In the fourth edition of ‘Product Insights’, we look at how users can efficiently track changes in consensus estimates during earnings season using Refinitiv Workspace for Analysts and Portfolio Managers.
One of the more time-consuming exercises is to systemically sort the companies that have reported earnings in the current earnings season period vs. those who have yet-to-report. Aligning this data can be challenging if we use periods such as FQ1 (current quarter). For example, we are currently in 22Q3 earnings season and FQ1 will show a mix of either a) current quarter estimates for a yet-to-report company (22Q3), or b) the next quarter estimate for a company who has already reported results in the current earnings season period (22Q4). This poses a challenge for Analysts and Portfolio Managers who are attempting to answer a typical question of ‘which companies in the Russell 1000 have seen the largest upward or downward revisions after reporting its results?’
Fortunately, Refinitiv Workspace provides an intelligent solution to this problem through the data library catalogue. There are specific parameters in the library that will automatically display data for companies that have already reported and for those who have yet-to-report.
Exhibit 1 shows the data library which allows users to pull in data from multiple data sets such as I/B/E/S estimates, ESG, Fundamentals, Ownership, Pricing and Reference, and many more. Users can mark a field as a favourite by clicking on the ‘star’ symbol at the bottom right for easy retrieval next time they access the library. The data library will remain the same no matter where it is accessed from (i.e., desktop applications such as Screener or via the excel add-in). In this product insight, we are using the Screener app (type ‘Screener’ in the toolbar).
To see how mean estimates have been revised, we use the ‘Mean Estimate Pct Change’ field which has numerous parameter that can be customized from the earnings measure itself (i.e., Revenue, EBITDA, EPS), the financial period (i.e., FQ1, FY1, FY2), and the window period (i.e., 7d, 14d, 30d, 90d). In this example, we will use EQ0 as the financial period which will display data for companies that have already reported results in the current earnings season period. EQ1 will display data for companies that have yet-to-report results in the current earnings season period. While EQ2 will display data for both reported and yet-to-report companies.
Exhibit 1: Refinitiv Data Library
Source: Refinitiv Workspace
While the ‘Screener’ application within Refinitiv Workspace is traditionally used for idea generation, it can also be versatile for monitoring analyst estimate activity for the entire index using the template feature where users can add columns of data from the data library. We use the Russell 1000 as our universe and incorporate numerous columns of data including EQ0, EQ1, EQ2, and other fields as shown in Exhibit 2. The screener app allows users to load any index (permission depending), ETF, Lipper Fund, or user-defined lists and portfolios.
The data is sorted by the largest downward earnings revisions for companies that have already posted results for the current earnings season period. Said differently, we can see how analysts have revised estimates for the following quarter once a company reported results.
Exhibit 2: Companies with Largest Negative Revisions Post-Results
Source: Refinitiv Workspace
For a better understanding of what each column of data represents, we outline field definitions as shown below:
- Number of Analysts (FQ1): Number of analysts who have submitted an estimate for the upcoming yet-to-report quarter.
- Preferred Earnings – Actual Surprise (FQ0): Earnings surprise % for most recent reported quarter, comparing actual EPS vs. consensus estimate on the day of reporting.
- Mean Estimate Pct Change (Preferred, EQ0, 90d): 90-day percent change in mean Preferred Earnings estimate for most recent reported quarter (i.e., how did estimates change heading into results. If a company reported results on Nov 1st, the 90-day calculation would compare the mean estimate from October 31st 90-days prior).
- Mean Estimate Pct Change (Preferred, EQ1, 90d): If a company has already reported in the current earnings season period (i.e., EQ0), EQ1 will display “NA”. However, if a company has yet-to-report in the current earnings season period, this value will be populated with the 90-day percent change in mean Preferred Earnings estimate for the upcoming yet-to-report quarter.
- Mean Estimate Pct Change (Preferred, EQ2, 90d): 90-day percent change in mean Preferred Earnings estimate for upcoming yet-to-report quarter (i.e., how have estimates changed after a company has reported earnings).
- Mean Estimate Pct Change (Revenue, EQ2, 90d): 90-day percent change in mean Revenue estimate for upcoming yet-to-report quarter (i.e., how have estimates changed after a company has reported earnings).
- Preferred Earnings – Predicted Surprise (FQ1): StarMine Predicted Surprise % for upcoming yet-to-report quarter (i.e., Predicted Surprise % = StarMine SmartEstimate vs. Mean Estimate, where SmartEstimate places a higher weight on more timely and accurate analysts. The Predicted Surprise is deemed significant when above 2% or below -2% as our research shows that the Predicted Surprise can accurately predict the direction of the earnings surprise correctly 70% of the time).
Putting this altogether using Alcoa Corp (AA.N) as an example, the company reported 22Q3 results on October 19th, posting earnings miss of 273.1% ($-0.33 actual vs. $0.19 mean estimate). Heading into the 22Q3 earnings print, EPS estimates were revised downwards by 69.2%. After the company reported 22Q3 results, the 22Q4 consensus EPS estimate was downgraded by -314.9% (values less than -100% occur when an EPS estimate turns from positive to negative), and the revenue mean estimate was downgraded by 15.4%. The company is expected to report 22Q4 results on January 17th, 2023 and has a positive Predicted Surprise of 11.0% which may surprise markets at the time of reporting.
We show the same data in Exhibit 3 for companies that have seen the largest upward revisions post company reporting. For example, after American Airlines (AAL.OQ) posted its 22Q3 results, analysts reacted by raising the 22Q4 consensus mean EPS estimate upwards by 362.8% (+2.0% for revenue). Other airlines including United Airlines Holdings (UAL.OQ), JetBlue Airways Corp (JBLU.OQ), Delta Air Lines Inc (DAL.N), and Southwest Airlines Co (LUV.N) saw similar upgrades. This provides powerful workflow tools for Analysts and Portfolio Managers looking to parse through 1000s of companies during earnings season to identify thematic trends.
Exhibit 3: Companies with Largest Positive Revisions Post-Results
Source: Refinitiv Workspace
From the Screener app, a user can click on an individual company and interrogate the revision activity in more detail by visiting the ‘Detailed Estimates’ page as shown in Exhibit 4. Using Caesars Entertainment (CZR.OQ) as an example, we can see it has a 22Q4 EPS Predicted Surprise of 60.20% which is calculated by comparing the StarMine SmartEstimate of $0.21 vs. Mean estimate of $0.13. Furthermore, the mean EPS estimate has been revised upwards by 291.04% over the last 30 days.
Looking further down the page, we can see the individual analyst estimates which are sorted by SmartEstimate Weight. Analysts who are more accurate and timely receive a higher weight in the SmartEstimate. We also display ‘Bold’ estimates which are 5-star rated analysts who have an estimate greater than 5% above consensus (or -5% below) and are highlighted in bright green or red to alert users of this activity.
Exhibit 4: Detailed Estimates for Caesar Entertainment (CZR.O)
Source: Refinitiv Workspace
Conclusion
Refinitiv Workspace intelligently allows users to sort a large universe of companies into ‘reported’ vs. ‘yet-to-report’ during the current earnings season quarter which helps Analysts and Portfolio Managers determine analyst revision activity more efficiently. This provides powerful workflow tools for Analysts and Portfolio Managers looking to parse through 1000s of companies during earnings season to identify thematic trends.
Previous Product Insights
Refinitiv Workspace is a complete solution for research and analytics. It places the most comprehensive market information, news, analytics, and trading tools available into a desktop.
Get unique value-add analytics and predictive financial modeling, dedicated to making investment research smarter with Refinitiv StarMine data.
Refinitiv I/B/E/S Estimates are a market leader, boasting 200+ metrics and indicators across 15 industries. Find more information on our estimates data.
