Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (4 results)

  • Thesis
  • Peer Reviewed
UC Berkeley Electronic Theses and Dissertations (2024)

Public procurement, contributing to total GDP, not only impacts economic growth and development by shaping market dynamics and competition but also plays a key role in promoting fairness and mitigating corruption within public sector transactions. This dissertation dives into public procurement, focusing on the challenges of corruption, its detection, and its impact on businesses and the economy. In Chapter 1, I reveal the prevalence of corruption in scoring auctions in public procurement in China, design a model-based tool to detect this corruption, and discuss policies to reduce it. In Chapter 2, inspired by complaint data, I propose another method to detect corruption in close games, which complements the first chapter, and I also discuss how to better design a complaint system to curb corruption. In Chapter 3, I link public procurement data to universal firm-to-firm transaction data to study the direct and indirect effects of participating in the public procurement supply chain through the propagation of production networks. Together, these chapters provide a comprehensive understanding of public procurement's role in shaping economic outcomes and propose targeted strategies for reform.

My first chapter proposes a method to screen out scoring rule manipulation corruption in scoring auctions in public procurement and discusses the policies to reduce corruption and increase transparency. I start the chapter by documenting that corruption is widespread in scoring auctions. Procurement officers can collaborate with firms to manipulate scoring rules, favoring predetermined winners, while corrupt firms orchestrate non-competitive bids from others to meet minimum bidder requirements. Drawing from extensive data on public procurement auctions in China, I introduce a model-driven statistical tool to detect this specific form of corruption. The findings indicate a corruption rate of approximately 65%. A procurement expert evaluation audit study confirms the test's validity, revealing a 91% probability that experts identify suspicious scoring rules when the test signals potential corruption. I also link procurement data to comprehensive firm data to examine the distortions caused by corruption. I find that local and state-owned firms, as well as less productive ones, are more favored in corrupt auctions. Lastly, I explore policy implications from the anti-corruption campaign, as well as the counterfactuals by estimating a structural model, concluding that general corruption investigations may be insufficient to address deeply ingrained corrupt practices in the long run whereas implementing anonymous call-for-tender file evaluation could significantly improve social welfare.

My second chapter complements the first, which focuses on the unreasonably large score gaps between winners and losers. It examines the ex-post complaint dataset, proposes a method to detect corruption in close-game cases in scoring auctions, and discusses how to better design the complaint system. The complaint system enables reporting of potential corruption and collusion in public procurement auctions, offering insights not visible to outsiders and facilitating corruption detection. In this chapter, I have gathered a dataset of complaints from China's public procurement system. Based on the patterns observed in the ex-post complaints, where the price bids of winners are much higher than those of the complainants, I applied the Regression Discontinuity Design (RDD) to detect corruption. My findings indicate that, contrary to competitive cases where winners and losers are chosen at random, in complaints, winners tended to submit prices that were, on average, 5% higher than those of the losing complainants. This suggests that at least 20% of the auctions in the complaint dataset were corrupt. When extending this methodology to the entire public procurement auction dataset, it appeared that 13% of the auctions in close-game scenarios were corrupt. To explore the low rate of complaints, I developed a model to investigate the decision-making process behind the lodging of complaints, with a specific focus on those bidders who lost by a narrow margin, and conducted a counterfactual analysis. I found that protecting whistleblowers by concealing their names can increase the reporting rate, and that random auditing by the financial team does not crowd out the functionality of the complaint system.

In my third chapter, coauthored with Ming Li and Wei Lin, we study the direct and indirect effects of participating in the public procurement supply chain through the propagation of production networks, utilizing tax data that tracks firm-to-firm transactions in China. To quantify the effects of directly winning public procurement contracts on firms, we employ an event study design. Our estimates reveal that firms winning public procurement contracts experience increased purchasing activities and gain additional non-public procurement contracts in the following months without crowding out effects. Moreover, these spillover effects in the non-public sectors originate from competitive procurement projects, rather than potentially corrupt ones. We then use a model-based method to measure the total ratio of public procurement contracts to sales through both direct and indirect channels, using the complete firm production network. We find that although only 0.5% of firms directly participate in the public procurement supply chain, a greater number are involved indirectly. Using these total ratios, we explore the effect of public procurement on firm revenue through network propagation. Without considering the indirect channels through the production network, we would underestimate the role of public procurement demand.

    Cover page: Essays on Public Procurement and Firms in China
    • Article
    • Peer Reviewed
    LBL Publications (2023)
    Entry into parenthood is a major disruptive event to travel behavior, and gender gaps in mobility choices are often widened during parenthood. The exact timing of gender gap formation and their long-term effects on different subpopulations are less studied in the literature. Leveraging a longitudinal dataset from the 2018 WholeTraveler Study, this paper examines the effects of parenthood on a diverse set of short- to long-term outcomes related to the three hierarchical domains of mobility biography: mode choice, vehicle ownership, spatial mobility, and career decisions. The progress of the effects is evaluated over a sequential set of parenting stages and differentiated across three subpopulations. We find that individuals classified as “Have-it-alls”, who start their careers, partner up, and have children concurrently and early, significantly increase their car uses two years prior to childbirth (“nesting period”), and they then relocate to less transit-accessible areas and consequently reduce their reliance on public transportation while they have children in the household. In contrast, individuals categorized as “Couples”, who start careers and partnerships early but delay parenthood, and “Singles”, who postpone partnership and parenthood, have less pronounced changes in travel behavior throughout the parenting stages. The cohort-level effects are found to be driven primarily by women, whose career development is on average more negatively impacted by parenting events than men, regardless of their life course trajectory. Early career decisions made by women upon entering parenthood contribute to gender gaps in mid- to longer-term mobility decisions, signifying the importance of early intervention.
      Creative Commons 'BY' version 4.0 license
      • Article
      • Peer Reviewed
      LBL Publications (2023)
      What makes you hold on to that old car? While the vast majority of household vehicles are still powered by conventional internal combustion engines, the progress of adopting emerging vehicle technologies will critically depend on how soon the existing vehicles are transacted out of the household fleet. Leveraging a nationally representative longitudinal data set, the Panel Study of Income Dynamics, this study examines how household decisions to dispose of or replace a given vehicle are: 1) influenced by the vehicle’s attributes, 2) mediated by households’ concurrent socio-demographic and economic attributes, and 3) triggered by key life cycle events. Coupled with a newly developed machine learning interpretation tool, TreeExplainer, we demonstrate an innovative use of machine learning models to augment traditional logit modeling to both generate behavioral insights and improve model performance. We find the two gradient-boosting-based methods, CatBoost and LightGBM, are the best performing machine learning models for this problem. The multinomial logistic model can achieve similar performance levels after its model specification is informed by TreeExplainer. Both machine learning and multinomial logit models suggest that while older vehicles are more likely to be disposed of or replaced than newer ones, such probability decreases as the vehicles serve the family longer. Pickup trucks and sport utility vehicles are less likely to be disposed of or replaced than cars, and leased vehicles are more likely to be transacted than owned vehicles. We find that married families, families with higher education levels, homeowners, and older families tend to keep their vehicles longer. Life events such as childbirth, residential relocation, and change of household composition and income are found to increase vehicle disposal and/or replacement. We provide additional insights on the timing of vehicle replacement or disposal, in particular, the presence of children and childbirth events are more strongly associated with vehicle replacement among younger parents.
        Creative Commons 'BY' version 4.0 license
        • Article
        • Peer Reviewed
        LBL Publications (2024)
        This paper documents the design and application of ATLAS (Automobile and Technology Lifecycle-Based ASsignment), a comprehensive household vehicle transaction and technology adoption micro-simulator in the San Francisco Bay Area. ATLAS evolves the fleet mix of individual households by simulating the vehicle transaction and choice decisions in response to co-evolving demographics, land use, and vehicle technology simulations. While most existing literature has focused on the aggregate clean vehicle uptake, this paper differentiates distributional effects and decomposes the underlying mechanisms across heterogeneous sub-populations of households. Using scenarios and sensitivity simulations that vary vehicle technology and policy assumptions, we find that Zero Emission Vehicles (ZEVs) penetrate into higher income groups at a faster rate than into lower income groups, which is intuitive and aligns with expectations. Interestingly, the relative income disparity in ZEV ownership shrinks over time across all scenarios, with a ZEV mandate coupled with declining battery cost leading to the greatest reduction in disparity of ZEV ownership by 2050. Federal, state, and local financial incentives influence the redistribution of ZEV uptake across income groups and contribute to narrowing income disparity. Vehicle transaction frequency and new versus used market dynamics are found to be important factors contributing to the income disparity.
          Cover page: Technology progress and clean vehicle policies on fleet turnover and equity: insights from household vehicle fleet micro-simulations with ATLASCreative Commons 'BY' version 4.0 license
          Top