Peiyi Jin

Welcome to my website!

I'm a Ph.D. candidate in the Department of Economics at National University of Singapore (NUS), currently on the 2025/26 academic job market. I'm supervised by Prof. Sumit Agarwal.

Research Interests: Fintech, Digital Economy, and Household Finance.

Working Papers

I. Blockchain

Abstract:
This paper investigates how tax-planning strategies affect market liquidity and credit risks in Decentralized Finance (DeFi) lending. Using an exogenous tax shock on cryptocurrency gains and millions of transactions, we show that tax-motivated borrowing to defer capital gains taxes significantly reduces liquidity—particularly among stablecoin borrowers, high loan-to-value borrowers, and those holding assets long-term. Instrumental variable estimates indicate that tax-induced illiquidity more than doubles defaulted loan values. Robustness checks confirm these effects among highly tax-sensitive borrowers. The findings highlight implications for market stability, tax revenue forecasts, and the regulation of digital asset taxation.

Abstract:
This paper investigates whether cryptocurrencies have become a new conduit for laundering diverted foreign aid. Using World Bank aid disbursement data from 2018 to 2024, linked with forensically tagged on-chain Bitcoin transactions and off-chain exchange activity, we document systematic surges in crypto transactions for anonymous wallets after disbursements, especially on exchanges located in tax-haven jurisdictions. A one-standard-deviation increase in lagged aid is associated with a 0.51 log-point rise in anonymous transactions on tax-haven exchanges—approximately a 66% increase—concentrated in newly created wallets and fading within two quarters. Network analysis reveals a real-time laundering pattern: funds flow through regulated platforms, then through mixers and tax-haven exchanges, mirroring the classic placement, layering, and integration stages. Off-chain data confirm spikes in transactions on suspect, lightly regulated platforms. To address endogeneity in aid allocation, we use an IV strategy based on historical aid shares interacted with governance quality. Overall, our findings suggest that cryptocurrencies are increasingly used for offshore banking in concealing aid diversion. Our study shows how blockchain forensics can trace hidden financial flows and offers new tools for anti-corruption and crypto regulation.

II. Fintech Lending

Abstract:
This paper studies how privacy regulation reshapes credit market equilibria by altering the balance between ex-ante screening and ex-post monitoring. Exploiting Google’s 2019 ban on FinTech lenders’ access to Indian borrowers’ call detail records (CDR) in a regression discontinuity design, we show that loan applications rose by 4%, consistent with borrowers valuing privacy. At the same time, approval rates fell by 25% with no change in default risk, as lenders substituted away from CDR-based monitoring toward stricter upfront screening. The resulting exclusion was concentrated among new-to-credit, lower-income, younger, and socially disadvantaged borrowers. Tracking rejected applicants in credit bureau data, we find their probability of accessing formal credit fell by 4–5 percentage points even four years later. For lenders, lost profits amounted to roughly 30% of potential earnings. Together, these findings reveal a central trade-off of digital privacy regulation: while enhancing consumer welfare on the demand side, it can unintentionally deepen financial exclusion and reduce lender profitability on the supply side.

Abstract:
This paper examines the impact of credit data sharing among competitive banks of different sizes in open banking. Analyzing data from three predecessors of Bank of America, we find that information sharing enhances predictive capabilities and increases market lending profit as the network expands. Banks that share loans spanning a wider range of collateral drive most of the predictive gains. However, competition creates unequal benefits, with smaller banks gaining while the largest bank loses borrowers and profits. These results underline the importance of effective bargaining for cooperative sharing. We also explore the Nash equilibrium for optimal outcomes in a competitive lending market.