Jillian Mascelli and Megan RoddenThis paper analyzes the risks posed by future-state quantum computers, specifically the “harvest now decrypt later” (HNDL) risk. We review foundational concepts of quantum computing to address the present and ongoing threat of HNDL to currently protected data. We use the Bitcoin network as an illustrative example to study the implications of HNDL for distributed ledger cryptocurrency networks that rely upon traditional cryptography.

Using novel data on sectoral safe asset positions in 21 advanced economies since 1980, we document the central role of the foreign sector in the market for safety and its macroeconomic implications. We show that safe asset holdings have expanded significantly relative to GDP, driven by rising net holdings of the foreign sector and accommodated by increased issuance from the financial and public sectors. Furthermore, fluctuations in safe assets are almost exclusively driven by the foreign and financial sectors, with close links between the two.

Jeffrey S. Allen and Max S. S. HatfieldWe examined the performance of four families of large language models (LLMs) and a variety of common fuzzy matching algorithms in assessing the similarity of names and addresses in a sanctions screening context. On average, across a range of realistic matching thresholds, the LLMs in our study reduced sanctions screening false positives by 92 percent and increased detection rates by 11 percent relative to the best-performing fuzzy matching baseline.

Yesol Huh and Matthew Vanderpool KlingThis paper introduces parallel trends forest, a novel approach to constructing optimal control samples when using difference-in-differences (DiD) in a relatively long panel data with little randomization in treatment assignment. Our method uses machine learning techniques to construct an optimal control sample that best meet the parallel trends assumption. We demonstrate that our approach outperforms existing methods, particularly with noisy, granular data.