The technical workings of the financial market are already esoteric by themselves, but compounding that complexity are the underlying social phenomena that shape decision-making in markets and organizations. Yet, due to the tangible ripples emanating from market activities, understanding these social phenomena is a crucial endeavor.
It is this endeavor that Abdullah Shahid probes. Shahid is a doctoral student in the Department of Sociology, having previously received bachelor’s and master’s degrees in Bangladesh and acquired industry experience in corporate consulting and banking, micro-finance, and social impact evaluation. Shahid currently has three research projects that proceed in two directions: the social aspect that influences valuation and innovations in the financial market’s legal language.
Stock Markets as Social Institutions
Shahid’s work on social valuation emerged from the backdrop of anomalies found in market outcomes that defy the efficient market hypothesis—the idea that markets, given time, will always adjust to the most rational and efficient outcome. This idea stems from the modern trend of rationalizing all aspects of life by valuating, or assigning value, to things. In the financial market, this translates to the attempt to valuate financial assets based on its market risks and make decisions based on such risk valuations.
However, the persistent, systematic anomalies observed in what should have been rational markets begged a more nuanced explanation. Many have been put forth, from norms to neurosciences.
Shahid concentrates on social institutions. “By institutions I mean broader social settings that people are situated in,” says Shahid. “They can be schools, colleges, parents, culture, and such that in some way constrain our behavior. That’s where I’m coming from: how institutions shape valuations in society.” For context, currently he is analyzing the stock markets, but in the future, he hopes to extend his work to studying other social contexts, such as riots and genocides, which is motivated by his past consulting work in Rwanda.
Shahid has two projects on social valuation. The first, in collaboration with Professor Rajib Hasan (University of Houston, Clear Lake Campus), is the study of what he calls experts’ limited attention. The subjects of this study are sell-side analysts of investment firms. These analysts act as interpreters of market events and provide recommendations to clients in buying or selling stocks.
Unlike many previous research approaches that suspect the integrity of analysts, Shahid contends that the analysts are more constrained by their limited ability to fully analyze all the relevant aspects of the stock market.
“There have been so many regulations around these analysts that it’s very difficult to [please higher-ups and, therefore, be biased],” says Shahid. “Instead, the folks that I interview reveal that the biggest inefficiencies come from having so much work in a day.” This is due to fierce competition among investment firms, forcing analysts to interpret market events quickly regardless of how numerous they may be.
Documenting the struggle of overworked analysts is not enough. Shahid has to show that stock markets do remain irrational over time because of this context-bound constraint of limited attention. To do so, he studies a 13-year period of market recommendations produced by a pool of about 10,000 analysts from 5,136 firms. He compares each recommendation with the actual performance of that industry, taking into consideration the amount of work needed to be done by the analyst during that period. “It was a nightmare [to study the data],” Shahid says, but the results have been promising, lending credence to the limited attention theory.
Status and Irrational Market Outcomes
For Shahid’s second project—in collaboration with David Strang, Sociology, and Hasan—he studies the effect of status association in causing irrational market outcomes. Status association is the social phenomenon whereby a person’s or party’s abilities are overrated or underrated by others because of that person’s association with a more capable person or party.
“We also found that even though your quality of recommendations got worse over time, people still take your recommendations more seriously because your colleague or company received awards.”
According to Shahid, in the context of the stock markets, analysts generally have three different sources of status associated with them. One source is the firm’s name for which they work. Another source stems from having been awarded for doing an excellent job. And the third comes from an association with colleagues who have been awarded.
To incorporate these three sources in his study, Shahid tracks all the awards given to Wall Street analysts, such as those given by industrial magazines. His sources extend back into the 1970s. He constructs measures by comparing the number of awards an analyst received and that of his colleagues who worked in completely different industries.
“We found that if you are associated with certain investment banks or your colleague got an award, people actually react more to your recommendations,” says Shahid. “But that’s not the end. We also found that even though your quality of recommendations got worse over time, people still take your recommendations more seriously, because your colleague or company received awards.”
Shahid compares this superstitious learning to that often found in rural villages, evoking examples of magic amulets or religious leaders who are wrongly but persistently associated with healing miracles.
Shahid’s third project tackles legal language in the financial industry. In collaboration with the Cornell Institute for the Social Sciences' Creativity, Innovation, and Entrepreneurship project, Shahid asks whether our legal language might affect the way we value assets. He collaborates with the legal language team: Diane Burton, Industrial and Labor Relations; Charles Whitehead, Law; Fedor Dokshin, Sociology PhD student, and Strang. His focus is the collapse of the United States housing market. He collects and reads all the mortgage-backed securities documents available since 2003 in the United States and tracks down the lawyers involved in drafting the securities.
Although the project is still at an early stage, Shahid has already found intriguing patterns. He says, “Even though the products have different levels of risks, the lawyers keep on copying and pasting the same agreements, simply because they are issued by the same company. They pay no attention to the content of the product.” He gives the analogy of a food company that manufactures ketchup and soy sauce and applies the same label to both, expecting no one to actually read the labels.
In order to study the treasure-trove of legal agreements, Shahid utilizes machine learning to convert texts into analyzable data and to prevent subjective reading by humans. He appreciates this ability at Cornell to easily transcend traditional boundaries of scientific fields of study.
“That’s the best thing about Cornell: you can be interested in neuroscience and just go sit in a neuroscience class. Professors welcome you! And people come from so many different backgrounds that the types of questions people ask are amazing. It makes you go, ‘I didn’t think about that!’” He credits the university for accepting his more applied background, which is unlike more traditional researchers in his field.
“If—and I’m not exaggerating—but if I had to start over… I would still have chosen Cornell. If I am ever curious about something, at Cornell I can meet that curiosity. It gives me that power,” he says.