Econlinguistics combines the fields of economics and linguistics. Using techniques from machine learning, natural language processing and econometrics, this emerging area of research is concerned with the economic impact of spoken and written language.
Economics is the study of the production, distribution and consumption of goods and services, which takes place in a complex system of verbal and written agreements. Econlinguistics aims at shedding light on the natural intersection of economics and linguistics. As such, econlinguistics is heavily concerned with the transmission of information between agents in an economic system.
Econlinguistics research group
The Econlinguistics Research Group (ERG) was established 2019 in order to foster interdisciplinary research with a focus on the economics of language. As such, the group consists of data scientists, financial economists and researchers from related fields. Please direct any inquires towards firstname.lastname@example.org
It is relatively easy for us humans to detect when a question we asked has not been answered - we teach this skill to a computer. Using a supervised machine learning framework on a large training set of questions and answers, we identify 1,027 trigrams that signal non-answers. We show that this glossary has economic relevance by applying it to contemporaneous stock market reactions after earnings conference calls. Our findings suggest that obstructing the flow of information leads to significantly lower cumulative abnormal stock returns and higher implied volatility. Our metric is designed to be of general applicability for Q&A situations, and hence, is capable of identifying non-answers outside the contextual domain of financial earnings conference calls.
A. Barth, S. Mansouri, F. Woebbeking and S. Zörgiebel. How to Talk down Your Stock Returns, 2019.
Available at SSRN.
We analyze how senior managements' willingness to orally convey information in earnings calls affects firms' stock returns. Using a novel "Blathering" metric for managements’ unwillingness to share precise information by ‘beating around the bush’, we show that market participants perceive imprecise information as bad news. Firms experience significantly lower cumulative abnormal returns following their earnings calls when managers blather more. This finding cannot be explained by time-constant firm characteristics or management style. Further, we investigate the motives behind blathering and observe that blathering is particularly pronounced when earnings management is more likely, when analysts' questions are tougher, and when last quarters' return on equity was poor.
Through permanent links, working with the glossaries on econlinguistics.org is as easy as:
import pandas as pd
dict = pd.read_table('https://econlinguistics.org/glossary.txt', sep=',')
glossary.txt (permanent link): This is the core glossary, containing 1,027 tokens (3-grams) with positive factor loads, which should serve most applications. Most importantly, this glossary is not specific to the contextual domain of earnings calls.