“Then why do we learn this stuff?” a student asked me.
We were discussing the meaning of “alpha” in the Single Index Model during a recent semester of my Portfolio Analysis course. The students had seen the concept in several classes before, but I was explaining that identifying positive alphas consistently is difficult if not impossible. When I pointed out that we expect the alpha in a portfolio of firms to average to zero, the student was troubled. We had spent the last several weeks examining efficient frontiers and discussing the advantages of diversification in reducing risk, but like many in attendance that day, the student still wondered what good any of these theories were if they could not guarantee returns.
“Good question,” I said.
Finance can be a dense subject for undergraduate and graduate MBA students. The formulas look intimidating. The underlying mathematics, daunting. And when so many of the seminal theories are decades old, the insights can seem distant and inapplicable today. Besides, as my student wondered, what do we care about any of it if we cannot guarantee higher returns?
“We observe abnormal returns, right?” I asked my class, and most quickly agreed that we do. “Then why don’t we just predict abnormal returns and pick our investments that way?”
I try to adhere to the schedule laid out in my syllabus. I believe students benefit from being able to rely on the structure of a course. I am accustomed to working with students who have significant family and work obligations, so I want them to be able to plan so they can succeed. However, when a class veers in directions I had not anticipated, I seize the opportunity to follow my students’ interests in the subject. I am more interested in my students having a clear understanding of the material and “learning how to learn” than following a strict schedule.
I do not know what asset pricing models my students will be employing when they begin their careers. I do not know exactly how they will immunize their bond portfolios. However, I try to ensure that they think critically about the models they are employing and that they consider advantages as well as limitations of whichever model we are examining. During my Ph.D. coursework in 2017, I guest lectured during a session of an Investments course. Afterward, a student asked me a question. During a recent job interview, he had been given a scenario in which the Capital Asset Pricing Model indicated a stock was undervalued, but analysts suggested it was overvalued. He was asked whether an investor should long or short the stock. He was unsure how to answer, and he asked if I knew what the “right” answer was.
I keep my answer to him in mind when I prepare and conduct every lecture. Unfortunately, I told him, I was not sure there was a “right” answer. I explained to him that I believed he was being given an opportunity to show that he understood the advantages and disadvantages of the CAPM and analyst forecasts. I try to make critical analysis the goal, not just memorizing the formulas.
In typical class meetings, I expect my students to spend more time working problems than I do. I recognize that these active learning approaches — in which students engage in class discussions or individual or group work — can be jarring, and even frustrating. However, recent research by Deslauriers, McCarty, Miller, Callaghan, and Kestin (2019) has affirmed my belief that students learn best when actively engaged with the material, whether they realize it or not. That same research found that students thought they learned more from lectures but performed better on exams of the material after classes that used active learning methods. Although I am willing to sacrifice points in my evaluations for improved student outcomes, my own students seem to recognize the benefits of my methods. My student evaluations have remained consistently high even as my classroom approach has emphasized more active participation and engagement.
I explain my motivation to my students, and I am vigilant to ensure that mistakes are not only allowed but also encouraged. I ask students to help each other as much as I help them. I ensure that these conversations are focused on exploration and understanding, and that they never feel like hazing. Being active participants in their education allows students to find their own examples and make their own connections, which I believe is critical to ensuring that all students feel connected and able to succeed. It also ensures that students are learning to communicate their knowledge, both orally and in written form even in subjects that are not traditionally associated with verbal skills.
I have attempted to incorporate the active learning ethos into all of my assignments, even when in-class meetings are not possible. Exams are open-ended and designed to elicit evidence of understanding beyond memorization. Grading is more labor-intensive, but I believe the benefit for my students is well worth my effort and more indicative of their abilities. Each of my classes also incorporates assignments that involve real company data and written reports designed to give students hands-on experience working with and presenting data. I have led several labs attached to the University of Texas at El Paso’s Managerial Finance and Portfolio Analysis courses, which take students through the process of conducting a sales forecast or estimating returns and risk to portfolios. We use data from WRDS, financial sites, annual reports, and Finra to get hands-on experience with Excel, as well as presenting data after it has been collected. These courses had in-classroom hours associated with them; however, the projects are video-recorded and accessible at all times and translated well to online-only courses in late spring and fall 2020. This experience served me well during the restrictions necessitated by the coronavirus pandemic, as I shifted the informational portions of my lectures to pre-recorded videos that could be delivered asynchronously. However, I also maintained optional livestreamed discussion sessions in which the students could pose questions about assignments and we worked through problems together. These were also recorded to allow any students experiencing difficulty logging on at the scheduled times access to the same instruction and experience. As I have worked through this difficult time, I have begun to incorporate some of my classroom engagement tactics in online-only settings and hope to have an even more active-learning environment online once students are able to more reliably and regularly engage synchronously.
Back in my in-person Portfolio Analysis course, my class was still pondering my abnormal returns question. I have a list of five to eight randomly selected students to whom I pose questions each class. The questions are typically based on the assigned readings up through that session, and often involve coming to the board to work problems and getting help from their peers. I use it as a commitment mechanism for students to keep up with the reading, and I incorporate a participation element into my grading. I posed the question directly to the next student on my list for the day.
“So why don’t we just predict abnormal returns and construct our portfolios that way?”
After a moment, she asked, “Can we predict abnormal returns?”
“Well, I can’t,” I said. “If I could, I wouldn’t be teaching, would I?”
I had not planned to address the efficient market hypothesis during this class, but I realized my student’s off-handed question was a way to connect another concept and explain some of the limitations of the model we were scheduled to discuss. I explained that profitable individual stock-picking, while not impossible, is very difficult to do consistently well. Before pursuing my Ph.D., I worked in media as a writer, editor and designer, and I am comfortable incorporating many methods of conveying information. For this example, I took to the chalkboard to draw the past price for two assets, one which had steadily declined and the other which had steadily increased. I asked the class which one I could expect to increase, assuming the market had priced these two assets correctly. Some guessed the past winner would continue to gain. Others guessed the past loser was due for a reversal. I asked follow-up questions with each guess. “If we think the price is going to go up, what is keeping it from being higher already?” “If the stock is undervalued and the price is ready to reverse, what happens when investors start to buy it up?”
The class soon concluded that we could not tell whether either stock would rise or fall if all available information was correctly priced into the market.
Tying it back to abnormal returns and the Single Index Model, I said, if we looked at the past returns we might detect negative alpha for the asset that had declined and positive alpha for the asset that had climbed. “But,” now focusing on the student who had posed the original question, “what can we say about future abnormal returns based on where prices are now?”
“Um, …” he paused. “Nothing?”
“Right!” I said. “To answer your question, we ‘learn this stuff’ so we know what models can tell us and what they can’t. And so that maybe one day one of you can figure out what we’re missing and can beat the market. … If you do, email me.”