Most of what we highlight in our blogs is extracting financial insights from text at scale: including trends in market share, pricing, and macro. We want to send you into the Christmas holiday with something different: our “Deception”analysis.

We used the Amenity Viewer to find companies where fundamentals appear solid but may have cracks beneath the surface. We did this by looking for both:

1. A positive and improving overall Amenity Score (standard fundamental analysis) from its most recent earnings call
2. An increase in events captured by our Deception model during that call

One company that our Deception analysis uncovered was Merck (MRK), where the company appeared unwilling to endorse some bullish analyst projections.

But before before going into the details and deeper into the world of our Deception analysis, we wanted to provide the proper context behind Amenity’s modeling to better encapsulate how incorporating our Deception model into your analysis can uncover valuable insights that are virtually impossible for a person to uncover without a text analytics platform.

Amenity Viewer's Deception Analysis Explained

Reading the body language of corporate management is a central part of the role of a fundamental analyst. It is vital in a face-to-face meeting, but what if you could get body language from text? On top of the hundreds of financial event classifications captured in Amenity Analytics’ core NLP model, the Text AI identifies linguistic patterns (“Events”) under the key driver, Deception, that may indicate evasive language, attempting to spin a negative, tension with the analyst community, etc.

Amenity Viewer's Deception Analysis in Action: Merck

The facts

Merck’s 3Q2018 earnings call registered an Amenity Score of 49, up from 31 the prior quarter.  The stock is also up 5% since then.  The Viewer also highlights Deception as the #1 Key Driver in 3Q, despite the overall positive trends in the fundamental commentary:

What do we make of this pattern?

1. First, we flag that our model captured the highest # of events in our “Deception” category in the same quarter as Merck’s highest overall Amenity Score, which measures standard financial analysis. That is the first trigger to dig deeper.
2. As with all things at Amenity, we view the score as just the starting point.  The real differentiation comes in the details: what topics were attached to these “Deception” events?
3. We see some of the non-answers relating to drug trials, regulatory studies, etc. That is completely logical and is exactly why digging beneath the score is important to cut through this noise.
4. Then we see a cluster of 3 events relating to the same topic: prospect for margin expansion.

Merck (10/25):

• "We see the growth of certain parts of our business like Oncology as being very helpful to driving that kind of margin expansion. But as it relates to being more specific and putting numbers around that and time periods at this time, we don’t know that that’s the right thing for us to do."
• "I don’t really think it makes sense for us today as we’re seeing this business grow and expand for us to try to say a specific number in terms of what the margin expansion will be. What I can tell you is that we will continue to work very hard to drive that margin expansion at the same time, making the right kinds of investments that we need to make to drive our growth over the next few years."
• "I don’t want to comment specifically on how we look relative to consensus."

What does this all mean?

It’s not data for the sake of data. By connecting the linguistic patterns to the underlying fundamental analysis, we can deduce there may be downside risk to the magnitude and timing of future margin expansion relative to consensus expectations.

‍

Join the Amenity Viewer Beta Program today to analyze earnings call transcriptions and enable you to spot outliers, identify critical insights, and understand key drivers.

‍

This communication does not represent investment advice. Transcript text provided by S&P Global Market Intelligence.

Copyright ©2019. All rights reserved.