September 17, 2024 • 57 mins
The balance between supply and demand has a significant impact on pricing and behavior in the trucking industry. During periods of excess capacity, known as soft markets, prices tend to fall as carriers are willing to honor existing contracts, as evidenced by high acceptance rates. However, in tight markets where demand outweighs supply, prices rise and carriers may prioritize more lucrative spot freight opportunities over contracted loads.
These market dynamics and the incentives that drive carrier and shipper actions were the focus of recent research conducted through a partnership between academia and industry. Angi Acocella, who has studied trucking procurement through her research at MIT, explored how carriers respond to shippers under different conditions. Her findings indicate that while maintaining relationships is important, carriers are primarily motivated by current market conditions when deciding which loads to prioritize.
While optimization models have proven useful in procurement, the assumptions made in early approaches underestimated the dynamics of real-world transportation. Deterministic frameworks that ignore variability did not translate effectively to practical applications. Newer techniques have emerged to instead impose constraints that reflect industry complexities. This enhances the strategic value of optimization beyond traditional deterministic, cost-focused objectives.
Network segmentation also deserves further study, as not all freight lanes are uniform. Factors like volume consistency, carrier types, and opportunity costs must be considered to tailor solutions. Dynamic contracts that fluctuate with market conditions could provide more flexibility than annual fixed-rate agreements.
To address the trucking industry’s many complexities, practitioners and academics should continue close collaboration. Analytics from such collaborations can provide insights for industry players ready to adapt to the constant changes they face. Maintaining perspective across networks while focusing on execution, not just planning, will be key to future success.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Dr. Matt Waller (00:10):
Welcome to the Matt Waller podcast, where we
look at success at theintersection of technology,
logistics, supply chain, retailand CPG, also known as the
retail value chain.
I want to clarify that thispodcast is distinct from my
responsibilities as a professorin the Sam M Walton College of
Business.
Nonetheless, it aligns with myaspiration to provide practical
(00:31):
insights to professionals andbusiness by showcasing companies
and people that can enhanceyour ability to manage, lead and
strategize and marketeffectively and the retail value
chain.
Before we dive into today'sexciting episode, I'd like to
thank our sponsor, new RoadCapital Partners.
New Road invests in proventechnology services and products
(00:54):
that serve unmet needs in themarketplace.
They look for companies insupply chain and logistics, as
well as consumer-orientedcompanies.
For more information, go tonewroadcpcom.
I would also like to disclosethat I am a strategic advisor to
New Road.
I'd also like to recognizepodcastvideoscom for the
(01:18):
services they provide for thesepodcasts.
I'm very pleased with theirservices and now, without
further ado, let's get into theexciting episode.
I have with me today ChrisKaplis and Angie Akechella, and
we're talking abouttransportation procurement,
(01:41):
particularly truckloadtransportation procurement, and
I have known Chris for about 30years, I think it is.
Chris is a senior researchscientist at MIT.
He's also the chief scientistat DAT Solutions.
He is an academic who hasstayed very involved with
(02:05):
industry his whole career and hehas his PhD from MIT.
And Angie also got her PhD fromMIT and she is now a
postdoctoral researcher atTilburg University in the
(02:29):
Netherlands.
And the reason I invited themto the podcast is, of course,
this podcast focuses on what Icall the retail value chain,
which includes supply chaintechnology, retail and marketing
technology and consumer goodscompanies.
(02:50):
So I read an article that theyhad published in an academic
journal called the Journal ofBusiness Logistics and I really
liked it and you'll see when youlisten to the podcast it's very
relevant to this podcast.
(03:10):
But they had an article theypublished in Journal of Business
Logistics called Research onTruckload Transportation
Procurement a review frameworkand future research agenda.
And one thing you'll see in ourdiscussion if you're a shipper
managing freight or if you're acarrier or if you're a technical
(03:31):
carrier or if you're atechnology company in supply
chain, you will want to listento this because at the end of
their article they talk aboutfuture research opportunities
and you'll hear us point out inthe interview that it is true
it's future research, but it's.
Also it can tell you wherecompetition is going, what kinds
(03:55):
of new technologies are goingto be developed, and so if
you're a technology company, itgives you an idea of what to
look at for new features,functions and capabilities, and
if you're a carrier or a shipper, it gives you an idea of where
(04:16):
the industry probably is goingto go.
But I also interviewed themabout an article that was in
Transportation Research, part E,called Elephants or Goldfish an
empirical analysis of carrierreciprocity and dynamic freight
markets.
And the question is do carriersact like elephants, meaning
(04:37):
they've got a long memory, or dothey act like goldfish, meaning
they don't have a memory ofwhat happened and so they just
optimize for that particularperiod of time?
I think you'll find itinteresting.
Again, chris has a long, long30-year experience in truckload
procurement and I think you'llfind it very interesting.
(05:02):
Thank you, chris and Angie.
Thank you so much for joiningme today to talk about some of
the research you've done intransportation.
Really appreciate it.
Chris Caplice (05:13):
Yeah, great to be here, thanks.
Angi Acocella (05:14):
Yeah, glad to be here.
Interesting for theconversation.
Dr. Matt Waller (05:18):
So just for people listening to understand,
I've known Chris for many yearsand I read an article in a
recent article that heco-authored with Angie called
Research on TruckloadTransportation Procurement a
(05:39):
review framework in futureresearch, and I really like it
and I typically don't likereview articles, but I like this
one a lot, and so I contactedhim and asked him if he would
mind talking about it and hesent me another article that I
also liked that was published.
The first one was published inJournal of Business Logistics.
(06:02):
He also he and Angie publishedone called Elephants or Goldfish
Question Mark an empiricalanalysis of carrier reciprocity
and dynamic freight markets, andthat was published in
Transportation Research Part E.
So I'm going to be talking tothem about that and thank you
(06:22):
both for your willingness totalk about that.
We'll get right into it.
The first thought I had wasyour paper in TRE is talking
about opportunistic behavior,and can you explain a little bit
(06:42):
about what is meant byopportunistic behavior and the
context of truckload freightcontracts?
Chris Caplice (06:51):
Yeah, maybe I can start by talking about the
market cycles, right?
So I think it's really importantto think about when we're in a,
you know, a softer market kindof what we've been seeing
recently, where you have a lotmore capacity than there is
demand for capacity, right, andso this is where you have prices
that are going down, you havevery high acceptance rates, so
(07:11):
your contracted carriers areaccepting a lot of your freight.
And then we have the flip sideof a tight market, right, a
constrained market, where youhave a lot more demand than
there is capacity for thatmarket.
You see prices going up, yousee acceptance rates going down.
And so when we think aboutopportunistic behavior
particularly in this study,we're looking at carriers
behavior we think about, youknow, if I have a contract for a
(07:33):
particular lane, am I going toaccept that contracted volume in
a tight market?
When I have, I'm constrained inmy capacities, maybe there's
higher prices elsewhere, or am Imaybe using the capacity that I
have and fulfilling higherprice, higher, you know, using
higher price load?
So am I, do I want to maintainthat contract?
Or am I acting a little moreopportunistically and, you know,
(07:55):
going for the higher pricedcontract, higher price chipmence
?
Angi Acocella (07:59):
Yeah.
So to add to Angie's point,everyone knows the truckload
market moves through thesecycles and the question that
really Angie honed in on thehonor dissertation but MIT is
reciprocity and the idea of who.
When is there an incentive todefect from the contract?
Because when the market's tightthe spot market is usually way
higher than the contract.
(08:20):
So the incentives to thecarriers to defect and say, well
, I'm not going to haul that,I'm not going to honor the
contract and go that way whenthe market's loose it's just the
opposite.
Well, we've been in since thespring of 22,.
The rate is spot rates belowcontract, so the incentive to
defect is at the shippers nowand they say, why should I honor
(08:40):
that contract when I go to thespot market and go cheaper?
So the incentive to defect thecontract flips depending on the
market.
And that's what Angie's greatwork really identified and
quantified at the impact.
Dr. Matt Waller (08:54):
So it really is an interesting topic and
everyone's aware of it and yourresearch question is really good
, and so could you talk a littlebit about the methodology you
use to do the empirical analysisof reciprocity?
What kind of metrics orindicators did you look at to
(09:18):
determine whether carriersresponded to past behaviors or
shippers?
Chris Caplice (09:26):
Yeah, so we were actually really lucky to have a
really detailed data set oftransactions.
So we had data from 2015 to2019.
And essentially we have all ofthe transactions for about 70
different shippers over thattimeframe.
For the continental US,specifically, full truckload,
and so we have information onthe price, whether a carrier is
(09:48):
contracted and if they acceptthat shipment.
That's key, of course,geographic regions, things like
this.
We end up with about 2 milliontransactions in that data set
and essentially we look at overthat time period.
First, we need to identify whenwas there a soft market, when
was there a tight market?
So 2017, 2018 was a very tightmarket and so we were really
doing a lot of this researchduring that time.
(10:09):
So we were really pinpointingthat timeframe.
So we define those marketperiods which we can talk a
little more detail with, andthen we look at what are the
characteristics of shippers andtheir contracted carriers, where
their performance, theirbehaviors during the soft market
.
Then we look at that tightmarket between those same
shipper carrier pairs and say,well, does that historical
(10:32):
relationship between the shipperand carrier in the soft market,
when, as Chris said, when thepower is more in the shipper's
hand, does that translate into acarrier in a tight market
prioritizing that shipper'sfreight.
And so we looked at things likein that soft market, what is the
contract price competitivenessof shipments?
So, essentially, relative to alane benchmark, how are those
(10:56):
shippers pricing those contractswhen maybe they don't have to
be too high, right, so we lookat contract price
competitiveness.
We look at the consistency andthe volume that's going on those
lanes.
So, is a very consistent lane?
Is it very volatile?
Is it a very infrequent lane?
We looked at things like thedwell time at facilities.
(11:16):
Right, so?
Is a shipper delaying carrierover time at its facilities and
then impacting its ability toget to follow on shipment?
So we looked at both thosethings in a soft market and then
also then in a tight market.
How are carriers responding tothose things?
Angi Acocella (11:32):
This is something that, matt, you enjoy, as well
as having a strong relationshipwith industry and having access
to this data.
So this was a partnership withCH Robinson, specifically the
TMC division, with the greatSteve Rates was leading that,
who was just retiring.
I think today might be his lastday at CH Robinson, but because
of him he kind of steered andenabled this kind of analysis
(11:55):
great analysis that Angie wasable to conduct.
So I encourage anyone listeningto this.
This is how you work withacademia.
Work with us to help you bringthe data and give us the
resources.
Then we can do this analysisthat brings insights that
industry by itself just doesn'thave the time to uncover, but
then they can use.
So that kind of makes sense.
(12:16):
But, matt, you've read thepaper but your thought going in,
did you think carriers weregoldfish or elephants?
And the difference is do theyremember things?
Elephants remember everything.
Goldfish have.
What a 10 second memory.
Yeah.
Dr. Matt Waller (12:30):
So I will answer that question because I
tried to guess before I read it.
But I know they're not theelephant in the room, so that
couldn't be it.
That was actually the firstmetaphor that came to my mind
and then I thought oh yeah,elephants remember a long time
(12:50):
into goldfish, don't?
But let me ask you a quickquestion before we get into that
.
But a quick answer and I'llexplain why in a minute.
Goldfish is what I expected so,and I'll tell you why in a
moment.
Before I do, you mentionedsomething about industry and
academia working together, andI'm glad you mentioned that.
(13:16):
Chris, and you have been sogood at that over the years,
from the very beginning.
I remember because when youpublished a paper from your
dissertation in Journal ofBusiness Logistics I don't
remember the year, but it seemslike was around the mid 90s I
read it and really liked it.
(13:36):
It was talking about carrierbid optimization I think that's
what we used to call it backthen, and I actually used.
I taught a master's classcalled Logistics Modeling and I
used that paper because Ithought, yeah, this is a long
(13:59):
time ago, obviously, but it was.
It's hard to find good papersout there that are strong
academically and extremelyrelevant at the same time, and I
could have a whole conversationon this, the academic, what it
takes to get published andgetting citations.
Angi Acocella (14:22):
You have all of researchers that just have to
cite each other and thereforetheir age index goes up.
But no one else reads it exceptfor those H number of academics
.
And then I don't know anypractitioners that read some of
the most respected journals.
I love some of thesetransportation sciences.
They're great journals.
No one reads it outside thesmall collection of research.
Dr. Matt Waller (14:47):
Right and it seems like I would have
estimated seriously back in the90s when I started as a
professor in 93, and, as youknow, I've always been linked to
industry and done research andcollaboration with industry.
It's one of the reasons why Ifollowed you.
But you're talking aboutprofessors citing one another.
(15:11):
You see that more in somedisciplines than others.
I agree.
I really got to see thatclearly when I spent my eight
years as dean because I wasdoing promotion and tenure
review for all these eight.
We have eight academicdepartments in our college and I
(15:31):
saw like in finance they don'tcite each other much.
In information systems they gooff the deep end on citing each
other.
Have you noticed that?
Angi Acocella (15:42):
I think there should be a limit on the number
of citations you're allowed toinclude in an article, say 10.
You can only do 10.
Unless you're doing a reviewarticle, right?
Oh yeah, Then that's what it'sthe purpose of it.
So instead of people doing theresearch themselves, they can
look at our bibliography.
That's the goal.
Dr. Matt Waller (16:00):
But your call that you made just here a minute
ago is really good, and most ofthe people listening to this
are practitioners, not academics, I think.
At least that's been true up tonow, and so You're right.
I mean, I think that there's agrowing number of academics who
(16:23):
want to work with industry, andsometimes it's hard to get data,
and so if industry collaboratedwith them a little more, it
would probably solve thisproblem to a large degree.
Angi Acocella (16:36):
And I think it's certain areas there's some
people in like pure inventory,that are tied to.
Like Sean Willems does stuff.
Everything he does is with thecompany, but in our, in my area,
in Angie's area, justin Millerdoes a lot of his stuff is all
applied.
Alex Scott, down at where is henow?
Tennessee, tennessee, very muchapplied, and so I think there's
a pocket of this out there, butthen there's a lot of stuff you
(16:59):
know where it's.
Just that's right.
Yeah, different audience.
Dr. Matt Waller (17:03):
For those listening, Jason Miller is at
Michigan State and I'll tell you, Chris, the only person I look
at their post every day onLinkedIn is Jason Miller and he
posts every day.
And it's great isn't it?
Angi Acocella (17:22):
Yeah, yeah, good stuff.
Angie, you're really sayingbetter than his posts is his
mustache.
Yes, I love that.
Chris Caplice (17:30):
I was going to add that, yeah, no, it's okay.
I was going to add that youknow one of the the also the
benefits of working withcompanies for for, hopefully for
them, but for us as well.
You know it's partially ourresponsibility as the
researchers one to get thepapers out there, but you know
Jason's really good at this atthen also creating, you know,
publications that are digestibleby industry, right.
So so that's what some of thefun part for us as well.
(17:51):
As you know these types ofpodcasts or LinkedIn posts or
blog posts.
You know there's so much morethat can be done with, with
research that's that's sponsoredby or partnered with industry,
and it's part of ourresponsibility to kind of help
get that out there.
Dr. Matt Waller (18:04):
Well, I'm glad you're doing that and again, I
think it's one of the reasons Ienjoyed reading your article.
But but you know, right now, infact, I'm doing some research
on omni-channel retail supplychain management and, and so I
did a literature review onco-authoring with three other
(18:24):
faculty members who arewonderful people that are very
involved in industry and and I.
We identified the journals thatneeded to be included and I
took one of the journals of itand it's a top journal, and they
actually had a large number ofpublications on omni-channel.
(18:47):
Surprisingly, I was surprisedhow many of them used strictly
analytical models.
You know, assume there'shomogeneous buyers, right,
whatever you know had, had, had,had a linear decreasing demand
focus and all of that.
And they get counterintuitiveresults right, which is desired
(19:14):
many times.
And I'm thinking I've beeninvolved in omni-channel
consulting and and and and workand research for since it began
and I can tell you some of theresults of these papers are
absolutely wrong.
There's nothing.
(19:34):
Yeah, they are counterintuitive, but they're they're wrong.
I I can't even imagine asituation.
You're an amazed man.
Angi Acocella (19:42):
I can't believe that.
So I we joke that it's.
A lot of these papers areTrojan.
They use trucking as a Trojanhorse.
They say this is why this isimportant.
It's a big industry, blah, blah, blah.
And here let me use this mathto prove a point that doesn't
make sense.
Let me assume away all theinteresting part.
Chris Caplice (19:59):
Assume deterministic demand yeah.
Angi Acocella (20:03):
I remember what text part is yeah, go ahead.
No, go ahead.
When you, when you interviewpostdocs we interview a lot of
postdocs coming in and you mustas well it's just depressing
sometimes when they will talkand they pretty much will sue
everything away.
They come with this greatsolution, like we had one for
doing optimal LTL network design, and we said, really this seems
(20:25):
pretty interesting and wasoptimal for a two terminal
system.
It's like okay, that's, youknow, what do you do with that?
But you know, but they havecontributions in other areas.
Math is improving andeverything, but to me it's not
applied research.
And if you're doing somethingin supply chain, it should be
applied.
We're not a theoretical science.
Dr. Matt Waller (20:44):
I agree, you know, if you think about it,
with some of these.
When I was a doctoral studentat Penn State, they asked me to
do an exec ed program for somecompany on forecasting sales
(21:05):
forecasting and I did it.
It was a good lesson for me.
Angie, you mentioneddeterministic demand.
I said let's start out withdeterministic demand.
And they're like well, whywould you need to forecast it?
That's a good point, but at anyrate, so to answer your
(21:26):
question, chris, yeah, I, myfirst thought of the metaphor
when I saw elephant was elephantin the room.
I thought, well, that can't beit.
I don't know how that wouldapply here.
Everyone knows thattransportation is out there.
There's so much of them.
If you look at truckload.
I could be wrong about this,but I think truckload
transportation is around in theUnited States around 700 billion
(21:51):
.
Angi Acocella (21:52):
Transportation.
At trucking it's fulltruckloads about 400.
Okay, and private is about 400,and so it's about 800.
80, 90 billion.
Dr. Matt Waller (22:01):
Okay so, but yeah, I thought, regarding the
question about reciprocity, I sothe reason I thought it would
probably be goldfish if I had toguess wasn't probably because
(22:21):
of the right reason.
But my experience in working inindustry is and any
organization is that there areshort memories.
Now in academic organizationswhere you have professors that
stay their whole careers, likeone of your co-authors who on
(22:42):
the JBL paper, yoshi YoshiSheffi, I think.
He's been at MIT for almost 50years.
If I'm not mistaken, he came inthe late 70s.
Yeah so you know.
But but in most industry andindustry in general, there's a
(23:04):
lot of turnover, usually notalways.
Angi Acocella (23:07):
There's exceptions to that, obviously,
but you see, there's a lot ofturnover in shippers in that
role, because they'll they'llrotate through, so someone will
be in there or they'll move toanother company.
But it's funny, if you talk inindustry, everyone says oh yeah,
we have a long memory.
The, the perception, the commonwisdom for carriers is that,
yeah, we don't forget, you'reexactly right.
(23:29):
Is that they're?
They're little goldfish.
They don't remember.
They might remember, but theydon't act on it.
Well, angie, do you want tocharacterize that?
Chris Caplice (23:37):
Yeah, sure, yeah.
So I mean that was also some ofthe motivation of this study
was, you know, well, you knowthat's there's a lot of
conversation around this.
Let's put some numbers to it,right.
And so, yeah, we were.
We were looking at, well, youknow, is in that tight market,
are carriers responding?
Are they, are they prioritizingshippers that had behaved
better in the past?
Turns out they're not right.
But but the way that we'reunderstanding and explaining it
(24:00):
is that you know it's not thatthey're not prioritizing a
shippers good behavior.
It's really that they'refocusing on what are they
currently doing in this currentmarket condition, because the
reality is they, they arecapacity constrained, they have
businesses to run right.
So they're they're looking forshippers and saying, okay, if
you have good performance now inthis tight market, I'll work
with you.
So it's not saying, you know,shippers shouldn't care about
(24:21):
the relationship.
It's just that the reality iscarriers have constraints,
capacity constraints andbusiness constraints.
Angi Acocella (24:28):
They can't afford to be elephants?
Yeah, absolutely.
But the question is, the realinteresting question is what
about shippers?
Yeah, are they elephants or arethey?
Dr. Matt Waller (24:38):
goldfish.
I would have guessed they'regoldfish.
Angi Acocella (24:43):
We don't know yet , because Angie hasn't finished
the work, that is, it's allblack and all finishing up this
station.
Chris Caplice (24:50):
Well, it's some other papers that came out too.
Dr. Matt Waller (24:53):
I just don't think I understand right.
I don't know what their memoryactually is, but when you look
at companies, that thesecompanies will try the same
things over and over and fail.
I've seen this in inventorymanagement in particular, which
is a little more where I'minvolved.
(25:15):
But in inventory management,you see cycles of companies
trying to implement things thathave already been tried and
failed, like using a DRPapproach to store item inventory
management, but there's a bunchof examples of this.
Now there's all these companiescoming out saying, well, we
(25:37):
have AI, I don't care what youhave, you can't forecast demand
for head and shoulders shampooat one particular store for one
particular day.
It doesn't matter how much AIyou have.
But you see these kind ofsituations where companies are
(25:59):
oh, and also, I think sometimesconsultants will repackage
things like taking some of thequality management, repackaging
it as re-engineering and thenrepackaging.
They keep repackaging as amillion different things.
Angi Acocella (26:14):
I'm shocked.
I'm shocked and amazed thatthat happens.
That's where the term 4PL camefrom right, because people
already had a 3PL.
Let's make a and I'm waitingfor 5PL.
We'll just say NPL for ingeneral, npl, pl plus 1.
Dr. Matt Waller (26:30):
There you go.
But the other thing I thinkabout on this topic that it
really is buyer beware forwhatever you're talking about.
To your point about carriersnot having a memory or, in
general, just not having, I mean, it doesn't make economic sense
(26:53):
for them to behave like theyhad a memory, right?
Angi Acocella (26:57):
All your grudge rarely benefits.
Dr. Matt Waller (26:59):
Yeah, and if you think about it, I mean,
there's no way they canguarantee that they're going to
have capacity in the right placeat the right time based on a
contract that they were awarded.
It's just not possible.
Angi Acocella (27:18):
And so, going back to the, I'm bringing in the
larger paper.
One of the reasons why wewanted to do this is one where
I've been working in this area along time and Angie's just
getting into it and it's great.
We wanted to kind of tie thingstogether.
But the paper that you referredto, that I wrote back in the
90s, based on my dissertationusing combinatorial optimization
(27:38):
for truckload procurement.
One of the things that made mewant to do this review is
because people still quote thatand use that as a model to go
deeper and refine the models,when in fact, what I've
discovered over the last 10years is that I was dead wrong
and that it doesn't work.
The actual auctions for gettingeconomies of scope and truckload
procurement fails no one usesit.
(27:59):
I actually did a study and sawthat if it's shippers rarely use
it even though it's in thetools Jagger and all those other
tools great tools, good maththey rarely use it.
When they do use it, mostcarriers don't reply back with
package bids and if they doreply pack with package bids
they tend not to win.
And if they do win, it nevermaterializes.
(28:20):
And so the flaw that I had thatmy old dissertation was based
on was that everything wasdeterministic and once you do
this strategic plan, you put itin place and boom, you're done,
when in fact there's so muchvariability and dynamicism for
the underlying for truckloadprocurement that it just failed
miserably.
And so it was a very humblingexperience, and that's why I
(28:43):
wanted to restate but I stillget asked to referee journal
articles that come in and assumethat once you have a strategic
plan, that it works and whatthey focus on is some little
piece of the algorithm to makeit that much more optimal in
planning.
So that's kind of the reasonwhy we kind of did this and
Angie just kind of brought itall together under one unified.
(29:04):
I want to come back.
It all came from my mistake.
Dr. Matt Waller (29:07):
I want to come back to that, but it was an
important mistake actually.
It had to be done.
I own that.
I own that mistake.
Chris Caplice (29:14):
I mean that's how it's a lesson for PhD students.
Dr. Matt Waller (29:16):
Right that their dissertation doesn't have
to change the world Well, Ithink his dissertation did make
a big change, but we have to dothe best we can with what we
have and over time it builds onit and you come up and reading
your and a little bit I want toget to your future research
(29:37):
section, because you do talkabout uncertainty, which I was
happy to see.
But before we do, back to yourpoint about the optimization
piece, so back in and you maynot remember this, chris, but I
started a company called MercariTechnologies back in the 90s
(29:59):
and went through three rounds ofventure capital.
But my idea was let's optimizethe shelf layout, take it into
account, things like case pack,quantity, inventory, carrying
costs, expect a number of unitsout of stock over a
(30:21):
replenishment period, et cetera,et cetera.
Right, and that's what I built.
And so my idea was that asupplier and a buyer could oh,
we had the first rendering of aretail shelf over the internet
using, because the World WideWeb had just came out, and we
(30:42):
created a rendering of that towhere a buyer and a seller could
collaborate and see whathappens when they make changes
and re-optimize.
That part turned out to be morevaluable than the optimization.
Nobody used the optimization andwell, some did.
(31:04):
Actually there were situationswhere they used it.
But I remember I went to talkto a partner.
One of the New Road CapitalPartners founders is a guy named
Doug Dane.
He used to be the chiefmerchandising officer at Walmart
and I don't know what year itwas I want to think it was 98 or
something like that.
I went to him with this.
(31:26):
We had a half hour meeting setup that lasted five minutes and
he said, matt, I would not let ablack box lay out the shelf for
me.
And now I really understand why.
So to your point, chris, right,but I think these kinds of
(31:48):
things are important to advancedknowledge.
Angi Acocella (31:50):
Oh, absolutely Absolutely, Because you don't
know where the benefit's goingto be.
So the real benefit usingoptimization in procurement
makes a ton of sense, but notfor the reasons we developed it.
I developed it because you wantto lock in these package bids,
combination bids and everything,but it's really good at forcing
side constraints, which youcouldn't do before the auction
(32:10):
by itself.
So it's not a pure auction, butyou can have these side
constraints and also with theobjective function, you can
favor or disfavor non-financialfactors.
So it's a huge benefit.
Just like what you described,the benefit was rendering over
the web.
The benefit for this was reallyin bringing optimization into
procurement, just like Slack.
When you use the software Slack, I think from gaming.
(32:32):
It was the gaming tool.
The game failed, but the toolthey used to develop it became
Slack.
So the unintended benefits aresomething.
You never know where it's goingto come from.
Dr. Matt Waller (32:42):
Well, that's such a good point and we see
this in so many differentindustries.
So I guess for because a lot ofearly stage technology
companies listen to this it's animportant lesson to know that.
This is why you don't just giveup or keep pushing something
(33:04):
that's not doing.
You just need to changedirections a little bit and do
what customers really want.
So, angie, I liked yourresearch and the research you
and Chris did on elephants andgoldfish, and I know you're
(33:24):
going to be looking next atshippers.
But what else would you like totell us about that that we
haven't covered?
Chris Caplice (33:33):
Yeah, I think the main thing is what can shippers
do about this, and especiallyas we're going into potentially
a tighter market sometime in thenext few quarters?
And the way I position it isyes, we typically run an annual
bid and you want to lock in aprice.
What this really shows is thatyou can't necessarily do that.
(33:54):
You need to stay on top of howthese contracts are trending
relative to the market and thinkabout updating prices as you go
into a tight market if you wantto keep getting carrier
acceptance.
So that's kind of the maintakeaway.
It's not necessarily thatrelationships are not important,
but you also can't just besetting prices and forgetting
about them.
Dr. Matt Waller (34:13):
That's great.
So when is your research on theshippers going to be finished?
Chris Caplice (34:19):
That's the question you should not be
asking.
Angi Acocella (34:25):
She decided to finish and get a dissertation
instead of doing that.
But we have the data now andwe're in a pretty loose market,
so maybe that's on the plate,but yeah, it is on the plate.
Dr. Matt Waller (34:39):
So we know that carriers behave as goldfish and
you have the data to find outif shippers act as carriers or
goldfish.
And I know you haven't donethat research yet, but you're
going to.
What do you expect to find?
I'm just curious.
Chris Caplice (34:59):
I think shippers are still going to be goldfish
right, so they're still runningbusinesses.
Very often they're looking forlow cost.
They're choosing a low costcarrier.
I think there's still, ofcourse, other things that
they're looking for in terms ofperformance from carriers, but
at the end of the day, veryoften we hear that shippers are
trying to get the lower cost.
(35:20):
Another thing that we find isthat, in these softer market,
shippers may try to pad theirbudget a little bit more, and so
there may be some kind ofgamification of the process that
I would be interested inlooking into.
Angi Acocella (35:33):
I will disagree with Angie on this.
I think they will be moreelephantish not pure, but I
think because they can behavedifferently than carriers.
Carriers can't.
It's a buyer supplierrelationship, so suppliers can't
really pick their buyers, butbuyers can sure pick their
suppliers, and so the whole ideais they could just drop the
carriers.
(35:54):
So I think it's a differentidea of reciprocity, so that
shipper could say you know whatthat carrier screwed me over.
I'm just not going to use thembecause there's so many fish in
the sea.
There were what 200 plusthousand truckload carriers and
of those even even 10,000 thatare of substance.
So I think there'll be moreelephanty.
Chris Caplice (36:13):
But I would say that's even more of a myopic
kind of behavior, right?
So if they're seeing thatcarriers are not performing,
then they're either still beingkind of in the moment making
decisions.
So, yeah, I think we'll let thedata decide.
Yeah, yeah.
Angi Acocella (36:27):
Yeah.
So the question is do theyreciprocate by not even using
those carriers anymore?
Chris Caplice (36:34):
That's true.
Yeah, that's how we have tolook at it.
Yeah, yeah.
Dr. Matt Waller (36:36):
That's true.
I anticipate, based on myexperience, that one other
variable that might affect howshippers behave, whether they're
elephants or goldfish, is howthe percentage of the total cost
that transportation represents.
The reason I say that is I knowI work with some shippers that
(37:03):
have where truckloadtransportation is a high
percentage of their cost.
They've got to behave kind oflike a carrier in that way.
Right, they don't really have achoice, they've got to pick the
low cost.
There's others thattransportation is not a big
percentage and they really arefocused on making sure they get
it on time.
Those shippers tend to developgood relationships with carriers
(37:28):
.
Carriers love working with themand so, anyway, that's
something that crossed my mind.
Angi Acocella (37:39):
And you brought up a really good point and ANGE
has done a lot more work on this.
It might not be across theentire shippers network.
They're freight network, they'repieces of the network that are
different and that's one of thebig things that I think is an
open, an opportunity for futureresearch is the segmentation and
understanding, because everytraffic lane is not the same
right.
Some of the pigs are more equaland so if you look at some lanes
(38:01):
that are high volume, you know,like Walmart, dc to store and
all that that is such highvolume, they're going to use
their own fleet.
At the other extreme, you mighthave a shipper that has two
loads a year on a lane and thepractice has been get a contract
rate on that just in case, andit's a bad strategy, and so
we're seeing a shifting of that.
And this is what ANGE has beendoing more work on is how do you
(38:24):
segment your network, becausethey might be more elephantish
on some areas, but down on thedynamic side I can be a goldfish
really, because the damage isso low, because it's a
transaction, it's not a longterm relationship.
But you want to add any more tothat, because this is what
you've kind of been looking at.
Chris Caplice (38:41):
Yeah.
So you know, there's variouscharacteristics of the network
that I'm finding in my researchthat shippers need to look at,
right.
So a lot of times we'll thinkabout volume.
But it's not just volume, right, it's volume, it's consistency
of volume and frequency ofvolume.
It's a type of carrier whereyou're working with.
Do you typically want to use anasset carrier versus a broker?
They offer different advantages, right.
(39:02):
So there's a lot of differentthings.
Another thing we're seeing iswhat we're calling ghost lanes,
right, where there's contractsset up that never get used.
So why are we actually, why arewe investing in setting up
those contracts?
In fact, there are additionalcosts as repercussions for those
things.
So you know, there's a lot ofdifferent ways that we can think
(39:23):
about whether a shipper isgoing to be an elephant versus a
goldfish.
Another thing is also, you know, a limitation of potential
future research.
So I have great data in that, Ihave a good sense of the
shippers network, right.
But what I don't see is acarrier, the whole carrier's
network.
I have to kind of infer, basedon how much volume or how much
revenue is transacting betweenthe shipper and carrier, kind of
(39:44):
how important at least thatcarrier is to the shipper, but
the flip side of how importantis that shipper to that carrier,
I don't know.
So maybe there's also a call tocarriers that might be
interested in working with usLike this is data that I have
not had access to, and I thinkit'd be really interesting to
get a sense of a carrier'snetwork.
How are they responding toshippers' behaviors?
Dr. Matt Waller (40:06):
That's a great point.
You know, and one of the thingsyou know, chris, when your
paper first came out and I readit, I and I understand the
combinatorial optimization maynot work, but this idea that you
know a wording, you knowcarriers are bidding on lanes
(40:28):
and they bid based on what'sbest for them.
Of course, there's a lot ofuncertainty.
They don't know what's best forthem for sure, but there's some
element they do.
They do know.
And then the shippers, you know, are awarding these lanes and
they're getting better pricesbecause of that, better rates.
(40:50):
And I thought what a neatexample of a Pareto optimal
solution, right, whereeveryone's being made better off
, the carriers and the shippers.
And so this leads into the nextthing I wanted to talk to you
all about, because, yeah, that'sthe ideal, that's what we'd
like to see, a Pareto optimalsolution.
(41:14):
But when there's lots ofuncertainty, it's hard to
achieve that because no oneknows.
But I've always wondered if youknow, if you look at carriers,
shippers don't often consider avery big set of carriers for
(41:35):
their transportation.
A lot of times they do, whenthey work with brokers and
various things.
Sometimes they do, but there isstill a lack of information.
So it's not just uncertainty,it's information asymmetries and
lack of information that arealso leading to this, and I'm
(41:56):
just thinking gosh, this needsto be addressed.
If you think about thepotential impact on our economy,
the ability to make ourcarriers and shippers more
competitive, right, it justseems like a great problem to
solve, and so when I was lookingthrough your paper that was in
(42:17):
Journal of Business Logistics,your review paper you had one
section.
I liked the whole thing, it wasreally good, but I particularly
liked your section on thefuture research questions that
you came up with, and I'd liketo talk about that a little bit,
(42:41):
but before I do, angie, isthere anything else you'd like
to say about this?
Elephants and goldfish paper?
Chris Caplice (42:51):
No, stay tuned for the flip side, I would say.
Dr. Matt Waller (42:54):
Okay, I'm looking forward to it.
You've got me hooked, so now Iwill be following your research.
There we go Now in yourliterature view.
So for those of you that don'tknow and I won't go into the
details on this but Chris andAngie have a very rigorous
(43:19):
review article on transportationprocurement and it's hard to do
a good review article.
Angie isn't in a lot of work.
Chris Caplice (43:33):
Many hours of work.
Angi Acocella (43:34):
yes, I mean none of can go in your dissertation.
Chris Caplice (43:38):
Exactly, this is not part of my dissertation.
Dr. Matt Waller (43:41):
Well, yeah, you could.
So people who haven't done aliterature review as a paper
wouldn't know that you have tocome up with a methodology for
determining which journalsyou're going to look at and
which articles you're going tolook at, and there's all kinds
of tools out there to help you.
One example is Google Scholar.
(44:02):
For those of you everyonelistening probably uses Google,
but if you type in GoogleScholar, it gives you access to
academic articles and you cansearch by year.
You can search by keywords,phrases, you can search by
timeframe, authors, journals, Idon't remember.
(44:24):
There's a long list of thingsyou can do.
Angi Acocella (44:26):
So different than when we were students in the
90s, where you'd have to go tothe Dusty Library.
Chris Caplice (44:33):
I recognize how spoiled I am.
Dr. Matt Waller (44:36):
Chris is right.
I mean, we used to have to goto the libraries and just look
through the journals page afterpage.
It was terrible.
It's still so.
Now that part's easier.
But there's 100 times morearticles out there, so you've
got more volume to come through.
So you have to have asystematic approach to figure
(44:59):
out which articles to include.
And then you have to go throughthe articles and it gets hard
because some of the articlesaren't very good and they're
hard to read and scammed.
And then the other articles aregreat and you can get caught up
in them to where you spend toomuch time reading the articles
that are fun to read.
(45:19):
So, Angie, could you tell us alittle bit about how you did
that?
Chris Caplice (45:24):
Yeah, and I will also say, maybe, before I get
into the details, one thing thatwe find is also in this field
so we're supposed to be lookingat truckload transportation,
procurement, right, so first youhave to also scope what are you
actually looking for?
Right?
So we're not looking at routing, we're not looking at
sustainability, those types ofthings.
One thing that we find is thatthis type of research comes from
(45:45):
many different fields, right,so you can't just look at the
traditional transportationjournals, you can't just look at
operations journals, right, wehave, you know, economics, we
have law policy, right, sothat's also an additional
challenge here.
And so what we did is we came upwith, or we didn't come up with
, but there are lists ofjournals, kind of top journals
in certain fields, right, sothere's a set for the economics
(46:08):
field, there's supply chain andtransportation, and so
essentially, we identified oneof the top journals or lists of
journals that we should startwith, right?
So then, as you said, in GoogleScholar you can also search by
only looking this particularjournal.
And so we came up with thatlist of journals and a set of
keywords related to truckloadtransportation and procurement,
(46:30):
buyer, supplier relationships,things like this.
And then, yeah, just did amassive search in Google based
on these journals, and then wekind of also took an iterative
approach.
So, you know, maybe we find inone of those really good
articles that there's a paperthat's cited that's not part of
one of those journals, but itlooks like it's high quality, it
touches on the topics thatwe're looking in, and so then we
(46:53):
can go back and look at thatpaper and include it if it has
some criteria based on, you knowagain, topic but also quality.
Dr. Matt Waller (47:02):
Great, Chris.
Do you want to add anything tothat?
Angi Acocella (47:09):
Well, agey did all the work.
There's window dressing, no,but out of this, what's really
interesting, the purpose ofdoing a review like this is one
is to prevent other people fromhaving to do it and hopefully
saying, ok, here's where westand, now, stand on these
shoulders and go forward.
But it also identifies the gaps, and I think the gaps that we
identified are the future.
(47:30):
Opportunities really arise fromthe weaknesses of the
approaches.
Because you asked a second ago,you know Pareto, optimal and
all that.
No carrier likes a bid.
You just don't like to be bidout because you're commoditized,
right, you're turning into acommodity, and so every carrier
tries to say, oh, we're not thesame, we do.
You know they try todecommoditize themselves and
(47:51):
shippers, if you're the buyer,is trying to put everyone apples
to apples and try to do that.
So no one likes the RFPs.
But and you said aboutinformation, it is truly
asymmetric, right, they eachhave very different information
and the auction is the best wayof.
You know discovery, right, youdiscover where the where the
best fits are, and that'shopefully why, why it works.
(48:13):
But the challenge is that thething being bid out is based on
their forecast of the future,and it's really everyone uses
history and so the idea if I geta lot of detailed information
that's overfitting on the pastand it doesn't mean the future
is going to be that, becauseI've seen some bids or
procurement events where theygive day of week, week of year,
(48:34):
exact, detailed, but the whatwe're finding now is we look at
longitudinal studies, 50% of thelanes that have less than six
loads a year on it don't show upagain.
So you're mapping really tightlyin this.
You're setting up a strategy onsomething that's ephemeral and
is going to change for a lot ofcases, and so that's kind of the
(48:55):
opportunity we see for thefuture is that don't put all
your expertise in the strategicplanning.
Bill Dragher, when he was atUber Freight, had this phrase
that you know execution, trumpstrategy.
It dominates, and I agree withthat 100% Because it used to be
when we were there.
You optimize up here strategybecause that's the only place
you could do it, and now we'reseeing much more smartness,
(49:19):
expertise in the execution, andthat's what we're kind of saying
.
You need to move down thatpipeline to execution.
That's where you need to bedynamic and do your optimization
Excellent.
Dr. Matt Waller (49:29):
Yeah, that's a great description and you know
to your point I mean.
That's why I wanted to talk alittle bit about the gaps that
you identified.
The future, because you posethis as future research, and I
think that you should, becauseit's an academic article, but
the other side of it is.
(49:49):
It shows where there'sopportunity for entrepreneurial
activity Absolutely Right.
And so, for those of youlistening to this, one of the
benefits of being familiar withthe review article in the area
you're doing business in isbecause it gives you a window
into one, what's probably goingto come out in the future.
(50:12):
Two, what your competitors aregoing to probably start working
on.
And three, what are yourbiggest opportunities here.
So, again, angie and Chrisposed this as future research,
but I think that it's more thanthat.
Angi Acocella (50:31):
I agree.
And so Angie spent the last twoyears in Europe.
And I'm curious, angie, whatare you seeing for the
opportunities, because you'vebeen talking about this.
You're working with companiesover there.
What do you see as the bigopportunities in Europe?
And then I'll respond what I'mseeing in the US.
Chris Caplice (50:44):
Yeah, so a big thing we're seeing is in this
digitization space, digitallogistics platforms, matching
things like this.
There's a lot of money pouringinto it, there's a lot of
identifying potentialefficiencies that can be gained,
and that's at all levels.
That's not this research, and alot of my research is focused
on truckload, but at ports, soI'm working with some of the
(51:08):
ports here.
So just on handling containers,but documentation, can we get
companies more into the digitalspace?
So that's one of the big onesthat I see.
And then also, I think the USis a little bit more advanced
than Europe in terms of havinggood benchmark pricing.
(51:29):
So a lot of the companies thatI work with also don't really
know how to price well becausethere's not, at least at this
point, a lot of informationabout what are the right, what
are benchmark prices for certaincorridors.
I would say those are the twobig ones that come up a lot for
opportunities for practitionersas well.
Angi Acocella (51:47):
So let me respond to that, because I'm also, in
addition to being at MIT, I'malso a DAT chief scientist.
They bought chain analytics,which we pioneered the
benchmarking of the US truckloadrates and it's pretty big now.
We used to have Europeanoperations but we found nothing
against.
Europeans love it, they arejust really cheap.
(52:08):
We could not get a benchmarkingservice sustainable over there.
We tried for three years and itwas so hard.
And it's such an interestingmarket because, yeah, it's the
EU, but talk about all thedrivers come from Poland, right,
and they put up every countryputs up barriers, even though
it's one big union.
There's all these differentthings that we don't have in the
(52:28):
US and you know Texas doesn'tcheat drivers from Florida out,
so it's a different kind ofdynamic.
But in the US the big thingthat I'm seeing is two big
things.
One is the portfolio.
Shippers are starting torecognizing that.
You know you need to have aportfolio of relationships and
that's why brokers when Istarted in the industry in the
90s, broker was a four letterword.
(52:49):
No one wanted to use a broker.
They were the carrier of lastresort.
Now they see where they fit in.
They fit in a niche for thosesparse lanes and that long tail
of carriers.
It's a perfect match there.
But the other thing that I'mseeing is the use of dynamic
contracts, because we've beenthrowing around the term
contract rates and everything,and if you're not in this
industry, you think a contract'sa contract.
(53:11):
But truckload contracts are alittle different.
They're binding in price butnot in the volume the shipper
guarantees or the capacity thecarrier guarantees.
There's wiggle room.
It's the only industry Mattmaybe you know another where
adherence to contract is aservice metric.
It's not violating a contract,it's like a metric, like on time
for level of service.
(53:33):
So it's different.
And so what I'm seeing is thereare more dynamic contracts that
are saying, ok, we won't set theprice, we'll have the price
tied to an index.
Maybe it's every day, itchanges, and then I'll be on the
guardrails around that, so Idon't need to have a set price
ahead of time.
Maybe I can have a dynamiccontract that's dynamic in terms
(53:53):
of the capacity provided day today and the price that's
provided day to day.
I don't need to set the pricefor a year out, and I'm seeing
that for some of the lanes.
Again, that goes to theportfolio where you go dedicated
, traditional contract, dynamiccontract, pure spot and maybe
some nuances in between.
So that's where I see the realdevelopment going and the real
opportunity.
You mentioned time to start upa company, matt.
(54:16):
No kidding, do it.
Just give me some of that money, give me some investment money,
we'll do something amazing.
Dr. Matt Waller (54:23):
You're right, that is a big opportunity.
You mentioned chain of letticksand of course it's a small
world out there.
Of course, mike Kelgore, one ofthe founders of chain of
letticks, and I were friends ingraduate school.
He was in the master's programwhen I was in the PhD program
and we became friends and he wasinterested in network
(54:46):
optimization and stuff like thatback then.
But it's been really fun to seehow you and he and others I
don't know, did you know JeffZeroyah very well?
Angi Acocella (55:00):
Yes, oh, I know Jeff, jeff, it's a funny story
about Jeff.
When I used to givepresentations on this at CSCMP
and he was at I2 at the time, Iwas with a company called
Logisticscom what it was calledit's called PDCG.
Jeff would always be in thefricking front row taking notes.
You're taking it straight backto I2.
I knew you are, so I gave him ahard time, but then you gave
(55:21):
him a joint chain of letticks.
But I have to say for Mike.
Mike is a great podcast guest.
He has so many good stories, somany good stories.
Dr. Matt Waller (55:29):
And he's a good storyteller too.
He is, yeah Well, I'll have to.
I haven't talked to him for awhile, but this whole industry
is pretty small it is.
But thank you both for takingtime to visit with me today, and
, angie, I think you've got agreat mentor there.
(55:50):
And Chris and Chris, thank youfor continuing to publish
research that I enjoy reading.
I appreciate that it's kind ofa rare thing for me, so keep up
the good work, thanks, thanks.
Angi Acocella (56:03):
I have to say, if you look at elephants and
goldfish in academia, we areelephants.
I remember.
Dr. Matt Waller (56:11):
We are elephants.
Angi Acocella (56:12):
Academics are elephants Memory.
Dr. Matt Waller (56:15):
Yes.
Angi Acocella (56:16):
That's true.
I wish I was a goldfish, butI'm sad.
Dr. Matt Waller (56:21):
Well, it was great to visit with you all.
Take care.
Angi Acocella (56:23):
Thank you.
Dr. Matt Waller (56:24):
Thanks.
Thanks, Matt.
If you're finding value in thispodcast, we greatly appreciate
your support by subscribing toour YouTube channel.
Additionally, following us onApple and Spotify and leaving up
to a five-star review would beimmensely helpful.
We welcome any feedback orquestions related to the podcast
, as well as suggestions forfurther topics and guests.
(56:46):
You can leave your comments onour YouTube channel and rest
assured that I will read eachand every one of them.
Please also take a moment tocheck out our podcast sponsors,
as they play a critical role inkeeping this podcast running.
For more information onspecific topics, timestamps or
links to articles mentionedduring the podcast, head over to
(57:08):
mattwallarpodcastcom.
Welcome to the Matt Waller podcast, where we
look at success at theintersection of technology,
logistics, supply chain, retailand CPG, also known as the
retail value chain.
I want to clarify that thispodcast is distinct from my
responsibilities as a professorin the Sam M Walton College of
Business.
Nonetheless, it aligns with myaspiration to provide practical
(00:31):
insights to professionals andbusiness by showcasing companies
and people that can enhanceyour ability to manage, lead and
strategize and marketeffectively and the retail value
chain.
Before we dive into today'sexciting episode, I'd like to
thank our sponsor, new RoadCapital Partners.
New Road invests in proventechnology services and products
(00:54):
that serve unmet needs in themarketplace.
They look for companies insupply chain and logistics, as
well as consumer-orientedcompanies.
For more information, go tonewroadcpcom.
I would also like to disclosethat I am a strategic advisor to
New Road.
I'd also like to recognizepodcastvideoscom for the
(01:18):
services they provide for thesepodcasts.
I'm very pleased with theirservices and now, without
further ado, let's get into theexciting episode.
I have with me today ChrisKaplis and Angie Akechella, and
we're talking abouttransportation procurement,
(01:41):
particularly truckloadtransportation procurement, and
I have known Chris for about 30years, I think it is.
Chris is a senior researchscientist at MIT.
He's also the chief scientistat DAT Solutions.
He is an academic who hasstayed very involved with
(02:05):
industry his whole career and hehas his PhD from MIT.
And Angie also got her PhD fromMIT and she is now a
postdoctoral researcher atTilburg University in the
(02:29):
Netherlands.
And the reason I invited themto the podcast is, of course,
this podcast focuses on what Icall the retail value chain,
which includes supply chaintechnology, retail and marketing
technology and consumer goodscompanies.
(02:50):
So I read an article that theyhad published in an academic
journal called the Journal ofBusiness Logistics and I really
liked it and you'll see when youlisten to the podcast it's very
relevant to this podcast.
(03:10):
But they had an article theypublished in Journal of Business
Logistics called Research onTruckload Transportation
Procurement a review frameworkand future research agenda.
And one thing you'll see in ourdiscussion if you're a shipper
managing freight or if you're acarrier or if you're a technical
(03:31):
carrier or if you're atechnology company in supply
chain, you will want to listento this because at the end of
their article they talk aboutfuture research opportunities
and you'll hear us point out inthe interview that it is true
it's future research, but it's.
Also it can tell you wherecompetition is going, what kinds
(03:55):
of new technologies are goingto be developed, and so if
you're a technology company, itgives you an idea of what to
look at for new features,functions and capabilities, and
if you're a carrier or a shipper, it gives you an idea of where
(04:16):
the industry probably is goingto go.
But I also interviewed themabout an article that was in
Transportation Research, part E,called Elephants or Goldfish an
empirical analysis of carrierreciprocity and dynamic freight
markets.
And the question is do carriersact like elephants, meaning
(04:37):
they've got a long memory, or dothey act like goldfish, meaning
they don't have a memory ofwhat happened and so they just
optimize for that particularperiod of time?
I think you'll find itinteresting.
Again, chris has a long, long30-year experience in truckload
procurement and I think you'llfind it very interesting.
(05:02):
Thank you, chris and Angie.
Thank you so much for joiningme today to talk about some of
the research you've done intransportation.
Really appreciate it.
Chris Caplice (05:13):
Yeah, great to be here, thanks.
Angi Acocella (05:14):
Yeah, glad to be here.
Interesting for theconversation.
Dr. Matt Waller (05:18):
So just for people listening to understand,
I've known Chris for many yearsand I read an article in a
recent article that heco-authored with Angie called
Research on TruckloadTransportation Procurement a
(05:39):
review framework in futureresearch, and I really like it
and I typically don't likereview articles, but I like this
one a lot, and so I contactedhim and asked him if he would
mind talking about it and hesent me another article that I
also liked that was published.
The first one was published inJournal of Business Logistics.
(06:02):
He also he and Angie publishedone called Elephants or Goldfish
Question Mark an empiricalanalysis of carrier reciprocity
and dynamic freight markets, andthat was published in
Transportation Research Part E.
So I'm going to be talking tothem about that and thank you
(06:22):
both for your willingness totalk about that.
We'll get right into it.
The first thought I had wasyour paper in TRE is talking
about opportunistic behavior,and can you explain a little bit
(06:42):
about what is meant byopportunistic behavior and the
context of truckload freightcontracts?
Chris Caplice (06:51):
Yeah, maybe I can start by talking about the
market cycles, right?
So I think it's really importantto think about when we're in a,
you know, a softer market kindof what we've been seeing
recently, where you have a lotmore capacity than there is
demand for capacity, right, andso this is where you have prices
that are going down, you havevery high acceptance rates, so
(07:11):
your contracted carriers areaccepting a lot of your freight.
And then we have the flip sideof a tight market, right, a
constrained market, where youhave a lot more demand than
there is capacity for thatmarket.
You see prices going up, yousee acceptance rates going down.
And so when we think aboutopportunistic behavior
particularly in this study,we're looking at carriers
behavior we think about, youknow, if I have a contract for a
(07:33):
particular lane, am I going toaccept that contracted volume in
a tight market?
When I have, I'm constrained inmy capacities, maybe there's
higher prices elsewhere, or am Imaybe using the capacity that I
have and fulfilling higherprice, higher, you know, using
higher price load?
So am I, do I want to maintainthat contract?
Or am I acting a little moreopportunistically and, you know,
(07:55):
going for the higher pricedcontract, higher price chipmence
?
Angi Acocella (07:59):
Yeah.
So to add to Angie's point,everyone knows the truckload
market moves through thesecycles and the question that
really Angie honed in on thehonor dissertation but MIT is
reciprocity and the idea of who.
When is there an incentive todefect from the contract?
Because when the market's tightthe spot market is usually way
higher than the contract.
(08:20):
So the incentives to thecarriers to defect and say, well
, I'm not going to haul that,I'm not going to honor the
contract and go that way whenthe market's loose it's just the
opposite.
Well, we've been in since thespring of 22,.
The rate is spot rates belowcontract, so the incentive to
defect is at the shippers nowand they say, why should I honor
(08:40):
that contract when I go to thespot market and go cheaper?
So the incentive to defect thecontract flips depending on the
market.
And that's what Angie's greatwork really identified and
quantified at the impact.
Dr. Matt Waller (08:54):
So it really is an interesting topic and
everyone's aware of it and yourresearch question is really good
, and so could you talk a littlebit about the methodology you
use to do the empirical analysisof reciprocity?
What kind of metrics orindicators did you look at to
(09:18):
determine whether carriersresponded to past behaviors or
shippers?
Chris Caplice (09:26):
Yeah, so we were actually really lucky to have a
really detailed data set oftransactions.
So we had data from 2015 to2019.
And essentially we have all ofthe transactions for about 70
different shippers over thattimeframe.
For the continental US,specifically, full truckload,
and so we have information onthe price, whether a carrier is
(09:48):
contracted and if they acceptthat shipment.
That's key, of course,geographic regions, things like
this.
We end up with about 2 milliontransactions in that data set
and essentially we look at overthat time period.
First, we need to identify whenwas there a soft market, when
was there a tight market?
So 2017, 2018 was a very tightmarket and so we were really
doing a lot of this researchduring that time.
(10:09):
So we were really pinpointingthat timeframe.
So we define those marketperiods which we can talk a
little more detail with, andthen we look at what are the
characteristics of shippers andtheir contracted carriers, where
their performance, theirbehaviors during the soft market
.
Then we look at that tightmarket between those same
shipper carrier pairs and say,well, does that historical
(10:32):
relationship between the shipperand carrier in the soft market,
when, as Chris said, when thepower is more in the shipper's
hand, does that translate into acarrier in a tight market
prioritizing that shipper'sfreight.
And so we looked at things likein that soft market, what is the
contract price competitivenessof shipments?
So, essentially, relative to alane benchmark, how are those
(10:56):
shippers pricing those contractswhen maybe they don't have to
be too high, right, so we lookat contract price
competitiveness.
We look at the consistency andthe volume that's going on those
lanes.
So, is a very consistent lane?
Is it very volatile?
Is it a very infrequent lane?
We looked at things like thedwell time at facilities.
(11:16):
Right, so?
Is a shipper delaying carrierover time at its facilities and
then impacting its ability toget to follow on shipment?
So we looked at both thosethings in a soft market and then
also then in a tight market.
How are carriers responding tothose things?
Angi Acocella (11:32):
This is something that, matt, you enjoy, as well
as having a strong relationshipwith industry and having access
to this data.
So this was a partnership withCH Robinson, specifically the
TMC division, with the greatSteve Rates was leading that,
who was just retiring.
I think today might be his lastday at CH Robinson, but because
of him he kind of steered andenabled this kind of analysis
(11:55):
great analysis that Angie wasable to conduct.
So I encourage anyone listeningto this.
This is how you work withacademia.
Work with us to help you bringthe data and give us the
resources.
Then we can do this analysisthat brings insights that
industry by itself just doesn'thave the time to uncover, but
then they can use.
So that kind of makes sense.
(12:16):
But, matt, you've read thepaper but your thought going in,
did you think carriers weregoldfish or elephants?
And the difference is do theyremember things?
Elephants remember everything.
Goldfish have.
What a 10 second memory.
Yeah.
Dr. Matt Waller (12:30):
So I will answer that question because I
tried to guess before I read it.
But I know they're not theelephant in the room, so that
couldn't be it.
That was actually the firstmetaphor that came to my mind
and then I thought oh yeah,elephants remember a long time
(12:50):
into goldfish, don't?
But let me ask you a quickquestion before we get into that
.
But a quick answer and I'llexplain why in a minute.
Goldfish is what I expected so,and I'll tell you why in a
moment.
Before I do, you mentionedsomething about industry and
academia working together, andI'm glad you mentioned that.
(13:16):
Chris, and you have been sogood at that over the years,
from the very beginning.
I remember because when youpublished a paper from your
dissertation in Journal ofBusiness Logistics I don't
remember the year, but it seemslike was around the mid 90s I
read it and really liked it.
(13:36):
It was talking about carrierbid optimization I think that's
what we used to call it backthen, and I actually used.
I taught a master's classcalled Logistics Modeling and I
used that paper because Ithought, yeah, this is a long
(13:59):
time ago, obviously, but it was.
It's hard to find good papersout there that are strong
academically and extremelyrelevant at the same time, and I
could have a whole conversationon this, the academic, what it
takes to get published andgetting citations.
Angi Acocella (14:22):
You have all of researchers that just have to
cite each other and thereforetheir age index goes up.
But no one else reads it exceptfor those H number of academics
.
And then I don't know anypractitioners that read some of
the most respected journals.
I love some of thesetransportation sciences.
They're great journals.
No one reads it outside thesmall collection of research.
Dr. Matt Waller (14:47):
Right and it seems like I would have
estimated seriously back in the90s when I started as a
professor in 93, and, as youknow, I've always been linked to
industry and done research andcollaboration with industry.
It's one of the reasons why Ifollowed you.
But you're talking aboutprofessors citing one another.
(15:11):
You see that more in somedisciplines than others.
I agree.
I really got to see thatclearly when I spent my eight
years as dean because I wasdoing promotion and tenure
review for all these eight.
We have eight academicdepartments in our college and I
(15:31):
saw like in finance they don'tcite each other much.
In information systems they gooff the deep end on citing each
other.
Have you noticed that?
Angi Acocella (15:42):
I think there should be a limit on the number
of citations you're allowed toinclude in an article, say 10.
You can only do 10.
Unless you're doing a reviewarticle, right?
Oh yeah, Then that's what it'sthe purpose of it.
So instead of people doing theresearch themselves, they can
look at our bibliography.
That's the goal.
Dr. Matt Waller (16:00):
But your call that you made just here a minute
ago is really good, and most ofthe people listening to this
are practitioners, not academics, I think.
At least that's been true up tonow, and so You're right.
I mean, I think that there's agrowing number of academics who
(16:23):
want to work with industry, andsometimes it's hard to get data,
and so if industry collaboratedwith them a little more, it
would probably solve thisproblem to a large degree.
Angi Acocella (16:36):
And I think it's certain areas there's some
people in like pure inventory,that are tied to.
Like Sean Willems does stuff.
Everything he does is with thecompany, but in our, in my area,
in Angie's area, justin Millerdoes a lot of his stuff is all
applied.
Alex Scott, down at where is henow?
Tennessee, tennessee, very muchapplied, and so I think there's
a pocket of this out there, butthen there's a lot of stuff you
(16:59):
know where it's.
Just that's right.
Yeah, different audience.
Dr. Matt Waller (17:03):
For those listening, Jason Miller is at
Michigan State and I'll tell you, Chris, the only person I look
at their post every day onLinkedIn is Jason Miller and he
posts every day.
And it's great isn't it?
Angi Acocella (17:22):
Yeah, yeah, good stuff.
Angie, you're really sayingbetter than his posts is his
mustache.
Yes, I love that.
Chris Caplice (17:30):
I was going to add that, yeah, no, it's okay.
I was going to add that youknow one of the the also the
benefits of working withcompanies for for, hopefully for
them, but for us as well.
You know it's partially ourresponsibility as the
researchers one to get thepapers out there, but you know
Jason's really good at this atthen also creating, you know,
publications that are digestibleby industry, right.
So so that's what some of thefun part for us as well.
(17:51):
As you know these types ofpodcasts or LinkedIn posts or
blog posts.
You know there's so much morethat can be done with, with
research that's that's sponsoredby or partnered with industry,
and it's part of ourresponsibility to kind of help
get that out there.
Dr. Matt Waller (18:04):
Well, I'm glad you're doing that and again, I
think it's one of the reasons Ienjoyed reading your article.
But but you know, right now, infact, I'm doing some research
on omni-channel retail supplychain management and, and so I
did a literature review onco-authoring with three other
(18:24):
faculty members who arewonderful people that are very
involved in industry and and I.
We identified the journals thatneeded to be included and I
took one of the journals of itand it's a top journal, and they
actually had a large number ofpublications on omni-channel.
(18:47):
Surprisingly, I was surprisedhow many of them used strictly
analytical models.
You know, assume there'shomogeneous buyers, right,
whatever you know had, had, had,had a linear decreasing demand
focus and all of that.
And they get counterintuitiveresults right, which is desired
(19:14):
many times.
And I'm thinking I've beeninvolved in omni-channel
consulting and and and and workand research for since it began
and I can tell you some of theresults of these papers are
absolutely wrong.
There's nothing.
(19:34):
Yeah, they are counterintuitive, but they're they're wrong.
I I can't even imagine asituation.
You're an amazed man.
Angi Acocella (19:42):
I can't believe that.
So I we joke that it's.
A lot of these papers areTrojan.
They use trucking as a Trojanhorse.
They say this is why this isimportant.
It's a big industry, blah, blah, blah.
And here let me use this mathto prove a point that doesn't
make sense.
Let me assume away all theinteresting part.
Chris Caplice (19:59):
Assume deterministic demand yeah.
Angi Acocella (20:03):
I remember what text part is yeah, go ahead.
No, go ahead.
When you, when you interviewpostdocs we interview a lot of
postdocs coming in and you mustas well it's just depressing
sometimes when they will talkand they pretty much will sue
everything away.
They come with this greatsolution, like we had one for
doing optimal LTL network design, and we said, really this seems
(20:25):
pretty interesting and wasoptimal for a two terminal
system.
It's like okay, that's, youknow, what do you do with that?
But you know, but they havecontributions in other areas.
Math is improving andeverything, but to me it's not
applied research.
And if you're doing somethingin supply chain, it should be
applied.
We're not a theoretical science.
Dr. Matt Waller (20:44):
I agree, you know, if you think about it,
with some of these.
When I was a doctoral studentat Penn State, they asked me to
do an exec ed program for somecompany on forecasting sales
(21:05):
forecasting and I did it.
It was a good lesson for me.
Angie, you mentioneddeterministic demand.
I said let's start out withdeterministic demand.
And they're like well, whywould you need to forecast it?
That's a good point, but at anyrate, so to answer your
(21:26):
question, chris, yeah, I, myfirst thought of the metaphor
when I saw elephant was elephantin the room.
I thought, well, that can't beit.
I don't know how that wouldapply here.
Everyone knows thattransportation is out there.
There's so much of them.
If you look at truckload.
I could be wrong about this,but I think truckload
transportation is around in theUnited States around 700 billion
(21:51):
.
Angi Acocella (21:52):
Transportation.
At trucking it's fulltruckloads about 400.
Okay, and private is about 400,and so it's about 800.
80, 90 billion.
Dr. Matt Waller (22:01):
Okay so, but yeah, I thought, regarding the
question about reciprocity, I sothe reason I thought it would
probably be goldfish if I had toguess wasn't probably because
(22:21):
of the right reason.
But my experience in working inindustry is and any
organization is that there areshort memories.
Now in academic organizationswhere you have professors that
stay their whole careers, likeone of your co-authors who on
(22:42):
the JBL paper, yoshi YoshiSheffi, I think.
He's been at MIT for almost 50years.
If I'm not mistaken, he came inthe late 70s.
Yeah so you know.
But but in most industry andindustry in general, there's a
(23:04):
lot of turnover, usually notalways.
Angi Acocella (23:07):
There's exceptions to that, obviously,
but you see, there's a lot ofturnover in shippers in that
role, because they'll they'llrotate through, so someone will
be in there or they'll move toanother company.
But it's funny, if you talk inindustry, everyone says oh yeah,
we have a long memory.
The, the perception, the commonwisdom for carriers is that,
yeah, we don't forget, you'reexactly right.
(23:29):
Is that they're?
They're little goldfish.
They don't remember.
They might remember, but theydon't act on it.
Well, angie, do you want tocharacterize that?
Chris Caplice (23:37):
Yeah, sure, yeah.
So I mean that was also some ofthe motivation of this study
was, you know, well, you knowthat's there's a lot of
conversation around this.
Let's put some numbers to it,right.
And so, yeah, we were.
We were looking at, well, youknow, is in that tight market,
are carriers responding?
Are they, are they prioritizingshippers that had behaved
better in the past?
Turns out they're not right.
But but the way that we'reunderstanding and explaining it
(24:00):
is that you know it's not thatthey're not prioritizing a
shippers good behavior.
It's really that they'refocusing on what are they
currently doing in this currentmarket condition, because the
reality is they, they arecapacity constrained, they have
businesses to run right.
So they're they're looking forshippers and saying, okay, if
you have good performance now inthis tight market, I'll work
with you.
So it's not saying, you know,shippers shouldn't care about
(24:21):
the relationship.
It's just that the reality iscarriers have constraints,
capacity constraints andbusiness constraints.
Angi Acocella (24:28):
They can't afford to be elephants?
Yeah, absolutely.
But the question is, the realinteresting question is what
about shippers?
Yeah, are they elephants or arethey?
Dr. Matt Waller (24:38):
goldfish.
I would have guessed they'regoldfish.
Angi Acocella (24:43):
We don't know yet , because Angie hasn't finished
the work, that is, it's allblack and all finishing up this
station.
Chris Caplice (24:50):
Well, it's some other papers that came out too.
Dr. Matt Waller (24:53):
I just don't think I understand right.
I don't know what their memoryactually is, but when you look
at companies, that thesecompanies will try the same
things over and over and fail.
I've seen this in inventorymanagement in particular, which
is a little more where I'minvolved.
(25:15):
But in inventory management,you see cycles of companies
trying to implement things thathave already been tried and
failed, like using a DRPapproach to store item inventory
management, but there's a bunchof examples of this.
Now there's all these companiescoming out saying, well, we
(25:37):
have AI, I don't care what youhave, you can't forecast demand
for head and shoulders shampooat one particular store for one
particular day.
It doesn't matter how much AIyou have.
But you see these kind ofsituations where companies are
(25:59):
oh, and also, I think sometimesconsultants will repackage
things like taking some of thequality management, repackaging
it as re-engineering and thenrepackaging.
They keep repackaging as amillion different things.
Angi Acocella (26:14):
I'm shocked.
I'm shocked and amazed thatthat happens.
That's where the term 4PL camefrom right, because people
already had a 3PL.
Let's make a and I'm waitingfor 5PL.
We'll just say NPL for ingeneral, npl, pl plus 1.
Dr. Matt Waller (26:30):
There you go.
But the other thing I thinkabout on this topic that it
really is buyer beware forwhatever you're talking about.
To your point about carriersnot having a memory or, in
general, just not having, I mean, it doesn't make economic sense
(26:53):
for them to behave like theyhad a memory, right?
Angi Acocella (26:57):
All your grudge rarely benefits.
Dr. Matt Waller (26:59):
Yeah, and if you think about it, I mean,
there's no way they canguarantee that they're going to
have capacity in the right placeat the right time based on a
contract that they were awarded.
It's just not possible.
Angi Acocella (27:18):
And so, going back to the, I'm bringing in the
larger paper.
One of the reasons why wewanted to do this is one where
I've been working in this area along time and Angie's just
getting into it and it's great.
We wanted to kind of tie thingstogether.
But the paper that you referredto, that I wrote back in the
90s, based on my dissertationusing combinatorial optimization
(27:38):
for truckload procurement.
One of the things that made mewant to do this review is
because people still quote thatand use that as a model to go
deeper and refine the models,when in fact, what I've
discovered over the last 10years is that I was dead wrong
and that it doesn't work.
The actual auctions for gettingeconomies of scope and truckload
procurement fails no one usesit.
(27:59):
I actually did a study and sawthat if it's shippers rarely use
it even though it's in thetools Jagger and all those other
tools great tools, good maththey rarely use it.
When they do use it, mostcarriers don't reply back with
package bids and if they doreply pack with package bids
they tend not to win.
And if they do win, it nevermaterializes.
(28:20):
And so the flaw that I had thatmy old dissertation was based
on was that everything wasdeterministic and once you do
this strategic plan, you put itin place and boom, you're done,
when in fact there's so muchvariability and dynamicism for
the underlying for truckloadprocurement that it just failed
miserably.
And so it was a very humblingexperience, and that's why I
(28:43):
wanted to restate but I stillget asked to referee journal
articles that come in and assumethat once you have a strategic
plan, that it works and whatthey focus on is some little
piece of the algorithm to makeit that much more optimal in
planning.
So that's kind of the reasonwhy we kind of did this and
Angie just kind of brought itall together under one unified.
(29:04):
I want to come back.
It all came from my mistake.
Dr. Matt Waller (29:07):
I want to come back to that, but it was an
important mistake actually.
It had to be done.
I own that.
I own that mistake.
Chris Caplice (29:14):
I mean that's how it's a lesson for PhD students.
Dr. Matt Waller (29:16):
Right that their dissertation doesn't have
to change the world Well, Ithink his dissertation did make
a big change, but we have to dothe best we can with what we
have and over time it builds onit and you come up and reading
your and a little bit I want toget to your future research
(29:37):
section, because you do talkabout uncertainty, which I was
happy to see.
But before we do, back to yourpoint about the optimization
piece, so back in and you maynot remember this, chris, but I
started a company called MercariTechnologies back in the 90s
(29:59):
and went through three rounds ofventure capital.
But my idea was let's optimizethe shelf layout, take it into
account, things like case pack,quantity, inventory, carrying
costs, expect a number of unitsout of stock over a
(30:21):
replenishment period, et cetera,et cetera.
Right, and that's what I built.
And so my idea was that asupplier and a buyer could oh,
we had the first rendering of aretail shelf over the internet
using, because the World WideWeb had just came out, and we
(30:42):
created a rendering of that towhere a buyer and a seller could
collaborate and see whathappens when they make changes
and re-optimize.
That part turned out to be morevaluable than the optimization.
Nobody used the optimization andwell, some did.
(31:04):
Actually there were situationswhere they used it.
But I remember I went to talkto a partner.
One of the New Road CapitalPartners founders is a guy named
Doug Dane.
He used to be the chiefmerchandising officer at Walmart
and I don't know what year itwas I want to think it was 98 or
something like that.
I went to him with this.
(31:26):
We had a half hour meeting setup that lasted five minutes and
he said, matt, I would not let ablack box lay out the shelf for
me.
And now I really understand why.
So to your point, chris, right,but I think these kinds of
(31:48):
things are important to advancedknowledge.
Angi Acocella (31:50):
Oh, absolutely Absolutely, Because you don't
know where the benefit's goingto be.
So the real benefit usingoptimization in procurement
makes a ton of sense, but notfor the reasons we developed it.
I developed it because you wantto lock in these package bids,
combination bids and everything,but it's really good at forcing
side constraints, which youcouldn't do before the auction
(32:10):
by itself.
So it's not a pure auction, butyou can have these side
constraints and also with theobjective function, you can
favor or disfavor non-financialfactors.
So it's a huge benefit.
Just like what you described,the benefit was rendering over
the web.
The benefit for this was reallyin bringing optimization into
procurement, just like Slack.
When you use the software Slack, I think from gaming.
(32:32):
It was the gaming tool.
The game failed, but the toolthey used to develop it became
Slack.
So the unintended benefits aresomething.
You never know where it's goingto come from.
Dr. Matt Waller (32:42):
Well, that's such a good point and we see
this in so many differentindustries.
So I guess for because a lot ofearly stage technology
companies listen to this it's animportant lesson to know that.
This is why you don't just giveup or keep pushing something
(33:04):
that's not doing.
You just need to changedirections a little bit and do
what customers really want.
So, angie, I liked yourresearch and the research you
and Chris did on elephants andgoldfish, and I know you're
(33:24):
going to be looking next atshippers.
But what else would you like totell us about that that we
haven't covered?
Chris Caplice (33:33):
Yeah, I think the main thing is what can shippers
do about this, and especiallyas we're going into potentially
a tighter market sometime in thenext few quarters?
And the way I position it isyes, we typically run an annual
bid and you want to lock in aprice.
What this really shows is thatyou can't necessarily do that.
(33:54):
You need to stay on top of howthese contracts are trending
relative to the market and thinkabout updating prices as you go
into a tight market if you wantto keep getting carrier
acceptance.
So that's kind of the maintakeaway.
It's not necessarily thatrelationships are not important,
but you also can't just besetting prices and forgetting
about them.
Dr. Matt Waller (34:13):
That's great.
So when is your research on theshippers going to be finished?
Chris Caplice (34:19):
That's the question you should not be
asking.
Angi Acocella (34:25):
She decided to finish and get a dissertation
instead of doing that.
But we have the data now andwe're in a pretty loose market,
so maybe that's on the plate,but yeah, it is on the plate.
Dr. Matt Waller (34:39):
So we know that carriers behave as goldfish and
you have the data to find outif shippers act as carriers or
goldfish.
And I know you haven't donethat research yet, but you're
going to.
What do you expect to find?
I'm just curious.
Chris Caplice (34:59):
I think shippers are still going to be goldfish
right, so they're still runningbusinesses.
Very often they're looking forlow cost.
They're choosing a low costcarrier.
I think there's still, ofcourse, other things that
they're looking for in terms ofperformance from carriers, but
at the end of the day, veryoften we hear that shippers are
trying to get the lower cost.
(35:20):
Another thing that we find isthat, in these softer market,
shippers may try to pad theirbudget a little bit more, and so
there may be some kind ofgamification of the process that
I would be interested inlooking into.
Angi Acocella (35:33):
I will disagree with Angie on this.
I think they will be moreelephantish not pure, but I
think because they can behavedifferently than carriers.
Carriers can't.
It's a buyer supplierrelationship, so suppliers can't
really pick their buyers, butbuyers can sure pick their
suppliers, and so the whole ideais they could just drop the
carriers.
(35:54):
So I think it's a differentidea of reciprocity, so that
shipper could say you know whatthat carrier screwed me over.
I'm just not going to use thembecause there's so many fish in
the sea.
There were what 200 plusthousand truckload carriers and
of those even even 10,000 thatare of substance.
So I think there'll be moreelephanty.
Chris Caplice (36:13):
But I would say that's even more of a myopic
kind of behavior, right?
So if they're seeing thatcarriers are not performing,
then they're either still beingkind of in the moment making
decisions.
So, yeah, I think we'll let thedata decide.
Yeah, yeah.
Angi Acocella (36:27):
Yeah.
So the question is do theyreciprocate by not even using
those carriers anymore?
Chris Caplice (36:34):
That's true.
Yeah, that's how we have tolook at it.
Yeah, yeah.
Dr. Matt Waller (36:36):
That's true.
I anticipate, based on myexperience, that one other
variable that might affect howshippers behave, whether they're
elephants or goldfish, is howthe percentage of the total cost
that transportation represents.
The reason I say that is I knowI work with some shippers that
(37:03):
have where truckloadtransportation is a high
percentage of their cost.
They've got to behave kind oflike a carrier in that way.
Right, they don't really have achoice, they've got to pick the
low cost.
There's others thattransportation is not a big
percentage and they really arefocused on making sure they get
it on time.
Those shippers tend to developgood relationships with carriers
(37:28):
.
Carriers love working with themand so, anyway, that's
something that crossed my mind.
Angi Acocella (37:39):
And you brought up a really good point and ANGE
has done a lot more work on this.
It might not be across theentire shippers network.
They're freight network, they'repieces of the network that are
different and that's one of thebig things that I think is an
open, an opportunity for futureresearch is the segmentation and
understanding, because everytraffic lane is not the same
right.
Some of the pigs are more equaland so if you look at some lanes
(38:01):
that are high volume, you know,like Walmart, dc to store and
all that that is such highvolume, they're going to use
their own fleet.
At the other extreme, you mighthave a shipper that has two
loads a year on a lane and thepractice has been get a contract
rate on that just in case, andit's a bad strategy, and so
we're seeing a shifting of that.
And this is what ANGE has beendoing more work on is how do you
(38:24):
segment your network, becausethey might be more elephantish
on some areas, but down on thedynamic side I can be a goldfish
really, because the damage isso low, because it's a
transaction, it's not a longterm relationship.
But you want to add any more tothat, because this is what
you've kind of been looking at.
Chris Caplice (38:41):
Yeah.
So you know, there's variouscharacteristics of the network
that I'm finding in my researchthat shippers need to look at,
right.
So a lot of times we'll thinkabout volume.
But it's not just volume, right, it's volume, it's consistency
of volume and frequency ofvolume.
It's a type of carrier whereyou're working with.
Do you typically want to use anasset carrier versus a broker?
They offer different advantages, right.
(39:02):
So there's a lot of differentthings.
Another thing we're seeing iswhat we're calling ghost lanes,
right, where there's contractsset up that never get used.
So why are we actually, why arewe investing in setting up
those contracts?
In fact, there are additionalcosts as repercussions for those
things.
So you know, there's a lot ofdifferent ways that we can think
(39:23):
about whether a shipper isgoing to be an elephant versus a
goldfish.
Another thing is also, you know, a limitation of potential
future research.
So I have great data in that, Ihave a good sense of the
shippers network, right.
But what I don't see is acarrier, the whole carrier's
network.
I have to kind of infer, basedon how much volume or how much
revenue is transacting betweenthe shipper and carrier, kind of
(39:44):
how important at least thatcarrier is to the shipper, but
the flip side of how importantis that shipper to that carrier,
I don't know.
So maybe there's also a call tocarriers that might be
interested in working with usLike this is data that I have
not had access to, and I thinkit'd be really interesting to
get a sense of a carrier'snetwork.
How are they responding toshippers' behaviors?
Dr. Matt Waller (40:06):
That's a great point.
You know, and one of the thingsyou know, chris, when your
paper first came out and I readit, I and I understand the
combinatorial optimization maynot work, but this idea that you
know a wording, you knowcarriers are bidding on lanes
(40:28):
and they bid based on what'sbest for them.
Of course, there's a lot ofuncertainty.
They don't know what's best forthem for sure, but there's some
element they do.
They do know.
And then the shippers, you know, are awarding these lanes and
they're getting better pricesbecause of that, better rates.
(40:50):
And I thought what a neatexample of a Pareto optimal
solution, right, whereeveryone's being made better off
, the carriers and the shippers.
And so this leads into the nextthing I wanted to talk to you
all about, because, yeah, that'sthe ideal, that's what we'd
like to see, a Pareto optimalsolution.
(41:14):
But when there's lots ofuncertainty, it's hard to
achieve that because no oneknows.
But I've always wondered if youknow, if you look at carriers,
shippers don't often consider avery big set of carriers for
(41:35):
their transportation.
A lot of times they do, whenthey work with brokers and
various things.
Sometimes they do, but there isstill a lack of information.
So it's not just uncertainty,it's information asymmetries and
lack of information that arealso leading to this, and I'm
(41:56):
just thinking gosh, this needsto be addressed.
If you think about thepotential impact on our economy,
the ability to make ourcarriers and shippers more
competitive, right, it justseems like a great problem to
solve, and so when I was lookingthrough your paper that was in
(42:17):
Journal of Business Logistics,your review paper you had one
section.
I liked the whole thing, it wasreally good, but I particularly
liked your section on thefuture research questions that
you came up with, and I'd liketo talk about that a little bit,
(42:41):
but before I do, angie, isthere anything else you'd like
to say about this?
Elephants and goldfish paper?
Chris Caplice (42:51):
No, stay tuned for the flip side, I would say.
Dr. Matt Waller (42:54):
Okay, I'm looking forward to it.
You've got me hooked, so now Iwill be following your research.
There we go Now in yourliterature view.
So for those of you that don'tknow and I won't go into the
details on this but Chris andAngie have a very rigorous
(43:19):
review article on transportationprocurement and it's hard to do
a good review article.
Angie isn't in a lot of work.
Chris Caplice (43:33):
Many hours of work.
Angi Acocella (43:34):
yes, I mean none of can go in your dissertation.
Chris Caplice (43:38):
Exactly, this is not part of my dissertation.
Dr. Matt Waller (43:41):
Well, yeah, you could.
So people who haven't done aliterature review as a paper
wouldn't know that you have tocome up with a methodology for
determining which journalsyou're going to look at and
which articles you're going tolook at, and there's all kinds
of tools out there to help you.
One example is Google Scholar.
(44:02):
For those of you everyonelistening probably uses Google,
but if you type in GoogleScholar, it gives you access to
academic articles and you cansearch by year.
You can search by keywords,phrases, you can search by
timeframe, authors, journals, Idon't remember.
(44:24):
There's a long list of thingsyou can do.
Angi Acocella (44:26):
So different than when we were students in the
90s, where you'd have to go tothe Dusty Library.
Chris Caplice (44:33):
I recognize how spoiled I am.
Dr. Matt Waller (44:36):
Chris is right.
I mean, we used to have to goto the libraries and just look
through the journals page afterpage.
It was terrible.
It's still so.
Now that part's easier.
But there's 100 times morearticles out there, so you've
got more volume to come through.
So you have to have asystematic approach to figure
(44:59):
out which articles to include.
And then you have to go throughthe articles and it gets hard
because some of the articlesaren't very good and they're
hard to read and scammed.
And then the other articles aregreat and you can get caught up
in them to where you spend toomuch time reading the articles
that are fun to read.
(45:19):
So, Angie, could you tell us alittle bit about how you did
that?
Chris Caplice (45:24):
Yeah, and I will also say, maybe, before I get
into the details, one thing thatwe find is also in this field
so we're supposed to be lookingat truckload transportation,
procurement, right, so first youhave to also scope what are you
actually looking for?
Right?
So we're not looking at routing, we're not looking at
sustainability, those types ofthings.
One thing that we find is thatthis type of research comes from
(45:45):
many different fields, right,so you can't just look at the
traditional transportationjournals, you can't just look at
operations journals, right, wehave, you know, economics, we
have law policy, right, sothat's also an additional
challenge here.
And so what we did is we came upwith, or we didn't come up with
, but there are lists ofjournals, kind of top journals
in certain fields, right, sothere's a set for the economics
(46:08):
field, there's supply chain andtransportation, and so
essentially, we identified oneof the top journals or lists of
journals that we should startwith, right?
So then, as you said, in GoogleScholar you can also search by
only looking this particularjournal.
And so we came up with thatlist of journals and a set of
keywords related to truckloadtransportation and procurement,
(46:30):
buyer, supplier relationships,things like this.
And then, yeah, just did amassive search in Google based
on these journals, and then wekind of also took an iterative
approach.
So, you know, maybe we find inone of those really good
articles that there's a paperthat's cited that's not part of
one of those journals, but itlooks like it's high quality, it
touches on the topics thatwe're looking in, and so then we
(46:53):
can go back and look at thatpaper and include it if it has
some criteria based on, you knowagain, topic but also quality.
Dr. Matt Waller (47:02):
Great, Chris.
Do you want to add anything tothat?
Angi Acocella (47:09):
Well, agey did all the work.
There's window dressing, no,but out of this, what's really
interesting, the purpose ofdoing a review like this is one
is to prevent other people fromhaving to do it and hopefully
saying, ok, here's where westand, now, stand on these
shoulders and go forward.
But it also identifies the gaps, and I think the gaps that we
identified are the future.
(47:30):
Opportunities really arise fromthe weaknesses of the
approaches.
Because you asked a second ago,you know Pareto, optimal and
all that.
No carrier likes a bid.
You just don't like to be bidout because you're commoditized,
right, you're turning into acommodity, and so every carrier
tries to say, oh, we're not thesame, we do.
You know they try todecommoditize themselves and
(47:51):
shippers, if you're the buyer,is trying to put everyone apples
to apples and try to do that.
So no one likes the RFPs.
But and you said aboutinformation, it is truly
asymmetric, right, they eachhave very different information
and the auction is the best wayof.
You know discovery, right, youdiscover where the where the
best fits are, and that'shopefully why, why it works.
(48:13):
But the challenge is that thething being bid out is based on
their forecast of the future,and it's really everyone uses
history and so the idea if I geta lot of detailed information
that's overfitting on the pastand it doesn't mean the future
is going to be that, becauseI've seen some bids or
procurement events where theygive day of week, week of year,
(48:34):
exact, detailed, but the whatwe're finding now is we look at
longitudinal studies, 50% of thelanes that have less than six
loads a year on it don't show upagain.
So you're mapping really tightlyin this.
You're setting up a strategy onsomething that's ephemeral and
is going to change for a lot ofcases, and so that's kind of the
(48:55):
opportunity we see for thefuture is that don't put all
your expertise in the strategicplanning.
Bill Dragher, when he was atUber Freight, had this phrase
that you know execution, trumpstrategy.
It dominates, and I agree withthat 100% Because it used to be
when we were there.
You optimize up here strategybecause that's the only place
you could do it, and now we'reseeing much more smartness,
(49:19):
expertise in the execution, andthat's what we're kind of saying
.
You need to move down thatpipeline to execution.
That's where you need to bedynamic and do your optimization
Excellent.
Dr. Matt Waller (49:29):
Yeah, that's a great description and you know
to your point I mean.
That's why I wanted to talk alittle bit about the gaps that
you identified.
The future, because you posethis as future research, and I
think that you should, becauseit's an academic article, but
the other side of it is.
(49:49):
It shows where there'sopportunity for entrepreneurial
activity Absolutely Right.
And so, for those of youlistening to this, one of the
benefits of being familiar withthe review article in the area
you're doing business in isbecause it gives you a window
into one, what's probably goingto come out in the future.
(50:12):
Two, what your competitors aregoing to probably start working
on.
And three, what are yourbiggest opportunities here.
So, again, angie and Chrisposed this as future research,
but I think that it's more thanthat.
Angi Acocella (50:31):
I agree.
And so Angie spent the last twoyears in Europe.
And I'm curious, angie, whatare you seeing for the
opportunities, because you'vebeen talking about this.
You're working with companiesover there.
What do you see as the bigopportunities in Europe?
And then I'll respond what I'mseeing in the US.
Chris Caplice (50:44):
Yeah, so a big thing we're seeing is in this
digitization space, digitallogistics platforms, matching
things like this.
There's a lot of money pouringinto it, there's a lot of
identifying potentialefficiencies that can be gained,
and that's at all levels.
That's not this research, and alot of my research is focused
on truckload, but at ports, soI'm working with some of the
(51:08):
ports here.
So just on handling containers,but documentation, can we get
companies more into the digitalspace?
So that's one of the big onesthat I see.
And then also, I think the USis a little bit more advanced
than Europe in terms of havinggood benchmark pricing.
(51:29):
So a lot of the companies thatI work with also don't really
know how to price well becausethere's not, at least at this
point, a lot of informationabout what are the right, what
are benchmark prices for certaincorridors.
I would say those are the twobig ones that come up a lot for
opportunities for practitionersas well.
Angi Acocella (51:47):
So let me respond to that, because I'm also, in
addition to being at MIT, I'malso a DAT chief scientist.
They bought chain analytics,which we pioneered the
benchmarking of the US truckloadrates and it's pretty big now.
We used to have Europeanoperations but we found nothing
against.
Europeans love it, they arejust really cheap.
(52:08):
We could not get a benchmarkingservice sustainable over there.
We tried for three years and itwas so hard.
And it's such an interestingmarket because, yeah, it's the
EU, but talk about all thedrivers come from Poland, right,
and they put up every countryputs up barriers, even though
it's one big union.
There's all these differentthings that we don't have in the
(52:28):
US and you know Texas doesn'tcheat drivers from Florida out,
so it's a different kind ofdynamic.
But in the US the big thingthat I'm seeing is two big
things.
One is the portfolio.
Shippers are starting torecognizing that.
You know you need to have aportfolio of relationships and
that's why brokers when Istarted in the industry in the
90s, broker was a four letterword.
(52:49):
No one wanted to use a broker.
They were the carrier of lastresort.
Now they see where they fit in.
They fit in a niche for thosesparse lanes and that long tail
of carriers.
It's a perfect match there.
But the other thing that I'mseeing is the use of dynamic
contracts, because we've beenthrowing around the term
contract rates and everything,and if you're not in this
industry, you think a contract'sa contract.
(53:11):
But truckload contracts are alittle different.
They're binding in price butnot in the volume the shipper
guarantees or the capacity thecarrier guarantees.
There's wiggle room.
It's the only industry Mattmaybe you know another where
adherence to contract is aservice metric.
It's not violating a contract,it's like a metric, like on time
for level of service.
(53:33):
So it's different.
And so what I'm seeing is thereare more dynamic contracts that
are saying, ok, we won't set theprice, we'll have the price
tied to an index.
Maybe it's every day, itchanges, and then I'll be on the
guardrails around that, so Idon't need to have a set price
ahead of time.
Maybe I can have a dynamiccontract that's dynamic in terms
(53:53):
of the capacity provided day today and the price that's
provided day to day.
I don't need to set the pricefor a year out, and I'm seeing
that for some of the lanes.
Again, that goes to theportfolio where you go dedicated
, traditional contract, dynamiccontract, pure spot and maybe
some nuances in between.
So that's where I see the realdevelopment going and the real
opportunity.
You mentioned time to start upa company, matt.
(54:16):
No kidding, do it.
Just give me some of that money, give me some investment money,
we'll do something amazing.
Dr. Matt Waller (54:23):
You're right, that is a big opportunity.
You mentioned chain of letticksand of course it's a small
world out there.
Of course, mike Kelgore, one ofthe founders of chain of
letticks, and I were friends ingraduate school.
He was in the master's programwhen I was in the PhD program
and we became friends and he wasinterested in network
(54:46):
optimization and stuff like thatback then.
But it's been really fun to seehow you and he and others I
don't know, did you know JeffZeroyah very well?
Angi Acocella (55:00):
Yes, oh, I know Jeff, jeff, it's a funny story
about Jeff.
When I used to givepresentations on this at CSCMP
and he was at I2 at the time, Iwas with a company called
Logisticscom what it was calledit's called PDCG.
Jeff would always be in thefricking front row taking notes.
You're taking it straight backto I2.
I knew you are, so I gave him ahard time, but then you gave
(55:21):
him a joint chain of letticks.
But I have to say for Mike.
Mike is a great podcast guest.
He has so many good stories, somany good stories.
Dr. Matt Waller (55:29):
And he's a good storyteller too.
He is, yeah Well, I'll have to.
I haven't talked to him for awhile, but this whole industry
is pretty small it is.
But thank you both for takingtime to visit with me today, and
, angie, I think you've got agreat mentor there.
(55:50):
And Chris and Chris, thank youfor continuing to publish
research that I enjoy reading.
I appreciate that it's kind ofa rare thing for me, so keep up
the good work, thanks, thanks.
Angi Acocella (56:03):
I have to say, if you look at elephants and
goldfish in academia, we areelephants.
I remember.
Dr. Matt Waller (56:11):
We are elephants.
Angi Acocella (56:12):
Academics are elephants Memory.
Dr. Matt Waller (56:15):
Yes.
Angi Acocella (56:16):
That's true.
I wish I was a goldfish, butI'm sad.
Dr. Matt Waller (56:21):
Well, it was great to visit with you all.
Take care.
Angi Acocella (56:23):
Thank you.
Dr. Matt Waller (56:24):
Thanks.
Thanks, Matt.
If you're finding value in thispodcast, we greatly appreciate
your support by subscribing toour YouTube channel.
Additionally, following us onApple and Spotify and leaving up
to a five-star review would beimmensely helpful.
We welcome any feedback orquestions related to the podcast
, as well as suggestions forfurther topics and guests.
(56:46):
You can leave your comments onour YouTube channel and rest
assured that I will read eachand every one of them.
Please also take a moment tocheck out our podcast sponsors,
as they play a critical role inkeeping this podcast running.
For more information onspecific topics, timestamps or
links to articles mentionedduring the podcast, head over to
(57:08):
mattwallarpodcastcom.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Crime Junkie
Does hearing about a true crime case always leave you scouring the internet for the truth behind the story? Dive into your next mystery with Crime Junkie. Every Monday, join your host Ashley Flowers as she unravels all the details of infamous and underreported true crime cases with her best friend Brit Prawat. From cold cases to missing persons and heroes in our community who seek justice, Crime Junkie is your destination for theories and stories you won’t hear anywhere else. Whether you're a seasoned true crime enthusiast or new to the genre, you'll find yourself on the edge of your seat awaiting a new episode every Monday. If you can never get enough true crime... Congratulations, you’ve found your people. Follow to join a community of Crime Junkies! Crime Junkie is presented by audiochuck Media Company.
Ridiculous History
History is beautiful, brutal and, often, ridiculous. Join Ben Bowlin and Noel Brown as they dive into some of the weirdest stories from across the span of human civilization in Ridiculous History, a podcast by iHeartRadio.