We are accustomed to splitting trading into technical and fundamental buckets. Both involve crunching data; one set includes market fundamentals and the other pure price data. Alternative data is a third bucket that is gaining traction.
The new paradigm of investment information, the stuff that makes hedge funds giddy and excited is alternative data. You probably have heard the folklore: how a hedge fund manager used satellite data to count cars in parking lots and made $75 million on his long position in a major retailer, or how private weather sensors and infrared data sensors predicted a healthy crop and higher yields weeks before the market dropped 14%.
These examples are just the anecdotal tip of an emerging iceberg of alternative data.
Alternative data refers to unique information content not previously known to financial markets, but nonetheless powerful to professional investors, according to an Opimas research paper written by Octavio Marenzi. “It is not available from traditional market data providers. By 2020, spending on alternative data and its associated infrastructure is expected to exceed $7 billion,” Marenzi noted.
Lee Ainsile, head of Maverick Capital, said, “About two years ago, Maverick started looking into alternative data sets that it had researched could help its fundamental investing process, and launched two funds internally to test it out.”
Citadel CEO Ken Griffin says, “Our ability to leverage big data effectively in our investment process is critical to our success as a firm.”
Research and consulting firm Tabb Group reported that the so-called “alternative data” market was worth about $200 million in the United States in 2016 and is expected to double in four years.
Kumesh Aroomoogan, co-founder and CEO of Accern, sums it up this way, “Quant hedge funds are buying as much data as they can.”
A recent study by Greenwich Associates found that more than 60% of traditional asset managers and nearly 75% of hedge funds are already using social media — a rich source of alternative data — as part of the investment process. The study also raised an important point: Social media driven news is just one form of data professional investors look to when aiming to profit in increasingly competitive financial markets.
But before we dive into alternate data, we should start by defining the old boring stuff we use in investment research - traditional data. Traditional data is price and fundamental data that is easy to find and use in the investment process. If you can find it on Google, Bloomberg or Reuters, then it’s considered traditional. Traditional data is company-reported financial data, fundamental data, price data and derived technical and sentiment indicators.
What is Alternative Data?
Alternative data is anything that is non-traditional data and used in the investment process. This data is typically obscure and not easy to find. The hot type of alternative data in the last decade was “Expert Networks” (group of professionals that provide specialized information and research). The new novel alternative datasets are all digital and can range from credit card transactions to mobile geolocation to satellite imagery to mobile app downloads, to name a few; a perfect fit for quantitative funds.
Greenwich Associates published its Alternative Data for Alpha report finding that, “nearly 70% of investors say that real-time market data provides them an edge. But with everyone using the same data to make decisions, determining how that shared resource can be used to gain an advantage is challenging.” This has led to huge demand of alternative data by fund managers.
This data is sourced in many ways, but usually is a byproduct or exhaust from company operations or government databases, that alternative data providers have turned into tradable information.
JP Morgan’s estimate of how much the investment management industries spend on alternative data is in the $2 billion to $3 billion range, and the number is expected to have double-digit annual growth. This spend includes acquiring datasets, building Big Data technology and hiring appropriate talent.
“Alternative data is driven by the desire to understand it. You may only find one substantial piece of information in the whole data dump and you will want to see if the information correlates with traditional data,” Louis Lovas, director of solutions for OneMarketData said in The Trade. “The multifactor models, moving beyond statistical models to machine learning is used to understand if this alternative data provides insight and then opportunity for alpha. While there has been a small height curve with social media, things related to transaction and satellite data are what’s coming next. Furthermore, we are mostly seeing clients interested in a hybrid cloud data hosting model. With the ability to mix and match the cloud infrastructure between private and public we expect the hybrid model to continue to grow in 2018, especially due to compliance and regulatory reasons.”
The Greenwich Associates study reported that investors are spending $183 million per year on alternative data sources. JP Morgan estimates that when hardware and analytic costs are combined with the actual cost of the data that the market for alternative financial data is closer to $2 billion. JP Morgan’s estimate of the investment management industries spend on Big Data noted above is in line with Big Data growth in other industries.
Why is Alternative Data Important Now?
Technology moves very fast. Cloud computing evolved making big data cheap to store. Previously, you would have spent millions of dollars to store data, but now data providers can store and process this data daily and with a small budget.
Faster computers and machine-learning software provided faster data processing for the new alternative data. This has allowed many asset managers to experiment and research with historical alternative data to find predictability.
MIT Professor Erik Brynjolfsson and principal research scientist Andrew McAfee speak of the “three Vs” of Big Data — volume, velocity and variety — noting that “2.5 exabytes of data are created every day, and that number is doubling every 40 months or so. A petabyte is one quadrillion bytes, or the equivalent of about 20 million filing cabinets’ worth of text. An exabyte is 1,000 times that amount (or one billion gigabytes).”
The growth is incredible, and if institutional players aren’t completely convinced, they are at least hedging their bets. Barry Hurewitz, global chief operating officer at UBS Group Research stated in Institutional Magazine: “We are trying to develop a new culture where both types of research [traditional analyst-driven security evaluation and new innovations in investment research] are deeply respected.”
Candace Browning, Bank of America Merrill global research head, says her firm has focused on increasing its use of big data analytics, citing, for example, the adoption of external data sets like company data from job search site Glassdoor.
If everyone is not sold, it has become so ubiquitous that even those doubters out there are compelled to collect datasets to prevent the perception of being left behind. According to Risk.net, Chief Technology Officer at quantitative fund Cantab Capital Tom Howat summed up one view at a recent conference: “There’s so much hype and hokum around this idea of alternative data that companies are just hoovering up these insane datasets and bundling them all together, and it’s a waste of time, actually. It’s a waste of my time.”
While that might be a minority view, “hoovering up” disparate data does not create an actionable strategy. Mark Robertson, senior portfolio advisor, NN Investment Partners, said in the Eagle Alpha report, “Opinions and emotions expressed in online content, from news articles, through blogs, forum posts, social media such as Facebook (FB) or LinkedIn (LNKD), to tweets, can provide a sense of market sentiment that can reinforce or even anticipate fundamental indicators, thereby helping make investment decisions.”
However, this can create a paralysis by analysis quandary. The key is in sorting data to give it relevance.
“Where that is important is sourcing data sets. Once you’ve identified an investment thesis and the relevant [key performance indicators] we’re trying to measure, we start the discovery process, which is basically a due-diligence initiative, where we will look at the data ecosystem,” System2 Founder Matei Zatreanu says. “You have the target company we’re interested in, and then we try to figure out who their partners are, who their customer are, what charities they donate to, what invoicing systems they use, what technology they use for their payment system.”
Point 72’s Chief Market Intelligence Officer Matthew Granade adds, “It is a real change from how investing used to work. If you want to understand what is going on with McDonald’s (MCD), you are going to have to look at credit card transactions data, you are going to look at geolocation data, at app downloads and a handful of other things. And suddenly you are going to have a very robust picture of how McDonald’s is doing and you are not going to have to talk to McDonald’s about that.”
Where’s the Alpha?
The Masters of the Universe hedge fund managers have made fortunes using traditional fundamental data. Recently, funds are having performance issues and are searching for new strategies to get an edge. Alternative data has provided that little extra bit of information to help them outperform and get ahead of the crowd.
Quinlan & Associates’ Alternative Alpha Report states, “In a world where traditional financial information is ubiquitous and where investment analysis remains largely homogenous, we believe alternative data provides a critical avenue by which active managers can look to stay relevant.”
Alternative Data Categories
UK- based alternative data broker Eagle Alpha breaks down alternative data into 24 primary categories and adds that new categories are being created every day (see “Types of alternative data,” page 29). The list of 24 (and growing) categories can be broken down it five general buckets (see “Bucketing Alternative Data,” page 32).
One of the main buckets is sentiment analysis. A July 2017 study by WorldQuant, “Discovering the Hidden World of Alternative Data,” noted that sentiment analysis is conducted by developing algorithms to identify positive or negative keywords in statements while taking into account user demographics and the posts’ context. The analysis is useful in determining new-product perception and brand reputation.
It also highlighted the role of “IoT” technology, which includes everything from irrigation devices that automatically maintain crops by monitoring public weather information and soil conditions to wearable technology that tracks in-store traffic.
The study notes that research firm Gartner forecasts there will be 20.4 billion connected devices in use worldwide by 2020, compared with 8.4 billion expected at the end of this year. Additionally, Cisco Systems reports that traffic from wireless and mobile devices will account for nearly two-thirds of total Internet traffic by 2021. However, IoT is still relatively unknown: Just 9% of the 996 respondents in a 2016 Penton corporate survey felt they were familiar with the technology, while the majority felt there was confusion surrounding IoT.
The study concludes, “There are no bounds on the type of data that can be generated. And whether it comes from a company’s supply chain, activity on an online platform or a sensor in a device, the data creates new types of intelligence.”
Tying it all Together
The Quinlan & Associates study states that the first step in understanding alternative data is to identify the most relevant data and choose the right data source to use in the investment process.
That sounds simple, but it isn’t. If there is one thing that should stand out from our analysis it is that there are tremendous streams of different kinds of data, which is growing at an exponential rate that may or may not be useful. Being able to corral those data streams, selecting the wheat from the chaff is a complex process that is key to extracting value from alternative data. But it is necessary. The Quinlan report adds, “It is our view that active managers who choose to continue operating using traditional data and analysis techniques are facing extreme margin compression.”
Jeffries’ report on Quantifying Intuition, “Mapping the Data Science Landscape in the Hedge Fund Industry,” argues that “we have reached a tipping point — data science is now, and will increasingly be, of strategic importance to all hedge funds in the coming years. This mainstreaming of data science solutions means managers, investors and counterparties should be informed about and conversant in the opportunities and challenges offered by these technologies.”
According to Tabb Group’s latest research, quantitative hedge funds have been at the forefront of creating alpha from this type of data, but increasingly fundamental hedge funds, sell-side firms, venture capitalists and even long-only asset manager firms are looking into use cases for alternative data sets.
Tabb Group goes on to say, “Advantages of the past, such as speed, are now available to most participants, leveling the playing field and making the capture of alpha more difficult.”
Both Greenwich and Tabb Group point out that the use of this data is expanding across the buy-side. While alternate market data does not guarantee success it is a tool used across the industry, it would be dangerous to dismiss it when nearly all of your competitors are working hard to extract value from it.
Greenwich points out that nearly half of hedge funds look to data to generate trading ideas, compared to just over 30% of asset managers, who generally make investment decisions utilizing corporate access and sell-side research rather than making data-driven predictions.
They go on to state, “Alternative data will only be alternative for so long, eventually becoming a core part of any portfolio manager’s toolkit once the aforementioned roadblocks are taken down.”
Quinlan’s Alternative Alpha Report states, “In a world where traditional financial information is ubiquitous and where investment analysis remains largely homogenous, we believe alternative data provides a critical avenue by which active managers can look to stay relevant.”
The report notes that asset managers access the same information and conduct similar analysis, which makes it more difficult to find a unique edge. It states, “Markets are increasingly becoming more efficient, limiting the opportunity for managers to identify mispriced assets.”
Data is only as valuable as what you can do with it. And hedge funds, institutional trading desks and individual traders have been squeezing all the value out of traditional fundamental and technical forms of data as possible for decades now. Most of the value of traditional data streams has been mined making it extremely difficult to extract trading alpha from it. Perhaps the greatest value of alternative data is that it is new. Many strategies to extract alpha from these datasets will fail, but the alpha is there to mine for those expert enough to exploit these new streams of data.
Daniel P. Collins contributed to this story.