Past performance does not predict future returns. You may get back less than you originally invested. Reference to specific securities is not intended as a recommendation to purchase or sell any investment.
While it is well-recognised that earnings calls are a crucial way for companies to communicate financial performance and outlook to investors and analysts, it is equally acknowledged that the details provided are not always accurate. From trying to hide bad news and exaggerating the good to avoiding answering difficult questions, executives all too often obfuscate the truth, making it difficult for investors and analysts to get to the reality of a company’s financial status. How can this be rectified?
One possible answer is the use of AI. By using techniques such as natural language processing (NLP) and sentiment analysis to scrutinise human language and behaviour, AI can uncover subliminal indications of confidence, or a lack of confidence, in a company's future performance.
What are NLP and sentiment analysis?
NLP uses complex algorithms to analyse human language to extract meaning, context and intent from text or speech. It performs tasks such as translation, summarisation, classification and natural language generation.
Sentiment analysis is a specific application of NLP that measures the positivity or negativity of text or prose to understand the hidden emotions behind the words. It can also detect other aspects of sentiment, such as polarity, intensity, subjectivity and tone.
By using a combination of NLP and sentiment analysis technologies, investors are analysing earnings calls via text or audio at scale to detect a variety of elements which may suggest the following.
NLP analysis:
-
Indirect answers to questions = evasion or uncertainty.
-
Exuberant words = overconfidence or exaggeration.
-
The use of fillers such as "um" or "uh" = hesitation or nervousness.
-
Qualifying statements such as "to the best of my knowledge" or "as far as I know" = doubt or lack of commitment.
Sentiment analysis:
-
By analysing different speakers, such as the CEO and the CFO, the software can identify discrepancies or inconsistencies.
-
Assessing the same speaker over time to identify changes or trends.
By using these techniques, it is possible for investors to gain a better insight into the true state of a company, helping them make more informed decisions.
An invaluable tool
One of the more notable players in the market so far is LiarLiar.ai, which employs a variety of techniques to detect lies, combining advanced AI with psychological insights. These include:
Micro Facial Expressions: The system analyses real-time video feeds to detect micro facial expressions that are often subconscious and can indicate deception.
Heart Rate Fluctuations: Using Remote Photoplethysmography, it monitors subtle colour changes in the face that correspond with heart rate fluctuations.
Body Language: It observes body language, including micromovements and gestures that may suggest dishonesty.
Voice Analysis: Monitors the voice for sudden shifts in tone and pitch, which can betray an individual's composure or reveal underlying stress.
Eye Movement Patterns: The tool monitors eye movements, such as decreased blinking or erratic gaze, which can be associated with lying.
Emotional Reader: Beyond detecting lies, it has an integrated feature that helps to understand the emotions the subject is trying to conceal.
There is no doubt LiarLiar.ai has the potential to be a valuable tool for investors by helping to verify statements and projections from management team earnings calls. By spotting inconsistencies, contradictions or signs of stress, LiarLiar can question the credibility of executives and help avoid costly mistakes.
Several companies and platforms have already been using AI for lie detection and sentiment analysis on earnings calls and have reported interesting findings and results. For example:
-
In 2023, a leading research and advisory firm, Gartner, published a report that examined how investors used NLP and sentiment analysis to detect possible cases of deception on earnings calls. The report also provided guidance for finance leaders on how to adjust their investor messaging and earnings call delivery to effectively communicate corporate performance and avoid any negative consequences.
-
A speech-to-text API provider, AssemblyAI, made a web app that provides sentiment analysis of earnings calls of companies that use their technology. It allows users to upload or stream audio files of earnings calls and generates a transcript with sentiment annotations for each sentence. It also provides a summary of the overall sentiment and the key topics discussed on the call.
As AI becomes more prevalent and sophisticated, companies are also becoming more aware and are preparing for the impact these technologies may have on their earnings calls. In a bid to brush up their act and not be caught off guard, companies are taking a number of approaches prior to earnings calls, including:
-
Hiring consultants or coaches to help them improve their communication skills and delivery.
-
Practicing their scripts and rehearsing their answers to potential questions.
-
Using tools or software to analyse their own transcripts and audio files and identify areas of improvement or risk.
-
Being more transparent and honest about their performance and outlook and providing clear and consistent explanations and evidence.
Going forward
AI is unquestionably changing the way we analyse financial information and will only become more sophisticated over time. The hope is that companies do more than try to outwit AI, and embrace the opportunity to enhance their credibility and trustworthiness by making better, more honest earnings calls. This in turn would enable analysts to make better recommendations for their clients and everyone else.
But one thing is for certain, AI is here to stay and we need to be ready for it.
Understand common financial words and terms See our glossary
Read, watch and listen to more insights from Liontrust fund managers here >
KEY RISKS
Past performance is not a guide to future performance. The value of an investment and the income generated from it can fall as well as rise and is not guaranteed. You may get back less than you originally invested.
We recommend any fund is held long term (minimum period of 5 years). We recommend that you hold funds as part of a diversified portfolio of investments.
All Liontrust Funds carry some degree of risk which may have an adverse effect on the future value of your investment. Therefore, before making an investment decision, you should familiarise yourself with the different types of specific risks associated with the investment portfolio of each of our Funds. There is no certainty the investment objectives of the portfolios or strategies mentioned in this document will actually be achieved and no warranty or representation is given, whether express or implied, to this effect.
The issue of units/shares in Liontrust Funds may be subject to an initial charge, which will have an impact on the realisable value of the investment, particularly in the short term. Investments should always be considered as long term.
DISCLAIMER
This is a marketing communication. Before making an investment, you should read the relevant Prospectus and the Key Investor Information Document (KIID), which provide full product details including investment charges and risks. These documents can be obtained, free of charge, from www.liontrust.co.uk or direct from Liontrust. Always research your own investments. If you are not a professional investor please consult a regulated financial adviser regarding the suitability of such an investment for you and your personal circumstances.
This should not be construed as advice for investment in any product or security mentioned, an offer to buy or sell units/shares of Funds mentioned, or a solicitation to purchase securities in any company or investment product. Examples of stocks are provided for general information only to demonstrate our investment philosophy. The investment being promoted is for units in a fund, not directly in the underlying assets. It contains information and analysis that is believed to be accurate at the time of publication, but is subject to change without notice. Whilst care has been taken in compiling the content of this document, no representation or warranty, express or implied, is made by Liontrust as to its accuracy or completeness, including for external sources (which may have been used) which have not been verified. It should not be copied, forwarded, reproduced, divulged or otherwise distributed in any form whether by way of fax, email, oral or otherwise, in whole or in part without the express and prior written consent of Liontrust.
