Latest / Freightvine / Andrew Sibold, IANA | Tracking the Intermodal Market

Transcript

Andrew Sibold, IANA | Tracking the Intermodal Market · Freightvine

• Chris Caplice: Hello and welcome to the Freightvine Podcast, your source for all things freight transportation. ⁓ Chris Caplice Chief Scientist at DAT Freight and Analytics, and today I'm joined by Andrew Siebold, Director of Economics for the Intermodal Association of North America, or IANA Now currently witnessing a unique market rebound where domestic intermodal volume is growing while international volume is decreasing. Andrew explains that this is a structural shift caused by rising truck rates. Hi, this is the market update for 6 August 2026. In DryVaN, we saw the change in active contract rates increase 1%. We saw spot rates ⁓ also increase about 0.4 percent. And the current level of replacement rates for dry van is 10.7 percent, which is very, very high, similar to what it was last period, actually. It's staying up there. And what that means is new contract rates are on average about 10 percent higher than the rates they're replacing, the retired rates. welcome to the Freightvine Podcast. Andrew Sibold: Hey, that's great to be here. Thanks for having me on. Chris Caplice: I've had some other people from ⁓ intermodal sides. I had Larry Gross on a couple of months ago, and that was always a fun conversation. But I'm glad you were able to make time to come in and ⁓ join the podcast. ⁓ before we get into the hard-hitting questions about intermodal, tell us a little bit about your career path to get to where you are now at IANA and what your current role is. The market gap between spot and contract for driving is 23 cents a mile, where spot is higher than contract. Year-over-year change in spot rates is 42.7%. Pretty, pretty high there. That's one of the highest year-over-year rates we've seen in a long time. change in ⁓ year-over-year for contract rates is about 14%. So as you'd expect, it's a little more muted. Temp we saw contract rates drop 1.5%. Spot rates also dropped 0.2%. dropping imports and a boom in data center construction. He actually notes something that I was surprised by that sensitive components like cooling racks are increasingly moving via intermodal because it's a much more secure option for high-value tech. He specifically noted how IANA's driver database and the interchange agreements help prevent fraud and theft at the driver level. also go into detail on IANA's latest metric, the Intermodal Volume Index or IVI. ⁓ Andrew Sibold: Okay, gotcha. Yeah, it was a bit of a of a long winding road. so I started out in the army as a armor officer. ⁓ while was there I got introduced to kind of the logistics side of the house just because, you know, tanks are kinda thirsty beast. They ⁓ they require a lot of food and fuel and ⁓ ammunition. so yeah, I started ⁓ by running kind of logistical operations while we were deployed to Eastern Europe. Chris Caplice: Yeah, yeah. Current level replacement rates for temp control is 9%. So on average, those contract rates are 9% higher than new ones over the ones that they're replacing. Market gap is 7 cents a mile, spot higher than contract. And year over year change in spot rates for temp control is 21%, ⁓ ⁓ year-over-year change in contract rates 6.8%. So roughly about half of what they are for the dry van. For intermodal, we saw contract rates increase 1.5%. Spot rates, although there's not many of them. And this modernized, de-seasonalized indicator can be used to describe the current situation across intermodal networks or within a region. It can also be used to provide a 12-month forecast using machine learning models and including macroeconomic factors like unemployment and diesel prices. Finally, we discussed that sticky nature of tariffs following the volatility of 2025. And you'll learn how the pull forward behavior from shippers has led to a pretty pronounced volume shift from the west coast toward East Coast and Gulf ports. Andrew Sibold: From there, I guess I kind of had a knack for it. but got out of the army ⁓ went to the University of Tennessee to study public policy and political science. got out from there and ⁓ worked with Federal Highway, where I was a ⁓ urban planner and ⁓ economic analyst. So I worked with a lot of the kind of freight planning efforts, a lot of the economic development efforts related to the bipartisan infrastructure law. Chris Caplice: Increase 6.1%. And the current level of replacement rates is negative 4.9%. So kind of the opposite of the rest of the other modes. New rates coming in are lower than those that are replacing. Market between spot and contract is 9 cents a mile for intermodal. And year-over-year change in spot rates is 14.8% and 6.4% for contract. Okay, finally, flatbed. We saw contract rates drop 2.2%. Spot rates also dropped 1.8%. Andrew Sibold: ⁓ ended up coming to D C and got caught up in the whole doge sweep, got promoted at the wrong time. but luckily kind of found my way to IANA and ⁓ been here ever since. ⁓ and I am Chris Caplice: So whether you're looking to understand how intermodal interacts with the trucking freight cycle or how autonomous trucking might impact the market, this conversation offers great insights into the economics of the intermodal market and the industry. following my conversation with Andrew, I'll present the truckload market update. So let's get started. So tell us a little bit about IANA. What is IANA and what is your what is your role there? 'Cause it's kind of a special role. Andrew Sibold: Yeah. Yeah. So it's the Intermodal Association of North America. basically we touch kind of everything within the Intermodal supply chain. So be that, you know, ocean liners to ⁓ to drayage trucks. not so much air, but you know, we do have like BCOs and stuff that use air components as well. and our whole job is to just kind of promote the industry. We provide a lot of data products, which is kind of my wheelhouse. I'm the ⁓ director of economics here. Chris Caplice: ⁓ replacement rates are 9.5% positive. And the market gap between spot and contract for flatbed is two cents a mile. Year-over-year change in spot rates is 23.5%. And for contract rates, it's 10.5%. Okay. active rates are mixed, right? Temp control and flatbed dropped a few points, drive-in and increased a few points. They're really kind of, they're bouncing around. And spot rates kind of followed that same thing. Andrew Sibold: And, you know, try to keep a pulse on what's going on in the in the wonderful world of freight and especially intermodal with ⁓ kind of exciting times we're living through right now. Chris Caplice: ⁓ same same sign as they did, but year over year spot rates are up across the board. ⁓ forty two percent, forty three percent for drive in that's a that's extreme. It's about half that much for flatbed and tem control. They're up like twenty percent. And even ⁓ intermodal is up ⁓ fourteen point eight percent year over year. Spot ra ⁓ contract rates are about half of each of those. So it's about half of the increase for each of the Yeah, no, it it has been. I mean, was kind of a a steady state for the last three years, probably probably end of twenty twenty two to the end of twenty-five. But then since then, twenty twenty six is definitely the cycle has started again. And so we're seeing it certainly on the truckload side. But let's let's talk about the intermodal side. You you had a ⁓ recent quote where you talked about or you you talked more about the rebound ⁓ in of intermodal in twenty twenty-six. Can you tell us more about that? Do you do you see this happening and and what Contract rates as the spot rates for each of the modes, which makes sense because it's more muted, doesn't hit hit the amplitudes that spot rates do. Replacement rates are high for everything except ⁓ you know, intermodal, like we talked about. digits almost for each of them. So this is a a bidding season. It's definitely ⁓ a seller's market, right? It's not a buyer's market, not a shipper's market anymore after three years. And ⁓ finally looking at the spot contract different, the market gap. How is it playing out? Andrew Sibold: Yeah, yeah. So as an economist, I will stay away from like the the causality, right? Because I I don't have enough data for that. But I will say that the data seems to kind of support the thesis, right? So, you know, it there's a bit of kind of unique factors going on that are kind of the stars are aligning for intermodal. So on the one side you have the the truck rates, as I'm sure you're well aware, have been going up, both due to the diesel prices. Chris Caplice: It's, you know, f anywhere from two to twenty-three cents a mile, it's still positive. So all signs are pointing that it's still a tight market. Capacity is still catching up to demand, which demand hasn't increased dramatically, but we're seeing that the market is starting to adjust and we'll see how long this lasts. But that's it for the truckload market update for sixth August, twenty twenty six. Andrew Sibold: as well as you know some of the crackdown on the labor side, kind of getting some of the the operators out of the market there. And the other big component of this is kind of the the import softness from tariffs. So a lot of intermodal obviously is international in character. and y ⁓ we a lot of lines that actually had a lot of slack in them at at almost the perfect time. So we had, you know less international coming in, but then that was able to fill with demand coming from domestic. So it was almost a a seamless transition from that, you know, Slack and international boxes coming in to to more domestic. and that's also been propped up a little bit by ⁓ almost kind of a mini boom, I want to say, in ⁓ in manufacturing. So think of like data centers, some of the the construction projects going on there. We've seen a lot of demand for domestic intermodal, which is kind of plug that gap that would normally be existing. Chris Caplice: Well let's let let's dive into that a little bit 'cause you you you you're raising a lot of things. Let's let's unpack some. One is the split between domestic and intermodal. ⁓ do you think this is just caused as a blip, a kind of a one time thingy, or do you think that's gonna be an ongoing structural difference? Andrew Sibold: So I think it will be ongoing. Whether it remains structural will kind of remain to be seen. But I definitely think for the rest of 2026 and likely into 2027, you're going to see domestic kind of lead the way for intermodal, which is is somewhat unique. Certainly since I've been here, it's the kind of the first time that's actually happened. but you know, it will kind of remain to be seen whether the the prices and kind of the labor market. support that domestic share shift ⁓ and whether I think like the consumer holds up overall is is the big wild card in all this. Chris Caplice: Yeah, can you can you give us a rough order of magnitude for numbers? How much is the domestic growing? Any any sense of that versus what the international is is dropping? Andrew Sibold: Yeah, yeah. So it used to be roughly like a city, sixty forty split in terms of international versus domestic. and you know, that would obviously vary depending on the different geographies. So, you know, with stuff coming into LA going to Chicago, that was almost all international. and then domestic was, you know, on shorter lanes. So think kind of going from Mexico to Texas, some of that reshoring, near shoring going on there. But Now we're seeing Chris Caplice: That would that would be called domestic, coming from Mexico? Andrew Sibold: Yeah, yeah, just because it's the size of the boxes is is kinda how we term it. So yeah, anything that's not in an international box is in a quote unquote domestic box, which is, you know, the fifty three foot container kind of makes that up. Chris Caplice: ⁓ okay. Okay. So so if something comes in on, you know, two forties or three forties and gets converted to two fifty threes, how is that counted? Andrew Sibold: Yeah, so for us that's counted as domestic because it it's in a domestic box. and Chris Caplice: It that it moves. So if it comes in whatever port and it goes fifty miles inland, they rip the the containers, they put ⁓ into two fifty threes. Don't don't you count three as in and for drayage and then two out as domestic? Is that how that goes? Andrew Sibold: Yeah, yeah. So for us, yeah, it's not so much about what's inside of the box, it's it's kind of the boxes themselves. ⁓ just because, you know, once it hits the rail, it's almost kind of like a black box 'cause it gets classified as freight of all kinds. So the only way we kind of have visibility of it is is based on the size of the container. Chris Caplice: Yeah. Okay. So I'm just curious because I'm gonna show my ignorance for a lot of this stuff. So in the backhaul, when the containers get sent back, so if something comes back empty, a forty gets sent to Kansas City, forty foot, and it comes back, does that count as two international moves? Andrew Sibold: Yes, but we do not count those as like revenue moves. so for most Yes, yeah. Chris Caplice: Well, the first one you would. The one that goes delivers full comes back empty, so Andrew Sibold: Yeah, yeah. So when the container's full, we count that as like a a revenue moves because it's moving, you know, full of stuff. But then the other ones are kind of repositioning, right? So it's the the empty containers making it where they need to go. And a lot of that is kind of we wash it out of the data, especially for a lot of our external audiences that aren't like intermodal practitioners. That's kind of what they care about, right? Is the full stuff, not necessarily the the technical repositioning of containers. Chris Caplice: So so which has more repositioning miles as a percentage of total miles, domestic or international? Andrew Sibold: You know, that's a good question. would say just because ⁓ I know, right? ⁓ so I don't have like exact data on that, but my intuition is that yeah, is probably international just because of the sheer number of containers. ⁓ yeah, yeah. Chris Caplice: I only ask good questions, Andrew. You should know this by now. That's all I do. One after the other. Mm-hmm. What's your intuition? Yeah. It's gotta Is there is there much coming? I know there's a lot of ballyhoo a couple of years ago about, you forty foot coming in and ag some ag companies are putting bags in and filling the forty foot to ship back out like soybeans or something. Is that still happening? Andrew Sibold: Yeah, yeah. So agricultural products is it's a major thing on ⁓ on intermodal that and ⁓ believe it or not, kind of like recycled metals. you wouldn't really think about putting those in a container, but I guess, you know, there's there's clearly a benefit there. ⁓ but yeah. ⁓ so the recycling, ironically it ⁓ Turkey is like a major metal hub. Yeah, yeah. So that's like Chris Caplice: Where are they going to? Do you know where Andrew Sibold: I think it accounts for the majority of of kind of our exports to Turkey as ⁓ is recycled metals. ⁓ so that's a fun little piece of trivia there. Chris Caplice: ⁓ that's interesting. And I know I did a project with a ⁓ with a ⁓ what I want to call fruit company years ago, years ago. And they ⁓ bring containers of bananas up from Costa Rica and whatnot, but what went down, they almost made more money on what they shipped out. And can can you guess what they were shipping back in their backhaul? Any guess? Going down to Latin America? What? No, cars, used cars. Andrew Sibold: ⁓ maybe paper? Paper? No? Cars, okay. Yeah. Chris Caplice: They'd put them in and they'd fill it up and they'd make more money on the backhaul than they did bring a banana set. Isn't that funny? I maybe that's changed. That was several years, 10, 15 years ago. ⁓ but you mentioned something else that I thought was really interesting. I didn't think that intermodal would be impacted by all the data center build out. What what is going on intermodal for that? Andrew Sibold: Interesting. Yeah. Yeah, yeah. So I think the majority of it, right, is gonna end up naturally on kind of flat cars for for rail, but you do have these kind of sensitive components that are are really expensive and intermodal is a pretty secure way to ship that. so think of like the cooling racks, the kind of you know, nuts and bolts inside of the data centers. A lot of that is moving on intermodal, ⁓ just because it's it's secure and it's kind of highly transportable. Chris Caplice: That's really interesting. I I hadn't thought of that before 'cause I always think of, you intermodal going on rail. It's a little more ⁓ slightly higher damage rate than ⁓ than like an air ride truck. But as far as theft, not not many ⁓ people on horses are stopping the trains these days to to hold up. Andrew Sibold: Yeah, yeah, I guess you don't really have the great train robbery happening much anymore. Chris Caplice: I had thought that it is I imagine that in fraud probably is lower too, I'm guessing, 'cause it's tough you have to divert the whole training. Andrew Sibold: Yeah, yeah. And intermodal it's a it's a pretty secure way to ship because so especially for what IANA does, we ⁓ the other side of our like house, if you want to call it that, is the UIAA, which is the I'm gonna test myself the uniform interchange intermodal interchange agreement. So basically since there are multiple handoffs of boxes between truck rail and then back onto truck, we ta we reduce a lot of the friction of that by having a driver database of like secure drivers who are up on insurance, you know, all the T's are crossed and I's are dotted. so when they go to pick up, it's actually like a a secure person coming to pick it up. So I think that is also a lot of the the appeal for intermodal for some of those higher end products. Chris Caplice: So so that's done at the driver level, not just the company. Andrew Sibold: Yeah, yeah, down at the driver level, ⁓ for who's both taking it to there and then picking it up on the train for that last mile delivery. Chris Caplice: I think ⁓ the recent Montgomery ruling in the Supreme Court, you know, where brokerage or third parties are equally liable if there's some damage or anything like that, is driving companies to to get more driver specific. Cause up until this point it's more like the carrier. You you focus on the carrier and then you assume the drivers, but there's more with the liability on the line. Seems like there's more interest in having that one on knowing specifically which driver and having information on. Andrew Sibold: Yeah. Yeah. Yeah. That's huge. And especially so, you know, we have like an electronic system so that when they go to pick up at the terminals they kind of get scanned in. so that's that's one of the big kind of fraud prevention things that I think Intermodal has that, you know, might not exist elsewhere. Chris Caplice: That makes sense. That makes sense. So something else that's come out recently your intermodal volume index. And this is something I don't can you tell us a little bit about that, when it was introduced, why and what what is it what is it showing? Andrew Sibold: Yes. Yeah. Yeah, yeah. So I'm sure as you're well aware, freight data is kinda hard to come by and and when you do get a good data source, it's almost as good as gold. ⁓ so we basically take our ETSO data, which is our equipment type size and ownership, which is you know what I was talking about that, you know, who owns the box and who's registered with it. we take that data and we kind of verify it with TTX, which is is one of our data partners. ⁓ and the class one railroads to kind of get a picture of what exactly is moving in intermodal specifically and you know the size of the containers, where it's going, things like that. you know, before it was kind of a a low tech operation, right? Like a lot of spreadsheets kind of getting handed off. but part of me coming on board with it with more of the economics and data science background is to, you know, almost modernize it. So The IVI is kind of our our flagship number that we want to have out there so we can kind of communicate at a glance, you know, the number is this, and that people can then kind of intuitively know, okay, that means we're up four percent based on the baseline, which is 2017 to 19. so that was kind of the thought process behind it. ⁓ and then we've built Since then, a lot of indicators and a lot of kind of prediction models that use that same base number in the prediction. So eventually we're gonna build out a whole suite of indicators based on the IVI numbers. Chris Caplice: So DAT is built on data, right? You can't build data without DAT. ⁓ and so we do a lot of stuff on this as well, looking for the the key KPIs metrics to describe data. Cause you you know this data's messy and you want to have something that gives you an indicator of direction, magnitude, whatever, all those things. So give us a little more detail on what IBI measures. It sounds initially it's a descriptive metric where it's describing is it pure volume level across the whole network, or do you do it by region? Andrew Sibold: Yeah. Chris Caplice: Do you do a proportional as an index or how how do you do that? Andrew Sibold: So Yeah, yeah. So we ⁓ currently what we release publicly is the the national number. Now for our kind of data subscribers, we can drill it down to different lanes. So we have different regions of the country that we have that we've you know arbitrarily picked. and then we can show geographically like what's moving where between these different lanes. so the IVI kind of gives a an equalizing snapshot of all of those things. ⁓ and Because Enermodal's like largely contractually based, we have a lot of seasonality in our data. So one of the big things with IBI is we have, you know, removed a lot of the seasonality using kind of the the REMA seats ⁓ adjustment that the Census Bureau does. So, you know, when we're describing a trend increasing, it's actually a trend increasing. It's not just, you know, July's our peak month. So yes. Chris Caplice: You de-seasonalize. Yeah. That let me ask you a question about that, because truckload, we'd like to think it's seasonal, but for the last, I don't know, six, seven years, it hasn't been. ⁓ the the pandemic squashed any seasonality. And when you had the real loose, as far as the price seasonality, if I look at it like price goes up, which is kind of proxy for volumes in in trucking. ⁓ and so we found that it's it now it's starting to reappear itself, but it's pretty slight, the seasonality. How dramatic is the seasonality? nationally for intermodal. ⁓ if you look at like a a standard month is one is the peak month like 1.2 20% higher and is the dearth eight twenty percent lower to point eight? How how how far are those ranges for seasonality at the national level? Andrew Sibold: Yeah, yeah. So magnitude wise it's not there isn't that much change. ⁓ I usually you would have kind of the you know, the doldrums of the year would be January, kind of the holiday hangover almost. ⁓ and that would be, you know, roughly, you know, if we indexed it right, we'd use that as like a a ninety eight, based on the baseline. So maybe Yeah, yeah. Like it's only like two percent below. And then like our peak historically is only like four ⁓ Chris Caplice: L that's it. Two percent below. Andrew Sibold: four percent high. so there is seasonality but but the magnitudes are are pretty different because, you know, it's very contractually based. So it's not like people are trying to to capitalize on on price changes. Chris Caplice: Yeah, and it seems like it probably ⁓ it moves slowly because it's contractual. So if you look at spot, there'sn't a big intermodal spot. We report intermodal spot, but it's so tiny. But you could look at truckload spot and fuel, that probably is a great indicator for you of when volume's going to increase. Have you have you looked at that? And do you have any insights on the correlation between fuel price, diesel number two, versus ⁓ volume on intermodal? Andrew Sibold: Yeah, yeah. So I actually did like a a quick study of that, ⁓ you know, with the the price spike from the Iran conflict and actually found so I'll caveat it by saying, you know, we haven't quite seen like such a a large jump before in such a short amount of time. But, you know, all things considered, when we've looked at other price spikes and drops, intermodal it's about like a three month lag before it starts showing up in the data of people switching over. But this time around it was actually a lot faster. And I think a lot of that it could be attributed to, you know, the slack that was present at the time from the the lower imports. And then also just the, you know, ⁓ ⁓ that happened in truckload rates to where, you know, people were willing to to pay ⁓ ish for intermodal just because they could keep the support that open ⁓ into the future. And we actually saw that in the data where when people switch over to intermodal, even as prices drop, they tend to stay with intermodal ⁓ until there's a service disruption for intermodal almost. Chris Caplice: That's interesting. I was gonna ask that because we find in trucking is you know, prices go up like a rocket and down like a feather. You know, they kind of they drag and because especially contract, because once you enter in, it it it'll it'll bleed out over time. Do you f you find the same thing that people tend to the the ⁓ when when ⁓ fuel price goes up, you see a different reaction. It's not opposite when price goes down. The big difference is speed of driving. Andrew Sibold: ⁓ yeah, it's not quite as drastic, I think, as trucking in terms of like the depth that it reaches of people leaving Intermodal. for Intermodal a lot of the competition is just based on on the length of haul. so you know, it ⁓ one, takes a little bit longer and two, the the math works out better the longer you haul it, right? So thinking going from LA to Chicago, that was historically the big route, right? ⁓ Chris Caplice: Right. Right. Andrew Sibold: But now we've actually seen Intermodal able to compete on the East Coast where the length of Paul is a lot shorter. and again, I can only attribute that to really like price differences ⁓ between intermodal and trucking. so yeah. Chris Caplice: Yeah, so I I when I had Larry on Larry Grosson a couple of months ago, he he told me his rule of thumb, which I think is great. And it's ⁓ it makes sense to consider intermodal for a shipper if at least from door to door distance for a shipper, at least two thirds of that could go on rail. So even if the length of hull is not, you know, several days, thousand miles, if you're close to ramps, loading and unloading ramps, it might be worth it. Have you seen anything like that? And being a being a formally trained economist, have you formalized Larry's rule? Andrew Sibold: ⁓ no, I have not done a hypothesis test on it quite yet. ⁓ it's definitely in my notebook of of things to do. but I will say, you know, there is for sure a price incentive for, you know, ⁓ hauls, typically higher weight goods to to move on intermodal, which I think is pretty intuitive. but for some reason, like I said, we're kind of seeing intermodal be competitive on these markets that Chris Caplice: Okay. Andrew Sibold: you know, weren't necessarily competitive before. And I think Chris Caplice: Why? What's happening? What's what's going on? Is there something structural? Is it just the price of fuel? 'Cause the fuel's come down quite a bit. So it might go right back up again day to day. Who knows? Andrew Sibold: Yeah, yeah. Yeah. Yeah. No, I think a lot of it is one, just capacity. So, you know, you had the big labor market disruption for trucking, ⁓ just from the sheer amount of of truckers that were removed from the from the market there. But you also have, I think, intermodal one just being more price you know, cheaper. ⁓ it's it's kind of entering a you know, ⁓ a good area in terms of being able to actually compete on price with trucking. Chris Caplice: Yeah, yeah. Andrew Sibold: Which I think historically hasn't been the case. and then also level of services. So I think a lot of especially on the Dre side, ⁓ you know, there's been a big push to increase service and increase kind of people's knowledge of Intermodal. ⁓ and I think that's been paying dividends as well. ⁓ in addition to that, because Intermodal is on ⁓ it's it's pretty ⁓ what do you call it? Chris Caplice: Right, right. Andrew Sibold: It doesn't break easily, so it's pretty robust. so I think coming out of COVID, there were a lot of folks that were highly interested in intermodal just because, you know, say there's a disruption of labor strike or something with truckers in California, ⁓ know, it's still able to get to where it needs to go based on different rail schedules and routing that can that can be done. So, you know, it's it's a little bit more robust, and I think that that's kind of paying off. Chris Caplice: That's I I'm surprised by that to be honest, because usually y there are no trucking companies that are unionized besides that one or two LTLs. I mean that yellow gone and all the stuff. So full truckload, I can't name one. Maybe maybe there's someone out there, but it I always thought it would be just the opposite, to be honest. The railroads are covered by multiple different unions, and any anytime we're seeing union negotiations or anything like that, there's always the multiple overlapping unions that have to approve it. Andrew Sibold: Yeah. Yeah, yeah, I guess I I should have caveated that with like we're North American intermodal, so Canada is, you know, the big trucker disruption there that happened a few years ago. yeah. So ⁓ Chris Caplice: Now that's let me let me go back to the metric, the in intermodal volume index. Because you talked about using it as a descriptive metric, which is always a great place to start. You know, what what are we doing? How do we do that kind of stuff? But you mentioned you're turning it into a predictive model and maybe eventually something prescriptive where you actually give advice or makes recommendations. What are you doing for the predictive side? What do you what do you who would use this and what would they use it for? Andrew Sibold: Yeah, yeah. So a lot of kind of ⁓ advisory services and investment banks are highly interested in it. ⁓ you know, folks who are looking to invest and also kind of have use it as a proxy for, you know, overall economic activity and and ⁓ as well as you know, BCOs. So like think your your Amazons, your Expo Logistics, things like that. so to kind of create that, we ⁓ took the IBI ⁓ ⁓ we created a ⁓ econometric model that takes into account different macro indicators. and using a machine learning model, we tested to see, you know, what had the greatest impact, what the the correlation coefficients were between, say, unemployment and, you know, intermodal freight volumes. ⁓ based on that, we selected the the most statistically significant ones ⁓ put that into our model and ⁓ you know, ⁓ created our predictions. So we're We're shooting about twelve months out. and you know, it'll also give us ideas of of what the, you know, biggest things impacting that. So whether it's the labor market or ⁓ diesel prices, things like that. Chris Caplice: So so your forecasting your your dependent variable is volume. Not not price. Not right. You probably stay at work. Andrew Sibold: Yes, yeah. No, yeah, yeah. We're ⁓ since we're a trade association, we'd we'd get in tr I've been explicitly told don't talk about price. Chris Caplice: Round upon. Yeah, yeah. So it's volume and and so are you looking at it at multiple levels? Do you look at national or you look at just specific corridors or regions? Andrew Sibold: Yeah, so where I have data, I try to get it down to the region. So think like diesel prices, right, where you have different markets reporting different prices. but a lot of the macros are just kind of by nature national. ⁓ so unemployment, things like that. Yeah. Chris Caplice: Sure. And monthly too. You can't go much if you're usually using any Fred data, it's monthly and a lot of them are lagging. Andrew Sibold: Yes, yes, for sure. And and that's been kind of one of the Chris Caplice: Sometimes they don't get reported for six months and sometimes things happen. Andrew Sibold: Yeah, there's there's a shutdown, right? ⁓ so yeah, that that's also actually one of the the things that we look to solve with intermodal, right? Because our historically the ETSO data was roughly a month lag. so we actually created ⁓ a now cast to kind of describe what's going on in the current month based on those macro indicators and also some more, you know, higher reporting volumes, so like AER data that comes out weekly. Chris Caplice: Yeah. So what what do you what's the difference between a now cast and a forecast, besides it sounding sexier? Andrew Sibold: Yeah. Yeah. So I now cast, you know, just takes kind of what's happening in the near term and rather than waiting the, you know, independent variables of the macro stuff going on, it mostly puts it as like when this has happened, this is what's happened almost immediately after. so you're trying to describe like one hump of the data versus, you know, multiple ripple effects in the future. Chris Caplice: Yeah. What I found is that that's complicated with this is ⁓ they're very much ⁓ recency have a lot of recency bias. And so when rates are going up, they're gonna go up forever. And they go down, they're gonna go down forever. It's very hard to get that ⁓ the cycle that we know exists. It exists for whether it's seasonality or business, it's been there. ⁓ but let me ask you another question about this. Are you familiar with the flow initiative that by DOT freight logistics optimization works that was put in through a several years ago? Andrew Sibold: Yes. Yeah. Yeah. Yeah, see yep. Yeah, yeah. I think we've ⁓ we've actually been on a few calls with them recently. so it kinda seems like maybe they're getting it up. Chris Caplice: I think they've changed the name because they ⁓ a lot of people got doged out of that as well. And I think it's been renamed and now it's the American supply chain I can't pronounce the word sovereign sovereignity initiative. A S C S C I I Is it the same thing? Are you involved in that or? Andrew Sibold: Yeah. Yeah. Yeah, yeah. So we have been on a few like listening sessions is what they call so ⁓ of it's related to kind of the disruptions from the Strait of Hormoes and, you know, what we're seeing in the data. but I hope that, you know, in the future, I think that was a a pretty good initiative and, you know, we can kind of partner with the federal government to hopefully produce maybe better data, ⁓ more relevant, you know, more recent data. I think that's kind of everyone's like large complaint in the data world is like, you know, it comes out so late that it's it's almost not useful in the for present. Yeah. Chris Caplice: Yeah. Yeah, it's it's funny. In truckings, especially there's like two types of models. I want to know what's the price of a truck right now versus longer term, what would what's a reasonable price for a contract for the next twelve months? And they're two totally different ways of doing it. ⁓ but you're right, there is a dearth of real time data for a lot of those models. Longer term, it's better than it was. There's a lot of dengue out here now. So it's shockingly. Andrew Sibold: Yeah, for sure. Chris Caplice: ⁓ let me switch topics a little bit and talk about tariffs. Twenty twenty five was huge for the tariffs, the uncertainties. I think the average was every nine days we had a new tariff announcement or modification. ⁓ that's slowed down a little bit since then. But how how have you seen tariff volatility affecting the shippers' confidence in committing to Intermodor? Andrew Sibold: Yes. Yeah. So in committing to intermodal, you know, I wouldn't say there's been too much disruption in that just because it's you know, when things come in on a box, it's easy to keep ⁓ on a box if it if it makes sense at all. but we did see a lot of pull forward behavior. So basically shippers would hear about, you know, a potential tariff announcement and then yeah, try to get things on the boat and, you know, about 14 days worth of sailing time from China to the US and people trying to front load because by the time the boat gets here, the tariff might be different. ⁓ but you know, I think a lot of that has leveled out, kind of as you said. It's it's calmed down. And I think, you know, there's also the idea, right, that Tariff announcements don't necessarily equal tariff enactments. so I think Yeah. Yes. Yeah. ⁓ Chris Caplice: That was a hard lesson. That was the 2025 lesson. And the the whole taco thing, you know, Trump always chickens out, whatever, it's a negotiation. It's not a yeah, that's it took a while, but I think companies have kind of maybe they've gotten used to it. Andrew Sibold: Yeah. Yeah. And I I think that's what more so than the actual terrorists themselves, I think that's what has kind of caused things to comp down as is people have just gotten used to the behavior and they're like, ⁓ you know, it's just an announcement. So Chris Caplice: Yeah. So we did a project up here recently at MIT where we looked at and one of our sponsoring companies wanted to know if a tariff of a certain ⁓ type of product happens for a certain country, how will that influence how those products are brought into the US? Whether it shifts from the East Coast to West Coast, ⁓ Canada, Mexico, and how that shifts. And they developed a pretty nice model for that. Have you seen a lot of shift away from the West Coast? Because a lot of this manufacturer, some of it has moved to Mexico. Which would definitely shift the import from a different make it more, I guess, domestic coming in from Mexico. Have you seen any shifts based on tariff changes? Andrew Sibold: Yes, quite a bit. So up until kind of the the diesel shock started showing up in the data, that was the number one cause of, you know, changes in our model. there was kind of a a big shift away from the West Coast to ⁓ East Coast and also the Gulf as well. the Mexico lanes, they're they're relatively smaller just because there historically hasn't been as much volume on them, but they have been operating like at at full capacity. ⁓ and I think that, you know, when you see investments in rail and intermodal, they'll likely be kind of along those lines with the whole Frenchhoring narrative. ⁓ yeah, we've seen a huge shift to ⁓ to the East Coast and the Gulf Coast just from the tariff volatility and kind of things moving away from China. Chris Caplice: You expect that to continue? Any any thoughts on Andrew Sibold: Yeah, so I anticipate it'll continue. certainly for the next couple of years. you know, if we have ⁓ what, three two, three more years of this administration. and ⁓ sure you're well aware, tariffs are sticky, right? So like once they get enacted, it's it's pretty hard to reduce them because they get worked into the bottom line in different industries and you know, they have lobbying groups like ⁓ associations that, ⁓ you know, fight for those things. So yeah, I don't ⁓ anticipate it. necessarily going back to what it was. now that's not to say that there won't be little shifts here and there, but I definitely see the East Coast making up a a larger chunk of imports and intermodal as well. I mean, I think you had what Savannah and Charleston are kind of neck and neck for always the fastest growing ports in the US. ⁓ so yeah. Chris Caplice: I know during the pandemic there was a huge shift 'cause of the ⁓ backlog on the west coast ports. ⁓ and I know Houston benefited and some on the southern part of the US, but I thought a lot of that volume went back to the West Coast right after the pandemic. And ⁓ Houston I think is still it it was a stickier. Andrew Sibold: Yes. Yeah, yeah. So post COVID there was a almost kind of return to normal on the West Coast, but like I said, since their the tariff announcements, there has been a pretty pronounced shift in the the east coast and likely to continue going forward. Chris Caplice: So let's shift to drivers, because I know I mean intermodal doesn't use as many, but you have it for draids, first mile, last mile. ⁓ how how tight is that capacity? You mentioned that your driver security or your ⁓ and authorization and everything is a lot tighter than generic over-the-road one-way trucking. ⁓ is that a bottleneck at all for the ⁓ driver shortage for the drage side of things? It's a really weird part of the whole trucking industry. Andrew Sibold: ⁓ yeah, yeah. So I yes and no. ⁓ there has been, you know, they've been impacted kind of like other folks have of, you know, driver issues with, you know, some of the labor side of things, the crackdowns. ⁓ but because intermodal is, you know, there's a little bit more of a barrier to entry for people to get into the, you know, quote unquote intermodal space, I think there's been less impact on intermodal because Typically these Dray Edge companies are are highly specialized and you know, work in Dre edge rather than than competing with ⁓ we'll do Dre this week and then, you know, long haul trucking the next. so I think yes, impacted, but not quite to the extent that that over the road trucking has been. Chris Caplice: So how would you characterize the drayage providers for this? Are they a lot of small folks? There's no big national. I know there have been some attempts at roll-ups of DRAEs, but it seems like it's so localized to the ports. It just seems like I I don't know, are there national players now or is it more a series of regional players? Andrew Sibold: Yeah, I think it's more regional players. ⁓ it's definitely a lot smaller scale than, you know, the big over the road kind of conglomerates. so yeah, it's it's more smaller, smaller companies, but n maybe less like independent truckers doing drayage. Chris Caplice: So smaller like fleets, no owner operators just drove that like five to hundred truck fleets. Andrew Sibold: Yeah, yeah. No, I wouldn't say no owner operators, but but certainly I think less than there is on on over the road trucking just because the the barrier to entry is higher. ⁓ so Chris Caplice: Yeah. So one one of the biggest challenges or the ⁓ with it with intermodal is that as you mentioned earlier, there are multiple trade-offs or handoffs rather. So you by definition, an intermodal move will have at least three handoffs, right? Or at least, you know, one to the to the truck that picks up, takes it to the ramp, onto the train, assuming nothing happens on the on the rail side of things, and then back to drayage. So that requires a lot of standardization. Andrew Sibold: Yes. Chris Caplice: How are there any barriers now to standardization? Are there is it is it totally done? Are there any are there still challenges to standardize different parts of Andrew Sibold: so I would say in terms of like the interchanges, the you know, what IANA's done has has largely smoothed out the frictions there. naturally there's still, you know, Murphy's law of of chaos, right? Any anytime things ⁓ change hands, there's that potential. But I would say that, you know, it's it's not frictionless, but it's certainly, you know, a lot less than it was. and don't know ⁓ Short of, you know, the push for like autonomous vehicles and things like that, ⁓ for like last mile delivery, I don't think that there's too much to be done there. you know, on kind of the the route optimization. Luckily with with railroads, things kinda you know, are kind of handled like up at the, you know, macro level with with scheduling and things like that. Chris Caplice: Yeah, so you brought up a A V, so I'll let's go a little deeper into that. Autonomous vehicles. One of the ⁓ there people are split between this. Some people think it, you know, robots will be all our drivers and you know, tomorrow. ⁓ others think it'll never happen. I'm kind of in the middle. It will happen, it's happening for some things. But what do you think the ⁓ the impact of autonomous vehicles for class eight vehicles will do to the intermodal market? Do you see it as robbing volume from what would have gone intermodal, or do you see it as complementing? How d how do you think about ⁓ autonomous vehicles, trucks, large class A trucks, not not the Waymo's? Andrew Sibold: Yeah. Yeah. Yeah. So it's it's kind of an interesting dynamic, right? Because you think about it with truck platooning ⁓ a lot of the autonomous vehicles where we can chain multiple trucks together on the highway and move things all at once, like it's almost like we're pushing to recreate the railroad, except on the road. so always thought of that as like, you know, that already exists. So like what's the Chris Caplice: Yeah, yeah. Andrew Sibold: what's the key driver there? And, you know, I think it's because the roads are subsidized by the government versus rails, which are typically owned by the railroads. so ⁓ don't think that there's going to be too much in-road in terms of like taking things away from intermodal because if in theory, right, like over the road will operate more like railroads. But the the competition I think will be ⁓ who gets the ⁓ the help in terms of maintaining their infrastructure and you know, road over the road has a huge advantage, right? Because the federal government and state governments pay for those roads. ⁓ Chris Caplice: Well, the eighteen sixties and you know, we can talk about ⁓ land grants and everything if you want to go back, talk about the Gilded Age. Come on. But no, I understand. I understand. But but one of the things, because I've had someone did some numbers and I I I can't quote exactly who we call who did this, but then if if they actually tried to move everything that moves intermodal onto the road, there'd be gridlock across every highway. That means so much volume moves on intermodal, it just it can't switch over. So I wonder if it's gonna Andrew Sibold: Yeah, yeah, yeah. Yeah. Yeah, robber barons and all. Yeah. Chris Caplice: But I did it at the margins. You know, those things that ⁓ maybe they just need to go a little faster and they're willing to pay a little more. Because I believe it'll be faster for a longer haul. I I I I'm not positive if you do this because ⁓ the transition. Andrew Sibold: Probably. Yeah. Yeah. For sure. And I think, you know, a a well rounded supply chain kinda has has need for all of it, right? Like there's there's really no shortage of stuff that needs to move. ⁓ and I think that, you know, while there is some competition, I don't think that there's necessarily as much as people might think. just because what moves on the different platforms is is almost like pre filtered, ⁓ just because of the nature of the different methods of moving it, right? So like You know, if I'm moving a a shipment of super supercomputers that I need like tomorrow, you know, I'm not gonna go with an remoto, I'm gonna go with truck, ⁓ or even air. so yeah. Chris Caplice: But it's it's shocking still to see how recent that development is. I know when I was coming up twenty years ago, there was railroads trucking companies hated each other, right? And 'cause they they were trying to eat each other's lunch. And I think shippers have finally everyone's finally realized that there's places for each and it just makes sense. I mean, there's a case study of Burlington Northern that gosh is forty years old now, that specifically talks about how to talk to shippers about what should go truck, what should go rail, you know, just the looking at the economics of it. It just it just kind of makes sense. ⁓ I I think there's less and and if you look at a lot of the large trucking companies like JB Hunt or Schneider, any of these guys, they will use intermodal dramatically because if something is good for an intermodal move, why would you use a truck for that? You might emit as the as that company as an IMC, but you don't need to do the move. It's just the economics just makes sense for intermodal. Andrew Sibold: Yeah. Yeah. Yeah, yeah, for sure. And, you know, like I said, each one kind of has their place place in the world. ⁓ and you know, I don't think I'd want to live in a world where everything's either moved by truck or by rail. I think having that that split is is important. ⁓ Chris Caplice: Well you can say that, right? Because you touch everything. you have the drays, trucks, the rail. I I'm sure if we talk to, you know, the Railroad Association or the ATA, they might have a different opinion. Yeah. We'll see how that goes. and then by law, I have to talk about AI. ⁓ so what are you seeing? What what first you know, you're you you're relatively nerdy, you deal with data. Can you define what AI means to Andrew Sibold: Yeah, yeah, yeah. We'll we'll see if I get in trouble for that take, but Yes. Chris Caplice: Diana, what what it means? What what is AI to you? Andrew Sibold: Yeah, yeah. So, you know, I think of AI as falling into, you know, three broad buckets, right? So you have the AI that, you know, exists in the physical world, which is, you know, like robot and remote sensing technology. then you have the AI I think that we're all more familiar with, kind of the the large language models and and some of the predictive stuff. ⁓ and yeah, yeah. Chris Caplice: Pattern matching. Yep, yep. Andrew Sibold: And then, you know, you have kind of the third ⁓ almost frontier bucket of the, you know, general autonomous ⁓ intelligence that, you know, is more in the realm of science fiction, I think, at this point. you know, I don't even think we really have a good good definition of what intelligence is. So ⁓ don't know how we'll know if we get autonomous, ⁓ completely autonomous intelligence. But ⁓ yeah, so For IANA specifically, ⁓ you know, as as I mentioned, we use a lot of machine learning. ⁓ just 'cause I I come from a data science background, so I'm I'm pretty comfortable with that. you know, it's it's kind of a game changer, I think, for ⁓ and for kind of the freight community overall, just because so much of what IANA does and some of our data is so seasonal that, you know, it lends itself quite well to machine learning to teach itself the patterns and you know, ⁓ Chris Caplice: Yeah. So so machine learning, a lot of times I hear that to me, ⁓ we use a lot of machine learning here at MIT as well. That kind of gets lumped in and called traditional AI, right? Machine learning slash AI, where it gets in whether it's ⁓ with many different techniques. But then generative AI is usually considered separate. And then there's reinforcement learning, which is even, you know, more interesting and starting to see more of that. That's more in the prescriptive side of things that we're seeing come out. Andrew Sibold: So yeah, it's it's been a huge thing. Yeah, yeah. Chris Caplice: But yeah, I was just at when I was last at Manifest, all the different software vendors, ⁓ everyone's doing AI. And that just means they use math. ⁓ you know, it's it's it's it's everywhere. Andrew Sibold: Yeah. Yeah. Yeah. Yeah. It's it it's a weird thing where, you know, like when people say AI but they just mean a large language model and then like you try to break it down and you tell them like all that's doing is just trying to predict the next word. It's not necessarily like understanding the the concept behind it. which I think is Chris Caplice: It's it's almost like back explaining the difference between optimization and simulation still. You know, they're they're different. They're both valuable, they're different tools, but that's that's our job to be able to explain it to people, the decision makers, that to make sure they understand. Andrew Sibold: Yeah. Chris Caplice: Well Andrew, thanks so much for joining me this afternoon. I appreciate it. I I I've learned a lot. I I still I'm still pretty ignorant on intermodal. I know it ⁓ only as a competitor trucking that I look at and it's more than that. And so I appreciate you shed. Andrew Sibold: Yeah, thank you. Yeah. Yeah, for sure. And if any one of your listeners is curious about intermodal, obviously feel free to reach out and we're having our our expo coming up the second week of September. So come on out to Long Beach. ⁓ Yep, yep. Intermodal Expo, ⁓ Long Beach, California. So September I think fourteenth. there's there's details on the INA website and lots of intermodal practitioners there and you know Chris Caplice: Plug it. Where's it gonna be? Where's where's ⁓ what what's it called? Expo twenty twenty six? Andrew Sibold: railroads, things like that. So ⁓ if you're in the area. Chris Caplice: I heard it's a new era for intermodal. Isn't that your slogan? Isn't that the theme? Andrew Sibold: It is. Yeah, that's the slogan, right? We're at ⁓ we're at an inflection point. We're we're in the zeitgeist. Chris Caplice: ⁓ there we go. There we go. All right. Thanks, Andrew. I appreciate you. stay tuned for the truckload market update.