Paul E. Soto

Paul E. Soto

Auto-generated profile, based on publicly available information

Paul E. Soto

Board of Governors of the Federal Reserve System (U.S.)

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 February 2025 in FEDS Notes
Verified
Journal Article
Closed access
LCLeland D. Crane
Leland D. Crane
MGMichael Green
Michael Green
PSPaul E. Soto
Paul E. Soto
Artificial Intelligence (AI) may be poised to raise productivity across various domains, including writing (Noy and Zhang 2023), programming (Peng et al. 2023), and research and development (Toner-Rodgers 2024; Korinek 2023). However, understanding the extent to which AI?and generative AI in particular?has been adopted as part of the production process remains an open question. This note reviews the extant surveys on AI adoption at both the employee and firm levels. Surveys of firms show a wide spread of adoption rates, ranging from 5 percent to about 40 percent. Surveys of workers show between 20 and 40 percent of workers using AI in the workplace, with much higher rates in some occupations like computer programming. While estimates of the level of AI uptake vary, measurement considerations partly explain the differences; more importantly, the available time series data all suggest rapid growth in adoption.

1 April 2023 in Finance and Economics Discussion Series
Verified
Journal Article
Open access
PDF file
PAPuriya Abbassi
Puriya Abbassi
RIRajkamal Iyer
Rajkamal Iyer
JPJosé-Luis Peydró
José-Luis Peydró
PSPaul E. Soto
Paul E. Soto
We study short-term and medium-term changes in bank risk-taking as a result of supervision, and the associated real effects. For identification, we exploit the European Central Bank’s asset-quality review (AQR) in conjunction with security and credit registers. After the AQR announcement, reviewed banks reduce riskier securities and credit supply, with the greatest effect on riskiest securities. We find negative spillovers on asset prices and firm-level credit availability. Moreover, non-banks with higher exposure to reviewed banks acquire the shed risk. After the AQR compliance, reviewed banks reload riskier securities but not riskier credit, resulting in negative medium-term firm-level real effects. These effects are especially strong for firms with high ex-ante credit risk. Among these non-safe firms, even those with high ex-ante productivity experience negative real effects. Our findings suggest that banks’ liquid assets help them to mask risk from supervisors and risk adjustments banks make in response to supervision have persistent corporate real effects.

1 February 2023 in FEDS Notes
Verified
Journal Article
Closed access
PSPaul E. Soto
Paul E. Soto
Supply chain bottlenecks can have significant impacts on the real economy. When firms experience shortages, shipping delays, or shutdowns, as occurred during the COVID-19 pandemic, they may be unable to produce, transport, and sell their products.

1 October 2025 in Management Science
Verified
Journal Article
Closed access
PAPuriya Abbassi
Puriya Abbassi
RIRajkamal Iyer
Rajkamal Iyer
JPJosé-Luis Peydró
José-Luis Peydró
PSPaul E. Soto
Paul E. Soto
We study short-term and medium-term changes in bank risk-taking as a result of supervision and the associated real effects. For identification, we exploit the European Central Bank’s asset-quality-review (AQR) in conjunction with security and credit registers. After the AQR announcement, reviewed banks reduce riskier securities and credit supply, with the greatest effect on riskiest securities. We find negative spillovers on asset prices and firm-level credit availability. Moreover, nonbanks with higher exposure to reviewed banks acquire the shed risk. After the AQR compliance, reviewed banks reload riskier securities but not riskier credit, resulting in negative medium-term firm-level real effects. These effects are especially strong for firms with high ex ante credit risk. Among these nonsafe firms, even those with high ex ante productivity, experience negative real effects. Our findings suggest that banks’ liquid assets help them to mask risk from supervisors and risk adjustments banks make in response to supervision have persistent corporate real effects. This paper was accepted by Victoria Ivashina, finance. Funding: J.-L. Peydró acknowledges financial support from the BBVA Foundation [2018 Leonardo Grant for Researchers and Cultural Creators], the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme [Grant Agreement 648398], the MCIU/AEI/FEDER, UE [Grant PGC2018-102133-B-I00], and the Spanish Ministry of Economy and Competitiveness, through the Severo Ochoa Programme for Centres of Excellence in R&D [Grant SEV-2015-0563]. The analysis, conclusions, and opinions set forth in the paper are solely those of the authors and do not necessarily represent the views of the Bundesbank, Eurosystem, Federal Reserve Board, or the United States. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2020.01714 .

1 May 2024 in Finance and Economics Discussion Series
Verified
Journal Article
Open access
PDF file
TCTomaz Cajner
Tomaz Cajner
LCLeland D. Crane
Leland D. Crane
CKChristopher Kurz
Christopher Kurz
NMNorman Morin
Norman Morin
PSPaul E. Soto
Paul E. Soto
BVBetsy Vrankovich
Betsy Vrankovich
This paper examines the link between industrial production and the sentiment expressed in natural language survey responses from U.S. manufacturing firms. We compare several natural language processing (NLP) techniques for classifying sentiment, ranging from dictionary-based methods to modern deep learning methods. Using a manually labeled sample as ground truth, we find that deep learning models partially trained on a human-labeled sample of our data outperform other methods for classifying the sentiment of survey responses. Further, we capitalize on the panel nature of the data to train models which predict firm-level production using lagged firm-level text. This allows us to leverage a large sample of "naturally occurring" labels with no manual input. We then assess the extent to which each sentiment measure, aggregated to monthly time series, can serve as a useful statistical indicator and forecast industrial production. Our results suggest that the text responses provide information beyond the available numerical data from the same survey and improve out-of-sample forecasting; deep learning methods and the use of naturally occurring labels seem especially useful for forecasting. We also explore what drives the predictions made by the deep learning models, and find that a relatively small number of words associated with very positive/negative sentiment account for much of the variation in the aggregatesentiment index.

1 June 2025 in Finance and Economics Discussion Series
Verified
Journal Article
Open access
PDF file
DCD. Leland Crane
D. Leland Crane
AKAkhil Karra
Akhil Karra
PSPaul E. Soto
Paul E. Soto
We evaluate the ability of large language models (LLMs) to estimate historical macroeconomic variables and data release dates. We find that LLMs have precise knowledge of some recent statistics, but performance degrades as we go farther back in history. We highlight two particularly important kinds of recall errors: mixing together first print data with subsequent revisions (i.e., smoothing across vintages) and mixing data for past and future reference periods (i.e., smoothing within vintages). We also find that LLMs can often recall individual data release dates accurately, but aggregating across series shows that on any given day the LLM is likely to believe it has data in hand which has not been released. Our results indicate that while LLMs have impressively accurate recall, their errors point to some limitations when used for historical analysis or to mimic real time forecasters.