Xinhao Dong

Using quarterly data for Chinese banks from 2008 to 2024, I compare the relationship between banks’ pre-reform deposit mix, reserve requirements, and market funding across pricing regimes. Among joint-stock banks, the fitted funding gradient falls from 1.144 to −0.024 percentage points per percentage-point increase in the reserve ratio. Limited common support and sensitivity to exposure trends prevent a causal interpretation of that regime difference.

We distinguish the urgency of spending needs from the liquidity services of assets using Treasury bills, commercial paper, and AAA corporate bonds. Our headline estimate puts the annual price of convenience at 5.5%, far above conventional spread measures; only about 23% of it is attributable to urgency.

Credit-card lines are commitments authorized before households use them. In U.S. large-bank data, monetary tightening lowers initial limits on new accounts relative to limits on open accounts, while total commitments and the number of open accounts do not decline. New authorizations are a small share of committed credit, so this margin changes the stock slowly. A household financing framework shows why removing credit at origination can reduce spending more than withdrawing the same amount from existing lines.

Using CHIP 2013 and 2018 data on 2,239 land-expropriated households in 15 provinces, I study how reported compensation form varies across provinces and expropriation cohorts. Before 2008, moving from the 25th to the 75th percentile of provincial fiscal pressure was associated with a 20 percentage-point lower probability of any continuing protection and a 17–19 percentage-point higher probability of cash only. These relationships are markedly weaker from 2008 onward, although imprecisely estimated. Across twelve consistently observed provinces, gaps in compensation form narrowed by about 38 percent.

Using age-60 eligibility around China’s 2016 pension reform, we find that pension access raises grandchild care by 8.2 percentage points. Grandchildren then spend fewer months with their parents, while complete paternal absence falls and shorter, more returnable work trips become more common. Remittances from pension recipients to adult children also decline, consistent with a shift in how families provide informal support.

We model costly traffic obfuscation and an adaptive censor’s blocking decision jointly. Exact unilateral best responses determine which obfuscation–blocking targets can survive repeated interaction; a common shortcut that checks only complete abandonment of obfuscation can misclassify whether a target is sustainable. A public-signal extension shows how missed detections and false alarms tighten these conditions.