Typology of Agency Models of Corruption

  • Andrei IvanovEmail author
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-31816-5_2787-1

Synonyms

Definition

The corruption is acts in which public power is used for personal gains in a manner that contravenes the rules of the game (Jain 2001, 73). It is the principal-agent model that is most often used as a methodological framework for modeling corrupt behavior. In the principal-agent model, the payoff to the principal depends on an action taken by the agent. The mainstream of the corrupt behavior modeling is model of bureaucratic corruption, which considers agent as potential bribee and examines its behavior for given regulation.

Introduction

From a great many definitions of corruption, we select the typical one: “The corruption is acts in which public power is used for personal gains in a manner that contravenes the rules of the game” (Jain 2001, 73). The above-cited author suggests that there are two forms of corruption, each occupying extreme positions on a scale of corrupt activities: “grand corruption” generally refers to the acts of the political elite by which they exploit their power to make economic policy and “bureaucratic corruption” refers to corrupt acts of the appointed bureaucrats in their dealings with either their superiors (political elite) or with the public (Jain 2001, 73–75, 2011, 3).

In this article, we will prove that despite of the fact that all corrupts can really be divided on political elite and appointed bureaucrats, this approach is not sufficient to construct the typology of the corruption models. To do so, we will use the approach, which is based on the applying of the principal-agent or agency model to the examination of corrupt behavior (Ivanov 2015).

It is the principal-agent model that is most often used as a methodological framework for modeling corrupt behavior: “Pathologies in the agency/principal relation are at the heart of the corrupt transaction” (Rose-Ackerman 2008, 330).

This model has been developed for describing processes in the private sector and understands the agency relationships as “a contract under which one or more persons (the principal(s)) engage another person (the agent) to perform some service on their behalf which involves delegating some decision making authority to the agent” (Jensen and Meckling 1976, 308). Accordingly, the principal faces the task of shaping a system of incentives for the agent, in which agent’s preference relation, defined on a corresponding set of alternatives, coincides with (or, in extreme case, equal to) preferences of the principal.

In turn, the starting point for modeling public sector processes is the assumption that to meet public needs the political elite (principal) delegates some decision-making authority to government agencies or other public entities or, other words, to appointed bureaucrats (agents). In this case, we can see that all core assumptions of the original principal-agent model are true or can be redefined very close to original ones. Actually, the agent takes an action that determines a payoff to the principal, the principal can readily observe the outcome but not the action of the agent, the agent’s preferences can be assumed to differ from the principal’s ones (asymmetry in preferences), and so on.

However, in contrast to the private sector, the applying of the principal-agent model in the public sector has its own specifics related to the fact that in a democracy the political elite, in turn, is an agent, who is elected by the populace for the achievement of social objectives. Thus, the ideal preferences in this case are not the preferences of political elite but society’s preferences, and we have some reasons to denote the society as a basic principal. Hence, there are three dramatis personae in the model, and this property can be considered as an attribute of public sector principal-agent model.

In the paper, the correspondence of the preferences of the principal and the agent to the basic principal’s preferences lays in the base of suggested typology of corruption models.

The Typology’s Basics

Let us consider some set of alternatives Ω, fix alternatives x and y from this set, and assume that for basic principal (principal, agent) alternatives, x is strictly better than y or indifferent to it: x \( \succeq \) BP y (x \( \succeq \) P y, x \( \succeq \) A y). The set of pair x and y (x, y ∈ Ω), for which basic principal considers x \( \succeq \) BP y, we will call the preference order (or preferences) of basic principal, defined on the set Ω, and denote it as \( \succeq \) BP. The preferences of the principal and the agent (\( \succeq \) P and \( \succeq \) A) are introduced the same way.

Assume that on the same set of alternatives, the preferences of the basic principal, the principal, and the agent are defined.

Definition 1

We call that the principal (agent) is mala fide if its preference order is different from the basic principal’s preference order, \( \succeq \) P\( \succeq \) BP (\( \succeq \) А\( \succeq \) BP), and bona fide if otherwise.

If we construct the typology of models of corrupt behavior or corruption models on the base of the Definition 1, we will find that theoretically five models can be created depending on the combination of the bona/mala fides of principal and agent.

The applying of given approach to the modeling of corrupt behavior is connected with the identification of bona fides of principal and agent. It appears that the first step of assessment is to determine the bona fides of the principal. Indeed, if the principal is bona fide, the vesting of agent with principal’s preference order will inevitably lead to the achievement of public objectives and, otherwise, will not allow of achieving them.

To determine the bona fides of the principal is necessary (Ivanov 2015):
  1. 1.

    To put forward assumptions about the properties of society’s preferences and build up a model of basic principal’s preference order

     
  2. 2.

    Then, based on the proposed regulation, to model the principal preference order

     
  3. 3.

    And, finally, to find out whether they match or differ

     

In the paper, if otherwise stated, we confine discussion to matching given typology of corrupt behavior models with public procurement issues. We will designate policy-makers, legislators, and other people and bodies, who are responsible for the public procurement regulation development and enforcement, as a principal, and contracting authorities (or public buyers), who are acting in the framework of given regulation rules, as an agent.

In the paper, we will use simplified way of the principal’s bona fide identification: we will say that principal is mala fide if the proposed regulation goes against international practice, which is concluded in the clauses of UNCITRAL Model Law on Procurement of Goods, Construction and Services (UNCITRAL Model Law 2011) (hereafter: Model Law) and bona fide if otherwise.

In turn, the agent’s bona fides can be identified on the base of accumulated enforcement practice.

From the five models, main hypotheses of which are given in Table 1 are well-known two ones: the model of bureaucratic (or administrative, according to the World Bank’s terminology) corruption (#2) and the model of efficient corruption (#3). In the next sections, we will consistently consider all models and give the rest of them absent titles.
Table 1

The main directions of corrupt behavior modeling

#

Principal

Agent

1.

Bona fide

\( \succeq \) P = \( \succeq \) BP

Bona fide

\( \succeq \) А = \( \succeq \) BP

2.

Bona fide

\( \succeq \) P = \( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP

3.

Mala fide

\( \succeq \) P\( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP (\( \succeq \) А\( \succeq \) P)

4.

Mala fide

\( \succeq \) P\( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP (\( \succeq \) А = \( \succeq \) P)

5.

Mala fide

\( \succeq \) P\( \succeq \) BP

Bona fide

\( \succeq \) А = \( \succeq \) BP

The Models with Bona Fide Principal

In this section, we assume that the discussed regulation lies in the framework of the clauses of Model Law, and, respectively, the principal is bona fide.

Conflict-Free Model

If the accumulated legal practice does not give us reasons to consider agents as mala fide, we obtain the model that is trivial in terms of the agency relationships (\( \succeq \) P\( \succeq \) А\( \succeq \) BP). Let us call this model the conflict-free one: agent has the opportunity to choose and is prone to selection of the optimal alternative for the society.

When the assumptions for conflict-free model are true, the researchers tend to focus on the study of the efficiency of public contracts, trying to identify the most exact sources of agency costs and assess their value (Laffont and Tirole 1993; Moszoro Spiller 2012).

Model of Bureaucratic Corruption

Assume that the law enforcement practice allows us to identify the existence of agents, who violate the rules and, possibly, policy of the regulation: \( \succeq \) A\( \succeq \) P. They are obviously mala fide: \( \succeq \) A\( \succeq \) P\( \succeq \) BP. Models based on the assumption of principal’s bona fides and agent’s mala fides (\( \succeq \) P\( \succeq \) BP, \( \succeq \) А\( \succeq \) BP) are called models of bureaucratic or administrative corruption.

Models of bureaucratic corruption are most frequently arisen in the studying of public procurement issues. Actually, in this case, the agent is endowed with a discretionary power and a certain budget to carry out procurement. In this situation, two of the three necessary conditions of corrupt behavior arise (Aidt 2003, F633): the relevant public official possesses the authority to design or administer regulations and policies in a discretionary manner, and this discretionary power can allow him the extraction of existing rents or creation of rents that can be extracted.

In the pioneer research, based on the assumptions of principal’s bona fides and agent’s mala fides, Rose-Ackerman examined the situation in which a private individual attempts to corrupt a bureaucrat in order to obtain a government contract (Rose-Ackerman 1975, 187). In this case, the agent is considered as a potential “bribee,” and the actual level of corruption is determined by how well the institutions governing the (corruptible) bureaucracy are designed (Aidt 2003, F635).

Modern studies of bureaucratic corruption develop ideas of Rose-Ackerman’s paper and are usually associated with the modeling agency costs and/or analysis of the specificity of the asymmetry of information between involved parties (Lambert-Mogiliansky et al. 2007).

The question is, is there some room for bureaucratic corruption in the country’s public procurement system? For developing countries, the answer gives the data of Business Environment and Enterprise Performance Survey (BEEPS), a joint initiative of the European Bank for Reconstruction and Development (EBRD) and the World Bank (http://data.worldbank.org/data-catalog/BEEPS).

Let us consider the World Bank data concerning the assessment of countries’ corruption level (http://www.enterprisesurveys.org/data/exploretopics/corruption). The data on Russian public procurement corruption, selected from the surveys BEEPS II–BEEPS V, look like as follows.

Thus, the data of Table 2 give the evidence of bureaucratic corruption existence in Russia and, moreover, let to assess the scale of it. The data dynamics is fit with the fact that intense economic reform (in Russia: beginning 1990–beginning 2000) is objectively accompanied by growth of corruption (Huntington 1968, 59).
Table 2

The assessment of the corruption level in Russian public procurement

Economy

Year

Percent of firms expected to give gifts to secure government contract

All countries

 

26.1

High income: nonOECD

 

10.6

Russian Federation

2002

28.9

Russian Federation

2005

30.1

Russian Federation

2009

47.0

Russian Federation

2012

30.9

The Models with Mala Fide Principal

Bureaucratic corruption model implicitly assumes that the political elite has developed regulatory rules relying solely on the interests of its principal, the society. At the same time, consideration of the political elite as an agent hired by the society naturally leads us to perceive politicians as “…maximizing agents who pursue their own selfish interest rather than as benevolent agents seeking to maximize aggregate welfare” (Grossman and Helpman 1994, 48). Corruption, directly related to activities of the political elite, was above called “grand corruption,” unlike “petty corruption,” which is treated in the bureaucratic model.

Thus, in the model of grand corruption, the principal is mala fide (\( \succeq \) P\( \succeq \) BP). There is another source of principal’s mala fides, his bounded rationality (Simon 1947, xxiv), caused by some specific features of country’s institutional environment.

In the modern information-transparent world, the corruption in the public procurement inevitably reflects in the mass media in the form of the numerous corruption cases and creates the demand on the strong anti-corruption policy, which had been formed in Russia by the beginning of 2000. However, due to lack of experience in the public procurement area and weak institute of Regulatory Impact Assessment, Government and Legislator had chosen tools, which are rarely used in international practice; in particular, the electronic price reverse auction was selected as the main procurement method. This selection is against international experience, since Model Law recommends applying electronic reverse auction (hereafter: ERA) to simple and off-the-shelf goods and standardized services only (Guide 2014, 226). Thus, we consider principal as mala fide (\( \succeq \) P\( \succeq \) BP) in spite of there are no reasons to suspect grand corruption in ERA promotion.

In the assessment of the legal act, involving the use of regulatory tool for which there is certain enforcement practice, improvement of the regulation rules, and, possibly, regulatory policy is heavily dependent on the specific of agent behavior.

If we reject the assumption of principal’s bona fides, assume him mala fide (\( \succeq \) P\( \succeq \) BP), and continue to consider mala fide agent (\( \succeq \) A\( \succeq \) BP), then, depending on whether the agent is prone to break the existing regulation (\( \succeq \) А\( \succeq \) P) or not (\( \succeq \) А\( \succeq \) P), we must distinguish between two types of models.

Model of Efficient Corruption

In the “queue model” and the “auction model” (Aidt 2003, F634), corrupt bureaucrats try to correct preexisting government failures. In these models, agent’s actions violate accepted rules of regulation that allows us to identify differences in preferences of the principal and agent (\( \succeq \) А\( \succeq \) P) and, correspondingly, the agency problem existence.

The models of this type, based on assumptions of mala fides of both a principal and an agent, form the class of “efficient corruption” models (\( \succeq \) P\( \succeq \) BP, \( \succeq \) А\( \succeq \) BP, \( \succeq \) А\( \succeq \) P) (Ibid, F633).

As an example of this kind of corruption, J. Nye viewed corruption of some factory managers in the former Soviet Union, which gave some flexibility to the centralized planning system (Nye 1967, 420), and Laffont and Tirole some instructions of the US Department of Defense (Laffont and Tirole 1993, 476).

It seems that in the case of an efficient corruption, the modeling of agent’s behavior must be primarily aimed at the identification and elimination of the sources of regulation’s inefficiency and, accordingly, to the conversion of efficient corruption into the bureaucratic one. In this case, the modeling is aimed at a changing of both regulatory legal acts and regulatory policy.

Model of Totalitarian Corruption

Nevertheless, the principal can create a system of incentives for the agent, which prevents him from any actions in opposition to existing regulation. This model (\( \succeq \) P\( \succeq \) BP, \( \succeq \) А\( \succeq \) P) we suggest to call a model of totalitarian corruption.

After the theoretical background of totalitarian corruption had been discovered, the case of it was found in the Russian regulation of single-source procurement. Following the Model Law propositions, the modern Russian public procurement law – Federal Law #44 “On the contract system in the procurement of goods, works and services for state and municipal needs,” which came into force from 2014 – permits single-source procurement for small contracts (up to 100,000 rubles, 1 ruble is about US $0.015 on July 2017) but only if the total amount of such purchasing is not more than 2,000,000 rubles.

Thus, the contracting authority that has spent the quota and need to buy something very cheap has to apply open tender (ERA or contest) to do it. And actually, the Russian unified information system in the field of procurement gives many examples of very small purchasing by means of the ERA, including contracts for one ruble! In this case, the principal is obviously mala fide and agent mala fide too, because she has no chance to avoid of open tender providing.

Thus, in the case of a totalitarian corruption, the modeling is aimed at the identifying of what underlies the ineffective regulation: grand corruption (Jain 2001, 73–74) or bounded rationality. It should result in a changing of regulatory policy and practices of regulation, especially in terms of the expansion of discretionary powers and responsibilities of agents.

Quasi-corruption Model

Let us consider the last model from Table 1 (#5), which is based on the assumptions of principal’s mala fides and agent’s bona fides (\( \succeq \) P\( \succeq \) BP, \( \succeq \) А\( \succeq \) BP).

Definition 2

Bona fide agent’s actions violating the rules of regulation created by the mala fide principal will be called quasi-corrupt behavior. The model, which examines bona fide agent’s behavior in institutional conditions created by mala fide principal, will be called quasi-corruption model.

It follows from the Definition 2 that in conditions of quasi-corruption agents have discretionary power broader than in totalitarian case. Therefore, analysis of the applying of this power may enable us to determine the main directions of the changing of regulatory policy and, respectively, regulation rules.

The question is how it is possible to identify fact of quasi-corrupt behavior. Actually, the quasi-corruption model was constructed on the survey of Russian public procurement in 2011.

In the Russian Federation in 2011, the open outcry auctions were replaced completely with ERA for the all types of public procurement bodies. The government was convinced that applying e-auctions would help suppliers to become involved in the procurement process, to ensure reducing corruption, and to hinder the possibility of collusion by suppliers, which would subsequently lead to improved competition in auctions and increase the size of price reductions.

However, the reaction of the system was unpredictable: the competition in auctions had significantly decreased, and, moreover, according to the data of the Federal Antimonopoly Service, at least 60% of e-auctions in that period did not take place due to the receiving of one bid only.

Thus, the sharp decline in competition in auctions and the significant number of failed auctions cannot be explained by anything other than unscrupulous actions by public buyers restricting competition in favor of a preselected “favorite.”

At the same time, the different surveys, including the BEEPS survey, showed that the level of corruption in the Russian Federation, in total, and in public procurement, in particular, was significantly lower (Table 2). One possible hypothesis for explaining identified inconsistencies involves assuming the existence of “quasi-corrupt” behavior of the contracting authority. In this hypothesis, competition may be restricted by both, the mala fide and bona fide public buyers, to avoid of attributive auction risks (Ivanov 2015). The first seeks to obtain bribes (efficient corruption); the second tries to obtain effectiveness in public procurement (quasi-corruption).

Conclusion

As it was mentioned above, A. Jain, trying to develop the typology of corruption models, offers to dispose the cases of corrupt behavior in between bureaucratic corruption and grand corruption – two extreme forms, restricting the scale of corruption activity. We have seen that this linear approach is not quite satisfied for constructing the typology due to, in particular, different forms of models connected with grand corruption.

The introduction of quasi- and totalitarian corruption models allows us to complete the construction of a typology of corrupt behavior models, which is based on the methodology of the agency relationships (Table 3).
Table 3

The typology of agency models of corruption

#

Principal

Agent

Model title

1.

Bona fide

\( \succeq \) P = \( \succeq \) BP

Bona fide

\( \succeq \) А = \( \succeq \) BP

Conflict-free model

2.

Bona fide

\( \succeq \) P = \( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP

Bureaucratic corruption

3.

Mala fide

\( \succeq \) P\( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP (\( \succeq \) А\( \succeq \) P)

Efficient corruption

4.

Mala fide

\( \succeq \) P\( \succeq \) BP

Mala fide

\( \succeq \) А\( \succeq \) BP (\( \succeq \) А = \( \succeq \) P)

Totalitarian corruption

5.

Mala fide

\( \succeq \) P\( \succeq \) BP

Bona fide

\( \succeq \) А = \( \succeq \) BP

Quasi-corruption

In the paper, we have complemented grand corruption as the source of inefficient regulation with the bounded rationality and demonstrated that based on the agents’ reaction, grand corruption (and bounded rationality as well) can bring one of the three corruption forms: efficient, totalitarian, or quasi-corruption.

On the one hand, the introduced typology let us construct a corresponding public policy aimed at the prevention of some type of corruption and substitution of some types of corruption with other ones. On the other hand, the introduced corruption models (quasi-corruption and totalitarian corruption) point out the new area of corrupt behavior examining and give us the chance to forecast the consequences of public policy and regulation improvement in a more detailed way.

Cross-References

References

  1. Aidt T (2003) Economic analysis of corruption: a survey. Econ J 113(November):F632–F652. Royal Economic SocietyCrossRefGoogle Scholar
  2. Grossman G, Helpman E (1994) Protection for sale. Am Econ Rev 84(4. (Sep.)):833–850Google Scholar
  3. Guide to Enactment of the UNCITRAL Model Law on Public Procurement. – UN, N-Y, 2014Google Scholar
  4. Huntington S (1968) Political order in changing societies, 7th printing. Yale University Press, New Haven. 1973Google Scholar
  5. Ivanov A (2015) Quasi-corruption in public procurement: the case of Russian Federation. In: Corruption, economic growth and globalization, Routledge studies in the modern world economy. Routledge, London, pp 112–130Google Scholar
  6. Jain A (2001) Corruption: a review. J Econ Surv 15(1):71–121CrossRefGoogle Scholar
  7. Jain A (2011) Corruption: theory, evidence and policy. J Inst Comp 9(2, Summer)Google Scholar
  8. Jensen M, Meckling W (1976) Theory of the firm: managerial behavior, agency costs, and ownership structure. J Financ Econ 3:305–360CrossRefGoogle Scholar
  9. Laffont J-J, Tirole J (1993) Theory of incentives in procurement and regulation. The MIT Press, Cambridge, MAGoogle Scholar
  10. Lambert-Mogiliansky A, Majumdar M, Radner R (2007) Strategic analysis of petty corruption: entrepreneurs and bureaucrats. J Dev Econ 83:351–367CrossRefGoogle Scholar
  11. Moszoro M, Spiller P (2012) Third-party opportunism and the nature of public contracts, NBER working papers 18636. National Bureau of Economic Research, IncGoogle Scholar
  12. Nye JS (1967) Corruption and political development: a cost-benefit analysis. Am Polit Sci Rev 61(2 (June)):417–427CrossRefGoogle Scholar
  13. Rose-Ackerman S (1975) The economics of corruption. J Public Econ 4:187–203CrossRefGoogle Scholar
  14. Rose-Ackerman S (2008) Corruption and government. Int Peacekeeping 15(3, June):328–343CrossRefGoogle Scholar
  15. Simon H (1947) Administrative behavior. Macmillan, New York. (2nd edn, 1961)Google Scholar
  16. UNCITRAL Model Law on Procurement of Goods, Construction and Services with Guide to En-actment (2011)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Graduate School of ManagementSaint Petersburg UniversitySaint PetersburgRussia