Mark Odendahl: Welcome back to Industries in Motion podcast, brought to you by RBC Capital Markets. My name is Mark Odendahl, and I'm head of US capital markets research here at RBC. Thank you for coming back to the podcast. As you know, this podcast is where we explore what's new and what's next in today's fast moving markets and industries to help you stay ahead of the curve. Please listen to the end of this podcast for important disclosures.
Today we have a special guest, Mike Tran. Mike is part of RBCs macro product. I'll step back and just point out that in the US we have six market strategists across US equity strategy, economics, commodities, ESG, and right now in the market macro drivers are moving the market daily, so the time is now for RBCs macro product, and it's really exciting to bring a leader among that macro team on our podcast today, Mike Tran. Now, Mike Tran is part of our commodity strategy team, but the reason he's on this podcast today is over the last year or so he's expanded his product, alongside RBC Elements, which is our in- house data strategy team, to create a new data science research product called Digital Intelligence Strategy. Mike, welcome to the podcast. Congrats on the growth of DIS, and we're all really excited to hear about it.
Mike Tran: Hi, Mark. Thanks for having me.
Mark Odendahl: Mike, we recently crossed the one year anniversary of launching Digital Intelligence Strategy, DIS, here at RBC. First off, thank you. Second, can we start to educate the listener on DIS and what exactly is the new strategy?
Mike Tran: We're really excited to have just crossed the one year mark since launching our Digital Intelligence Strategy research platform. I would argue that while building and scaling DIS has been a massive undertaking, I would also say that over the course of the past year, global markets have certainly been very fertile ground for building such a platform. Now, there's no shortage of market problems to tackle. For example, as we come out of a pandemic, the reopening trade has certainly been very topical, you add in inflationary pressures at multi- decade highs, a gummed up supply chain at historic levels, record gasoline prices and surging food prices, and on and on and on, this market has really been ripe for injecting a new, a deeper degree of innovation into traditional Wall Street research. So as you say, what is Digital Intelligence Strategy? DIS is effectively thematic macro research powered by data science and alternative data.
We work alongside RBC Elements, our in- house data science team, to really harness the power of big data to unearth actionable items and ideas by quantifying market inflection points that previously or historically were difficult to quantify. And what's most exciting is that we look across sectors, across industries and across geographies, and we collaborate with our equity sector analysts to generate ideas and write published research for our clients. If you asked how is this different than traditional research? Well, sell- side research historically was focused on forecasting the markets. And while that remains the primary objective, of course, we think that investor focus has evolved over recent years, particularly since the pandemic. For example, in my other role as an energy strategist, the focus was always to try to forecast what global energy consumption would look like two years out or five years into the future. But when the pandemic hit in March, April, May of 2020, I can tell you that investors didn't care what energy demand looked like two years from now.
They cared about what oil demand looks like two hours from now or two days from now. So unbeknownst at the time, we started rebuilding our models to scrape and count flight patterns for every major airport across the planet. We tracked traffic and traffic congestion for major cities across the globe to really try to understand how jet fuel and gasoline demand was changing in real time. So in short, we really discovered the power of nowcasting, and that has since evolved into what we call Digital Intelligence Strategy, where we use various forms of natural language processing or geospatial intelligence, and we'd like to incorporate more neural networks and other forms of big data into our workflow in our attempt to really try to understand what's happening on and to the earth in real time.
Lastly, I'll say that I think it's really important that when we ask our clients, " What do you find most valuable from sell- side research?" A lot of buzzwords come up. We often think that clients want to see an out of consensus call or a differentiated deep dive analysis. Now, all of those things are certainly true, but what do clients really, really care about? They really care about your level of conviction in a view. And is there anything more powerful than being able to stand in front a group of clients and share with them that the conviction to noise ratio is extremely high because we can see how various components in their respective markets are unfolding in real time.
Mark Odendahl: Thank you, Mike. And that nowcasting example I think hits home with a lot of people, so I appreciate that description of the product and agree that has led to many of the successful products that you had out over the last year. Could you elaborate on maybe one case or just in general how DIS is helping our clients?
Mike Tran: Sure. If you think about our guiding principle, which is to try to tackle the biggest problems in the market that historically were difficult to quantify, what are they? So I think the key idea here, Mark, is instead of thinking about what are the themes, it's more about thinking about the themes that are changing and how fast they're changing. For example, the consumer is strong until it's not, or supply chains are gummed up until they're not. So we really care about assessing the market inflection points. What we've done is we've built out many regularly updated indices to really help us understand market inflection points. For example, a major theme over the course of the past year has been understanding the reopening post pandemic and how societal behavior is changing in real time. Now, we have created our GOAT and our GOAL indices, this stands for our get out and travel and our get out and live indices, which really measure various components of consumer or societal mobility.
These indices would update every Wednesday ahead of the US market open on a webpage called the Next Generation Elements app. Now, as you can imagine, this metric was closely followed throughout the Delta and through the Omicron waves to really understand consumer mobility. Then this summer as concerns of COVID tapered, but inflation surged, many investors or many of our clients continue to return to this website every Wednesday morning as they used these indices as a proxy for how consumers were reacting to travel inflation and what demand elasticity looks like. If you think about another major theme that we've really tackled, it's the idea of supply chains, and these have been one of the major global storylines of the past year.
And while many market participants use anecdotes to try to understand the degree of supply chain and port chaos, i. e, many chartered helicopters to snap pictures of the ship backlog at the ports of, call it, LA and Long Beach, we use scores and scores of geospatial intelligence to really build a metric called the time of turnaround, or TOT, to understand the rate of change for how all of the major ports in the world that really matter, how those have changed and whether the ports were improving or getting worse in real time. Now, I'll give you a couple more examples that are really relevant in this space. Food has been a major sticking point for the economy.
Now, a number of months ago we wrote on food, food inflation, food scarcity. We also launched what we call our food input index that tracks many different input costs that go into food and grocery. Logic would suggest that if food input costs fall as they have over recent months, then grocery prices should also fall. So this is a big leading indicator for how we think about the direction of grocery pricing and the impact on the consumer. Lastly, what I'll highlight is a use case in energy. Now, gasoline prices hit record highs in North America over the summer and we were able to use several layers of alternative data to better understand demand elasticity. So what do I mean here?
The first layer of alternative data was that we used geospatial analytics to draw areas of interest around 135, 000 gas stations to monitor movement in real time. The second is we use connected car data, this is IOT data, to monitor how clusters of vehicles were moving throughout each of the identified gas stations so that we know how many cars were pulling up to the gas station to fill up, again, in real time. The third was that we were able to overlay gasoline expenditure data to understand how much people were spending during each gas station visit. By overlaying all three of these data sets, we were able to augment a view of how consumer behavior was changing and various price points for retail gasoline. Now, we estimate that gasoline demand destruction came in at about 7. 9% over the course of the summer because we saw record pump prices, but we also think that demand is fairly elastic given that we've already seen a meaningful rebound over recent weeks as gasoline prices have eased.
Mark Odendahl: Mike, that's great, and just some wonderful examples here how we're helping clients. Could you dig into some of the tools that you look to use and that you have used to develop this DIS project?
Mike Tran: We use multiple different layers of big data, alternative data to really help to shape our views. And like I mentioned, we work really closely with our great partners at RBC Elements, which, again, is our in- house data science team, where we work with them to curate mountains and mountains and mountains of data to try to understand what potential answers we could use to shape how we think about major problems in the global market. So we use a lot of geospatial analytics. This is really interesting in terms of understanding. This is cell phone pings, this is connected vehicle data, car data to try to understand how clusters of people are moving around in real time. Naturally, this gives a very good signal to noise ratio. And what I mean by that is, for example, if we're trying to understand the health of the consumer, one exercise that we've done over the course of the past year is we drew areas of interest around several dozen of the biggest shopping centers in America.
How many cell phone pings are going through each of these shopping centers in real time? Now, naturally, if there's an above average amount going through these shopping centers, you would be able to suggest that, well, the retail or consumers continue to be quite strong, and vice versa if you see a significant dip. So this is how we use geospatial analytics. Another example of some of the tools that we use are natural language processing. We write algorithms to funnel through hundreds and thousands of different documents to pull out different keywords to try to understand or augment how we should think about a view or which keyword searches are really important in terms of shaping our view for specific sectors or specific markets using neural networks to really augment how historically we've built models and as these models, as these neural networks take in more and more data in real time, they get smarter over time in terms of producing their output.
So our models effectively become more intelligent with more and more data that we take in. And like I mentioned, we're in very much an unprecedented period right now in terms of global markets. And so at that point I'll often ask and I'll say, " Look, if the world is changing so quickly and the world is moving in such unprecedented fashion, why are we on Wall Street still trying to solve evolving problems with the same tools and models that we used prior to the pandemic?" So our nowcasting work is really dedicated to helping clients call inflection points in a fast moving market. We're not economists, we're not equity analysts, but instead our goal is really to try to provide the additional nowcasting lens or what we like to think is an additional layer of intelligence into modeling traditional financial markets.
Mark Odendahl: I thought one of the big products this year was you able to take a look at a major blind spot for the market and the economy as it relates to global supply chain. As you've said throughout this conversation, you've talked about real time data led, could you walk us through how you were able to accurately predict that these supply chain or logistics issues that we were suffering from in the US and how you were able to predict them taking longer to clean up?
Mike Tran: Needless to say, like you suggest, Mark, global supply chains, port congestion was really one of our most popular focus points of the past year. The client engagement was very, very strong. Now, as a recap, what we did was we used geospatial analytics to draw geofences around nearly two dozen of the world's largest and most important shipping ports. Additionally, we're able to monitor ship transponders for all of the container ships across the world. Now, by layering those two types of alternative data, we're able to track when each ship would enter our geofence at each port. And as such, we were able to really measure when a ship would enter our geofence at each port, discharge, and then exit our geoboundary as well. So in short, we're able to measure the rate of change at which the logistical chaos was either improving or getting worse at each port across the world.
Now, in short, global supply chains are much improved and we anticipate global port congestion to normalize, although in a non- linear way, by early to mid- 2023. Now, as mentioned, we created our port time of turnaround metric, which, again, measures port congestion at a couple of dozen of the most influential ports. Now, arguably, the most important port that has gotten the most airplay has been the ports of Los Angeles and Long Beach. Now, these two ports have improved a ton from peak congestion, but what we're seeing is the ports still struggle to return to pre- COVID levels of efficiency. Currently what we're seeing is the time it takes to turn around one ship at the port of LA and Long Beach is clocking in at 5. 7 days. This is improved, or down, from 8. 2 days at the worst seen late last year, earlier this year, but the pre- COVID efficiency levels for the Port of LA and Long Beach are about 3. 5 days to turn a ship around our geoboundary.
So we're still a ways away from there. What we've been saying the entire time is that the worst in supply chains peaked around the turn of the calendar a year, and prices are certainly off the highs, but shipping is still about three times higher than pre- COVID levels. The bottom line here is there's a fatter tail in getting back to normal. The simple rule thumb of supply chains normalizing as we shift from spending on goods to services was a very simple framework and proved to not be true. So the time of turnaround metric really helped a lot of clients in terms of understanding the rate of change and whether or not things were getting better or worse with supply chains. We are on the path towards global port congestion normalizing, but it'll be a non- linear path and it'll likely be an early 2023 story, Mark.
Mark Odendahl: Thank you. So with a year on the books, how do you see this product evolving over the next year, so in year two, and in the future years?
Mike Tran: Look, it's a great question. Given how tumultuous global markets are, there certainly will not be any shortage of major market themes that will take a big data lens to try to solve. Now, in collaboration with RBC Elements, we have created what we call our next Generation Elements web portal. This is an interactive website that updates regularly with some of our big data applications that clients can really utilize and look at how data updates change on a regular basis for a variety of market teams. This is where investors can see our get out and travel and our get out and live indices, or our city level TSA functionality. Now, we have found this platform to be a superb website that clients can really engage with on a regular basis, and we're really planning on adding many more tools to that web portal to help service clients over the past year.
The last thing I'll say here is that given that we have a wide mandate that could be positioned to tackle many different corners of the market, we would all encourage all of you to reach out and connect with us on any market problems or puzzles that could be solved by data science, and we'll certainly take that back to the lab and try to tackle it. So in short, we would love to hear from clients on the most pressing and biggest market blind spots for you, the audience, and what you're really thinking about, and we'll work on publishing research that's really most relevant to all parties.
Mark Odendahl: Mike, thank you very much. This is just an excellent example of the innovation that we have continued to produce at RBC Capital Markets. It shows a tremendous intellectual curiosity on your part to expand on these RBC Elements data sets that we collected during the pandemic and to create this new project. So congratulations on a year with this product, and we're all very excited about the years to come. Thanks for joining the podcast today.
Mike Tran: Thanks, Mark. Such a great opportunity and a pleasure to be with you today.
Mark Odendahl: Thank you very much. It was exciting to hear about it and your plans going forward. What else lies ahead in today's ever evolving markets and industries, we'll be keeping track right here on Industries in Motion. Until then, thank you for joining us on this episode recorded September 19th, 2022. Please be sure to subscribe to Industries in Motion wherever you listen to your podcast. Like Mike said, if you're interested in continuing this conversation or interested in more information about DIS, please contact your RBC representative directly. You also can visit our website at www. rbccm.com\ industriesinmotion for further insights. Thank you for your time today and look forward to talking to you in the future.
Speaker 3: This content is based on information available at the time it was recorded and is for informational purposes only. It's not an offer to buy or sell or a solicitation, and no recommendations are implied. It is outside the scope of this communication to consider whether it is suitable for you and your financial objectives. For disclosures, please visit www. rbccm. com/ disclosure.