Federal Reserve analysts have published a paper describing what they call the Twitter Financial Sentiment Index, or TFSI. The tool aims to gauge how investors and consumers feel by tracking social media posts about finances and credit markets. The Fed stresses that the document’s conclusions are tentative and preliminary.
A financial advisor can help you build a long-term investment plan.
What Is the Twitter Financial Sentiment Index?
The TFSI is a new tool in development by a group of economists at the Federal Reserve. While preliminary and, as the authors stress, still tentative, this tool measures investor sentiment and the consumer marketplace based on information gathered from social media posts.
The research is titled More than Words: Twitter Chatter and Financial Market Sentiment, written by Federal Reserve economists Travis Adams, Andrea Ajello, Diego Silva and Francisco Vazquez-Grande. It’s part of a discussion series run by the Federal Reserve in which economists explore new ideas, so this research doesn’t necessarily reflect the positions of the Federal Reserve or Board of Governors themselves.
In this case, the economists behind the TFSI wanted to explore whether “social media activity [can] carry any meaningful signal on credit and financial markets’ sentiment.” Essentially, when people post about the market online, does this accurately reflect their opinions? And furthermore, can economists pull any useful data from what is, effectively, a massive, real-time survey?
To answer that question, they turned to Twitter. The economists built a real-time sentiment index that pulled more than four million single tweets from 2007 to April 2023, searching specifically for posts that contained words and phrases pulled from financial and market dictionaries. So, for example, their system might flag a tweet with the phrase “bonds” or “assets” to include in the index.
Using a natural language processor, which is software that analyzes text for what the author intended to communicate, the index gives each tweet in its database a positive or negative flag depending on how the post talks about the market. Then, in aggregate, the index produces an overall current sentiment of the market. If most of the recent tweets are talking about the market confidently, the TFSI registers a positive sentiment. If lots of people are tweeting about selling or hoarding cash, the TFSI registers a negative sentiment.
The authors say that this binary approach of positive and negative works better than trying to assess how positive or negative a given tweet seems. Their goal isn’t to judge the strength of an individual poster’s emotions, but rather to judge the overall emotional state of the market at large. And, they say, it appears to work.
What Exactly Does the TFSI Measure?
Among other sources of data, economists rely heavily on surveys and price trends to make predictions and policy assessments. Surveys like the famous University of Michigan Consumer Sentiment Index gather data by directly asking people about their financial situation and choices. Price trends measure the current prices in a market and compare them to historic patterns to make predictions about what will happen next. In both cases, economists effectively look for massive amounts of data from which to pull trends.
The TFSI takes a similar approach. It is, in effect, an always-on survey, in which the authors look for patterns in how people talk about their finances and market issues. As with all matters related to social media, though, the question is whether this information is reliable. When people post on Twitter, does it reflect their true position? According to the economists involved, the answer is yes.
What the TFSI Reveals
“We find,” wrote the authors of the index, “that the Twitter Financial Sentiment Index (TFSI) correlates highly with corporate bond spreads and other price- and survey-based measures of financial conditions.”
The authors also state that they “document that overnight Twitter financial sentiment
helps predict next day stock market returns. Most notably, we show that the index contains information that helps forecast changes in the U.S. monetary policy stance: a deterioration in Twitter financial sentiment the day ahead of an FOMC statement release predicts the size of restrictive monetary policy shocks. Finally, we document that sentiment worsens in response to an unexpected tightening of monetary policy.”
Among other correlations, they say, the TFSI has a few key uses.
First, it is quite adept at predicting next-day stock market returns. Strong real-time sentiment tends to correlate with gains in the next 24 hours, while a negative sentiment tends to precede losses. “This fact,” the authors write, “speaks to the ability of tweeted sentiment to reflect information that will later be included in stock prices once U.S. markets open.”
Second, and of more interest to economists, the TFSI “correlates highly with market-based measures of financial sentiment.” This includes indicators like bond and corporate bond spreads, as well as survey-based metrics like the Michigan sentiment index.
Potential Uses of the TFSI
This makes the new index a potentially useful tool for monetary policymaking. Based on how people discuss monetary policy and financial sentiment, the authors suggest that the TFSI can help “predict the size of restrictive monetary policy shocks.” In other words, it can “predict the market reaction around the FOMC statement release. We also find that the TFSI worsens in response to an unexpected tightening in the policy stance.”
Essentially, it can help the U.S. central bank measure how much the economy will slow down after it reduces the money supply (typically by raising interest rates).
Of course, there are limits to even the best tools. The TFSI is a new metric, and as such its results are still preliminary. It remains to be seen whether this will remain a valuable tool, especially once social media posters can access the TFSI itself, which could create a sort of feedback loop where index results begins to influence the index’s underlying data.
And the TFSI is a linear tool. It can signal whether people feel good or bad about the market, and the strength of that general sentiment but doesn’t provide context or lateral details such as whether they feel good about some issues and negative about others.
Still, in its early applications, it looks like the TFSI might have found a use for social media after all.
Federal Reserve analysts have developed a tool for gauging investor and market sentiment around the bank’s policies and pronouncements. Called the Twitter Financial Sentiment Index, it measures the economy by listening to millions of tweets. According to the paper, the tool “helps predict the size of restrictive monetary policy surprises, while it is uninformative on the size of easing shocks,” when the FOMC eases its federal funds rate.
- A financial advisor can help you build a comprehensive investing plan. Finding a financial advisor doesn’t have to be hard. SmartAsset’s free tool matches you with up to three vetted financial advisors who serve your area, and you can have a free introductory call with your advisor matches to decide which one you feel is right for you. If you’re ready to find an advisor who can help you achieve your financial goals, get started now.
- Investors use a wide range of metrics and indices to make their decisions. Whether you’re buying assets that you’ll hold for years to come or looking to make a profit day trading, it’s worth familiarizing yourself with some of the most common, like the Consumer Price Index.
Photo credit: ©iStock.com/Douglas Rissing, ©iStock.com/Stockfoo, ©iStock.com/Caner CIFTCI
Eric Reed is a freelance journalist who specializes in economics, policy and global issues, with substantial coverage of finance and personal finance. He has contributed to outlets including The Street, CNBC, Glassdoor and Consumer Reports. Eric’s work focuses on the human impact of abstract issues, emphasizing analytical journalism that helps readers more fully understand their world and their money. He has reported from more than a dozen countries, with datelines that include Sao Paolo, Brazil; Phnom Penh, Cambodia; and Athens, Greece. A former attorney, before becoming a journalist Eric worked in securities litigation and white collar criminal defense with a pro bono specialty in human trafficking issues. He graduated from the University of Michigan Law School and can be found any given Saturday in the fall cheering on his Wolverines.