Corruption risk indicators in

public procurement. What we have, should have, and what to do with them

Mihály Fazekas University of Cambridge and

Government Transparency Institute

misi.fazekas@gmail.com

2017.04.12. 1

Using indicators to inform policy

and measure progress on anti-

corruption, DG HOME, 31/3/2017

This project has received funding

from the European Union’s

Horizon 2020 research and

innovation Programme under

grant agreement No 645852

Three main topics

1. State of public procurement data in

Europe & what needs to be done

2. Overview of corruption risk indicators &

why we trust (some of) them

3. Selected conclusions & policy advice

2017.04.12. 2

Why public procurement?

1. A lot of money involved~1/3 of gov’t spend

2. Crucial role in development (e.g. capital accumulation)

3. Indicates the broader quality of institutions

4. Very corrupt (at least perceivet to be...)

5. And of course: LOTS OF DATA

2017.04.12. 3

I. Public procurement data

2017.04.12. 4

The DIGIWHIST data template

• Public procurement data

• Company data: registry, financials, ownership

• Political officeholder data

• Treasury accounts of public organisations

2017.04.12. 5

Where is that data? Minimum threshold for publication supplies and services (EUR), 2015

2017.04.12. 6

2017.04.12. 7

Where is that data?

Covering the full

tender cycle

2017.04.12. 8

Where is that data?

Knowing the

actors

Administrative error: missing information

Average % missing information (13 mandatory fields), 2009-2015, TED data

2017.04.12. 9

Where is that data?

0% 5% 10% 15% 20% 25% 30% 35%

SK

IS

RO

HU

CZ

LV

PL

EE

SI

HR

LT

CY

AT

GR

BG

BE

EU AVG.

IT

DE

DK

UK

NL

ES

PT

FR

LU

CH

FI

SE

II. Proxying corruption

2017.04.12. 10

Corruption definition

In public procurement, the aim of corruption is to steer the contract to the favored bidder without detection. This is done in a number of ways, including:

– Avoiding competition through, e.g., unjustified sole sourcing or direct contract awards.

– Favoring a certain bidder by tailoring specifications, sharing inside information, etc.

See: World Bank Integrity Presidency (2009) Fraud and Corruption. Awareness Handbook, World Bank, Washington DC. pp. 7.

Note the difference from legal definitions

2017.04.12. 11

Conceptualizing public

procurement corruption indicators

April 12, 2017 12

Contracting

body Supplier Contract

Particularistic tie

Tendering Risk Indicators

(TRI)

Supplier Risk

Indicators (SRI)

Contracting Body

Risk Indicators

(CBRI) Political

Connections

Indicators (PCI)

Source: Fazekas, M., & Kocsis, G. (2015). Uncovering High-Level Corruption: Cross-National Corruption Proxies Using Government Contracting Data. GTI-WP/2015:02, Government Transparency Institute, Budapest.

Corruption proxy building approach

1. Clear definition of corruption

2. Dictionary of corruption technologies

3. Statistical modelling of corrupt contracting

4. Indicator validation: triangulation

Corruption risks are only approximated! 2017.04.12. 13

(BAD) Alternative approaches

• Naive summation of expert-suggested red

flags

• Arbitrary cut-points

• Fixation on selected indicators

• Others which might work: PCA, SEM,

machine learning

2017.04.12. 14

Selected examples

1. Single bidding on competitive markets

2. Swings in company market shares when

governments change

3. Supplier anomalies

2017.04.12. 15

2017.04.12.

Modelling corrupt

contracting: single bidding

16

Distribution of contracts according to

the advertisement period

Probability of single bid submitted for contracts

compared with the market norm of 48+ days

Source: EU’s Tenders

Electronic Daily (TED),

Portugal , 2009-2014

Single bidding

Tight deadline

2017.04.12. 17

Macro validity:

Corruption perceptions & single bidding

Single bidding correlates with subjective indicators of corruption

Source: Fazekas, M., & Kocsis, G. (2015). Uncovering High-Level Corruption: Cross-National Corruption Proxies Using Government Contracting Data. GTI-WP/2015:02, Government Transparency Institute, Budapest.

Micro validity:

Number of bidders & prices • Price savings by the number of bidders

• 543,705 contracts, EU27, 2009-2014

2017.04.12. 18

2017.04.12.

19

Surprise success

goes together with

procurement red

flags (CRI)

Politically driven company success: Hungary a paradigmatic case

Companies lose/win

surprisingly when

government changes

Hungary, 2009-2012

2017.04.12.

20

Few companies

lose/win surprisingly

when government

changes

UK, 2009-2012

Surprise success

sometimes goes

together with

procurement red flags

(CRI)

Politically driven company success: UK exception to the rule

Selected supplier risks

• Tax haven registration

• Young companies/company age linked to

government change

2017.04.12. 21

Tax havens & procurement corruption (TRI)

2017.04.12. 22

• Tax havens (Financial Secrecy Index) higher corruption risks (single bidding, Corruption Risk Index)

• EU28, 2009-2014

SRI: Expected success of companies by

age

1. Gradual

build-up of

contracts

2. Natural

fluctuation

over time

2017.04.12. 23

SRI: Observed success of companies at

‚specific’ ages

2017.04.12. 24

1. ‚Just’ founded

companies

2. Companies

founded

under party

last in power

Hungary, 2010

Source: Fazekas, M., Lukács, P. A., & Tóth, I. J. (2015). The Political Economy of Grand Corruption in Public Procurement in the Construction Sector of Hungary. In A. Mungiu-Pippidi (Ed.), Government Favouritism in Europe. The Anticorruption Report 3 (pp. 53–68). Berlin: Barbara Budrich Publishers.

Limitations

• Data, data, data!

• You get what you measure: no general

indicator of corruption!

• Only lower bound estimate: sophisticated

actors can avoid detection (e.g. cartels)

• Variance-driven: corruption is deviation

from a norm

2017.04.12. 25

III. Some conclusions

2017.04.12. 26

Objective proxies are better in tracking trends

• One example: Hungary 2009-2012 – ‚Something has changed’

– WGI CoC reports NO CHANGE (improvement not sign.)

– CRI reports INCREASING RISK

2017.04.12. 27

Objective proxies can surprise you

2017.04.12. 28

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)

Panel-corrected standard errors Municipality-clustered standard errors

Static AR1 LDV Static LDV

New ruling

party

-2.197* -2.075* -2.159* -2.251* -2.191* -2.101* -2.035* -2.181** -1.991** -2.033** -2.110** -2.186* -2.084*

(1.248) (1.253) (1.233) (1.165) (1.189) (1.181) (1.171) (0.933) (0.856) (0.908) (1.010) (1.137) (1.247)

Voteshare,

governing party

0.0713** 0.0808** 0.0790** 0.168*** 0.169*** 0.0735** 0.162*** 0.00574 0.224*** 0.0813 0.147* 0.00474 0.214*

(0.0357) (0.0403) (0.0403) (0.0616) (0.0618) (0.0297) (0.0594) (0.0564) (0.0556) (0.0829) (0.0887) (0.113) (0.109)

Observations 1,870 1,870 1,870 1,870 1,870 1,870 1,870 1,544 1,544 1,901 1,901 1,576 1,576

R-squared 0.276 0.289 0.289 0.293 0.293 0.322 0.342 0.284 0.302 0.283 0.298 0.333 0.350

Number of mc 278 278 278 278 278 278 278 270 270

Year FE NO YES YES YES YES NO YES NO YES NO YES NO YES

Party FE NO NO NO YES YES NO YES NO YES NO YES NO YES

Additional

control(s) Population

Median

income All All All All All

Sweden: new governments decrease corruption risks

Objective proxies can surprise you

2017.04.12. 29

UK: entrenched parties increase prices

Objective

proxies go

beyond the

mere more or

less corruption

2017.04.12. 30

Types of corruption risks in CEE &

Eu Funds’ corruption risk effects

2017.04.12. 31 Source: Fazekas, M., & King, L. (2016) Perils of development funding? The tale of EU Funds and

grand corruption in Central and Eastern Europe. Regulation and governance. Under review.

Pointers for short term action

• Get data up to the mark

– Make sure laws are implemented: data quality

– Publish structured data (csv, etc)

– Make sure variables relevant for corruption

flagging are captured (IDs, final prices, etc)

See: https://opentender.eu/blog/2017-03-towards-more-transparency/

2017.04.12. 32

Key

variables

to capture

in PP

systems

2017.04.12. 33

Variable group Variable Included in the announcement

call for tender contract award contract implementation

Buyer Buyer’s name ● ● ●

Buyer’s department/office ● ● ●

Buyer’s unique ID ● ● ●

Buyer’s address ● ● ●

Buyer’s type ● ● ●

Bidder / bids Bidder’s name ● ●

Bidder’s unique ID/tax ID ● ●

Bidder’s address ● ●

Number of bids submitted ●

Number of bids excluded ●

Bid price (details on total and unit prices) ● ●

Exact time of bid submission ●

Bid type (winner/loser bid) ●

Beneficial owners ● ●

Tender /

contract

Tender unique ID ● ● ●

Procedure type ● ●

Framework agreement (1st/2nd stage) ● ●

Award criteria ● ●

Threshold (below/above EU thresholds?) ● ●

Estimated price (details on total or unit prices) ● ●

Procurement type (service, supply, work) ● ● ●

CPV codes (% contract value per product) ● ● ●

NUTS code(s) of contract implementation ● ● ●

Status (cancelled, pending, etc.) ● ● ●

Dates Call for tender publication date ● ● ●

Bid submission deadline ●

Contract start and end dates ● ● ●

Publication date of contract award ●

Contract signature date ●

Publication date of contract completion ●

Subcontracting Subcontractor’s name and unique ID (tax ID) ● ●

Subcontractor’s share ● ●

Consortium Consortium members’ name and unique ID (tax ID) ● ●

Consortium members’ share ● ●

Contract

performance

Contract performance end date ●

Was performance according to the contract ●

Explanation in case of deferring from contract ●

Information on contract modification ●

Information on performance quality ●

Pointers for short term action

• Get data up to the mark – Make sure laws are implemented: data quality

– Publish structured data (csv, etc)

– Make sure variables relevant for corruption flagging are captured (IDs, final prices, etc)

– Link data (e.g. procurement to company)

• Introduce real-time monitoring and regular data use data is immediately actionable!

• Learn how to use indicators carefully

See: https://opentender.eu/blog/2017-03-towards-more-transparency/

2017.04.12. 34

Differences in the average number of bids submitted between non-restricted

and restricted procedures, OECD-Europe, 2013, c.value>58k eur

2017.04.12. 35

Beware of the context use of restricted procedures

2017.04.12. 36

Single bidder

ratio, TED, EU,

2009-2014

Monitoring can invalidate indicators Single bidding & organised criminality

For more information check out

digiwhist.eu/resources

2017.04.12. 37

Furher readings

2017.04.12. 38

Charron, N., Dahlström, C., Fazekas, M., & Lapuente, V. (2017). Careers, Connections, and Corruption Risks: Investigating the impact of bureaucratic meritocracy on public procurement processes. Journal of Politics, 79(1), p. 89–103.

Fazekas, M. & Cingolani, L. (2017), Breaking the cycle? How (not) to use political finance regulations to counter public procurement corruption. Slavonic & East European Review, 95(1)

Fazekas, M. & Tóth, B. (2017), Infrastructure for whom? Corruption risks in infrastructure provision across Europe. In Hammerschmid, G, Kostka, G. & Wegrich, K. (Eds.), The Governance Report 2016 . Oxford University Press, ch 11.

Fazekas, M., & Tóth, I. J. (2017). Corruption in EU Funds? Europe-wide evidence on the corruption effect of EU-funded public contracting. In J. Bachtler, P. Berkowitz, S. Hardy, & T. Muravska (Eds.), EU Cohesion Policy. Reassessing performance and direction. London: Routledge, ch. 13.

Rasmus Broms, Carl Dahlström and Mihaly Fazekas (2017). Entrenched parties and control of public procurement in Sweden. University of Gothenburg-Quality of Government Institute, manuscript

Fazekas, M. and Tóth, I. J. (2016). From corruption to state capture: A new analytical framework with empirical applications from Hungary. Political Research Quarterly, 69(2), p. 320-334

Fazekas, M., Cingolani, L., & Tóth, B. (2016). A comprehensive review of objective corruption proxies in public procurement: risky actors, transactions, and vehicles of rent extraction: GTI-WP/2016:03. Government Transparency Institute. Budapest.

Fazekas, M., Tóth, I. J., & King, L. P. (2016). Anatomy of grand corruption: A composite corruption risk index based on objective data. Eu. Journal of Criminal Policy and Research, 22(3), 369–397.

Fazekas, M. (2015). The Cost of One-Party Councils: Lack of Electoral Accountability and Public Procurement Corruption. London: Electoral Reform Society.

Fazekas, M., & Kocsis, G. (2015). Uncovering High-Level Corruption: Cross-National Corruption Proxies Using Government Contracting Data. GTI-WP/2015:02, Government Transparency Institute, Budapest.

Fazekas, M., Lukács, P. A., & Tóth, I. J. (2015). The Political Economy of Grand Corruption in Public Procurement in the Construction Sector of Hungary. In A. Mungiu-Pippidi (Ed.), Government Favouritism in Europe The Anticorruption Report 3 (pp. 53–68). Berlin: Barbara Budrich Publishers.

Fazekas, M. (2015). The Cost of One-Party Councils: Lack of Electoral Accountability and Public Procurement Corruption. London: Electoral Reform Society.