mechanisms. We show how interventions. business in each sector is characterised by Second, our results suggest a role for more bespoke inputs and uncertain outputs, which investment in sector-level monitoring are difficult to measure and monitor. As a agencies and watchdogs. As discussed at the result Indonesian firms in these sectors can opening of this report, the Jokowi and do routinely manipulate and conceal government has, overall, chosen to pivot what they spend and what they produce. away from patrolling and punishing Certainly, companies in other industries corruption, and instead emphasised would engage in similar tactics to try to avoid preventative measures such as cutting red taxation by hiding portions of their profits. tape and improving permitting processes. But construction and primary commodities Such interventions have probably reduced

6

firms’ exposure to bribery and rent-seeking.

Paying bribes in Indonesia:A survey of of survey Indonesia:A bribes in Paying But the government’s approach has also

drawn much criticism. A new law introduced

in late 2019, for example, reduces the investigatory powers of the KPK, and undermines the body’s independence by bringing it under the auspices of a politically appointed supervisory body. This strategy, Jokowi hopes, will relieve investors and firms from the burden of bribery, while avoiding what he views as politically disruptive investigations by the independent watchdog.

Our data, on the other hand, suggest that business corruption business firms would embrace more independent monitoring and enforcement. Indonesian business people trust the KPK, and in sectors most afflicted by corruption, firms are willing to invest in the work of an independent watchdog such as the KPK. Our results suggest, therefore, that Indonesia’s anti- corruption interventions should be designed at the sector level, should be led by an independent KPK, and should engage with the major business actors in each sector.