You are here: Home / Latest Research / Blogs and Articles / From corruption to state capture: a new analytical framework

From corruption to state capture: a new analytical framework

A recent paper published by the Corruption Research Center Budapest offers an exciting new take on how to analyse state capture; a term which is widely debated and used, yet has so far largely escaped a precise analytical definition and measurement.

The debate around state capture has so far centered on theoretical contributions while empirical studies were scarce and either relied on qualitative data lacking sufficient breath, or survey data lacking sufficient reliability. With the availability of reliable micro-level data on institutionalised grand corruption in public procurement (e.g. Fazekas, Tóth, and King, 2013a), scholars can begin to rigorously test theories of state capture and investigate the underlying actor networks and corrupt transactions. This blog describes the development of this analytical framework for the case of Hungary.

It is our starting point that state capture cannot be defined as the mere existence of widespread corruption in a given country. Rather, its essence lies in a distinct network structure in which corrupt actors cluster around certain state organs and functions. By analysing the distribution of corrupt transactions and clustering of high corruption risk actors, we can establish the degree of state capture. For example, it becomes possible to distinguish between localized and full-scale capture: while in the first case only some public and private organisations enter into a capture relationship which remains relatively separated from the rest of the government, in the second case captured organisations are linked to each other and a national-level elite controls them.

In order to develop a new analytical framework for gauging state capture based on micro-level contractual networks in public procurement and to demonstrate its usefulness, the following steps were taken: First, establish a robust measure of corruption risks in public procurement transactions focusing on binary relationships between issuers and suppliers; Second, construct a contractual network of organisations to demonstrate the non-random distribution of corruption risks; Third, systematically explore the impact of changing elite group internal composition (from decentralisation to centralisation following the landslide victory of conservative party Fidesz in 2010) on the network structure of rent extraction in public procurement in Hungary.

A robust measure of corruption risks

Corruption Research Center Budapest’s Corruption Risk Index (CRI) measures the probability that the principle of open access is violated in the process of awarding and performing public procurement contracts in order to serve corrupt rent extraction by a select few (Fazekas, Tóth, and King 2013a). In other words, it expresses the probability of issuers pretending that tenders are competitive as prescribed by law while restricting competition to award contracts to a well-connected bidder on a recurrent basis. CRI is a composite index of elementary corruption risk indicators capturing ‘corruption techniques’ such as tailoring eligibility criteria to fit a single company or using exceptional procedure types to limit openness of competition (Fazekas, Tóth, and King 2013b). It reflects a corrupt rent extraction logic where elementary corruption techniques are systematically used for restricting access and recurrently benefiting the same winner (for example leaving only a few days for bidders to prepare and submit their bids, making it more likely that only one company will participate in the tender – a company which is well-connected and got the information on the tender long before it was published). The composite indicator is constructed by linking public procurement process ‘red flags’ to restrictions of market access. The resulting CRI can take any value between 0 and 1, where 0 means minimal or no corruption risk and 1 means maximal corruption risk observed.

The contractual network of organisations

Using public procurement data, two types of actors were selected for network analysis: issuers and winners. Network analysis focusing on organisational networks is capable of capturing the most relevant means and structure of high-level rent extraction and corruption. In order to analyse the changing elite configuration resulting from the change of government in May 2010, two roughly equal time periods were selected roughly one and a half years before and after the government change.

The figures below depict the results, i.e. the total contractual networks of both periods, with red dots representing suppliers and green dots denoting issuers. Dot size reflects the number of contracts an organisation awarded or received, while the location of the dot on the graphs reflect how central the organisation is in the overall network. The width of the ties linking each organisation reflects corruption risks (CRI) of the contracts between those organisations. The large green circles in each figure highlight some clusters which are identifiable by visual observation with several sub-centres in 2009-2010 and one main centre in 2011-2012. It is apparent after a brief visual observation that the 2009-2010 period had a number of sub-centres, while such formations are much less common in 2011-2012. This changing structure may indicate that state capture has become more centralised with the change of government in 2010.

 Total contractual network, Hungary

Total contractual network, Hungary, 2009M1-2010M4Total contractual network, Hungary, 2011M1-2012M7
                                      2009M1-2010M4                                                                        2011M1-2012M7

Change in elite group – change in public procurement network structure

In order to identify captured, partially captured, and clean organisations, we clustered supplier and issuer organisations based on 1) how corrupt they are on average in their contracting activities (average CRI) and 2) how consistent they are in their corruption throughout different contracts (standard deviation of CRI). While theory suggested clean, fully captured and mixed clusters, the clustering algorithms  instead revealed four groups:  1) clean organisations: low average corruption with low variability of performance; 2) occasionally corrupt organisations: low average corruption with highly variable performance indicating that there are occasional deviations from the low corruption standard contracting practice; 3) partial capture: high average corruption with high variability indicating that there are still low corruption contracts, which nevertheless represent the deviation from a high corruption norm; 4) full capture: high average corruption with low variability of performance indicating that corrupt exchanges represent the norm in the organisation’s contracting practice.

In order to gauge the degree of state capture and to understand the nature of capture (i.e. localized vs full-scale capture), we removed all the non-captured organisations from the contractual network and checked to what degree fully/partially captured organisations form a coherent network.  Interestingly, captured organisations form a highly-connected network in both periods with a few isolated single nods and smaller networks of 2-5 dots. Such a highly networked nature of political corruption in Hungary makes any administrative fixes to corruption likely to fail: it seems highly impossible to break up the circles of corruption by removing a small set of corrupt organisations – something which is typically the result of investigations and targeted bureaucratic reforms (for example of how state-owned enterprises are regulated). Instead a big-bang approach to anticorruption that is simultaneously changing a wide array of regulations and replacing most key personnel is a more realistic strategy in terms of impact. In addition, the difference in the networks of the two periods points out that a connected network of sub-centres in 2009-2010 transformed into one centralised network. This change in network structure goes hand in hand with the increased centralisation of the governing party (Fidesz) suggesting that elite group configuration does influence the structure of corrupt rent extraction as well as state capture.

Contractual network of partially and fully captured organisations, Hungary

Contractual network of partially and fully captured organisations, Hungary, 2009M1-2010M4Contractual network of partially and fully captured organisations, Hungary, 2011M1-2012M7

                                        2009M1-2010M4                                                                   2011M1-2012M7

Red: partial capture of organisations; Black: full capture of organisations

With a lot more data collection starting in March with a project called DIGIWHIST covering 35 European countries from Norway to Georgia, many countries at risk of state capture will be compared and potential reform avenues identified.

 * Mihály Fazekas is the co-founder of the Corruption Research Center Budapest and ANTICORRP researcher. Ágnes Czibik is a researcher at the Corruption Research Center Budapest and ANTICORRP researcher.

References

M. Fazekas, I.J. Tóth, „From corruption to state capture: A new analytical framework with empirical applications from Hungary”, Working Paper Series: CRCB-WP/2014:01 Available at: http://www.crcb.eu/?p=718

M. Fazekas, I. J. Tóth, and L. P. King, 2013a “Anatomy of Grand Corruption: A Composite Corruption Risk Index Based on Objective Data”, CRC-WP/2013:02, Available at: http://www.crcb.eu/?p=274

M. Fazekas, I. J. Tóth, and L. P. King, 2013b. Corruption Manual for Beginners: Inventory of Elementary “corruption Techniques” in Public Procurement Using the Case of Hungary. CRC-WP/2013:01, Corruption Research Centre, Budapest. Available at: http://www.crcb.eu/?p=269

Author : By Mihály Fazekas and Ágnes Czibik

20 Jan 2015


Bookmark and Share

Document Actions

Filed under:
Add comment

You can add a comment by filling out the form below. Plain text formatting. Comments are moderated.

Our partner