Products
Media Analysis
LSEG MarketPsych Analytics
Rich sentiment data tracking news and social media in real time
LSEG MarketPsych ESG Analytics
ESG data designed to drive better investment decisions
LSEG MarketPsych Transcript Analytics
Earnings calls analytics for investing, research and risk management
Prediction and Monitoring
StarMine Media Sentiment Model
30-day prediction model for global equities using sentiment
MarketPsych Event Pulse
An AI-driven investment news feed
Advanced Solutions
Scoring Engine for Natural Text
An NLP-as-a-service platform for financial text analytics
March 06, 2016

The Trump Effect, Media Attention, and Stock Price Patterns

Bill Miller commenting on our new book, Trading on Sentiment: The Power of Minds Over Markets:

Human nature, the cycles of fear and greed, are unchanging and are the most fertile ground of sustainable competitive advantage. Until now, there has been no comprehensive guide to this almost inexhaustible source of potential profits.  Dr Richard Peterson is one of the world’s experts on human behavior in capital markets and he has written by far the best book to appear on the subject.  This marvelous book should be read by every serious investor.  It will pay dividends for years to come.
~ Bill Miller, Chairman and CIO of LMM LLC

Talks and webinars describing Trading on Sentiment:
1.  CJ Liu is giving a talk on currency-based sentiment strategies at the Sentiment Analysis conference in Singapore on March 8, 2016. 
2.  Dr. Peterson giving a free Thomson Reuters-hosted webinar on March 15th at 14:00-15:30 UTC & GMT,9:00-10:30am ET.  Registration here.
3.  Dr. Peterson is giving a talk for the CFA Institute in Los Angeles on Wednesday March 16th, 2016.
4.  Dr. Peterson is giving a talk for the CFA Institute in San Francisco on Wednesday March 23rd, 2016.
5.  Dr. Peterson is giving a talk in Budapest at the Global Derivatives conference on May 9th, 2016.


The Paris Hilton Effect

There's no such thing as bad publicity.
~ Associated with P.T. Barnum, the 19th century American showman and circus owner

Before the premiere of her first reality TV show - a show called The Simple Life, set on a farm - Paris Hilton was a little-known American socialite.  Little-known, that is, until someone released an amateur sex tape of her and her then-boyfriend three weeks before the premiere of the show.  

Despite the painful inanity of The Simple Life and the crassness of releasing a sex tape 3 weeks before her show’s premiere, Hilton became a media star and a business success who is now worth around $100 million (per Wikipedia).  

There's nobody in the world like me. I think every decade has an iconic blonde, like Marilyn Monroe or Princess Diana and, right now, I'm that icon.
~ Paris Hilton

Any publicity - even the moral outrage over a (probably deliberate) sex tape release - was good publicity for her brand.   Hilton’s strategy was later repeated by Kim Kardashian and most recently Donald Trump (minus the tape, so far).

If a strategy of grabbing media attention with morally outrageous acts boosts celebrity brands and sways voting patterns, might media attention to companies also boost stock prices? Today’s newsletter looks at the power of attention to drive stock returns.  While celebrities appear to be boosted by publicity - any type of publicity - studying the repeating effects of media attention on stock prices reveals more nuanced but similarly broad patterns over time.  But before diving into that, a quick plug for our book, launching this month!

The Popularity Effect


Academics and portfolio managers Roger Ibbotson and Thomas Idzorek recently popularized a conglomerate of stock price patterns they hypothesize are due to a Popularity Effect (written up here, from the Journal of Portfolio Management).  Their work is summarized here in a powerpoint, and it posits that in the short term momentum drives price higher as popularity increases, but over the longer term popularity weighs on stock prices.

The measures of popularity Ibbotson and Idzorek describe include share turnover (more popular companies have more turnover and, thus, liquidity), the value premium (stocks are more likely to be cheap if they are disliked), and the size premium (smaller stocks are more often overlooked and thus less popular).  In Ibbotson and Idzorek's view, the momentum effect - in which the past 3 to 12 months of a stock's price trend tends to continue into the future - is "probably not a premium but a slow market reaction to a transition state, such as a stock changing its characteristics or fundamentals over time."   According to Ibbotson and Idzorek, "momentum is not easily explained by a long-term, semi-permanent shift along one of the identified dimensions of popularity, but rather it appears to be a result of a shorter-term, self-enforcing popularity wave."

Are social phenomena outside of the financial markets also susceptible to such momentum?  This is the question facing the Republican's "Dump Trump" movement.
 

Trump Momentum

I could be a jerk and get a lot more publicity, but that's not who I am. 
~ Pete Sampras

Eventually momentum flames out.  Just as Paris Hilton’s star shone, so did it fade.  Thus far in the Republican nomination process, when Trump made an outrageous statement, his momentum towards the nomination, and his overall popularity, seemed to increase.

The media has devoted much more attention to Trump than to other candidates, in part due to the popularity of articles about his uncouth insults and bigoted statements.  It is not only the media - the public is similarly engaged, and Trump's last name is consistently the most searched of all U.S. Presidential in Google Trends over the past 12-months.  A comparison of searches for Trump (red) versus Clinton, Rubio, and Sanders is here and below.  Ted Cruz was not included because his last name was associated with too much noise ('Santa Cruz", etc...).



Trump’s insults and outbursts provoke moral outrage, but they also garner attention.  Provoking moral outrage paid off for Paris Hilton (and Kim Kardashian), and it is working for Trump so far.  As Irish comedian Brendan Behan noted:  “There's no such thing as bad publicity - except your own obituary.”  That said, a reference to his own penis size in the early minutes of this week's Republican debate, or more likely, the disavowal of his candidacy by Mitt Romney and other establishment Republicans, may have lost him supporters judging by the results of this weekend's primaries.

The popularity effect seems to bolster brands and provide momentum (e.g., Trump and Hilton), but at least in the case of stock prices, it eventually reverses and prices trend the opposite direction over annual periods (as Ibbotson and Idzorek described).  Trump experienced bursts of popularity in polls following crude public statements and actions, implying that there is also a short-term popularity effect. Each offensive statement added to his momentum. According to research by Thomson Reuters' Elijah DePalma, over short term periods the ebb and flow of social media attention significantly impacts stock prices as well.

Profiting from Social Attention


Elijah DePalma, Senior Quantitative Research Analyst at Thomson Reuters, collected the intra-day Reuters News alerts about mergers, acquisitions, and takeovers (code: MRG) for S&P 500 companies from June 2014 through May 2015. He then aggregated the relevant stock’s social media buzz (using our TRMI indexes) over a 6-minute window covering the minute of the news alert and the trailing 5-minutes. For the 1178 MRG events identified, he classified the strength of the social media buzz relative to its average level as low (966 events) or high (212 events). He classified an extreme level (49 events) of social media buzz as a subset of high buzz. The Buy-and-hold-average-return (BHAR) to be gained in the minutes before and after the MRG news event is depicted below:


The price impact of a news alert about a merger in the case of low social media buzz was low - approximately 15 basis points (0.15 percent). Higher levels of social media buzz preceded an average 40 basis point (0.40 percent) increase in the stock price. A significant average positive drift of 150 basis points (1.5 percent) occurred following the news events associated with extreme spikes in social media buzz (49 total events). The most significant price drift occurs over the 10 minutes following the news release. While this popularity effect is short term, it shows that the more attention is paid to a stock in social media when extremely positive news is released, the greater the subsequent (trade-able) price drift higher.

DePalma also has preliminary findings that the level of social media buzz around natural gas and crude oil news correlates with the price impact of relevant news over the following 1-3 hours (Crude Oil & Natural Gas).  Short-term momentum traders may benefit from buying on news events conditioned on the volume of attention being paid to them (measured via social media buzz).

Trees Falling

If a tree falls in a forest and no one is around to hear it, does it make a sound?

At MarketPsych we produce not only buzz metrics that measure how often a topic is discussed in the media about a trade-able asset, but we also extract complex meanings from text - an integral part of a process called cognitive computing.  One of our unique indexes measures all future-tense references in the media to a company’s earnings rising versus falling.  This index is called the earningsForecast index.

In order to study the predictive power of the earningsForecast TRMI, CJ Liu performed cross-sectional analysis of this index across U.S. The equity curve for the daily arbitrage is visible below.  Transaction costs were not included.  The (unrealistic) theoretical return following this strategy since 1998 is 150x.



Over 24 hours investors are typically late to jump into a stock with an extreme earningsForecast value.

Elijah DePalma also performed research on the earningsForecast index, and he found a significant attention effect.  DePalma examined all earningsForecast signals from the news about S&P 500 stocks over the period January 2012 to August 2015. He found 4,300 earningsForecast signals during this period. Following these signals, there was an average price drift over the following 15 business days. When he isolated the 1,000 signals (of the 4,300) that were associated with the lowest social media buzz, he noted significantly higher returns over that period. When a positive earningsForecast appeared in the news media with little associated social media buzz (approximately 500 samples), the stock price outperformed the overall stock market an average of 120 basis points (1.2 percent) over the next 15 days. For negative earningsForecast signals (approximately 500 samples), the outperformance was by approximately 80 basis points (0.80 percent). These results show that changes in earnings forecasts that are not noted by social media lead a 15-day drift in the stock price in the direction of the forecast, and they are improved by taking advantage of low buzz events.

DePalma's result indicates that momentum may be, as Ibbotson and Idzorek described, "a slow market reaction to a transition state."  

In summary, media attention to important events, such as merger announcements, boosts stock returns in the short-term (minutes).  However fundamental shifts reflected in changes in earnings forecasts are incorporated slowly into prices if they are not noticed, which creates momentum as others shift their attention to the change.  

Housekeeping and Closing

Democratic nations must try to find ways to starve the terrorist and the hijacker of the oxygen of publicity on which they depend.
~ Margaret Thatcher

In the current U.S. Republican nomination process, the momentum of outrageous statements has drawn attention to Trump.  And to Ted Cruz as well - his lone filibuster to shut down the U.S. government as a Senator similarly drew attention to his extreme moralistic stances. (And to be completely honest, our reporting on others' outrageous moral statements may have drawn more readership to this newsletter). The nomination of either Trump or Cruz will herald a new style of political campaigning in the U.S., and the moral shock approach is likely to be repeated going forward - it gains attention and votes, after all - unless the nomination system or media cycle themselves are transformed.  

We love to chat with our readers about their experience with psychology in the markets.  Please send us feedback on what you'd like to hear more about in this area.  We encourage you to check out our new book as well.

If you represent an institution, please contact us if you'd like to see into the mind of the market using our Thomson Reuters MarketPsych Indices to monitor real-time market psychology and macroeconomic trends for 30 currencies, 50 commodities, 130 countries, 50 equity sectors and indexes, and 8,000 global equities extracted in real-time from millions of social and news media articles daily.

Investing in interesting times,
The MarketPsych Team