Although financial markets tend to be explained largely in quantitative terms, human behavior still plays a major role in driving price action, says Sheba Jafari, head of technical analysis for Goldman Sachs’ Securities Division. Jafari, who looks at historical patterns to predict movements in markets, explains: “In my opinion, the mere fact that we have the existence of [asset] bubbles indicates that markets are still run by emotions -- fear, greed and hope.” Also in the episode, Jafari discusses the impact of AI and machine learning on trading decisions and her own unlikely path from film studies to finance.
This is an unofficial transcript meant for reference. Accuracy is not guaranteed.
This is exchanges Goldman Sachs, where we discuss developments currently shaping markets industries in the global economy, objects, Ewart, Global, have corporate communications here at the farm. The question of today's episode is: does human behavior move the markets? to answer that question were joined by she budget. Far ahead of the global macro technical analysis team in our securest vision. She but welcome to the programme. Thank you so much for having me so she would give us a little context on how markets would behave if there were completely rational. immediately and efficiently priced in every event or for starters, I would be out of a job. So the question really just calls upon the textbook old school theory of efficient market hypothesis and inefficient markka hypothesis. The claim is that all investors informed, rational decisions. You and I would way out events and data points identically, and
neatly adjuster portfolios. Accordingly, market movements aren't Early random and unpredictable prices fully reflect all available information and there is really no arbitrage. Profits remained in this perfect world. The economy distributes all resources in an optimal manner, so there's no alternative allocation, which can make one person better off without me, another person worse off. You studied this a bit yeah. What does the miracle Evans suggest about the efficient market hypothesis, our markets completely rational. I love this quote by kings. it can remain an irrational longer than you can remain solvent there really to problems with the efficient market hypothesis. Firstly, the fact that information is disseminated equally to every market prices and, secondly, and most importantly, that you and I will process that information in a similar, rational manner the same every time so will tackle the first problem, Leon
out of computers and the internet access information. All of that has helped to improve the speed by which we access information and that's efficiency, enhancement of no doubt about it, but on the contrary, technology doesn't always make markets more rational. In fact, I would argue that wild traditional inefficiencies of the past, such as delay, and access to information are diminishing were getting new types of inefficiencies. So let me give an example: We don't have the internet thirty years ago now we have probably far too many sources of information. These provide contradictory truce liable data points, noise and, as a result, information is asymmetrical, an imperfect not to mention
There's a number of structural constraints that impact agents and create pockets of inefficiencies, and I can be as basic as your accounting constraints. Regulations on one investors need to closer books at the end of the month or at the end of the quarter at the end of the year, and they will tend to trigger outsize moves people optimizing for different time horizons, different mandates, liquidity constraints. The list goes on and the way that I see it and if we kind of trend lay the question on its head, there are ancient and then there's aggregates of those agents The question is whether the agents are rational. I've, no doubt that you and I are rational human beings. The question is really whether or not the aggregates of those agents. The marketplace itself is irrational entity. Give us a couple examples of how human rights Two different events has an impact on trading decisions.
a very normal everyday analogy. Have you tried to make a restaurant looking in New York recently? I have so. I have about three ups on my phone. At the moment there is about twenty five thousand restaurants. In New York City in Manhattan era, and you would think that we wouldn't have a problem getting into a restaurant on a Saturday night problem. As I don't wanna go to any restaurant, I want to go to a restaurant that has a vibe that has a really good chef that has a good reputation, Let's imagine that you walked into a restaurant, that you made a to it's its enormous restaurant restaurant and it's completely empty others. Five waiters waiting to wait on you chances are you probably gonna turn around. quite right. There's something wrong with this wise yeah. Now of you. If you take the opposite, you're walking down the street you're normal neighbourhood and there's a new restaurant, that's just popped up and there's a cue out the door down the street and around the corner.
That would be intriguing to you, my need one or no. What is it about this restaurant that enticing people to want await an hour or even two hours, In my opinion, the mere fact that we have the existence of bubbles indicates that markets are still run by emotions, fear greed and hope, and just to name a few had the dot com bauble. We had housing, crisis, silver bubble, asian financial crisis, and just last year, Bitcoin and there are a number of reasons why a bubble can develop. In essence, it's a narrative whether its rationally irrational doesn't matter triggers a human and machine driven reaction which will impact the way there, one looks at their trading decisions and the other thing that I would mention about bubbles which a lot of times when people have this discussion, they don't really bring up this topic is that at the onset of a bubble, the perfectly rational thing to do is to. trade. The bubble right up, I mean just the triggers: do not stay longer than you
to greater full of very high, but let's be honest in hindsight, we would have all wanted to buy Bitcoin and he doesn't fifty right so connect that dynamic of bubbles or crowding, as it were, to the work you here at home in the technical analysis, what's the free Mark you use when you're lucky markets and what are you trying to figure out and predict about how human behavior makes the markets less efficient than they would be otherwise will sort of a technical analysis? The gold technical analysis is really to describe the aggregate behaviour of the market and the assumption behind technical analysis is that markets
oh specific patterns, and that those patterns are not only identifiable, but they can be predictable. The framework that I use personally is called Elliot wave theory, and there are two assumptions being made here. On top of this ended premise: behind technical analysis on Elliot Wave theory, the markets are suggested to move in a fractal manner and they respect the laws of the golden mean or so if a well defined ratios with respect to the fractals, if you ever read or heard about the professor in one Mendel brow, he demonstrated that there are natural forms in life than the world that are self similar at wave in the ocean is an all too different and structure pattern composition, whether it's a hundred
here's hired a couple inches branches of a tree, likely small versions of the tree itself, so Elliot waste theory put forth that similar to nature major bull market structures are no different than short term swings and daily fluctuations in the second assumption is even more interesting and I would argue, even more beautiful. It's discounts that the golden mean is everywhere, so the golden mean as this concept that we learned about in successive class some years ago and its basically a single ratio? Six when aid, specifically that defines the shape of rural pools ocean waves, anything a small, the growth of bacteria to the formation of the galaxies, the mapping of human life, the population, growth and effects. Alleyways theory suggests that the same law that shapes living creatures, organisms, Galaxy
is inherent in market price action which is intrinsically driven by human behaviour, and I would argue that this is true not only for human behavior but also machine behaviour. So when I started, my career, I often say that I'm an anthropologist of the market reaction function or of humans. Just then at it simply different Burma. Now that's change over the past five to seven years. I argue that I'm an anthropologist of human and machine behaviour. It doesn't necessarily mean that its less effective, it's just changed, machines and machine learning, and I and the data have all become much bigger factors in the market, but at least at some level- and you seem to see ass, this machines are always good as what they humans put into them. Sore genes mimicking human behaviour and, as a result, they prompted the same biases.
Or can somehow the machines remove an emotional element that makes trading less efficient. We go back to what we were discussing earlier, which is at the speed at which information is distributed. With the onset of machines, we ve achieved incredible speed. Immense data analytics we can solve complex problems out. Maybe we never be able to solve ten or fifteen years ago, not to mention the liquidity that its provided, the spread compression that wouldn't be possible without machines, the irony is that
While we are ruling out older inefficiencies- and there has been an emergence of new types of inefficiencies for one There'S- been an increase in flash crushes over the past ten years. These are happening in pretty liquid market, so S MP, may of two thousand ten: U S treasuries October, two thousand fourteen sterling and October two thousand. Sixteen and fix in February last year, one theory behind this increase in flash crashes is that machines are unable to process. Complex macro surprises the way that more fundamentally inform traders might so the responses to withdraw liquidity and even sometimes switch to demanding liquidity from the market rather than supplying it see. You have. This really interesting done
like, where, during periods of complex fundamental scenarios, its humans that game the informational advantage and machines that requests- and I thought it was interesting to re personally recently- that a two thousand and seven report commissioned by the Uk Government Office of Science explains how computer based trading environment may actually reduce the diversity of trading Elgar's, making them more v herbal to hurting the crowding they're, all the same exact. Exactly given that backdrop border our clients, the car. You're, interacting with what our questions are they asking now and Elsie. Some of these clients are, quite traders, some of them are human act of managers They cover every everyone, everyone, its hedge funds, real money, corporate spanks, I'd, say some of the questions that people are we ve been in a thirty year? Bond market rally to your yields have gone from seven percent less than two percent over the course of the past ten years. Lessened as we all know that it can come that much more than zero.
Sir wanting to know what that means for bonds. What that means for equities what that means for the macro environment, and on top of that, we are seeing the stark breakdown in correlations. bonds of rallied in an almost straight lines in September of last year and as an p drop twenty percent from October two December. Twenty five percent rally from December through two April, so you're classic, Economics textbook will tell you that higher bondsmen lower yields and therefore easy money which should stimulate the economy. but we ve seen a number of instances this past few years, where those standard relationships have broken down. So, in contrast, technical analysis focuses less on those equally brim theories and more on the context of the market and the aggregation of those markets. Psychologists, I'd, say one thing: people asking about which I think is interesting as Bitcoin two years ago, a year and a half ago is probably the most popular ever ask everyone is asking about. I guess I've got one or two clients, you guys know who you are, but
back a little bit, so these are the bubble, but its back here. So I saw how did you end up in this field and technical analysis? I had no What I wanted to be when I grew up my certainly in an ever say that I wanted to be a technical analyse. Why have you noticed technical and allows us to say I didn't I what do you mean artists and I guess to some degree in drawing lines for living right. I started school and film studies I dropped out of you see, lay at the age of nineteen to move to Scotland, and I asserted setting Social anthropology there, eventually started my career in a financial data company as a salesperson, glorified call assistant. Basically and the only reason why really join that company was purely because I was told that finance pays and lay needed to fund my lifestyle in London and then Social Anthropology doesn't give surprise right, and I honestly I couldn't relate to anything in the banking world. I didn't understand it interesting, make any sense to me so
somebody studied Anthropology hadn't. You start thinking about these issues. I came across this underground debate forum that was happening at the from that we're out and it happened in the basement and one evening they had to people sitting on stage. One of them was basically debating the philosophical standpoint from the view of a fundamental economist and the other person was debating from the side of a behavioral economist, and I listened and it just clicked. I could relate to this. I could understand ass. It was the first time that I thought everything that I'd studied up to that point. Suddenly. fit into this environment in film school. They actually teach you that there are only a few narratives that exist. Every story, every movie that has existed since Shakespeare to today can fit within those eight to ten narratives and are now, if you ve, ever heard of it and anthropologists by the name of current vodka, is also writer who talks about those narratives as having ups and downs and let you couldn't gruff those stories to reveal the shape of the narrative. So imagine us,
we need girl. Boyfriends love, discover something wonderful and then loses the chart. Crashes there's a period of depression. A trial and then he makes his way back out of the trough through scepticism and pain and hurts and gets the girl back and ends are better off from the experience of that sound familiar Let us see what he was excited about, as you think about the feel that you found an event it up in what he must, Said about in the future, the machines get faster and faster, but some of the fundamentals- probably don't change. What are you Looking out when you look to the future of technical analysis, historically, technical analysis has always been an art form. Most of the practitioners have always been discretionary. One thing that we're trying to tackle a Goldman Sachs as we're trying to ask the question of how do we bring sign start? How do we start to make this more quantitative? How do we go beyond just simple back tests? I do we work towards building a robust quantitative technical platform. So one thing we recently just done as we ve hired someone is.
come on to my team and actually work primarily on building that quantitative platform when using the resources that we have and machine learning and Argos, and I'm just super excited to be a part of that transition, because markets or explained in quantitative ways. I think a lot of people, think of them as scientific. You used the word art to describe some of what you do, what parts art, what parts I answer how the inner relate, and the only thing I ever look at is a price chart and I think a lot of people think that when they speak to me that I have these algorithms formulas the background. But what I'm really doing is I'm looking at a price lame? I'm looking for patterns that have existed in the past and am overlaying the Elliot Wave and algae witches cycles. Your traditional business cycle with a slight overlay, a slight difference in the fact that it has the ratios, the golden mean ratio. That is pretty much it and it's more of an art form in the fact that you're trying to understand or determine the behaviour of markets based on those patterns
thought interpretive drawing interpreting enjoying so to wrap up the absurd. Let summarised the central question: the episode in thirty seconds or less our markets driven by human behavior. Absolutely yes, the end, user of financial markets will always be you me moms pops, the Delhi and down the street, the start up. Bro the guy mining for Crypto. There's great parallels here with software in cars, does it matter to you if you're goober is a human or a self driving cars? Probably not the key here is at the end. User is always can be human. She, but thank you for joining me today. Thank you so much for having me it's been a pleasure. That concludes this, a sort of exchanges, Goldman Sachs thanks for listening and if you enjoy the show we hope you subscribe and Apple podcast. and leave a rating or a comment and, if like to hear more from Goldman Sachs experts, as was influential policymakers, academics and investors be sure,
check out our new podcast top a minor Goldman Sachs, hosted by Allison Nathan, a senior strategist in our firms, research, division. This part gas was recorded. Aren't you six? Two thousand nineteen, all price references and market forecasts correspond to the date of this recording. This podcast should not be copied distributed, published or reproduced in whole or in part. The information contained in this package does not constitute research or recommendation from any Goldman Sachs Entity to the listener. Neither Goldman Sachs nor any of its affiliates makes any represent
or warranty as to the accuracy or completeness of the statements or any information contained in this podcast, and any liability, therefore, including in respect of direct indirect or consequential loss or damage, is expressly disclaimed. The views expressed in this podcast, or not necessarily those of Goldman Sachs and Goldman Sachs is not providing any financial, economic, legal, accounting or tax advice or recommendations in this podcast. In addition, the receipt of this podcast by any listener is not to be taken as constituting the giving of investment advice by Goldman Sachs too that listener, nor to constitute such person a client of any Goldman Sachs Entity.
Transcript generated on 2021-09-18.