Blog — 14 Jun, 2023
Analyzing Sentiment in Quarterly Earnings Calls Q2 2023
By Nicholas Ephraim
Net Positivity
In Q2 2023, S&P 500 earnings call transcripts had an overall Net Positivity score of 1.13%, up 0.05% from Q1 2023, and up 0.18% when compared to Q4 2022. This shouldn’t be a surprise as the S&P 500 is up 10.75% this year, and the broader market seemed to weather the turmoil created by the banking crisis and inflationary pressure which has come down in the last few months. Consumer Discretionary had the highest Net Positivity score out of the individual sectors with 1.54%, increasing 0.14% from Q1 where the sector also claimed the top spot. The success of the sector was driven primarily by strong Net Positivity scores from Starbucks Corporation (NasdaqGS:SBUX) with 3.21%, Chipotle Mexican Grill, Inc. (NYSE:CMG) with 2.64%, and Yum! Brands, Inc. (NYSE:YUM) with 2.56%. After Consumer Discretionary, Communication Services was the next highest sector for Net Positivity with 1.39%, improving 0.13% compared to Q1 2023. The success of Communication Services came from a few constituent companies with high scores including Alphabet Inc. (NasdaqGS:GOOGL) with 2.25%, Comcast Corporation (NasdaqGS:CMCSA) and Netflix, Inc. (NasdaqGS:NFLX) both with 1.95%, and Meta Platforms, Inc. (NasdaqGS:META) with 1.90%. Following Communication Services was Consumer Staples with a Net Positivity score of 1.33% and Health Care with 1.28%. The sector with the lowest score of the quarter was Financials with 0.68%, which makes sense given the chain reaction started by the failure of Silicon Valley Bank. The lowest scored companies in the Financials sector included Loews Corporation (NYSE:L) with -0.67%, JPMorgan Chase & Co. (NYSE:JPM) with -0.65%, and Zions Bancorporation (NasdaqGS:ZION) with -0.52%. The next lowest sectors after Financials were Materials with 0.92% and Real Estate with 0.96%.
For Q2 2023, we also analyzed Net Positivity scores in the various sections of the earnings call transcripts. For the S&P 500 this quarter the Presenter Speech had a Net Positivity score of 1.53%, while the Answer section had a score of 1.09%, showing the Executives of these companies used less favorable language during Q&A compared to their prepared remarks. The sector with the least amount of variance between the two transcript sections was Communication Services, which received a Net Positivity score of 1.45% during the Presenter Speech and 1.44% during the Answer section. The sector with the largest variance between sections was Energy, where the Presenter Speech was 1.65% and the Answer Section was 1.00%.
Numeric Transparency
Overall, in Q2 2023, the S&P 500 had a Numeric Transparency score of 2.12%, significantly lower than it’s Q1 2023 score of 2.66%, and Q4 2022 score 2.27%. This score shows overall for the quarter firms in the S&P 500 had a lower use of numbers relative to words and were less transparent in their earnings call transcripts compared to previous quarters. Utilities was the most transparent sector of the quarter with a score of 2.65%, driven by favorable scores from Atmos Energy Corporation (NYSE:ATO) with 4.40%, Ameren Corporation (NYSE:AEE) with 4.29%, and NRG Energy, Inc. (NYSE:NRG) with 3.62%. Although Utilities’ Numeric Transparency score was down from it’s Q1 2023 number of 3.16%, it still was significantly higher than the other sectors this quarter. After Utilities, we had Information Technology with a score of 2.26%, and Financials with 2.23%. The least transparent sector this quarter was Communication Services with 1.70%, which is interesting given they were the second highest sector for Net Positivity. Some of the lower scores contributing to Communication Services were Netflix, Inc. (NasdaqGS:NFLX) with 0.93%, Paramount Global (NasdaqGS:PARA) with 1.00%, and AT&T Inc. (NYSE:T) with 1.09%. After Communication Services, Consumer Staples was the next lowest sector for Numeric Transparency with 1.74%, followed by Materials with 1.89%.
Language Complexity
In Q2 2023, the S&P 500 constituents had a Language Complexity score of 12.31, down from Q1’s score of 12.42, showing the language in this quarter’s earnings call transcripts were less complex and seen as more favorable. The individual S&P 500 sector with the least complex language this quarter was Industrials with 11.91. Although Industrials has had the lowest Language Complexity score over the past five quarters, this time it was followed very closely by Materials with 11.92. Some of the firms contributing to the more favorable sector score for Materials were Packaging Corporation of America (NYSE:PKG) with 10.10, Vulcan Materials Company (NYSE:VMC) with 10.13, and The Mosaic Company (NYSE:MOS) with 10.55. After Industrials and Materials was Energy with 12.00 and Real Estate with 12.10. The S&P 500 sector with the highest score for Language Complexity was Utilities with 13.32, contributed by more complex language in transcripts from firms such as Ameren Corporation (NYSE:AEE) with 16.21, NextEra Energy, Inc. (NYSE:NEE) with 16.01, and Alliant Energy Corporation (NasdaqGS:LNT) with 15.80. After Utilities, the sectors with the most complex language use were Communication Services with 12.58 and Health Care with 12.53.
Analyst Selectivity Ratio
For this quarter S&P 500 constituents had an Analyst Selectivity Ratio of 44.34%, on par with scores from previous quarters, meaning active analysts covering a company have about the same opportunity to ask questions on earnings calls as before. Overall, the sectors with the highest Analyst Selectivity Ratio were Materials with a score of 51.99%, followed by Industrials with 51.62%, and Real Estate with 49.22%. The sectors with the lowest ratios were Communication Services with 26.57%, Energy with 30.65%, and Utilities with 31.18%. Analyzing the variance between Analyst Selectivity Ratio quarter-over-quarter, Energy had the largest decrease going from 34.49% in Q1 to 30.65% in Q2. Health Care had the second highest decrease between quarters going from 50.04% in Q1 to 46.76% in Q2. The sectors with the biggest increase between quarters were Consumer Staples (+4.98%) and Consumer Discretionary (+3.37%).
If you would like to receive the next infographic, register your interest here.
About Transcript Sentiment Scores
Transcript Sentiment Scores use natural language processing to provide a way to look at earnings call transcripts in a quantitative fashion. S&P Capital IQ Pro provides net positivity, numeric transparency, language complexity, and analyst selectivity ratio metrics for transcripts at the total, speaker, and component level. S&P Capital IQ Pro provides Net Positivity, Numeric Transparency, Language Complexity, and Analyst Selectivity Ratio metrics for transcripts at the: 1) Total level, 2) Speaker level, such as Executive or Analyst, and 3) Component level, such as Presentation Operator message, Presenter speech, Question, or Answer. Scores for companies are based on the most recent transcript of an earnings call in the calendar quarter that the event occurs. Scores are typically available 3 hours after a transcript is published to CIQ Pro. Scores are most effective when used to determine trends in sentiment by comparing current scores for a company vs. the prior quarters’ scores of the same company.
- Net Positivity: The net positivity score is based on the ratio of positive to negative words from the Loughran & McDonald’s (LM) Sentiment Word Lists and compares that to the total number of words.
- Numeric Transparency: The numeric transparency score is the ratio of numbers to words. A higher value means more use of numbers relative to words, which signifies a higher level of transparency. This is considered more objective and precise and thus, more favorable.
- Language Complexity: We use the Gunning Fog Index as a proxy for language complexity. Each score can be interpreted as the number of years of formal education a person needs to understand the text on the first reading. A lower score denotes simpler language and is viewed favorably.
- Analyst Selectivity Ratio: An analyst selectivity ratio is the percent of active analysts covering the stock that are allowed to ask questions during the call. A higher value is viewed favorably, and scores will range from 0% to 100%.
Source: S&P Capital IQ Pro. Data as of June 2, 2023.References:Zhao, F. "Natural Language Processing – Part II: Stock Selection" (September 2018). Natural Language Processing – Part II: Stock Selection | S&P Global Market Intelligence (spglobal.com)Zhao, F."Natural Language Processing – Part III: Feature Engineering" (January 2020). Natural Language Processing – Part III: Feature Engineering | S&P Global Market Intelligence (spglobal.com)