Data Spotlight: Geolocation risk concentrations & more

Bloomberg Professional Services

Welcome to Data Spotlight, our series showcasing insights derived from Bloomberg’s 8,000+ enterprise datasets available on data.bloomberg.com via Data License.

In this edition we take a look at using geolocation data to assess risk when constructing energy-focused investment portfolios. We also look at how Bloomberg sentiment score can be used by investors to generate investment ideas.

Our previous editions of Data Spotlight can be found here: 

• Sustainability-Linked Bonds, ESG backtesting
• Alternative data – Transactions, fundamentals
• Merger arbitrage, inflation

1. Building a diversified portfolio with energy facilities location data

Thematic investing focuses on macro-level trends and builds portfolios around them. However, this strategy can expose investors to high concentration risk in specific regions, such as geographic risks. In this piece, we measure geographic diversification to mitigate potential risks.

Embracing geographic diversity within a thematic investment framework is crucial for managing risks and maximizing long-term returns. Chart 1 illustrates why investors should care how facilities from two companies are distant from each other to provide a good geographic diversification: typically in this example concentration of facilities in the same area (in this case: Germany) can expose assets to the same geographic risk factors.

For this analysis, we use The BloombergNEF (BNEF) Energy Physical Assets product that delivers the geolocation of ~130,000 hydrogen, wind, solar, and energy storage assets and provide production facilities for companies in various thematics. 

For illustration purposes, we consider how a portfolio manager investing in the Clean Hydrogen space in large cap companies (>100bn USD) can use this data to quickly understand potential geolocation risk concentrations. To build a diversified portfolio, she needs to take into account the distance between all the clean hydrogen assets of companies in the universe in order to limit the risk of too much exposure in the same area. 

For that, she measures the distance between all those assets and applies a classic max-diversification algorithm with a risk matrix that corresponds to distances between companies based on those assets. She finds the optimal portfolio according to that specific geographic diversification. Chart 2 provides an overview of the optimal portfolio composition as well as its 2-year risk return profile compared to a market-weighted portfolio of the same universe of companies (producers in the clean hydrogen space with at least 100bn USD market cap).

Theme: Clean Energy Investing
Roles: Portfolio Managers, Analysts, Strategists, Sustainability Investors
Bloomberg Dataset: BNEF Energy Physical Assets

2. Uncovering climate risks impacting companies’ facilities

Energy geolocation data can also be used to assess risk related to individual companies. The BNEF Energy Physical Assets product, includes location information for production, consumption, transportation, and storage assets.

For example, Chart 1 shows the breakdown of different types and operational status of physical assets of NextEra, a sustainable energy generation and distribution services provider. For its three clean hydrogen planned projects, two are production projects and one is a consumption project which makes or uses clean hydrogen with zero or low greenhouse gas emissions, using either renewable and nuclear electricity, or biomass and fossil fuels coupled with carbon capture and storage technologies.

BNEF Energy Physical Assets data can support clients in analyzing physical risk, performing competitive geographic analysis to track where companies are building or operating energy facilities.

Take NextEra Energy as an example, users could easily overlay a US strong wind risk hazard map to analyze if the company has any planned or active operating assets located in extremely high risk areas. In our calculation, about 10% of its planned or active operating assets are located in areas with a risk score larger than 90. This is particularly important as severe weather could cause disruptions to company’s operations (chart 2). 

Theme: Risk Management
Roles: Portfolio Managers, ESG Analysts, Risk Managers, Corporates
Bloomberg Datasets: BNEF Energy Physical Assets

3. Backtesting sentiment for equity markets

News stories are often used as a proxy for investor confidence in a particular asset. Bloomberg’s Company Sentiment And Market Moving News dataset employs advanced natural language processing techniques to provide a real-time numerical estimation of this confidence, enabling investors to distill vast amounts of unstructured textual data into quantitative signals that can inform trading decisions. 

A straightforward way to gauge the underlying benefits of these signals is to run financial backtests using the sentiment data on an index. The strategy here is simple: on each day we average the sentiment associated with all the companies in our universe and then create three baskets of stocks based on the value of this sentiment (High, Medium and Low). 

The thesis is that a basket of stocks with strong positive sentiment should outperform their negatively perceived counterparts, creating a long-short opportunity. Details of this backtest are as follows:

Universe: TSX 60 Index (Canadian equities)

Time Range: 2016 to 2023

Weighting Scheme: Equal weighted

Rebalancing Period: Daily

Strategy: Open-to-open (Receive sentiment file at 8pm ET, enter positions at market open the next day and rebalance at market open the following day)

We picked a Canadian index to show that the product had ample coverage beyond the names in your typical large cap US universe. As chart 1 shows, tens of thousands of news stories are available every year for the 60 companies of the Toronto index.

Chart 2 displays strong performance of the High Score group, generating 3.5x returns over the time period assessed. This is significantly better than the returns of the actual index, which roughly doubled over the same time period.

Since each tertile can be thought of as its own miniature long-only portfolio, a market neutral deployment of this strategy involves going long on the High Score group and shorting the Low Score group. Thus monetizing the difference in performance.

Bloomberg’s Company Sentiment And Market Moving News dataset is a valuable tool to inform quantitative trading strategies with granular, comprehensive insights. What’s more, the research in this blog was conducted on the raw sentiment scores, meaning there is plenty of room for more complex strategies that could generate new and unique results.

Themes: Quantitative Trading, Alpha Generation
Roles: Quantitative Research, Traders
Bloomberg Dataset: Company Sentiment And Market Moving News

Interested to learn more about our data offering? Bloomberg’s Enterprise Data business transforms the way customers extract value from data by providing the most comprehensive coverage and highest data quality in the industry. 

Bloomberg’s OneData approach makes all of the data easily discoverable on data.bloomberg.com, via Data License, instantly usable and delivered via APIs on the Cloud, to a customer’s hosting provider or data center. Our data management solutions aggregate, organize, and link your data to make it accessible through a diverse set of delivery mechanisms.