Bid-rigging causes a great harm to economy and society. The literature on the topic mainly aims at quantifying the economic damages caused by cartel agreements. Instead, there is poor research on how firms interact among each other to rig tenders. This is quite surprising, considering the co-offending nature of bid-rigging. Building on a judicial case study in the Italian public procurement, this work aims to assess which bidding behaviours are associated with bid-rigging. Data comprise 1,242 companies, 112 of which colluding, participating in 357 Italian roadworks average bid auctions awarded between 1999 and 2002. Using regression and social network analysis techniques, this study empirically tests the ability of different company-level measures in predicting whether companies are rigging public tenders. Among them, it explores the role of embeddedness and brokerage power in explaining cartel membership. The results of this study show that cartels exploit legal opportunities to collude (e.g., using temporary consortia), while diversifying their strategies: sometimes they massively bid in auctions, while in other cases they prefer to submit few but stronger bids to fight competition of non-colluding companies.
La collusione negli appalti procura un grave danno all'economia e alla società. La letteratura sull'argomento mira principalmente a quantificare i danni economici provocati dai cartelli negli appalti pubblici. Pochi studi hanno invece cercato di analizzare come le aziende del cartello interagiscono tra loro in questo settore per capire quali tecniche utilizzano per vincere gli appalti. Basandosi su un caso di studio giudiziario negli appalti pubblici italiani, questo lavoro ha lo scopo di determinare quali strategie collusive sono associate alla collusione. Il dataset comprende 1.242 aziende, 112 delle quali colluse, che partecipano a 357 aste di lavori aggiudicate tra il 1999 e il 2002. Utilizzando tecniche di regressione e analisi di rete, questo studio verifica empiricamente la capacità di diverse misure a livello di azienda nell’identificare imprese colluse. I risultati di questo studio mostrano che i cartelli sfruttano alcune opportunità insite nel contesto in cui operano per aumentare la loro probabilità di vincere (ad esempio, formando associazioni temporanee di imprese). I cartelli alternano le loro strategie collusive a seconda dell’asta: in alcuni casi, partecipano tutte insieme, mentre in altri casi preferiscono presentare offerte poche ma più forti per scoraggiare la partecipazione di imprese non colluse.
BIDDING BEHAVIOURS ASSOCIATED WITH BID-RIGGING IN THE ITALIAN PUBLIC PROCUREMENT
Carbone, Carlotta
2020
Abstract
Bid-rigging causes a great harm to economy and society. The literature on the topic mainly aims at quantifying the economic damages caused by cartel agreements. Instead, there is poor research on how firms interact among each other to rig tenders. This is quite surprising, considering the co-offending nature of bid-rigging. Building on a judicial case study in the Italian public procurement, this work aims to assess which bidding behaviours are associated with bid-rigging. Data comprise 1,242 companies, 112 of which colluding, participating in 357 Italian roadworks average bid auctions awarded between 1999 and 2002. Using regression and social network analysis techniques, this study empirically tests the ability of different company-level measures in predicting whether companies are rigging public tenders. Among them, it explores the role of embeddedness and brokerage power in explaining cartel membership. The results of this study show that cartels exploit legal opportunities to collude (e.g., using temporary consortia), while diversifying their strategies: sometimes they massively bid in auctions, while in other cases they prefer to submit few but stronger bids to fight competition of non-colluding companies.File | Dimensione | Formato | |
---|---|---|---|
tesiphd_completa_Carbone.pdf
accesso solo da BNCF e BNCR
Dimensione
4.81 MB
Formato
Adobe PDF
|
4.81 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/159866
URN:NBN:IT:UNICATT-159866