Measuring the Valuation of Liquidity with Penalized Withdrawals

We introduce penalized withdrawals from retirement savings accounts as a new revealed-preference tool to measure households' valuation of liquidity. This approach addresses key empirical challenges by providing a proxy for marginal utility without requiring assumptions on preferences or data on access to credit. Using U.S. administrative tax data from 1999-2018, we document three main findings. First, geographic differences, driven by local credit supply, account for over 30% of the nationwide variation in the valuation of liquidity across labor markets. Second, areas hit hardest by the Great Recession saw large increases in the valuation of liquidity, with local credit market spillovers explaining two-thirds of the effect. Third, Black households rely more heavily on penalized withdrawals, even after controlling for income and location, consistent with limited access to formal credit. Together, these findings highlight significant scope for welfare gains through more targeted safety-net policies and demonstrate the practical value of penalized withdrawals as a tool for monitoring liquidity needs.