Search

We construct a macroeconomic estimate of total AI production in the United States, combining inference and R&D/training activities with quality adjustments to account for algorithmic progress. We then develop a nascent framework for "AI GDP" that tracks the AI economy as a coherent whole, complementing traditional national accounts.

We study liquidity requirements in a framework with fire sales. The framework nests three common pricing mechanisms and produces the same observables. Absent risk-sharing considerations, the equilibrium is efficient with cash-in-the-market pricing; a liquidity requirement is optimal with second-best-use pricing; and a liquidity ceiling (i.e., a cap on liquid assets) is optimal with adverse selection.

Decentralized finance surged in popularity around 2020. We explore its value and limitations and highlight some potential regulatory concerns.

Do banks realize simultaneous trading losses because they invest in the same assets, or because different assets are subject to the same macro shocks? This paper decomposes the comovements of bank trading losses into two orthogonal channels: portfolio overlap and common shocks.

This article examines the incentives for banks to hold various assets on their balance sheets for use as collateral when the opportunity cost of doing so can be high. Focusing on the five-year period (2002-07) that preceded the financial crisis, it examines the choices made by financial institutions among the assets that are pledged as collateral in Canada's Large Value Transfer System. This serves as a baseline for collateral-management practices during relatively normal times. The results of this study are important for policy-makers, especially the Bank of Canada, which is concerned both about the efficient functioning of fixed-income markets and about the credit risk it ultimately bears in insuring LVTS settlement. The results suggest that relative market liquidity and market-making capacity are important factors in the choice of securities pledged as collateral in the LVTS.

This special issue discusses tools used at Bank of Canada to conduct current analysis - the collection and analysis of a broad spectrum of information to form a view of current economic activity. This analysis is a key input into the Bank’s monetary policy decision-making process.

We document that, across households, the money consumption ratio increases with age and decreases with consumption, and that there has been a large increase in the money consumption ratio during the recent era of very low interest rates. We construct an overlapping generations (OLG) model of money holdings for transaction purposes subject to age (older households use more money), cohort (younger generations are exposed to better transaction technology), and time effects (nominal interest rates affect money holdings).

We offer a theory of how inefficiently lax financial regulation could arise in a democratic society.

Many decentralized markets are able to attain a stable outcome despite the absence of a central authority (Roth and Vande Vate, 1990). A stable matching, however, need not be efficient if preferences are weak. This raises the question whether a decentralized market with weak preferences can attain Pareto efficiency in the absence of a central matchmaker.