Zirui Wang

Mortgage brokers promise to shop across lenders, yet rates on brokered loans are as dispersed as those on retail loans. Using novel U.S. mortgage data linking borrowers, brokers, and lenders, I study why similar borrowers in the same local market on the same day pay widely different rates despite using mortgage brokers. Broker firms exhibit persistent differences in the rates their borrowers pay. An IV design based on borrowers’ relative proximity to broker branches estimates that using a high-cost broker raises rate spreads by 10–12 basis points. Lender choice helps explain these differences: some firms systematically choose lenders that are relatively expensive and pay brokers more, consistent with steering. Steering practices differ across firms and persist within firms: borrowers who use firms with stronger past steering practices pay higher rates. Overall, the findings show how conflicts of interest can undermine the benefits of delegated search and raise household borrowing costs.

We evaluate two major types of crypto enforcement actions: policies targeting services and policies targeting users. First, sanctions against the Tornado Cash mixer resulted in an 80.4% decline in hacker illicit inflows and shifted their usage toward more traceable swaps and bridges. Second, overseas exchange acceptance of illicit funds declines only after the U.S. DOJ settlements. Third, OFAC sanctions against individuals are relatively rare and often after the funds have left the wallets. Fourth, after Tether freezes, criminals continue to use related wallets. Overall, anti-money laundering policies targeting services like mixers and exchanges appear more effective than policies targeting users.

Using novel property-level operating statements for multifamily properties, we document a large, widespread rise in insurance costs and study who bears them along three margins. Across locations, costs rise most where disaster risk, reinsurance exposure, and homeowners-insurance regulatory frictions are greatest. Between tenants and owners, rent increases initially passed nearly all the cost to tenants, but this response has weakened sharply, leaving owners to absorb larger, recent shocks through lower profitability and transaction prices. Across owners, larger and lower-portfolio-risk owners obtain favorable insurance pricing, and high-risk assets reallocate toward these owners. Repricing therefore shapes cash flows, asset values, and ownership.