This site has moved. My new homepage is chennoshen239.github.io - no longer updated here.
Default with Policy-Randomness Overestimation [PDF]
Abstract: Why do some sovereigns pay high spreads despite moderate debt? I build a quantitative default model where lenders overweight the dispersion of policy realizations (policy‑randomness overestimation, PRO). Within a single pricing operator, PRO pivots price/spread schedules around a state‑contingent threshold: the sovereign deleverages while defaulting at higher debt, raising average spreads and producing a "stability illusion" (lower volatility with higher premia). A rational‑inattention microfoundation, based on a mental‑model‑plus‑residual representation, endogenizes PRO level. It highlights two empirically salient channels (Argentina): (i) a credibility collapse that shifts attention away from fundamentals toward diagnosing instability, raising the effective temperature; and (ii) a sharp instability signal that increases perceived residual uncertainty and lifts the effective temperature on impact. Policy and information extensions show limited power of fiscal transfers, persistence from negativity‑biased learning, and welfare gains from transparency.