Search Results
Working Paper
Introducing a Framework for Measuring the Quantitative Benefits of Privacy-Enhancing Technologies
This paper reviews privacy-enhancing technologies (PETs) and explores their benefits when used to make traditional payment processes more private. PETs can decrease privacy risk by reducing the amount of sensitive information accessible to payment-processing personnel and systems. This paper proposes a framework for quantifying the risk-reduction benefits of PETs. This method can be used to calculate the amount of privacy-risk exposure that may be created by a set of payment activities, estimate the amount by which PETs can decrease that exposure, and compare that quantified benefit against ...
Speech
Elementary, Dear Data
Remarks at 2023 U.S. Treasury Market Conference, Federal Reserve Bank of New York, New York City.
Speech
Weighing the risks to the economic outlook: remarks at The Leo J. Meehan School of Business, Stonehill College, Easton, Massachusetts, September 3, 2019
It was an eventful August in the financial markets amid talk of additional tariffs and tax cuts, the falling 10-year Treasury rate, and volatility in stock prices. But the economic data and forecasts indicate a relatively good domestic economy.
Briefing
Discussing Data: The Elevator Pitch
Researchers and custodians of data are often tasked with explaining their data to unfamiliar audiences. There exists a knowledge gap that can be challenging, particularly when attempting to communicate research that uses complex or less well-known data. This paper describes an “elevator pitch” that can be used to quickly and efficiently present core characteristics of data. Specifically, it outlines seven core characteristics that can be used to create such a pitch. Additionally, it provides examples of how this compact description can be integrated into research papers, presentations, or ...
Will Digital Wallets Replace Cash?
Dialogue with the Fed attendees hear about the opportunities and challenges involved with digital wallets like Venmo.
Briefing
Discussing Data: Evaluating Data Quality
Data-driven organizations are becoming increasingly aware of data quality in order to mature their data activities. Data quality is of particular concern to research functions, but many existing frameworks are not well suited for use by researchers. This paper discusses existing data quality frameworks and focuses on one that meets the needs of research functions. A modified version of that framework is described along with details for use with a wide range of data.
Speech
Center for the REstoration of Economic Data: The Past Is Indeed Prologue
Philadelphia Fed President and CEO Patrick T. Harker, speaking at the inaugural conference of the Philadelphia Fed’s new Center for the REstoration of Economic Data (CREED), announced the Center’s commitment to bringing historical data to light by “taking unstructured, analog data and converting it into useful, accessible, and high-quality data” for public use.In discussing the Philadelphia Fed’s recent research on housing, Harker stressed the importance of “bringing old data to light” to find a “historical economic perspective,” and “to better inform our present ...
Discussion Paper
Monetizing Privacy with Central Bank Digital Currencies
In prior research, we documented evidence suggesting that digital payment adoptions have accelerated as a result of the COVID-19 pandemic. While digitalization of payment activity improves data utilization by firms, it can also infringe upon consumers’ right to privacy. Drawing from a recent paper, this blog post explains how payment data acquired by firms impacts market structure and consumer welfare. Then, we discuss the implications of introducing a central bank digital currency (CBDC) that offers consumers a low-cost, privacy-preserving electronic means of payment—essentially, digital ...
Discussion Paper
Insights from Newly Digitized Banking Data, 1867-1904
Call reports—regulatory filings in which commercial banks report their assets, liabilities, income, and other information—are one of the most-used data sources in banking and finance. Though call reports were collected as far back as 1867, the underlying data are only easily accessible for the recent past: the mid-1980s onward in the case of the FDIC’s FFIEC call reports. To help researchers look farther back in time, we’ve begun creating a complete digital record of this “missing” call report data; this data release covers 1867 through 1904, the bulk of the National Banking Era. ...
Working Paper
Evaluating Local Language Models: An Application to Bank Earnings Calls
This study evaluates the performance of local large language models (LLMs) in interpreting financial texts, compared with closed-source, cloud-based models. We first introduce new benchmarking tasks for assessing LLM performance in analyzing financial and economic texts and explore the refinements needed to improve its performance. Our benchmarking results suggest local LLMs are a viable tool for general natural language processing analysis of these texts. We then leverage local LLMs to analyze the tone and substance of bank earnings calls in the post-pandemic era, including calls conducted ...