Political Risk and Sentiment
Measures of firm-level political risk (PRisk) and sentiment (PSentiment) similar to Hassan et al. (2019).
At a glance:
Political Risk and Sentiment
- Coverage
-
Public firms, 80+ countries, 2002-present.
- Granularity
-
Earnings-call level (firm-panel export: one row per call).
- Source
-
LSEG earnings-call transcripts (English).
- Update schedule
-
Last data refresh Jan 12, 2026.
Update frequency:
Quarterly
- Export shape
-
Time-series aggregates; firm-panel call-level rows.
- Access
-
Free preview; exports depend on plan.
Overview
Political Risk (PRisk) and Sentiment (PSentiment) tracks when companies discuss political issues as a source of uncertainty -- and whether they frame those political developments as net positive or net negative.
The Curated Measure contains two core signals:
- PRisk: how intensely call participants discuss political matters in the context of risk or uncertainty. Higher PRisk means more of the conversation is devoted to political risks that could affect the firm.
- PSentiment: the net tone of political discussion (positive vs. negative) in those same political passages. Positive values indicate a more favorable framing; negative values indicate a more adverse framing.
In addition to the overall series, we provide topic-based PRisk that decomposes political risk into eight policy areas: Economy, Trade, Tax, Security, Institutions, Health, Environment, and Technology.
Common applications include:
- Identifying which firms are most exposed to political uncertainty at a point in time (and how that changes).
- Studying how political risk relates to investment, hiring, guidance, or volatility.
- Separating "politics is uncertain" (PRisk) from "politics is good/bad for us" (PSentiment).
- Pinpointing what kind of political risk is driving attention (e.g., trade vs. tax vs. regulation/institutions).
- Building sector- or country-level aggregates from firm-level signals while preserving firm heterogeneity.
Why it matters
Political events often matter most through firm-specific channels -- regulatory exposure, procurement dependence, licensing, cross-border supply chains, and industry-specific policy. Traditional measures tend to be slow-moving or too aggregate to reveal which firms are actually affected.
This dataset captures political risk as expressed by decision-makers and analysts in real time, namely what they choose to talk about when discussing outlook, operations, and constraints.
Explore the data
This measure includes 9 related series. Use the filter to find a series and open the series page for full query and methodology notes.
No series match your filter.
-
PRisk and PSentiment
Free downloads417views • 145downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Economy
130views • 18downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Environment
103views • 19downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Health
105views • 17downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Institutions
155views • 19downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Security
123views • 18downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Tax
100views • 20downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Technology
96views • 18downloads • Data through: Dec 31, 2025 -
Topic-based PRisk: Trade
211views • 27downloads • Data through: Dec 31, 2025
How to access?
Preview: Browse all series and view charts without an account. Downloads require sign-in .
Free plan
- series: time-series + panel included.
- Other series: 1 unlocks panel + time-series.
Standard plan
- Time-series: included for all series.
- Panel: series included; other series require 1 panel export credit .
Research / Enterprise
- Time-series: included for all series.
- Panel: included for all series.
Methodology and data
Methodology (conceptual)
This Curated Measure is based on the approach in Firm-Level Political Risk: Measurement and Effects by Hassan, Hollander, van Lent, and Tahoun and published in the Quarterly Journal of Economics in 2019:
-
Identify political language. We use a political-language classifier derived from training libraries to identify phrases -- weighted by an associated tf*idf score ** strongly associated with political discussion.
-
Construct PRisk (political risk). Political risk is measured by counting political-language instances that occur in conjunction with synonyms for risk or uncertainty, producing a firm-by-quarter (and call-level) measure of political risk intensity.
-
Construct PSentiment (political sentiment). Political sentiment measures the net tone (positive vs. negative) of political discussion, computed within the same political passages used for PRisk.
-
Topic decomposition. Topic-specific PRisk applies the same logic within eight political domains: Economy, Environment, Health, Institutions, Security, Tax, Technology, and Trade.
> Note: Hassan et al. (2019) measure political phrases and associated risk terms in a sliding +/- 10 word window. We measure it at the sentence level.
Data
- Earnings-call transcripts (English), 2002–present
- Public firms across 80+ countries
- Source: LSEG transcripts
- Exports: time-series aggregates and firm-panel call-level rows
References
Hassan, T. & Hollander, S. & van Lent, L. & Tahoun, A., "Firm-Level Political Risk: Measurement and Effects," The Quarterly Journal of Economics, 134(4) (2019)
NL Analytics. (2026). Political Risk and Sentiment [Data set]. NL Analytics. https://apps.nlanalytics.tech/curated-measures/political-risk/