Prices, Market Structure, and Sustainability in the Trucking Industry��

Alex Scott

ascott@utk.edu

Agenda

• Structure of the truck transportation market

​

• Modeling market prices in the truck transportation market

​

• Possible future contracting methods

​

• Sustainability (time permitting)

2

Truck transportation market

3

Transportation services

Customers

Trucks

Broker,

asset-based carrier,

or private fleet

Shippers

4

Source: CSCMP 31^st Annual State of Logistics Report (2020)

Market pricing

• Pricing is mostly relevant in the context of the “for-hire” truckload (TL or FHTL) market
• $300+ billion annual spend

​

• However, there are spillovers into:
• intermodal (truck-rail-truck) transportation
• the make-or-buy decision for shippers (decision to use a private (in-house) fleet or contract for a “dedicated” fleet)

5

For-hire truckload market

• Shipper uses a third party (carrier) to haul a load of goods to a customer

​

• Carriers can be asset-based or a non-asset-based broker
• Asset-based: Schneider National, J.B. Hunt Transport, Knight-Swift
• Brokers: C.H. Robinson, Coyote, Total Quality Logistics

6

Supply side is extremely disaggregated

• 30,000 registered brokers

7

Sources: 2020 Pocket Guide to Large Truck and Bus Statistics, FMCSA; https://ai.fmcsa.dot.gov/RegistrationStatistics/

​

Contracting methods

• Shippers contract with for-hire carriers in three ways

​

• Dedicated fleets – a fleet of trucks operated by a carrier but under the direction of the shipper
• Contract freight – a price negotiated at one point in time but valid for some extended period of time
• Most common for annual procurement auction, prices negotiated for one year
• Often renegotiated, also loads can be rejected by the carrier
• Spot freight – a price negotiated that is valid only for a load or small set of loads, typically occurring in the near future
• Best measure for real-time market conditions

​

8

Market pricing

• Contract freight is 80-90% of the FHTL market, spot 10-20%

​

• Spot prices are the best way to understand market conditions

​

• Contract prices are flat for extended time periods, negotiated based on volumes, and could be influenced by negotiations, past relationships, etc.
• Contract prices generally respond to changes in the spot market with a significant lag (3-6 months)

9

Estimating market conditions

10

Spot prices

• Historically, finding a good source of spot prices was hard to find

​

• Ideally want prices
• over a wide geographic region
• consistent over time
• from a wide range of carriers/brokers
• as many variables as possible to be held constant (e.g., dry van vs. flatbed, loading/unloading performance)

11

General approach

• We cannot observe market conditions directly, but we can observe most or all of the operational factors that affect the price of a load

​

• Lane that it travels on
• Lead time of load (time from when spot price requested to load pickup)
• Time and day effects
• Weather effects
• Carrier effects
• If multiple shippers, would want to know loading/unloading performance

12

General approach

• Given prices and the operational characteristics of a load, we can then estimate market conditions

13

Source: https://www.mhlnews.com/transportation-distribution/article/22051182/a-realtime-method-to-estimate-trucking-market-conditions

General approach

• If we have spot and contract prices, then define the dependent variable “bid premium” as the spot price / contract price on a lane
• the coefficients in a regression would be interpreted as a percentage change in price relative to contract price (could also use log of spot price, or just spot price)

​

• Include all observable variables

​

• Latent market conditions in the error term; include time fixed effects to capture this

14

General approach

• For carrier i in spot auction j:

15

Source: Carrier Bidding Behavior in Truckload Spot Auctions, Journal of Business Logistics 39(4), Scott 2018

​

Data source

• Partnered with a large manufacturer
• Hundreds of millions of $ a year in FHTL expenditure
• Manufacturing plants across the U.S.
• Standard product, dry van trailers
• Held tens of thousands of spot auctions every year

​

• Every invitation to a carrier to an auction from 2012-2015
• 3.4 million invitations
• 425k bids from more than 100 carriers/brokers
• >100k auctions

16

Insights

1. Spot price index is highly correlated with other measures of TL market

​

1. Tender rejections are highly correlated with spot prices
• An inefficiency of static contract prices that do not adjust to the market

​

• Different types of carriers have vastly different spot strategies

​

1. Bids fluctuate in-line with one another despite no central market
• Small asset-based carriers are the exception
• Regional indices are positively correlated

17

Spot price index and other measures

18

Source: https://www.mhlnews.com/transportation-distribution/article/22051182/a-realtime-method-to-estimate-trucking-market-conditions

Prices and rejections are correlated

19

Source: Service Refusals in Supply Chains, Transportation Science 51(4), Scott, Parker, and Craighead 2017

​

Different carriers have different strategies

20

Source: Carrier Bidding Behavior in Truckload Spot Auctions, Journal of Business Logistics 39(4), Scott 2018

​

Different carriers have different strategies

21

Source: Carrier Bidding Behavior in Truckload Spot Auctions, Journal of Business Logistics 39(4), Scott 2018

​

Different carriers have different strategies

• Brokers bid often and higher

​

• Asset-based carriers (with no brokerages) bid infrequently but lower

​

• Bidding strategies reflect their positions in the market
• Brokers provide service (find capacity) but at a cost
• Asset-based carriers infrequently use the spot market and are less sophisticated with their pricing

22

Different carriers have different strategies

23

Win Percentage

Market Prices

Bids correlate with one another

24

Source: Carrier Bidding Behavior in Truckload Spot Auctions, Journal of Business Logistics 39(4), Scott 2018

​

Regional prices correlate with one another

25

Regional prices correlate with one another

26

The market reacts to supply and demand

• When conditions are tight (demand is high relative to supply):
• carriers reject loads more often at static prices when spot prices increase
• (nearly) all carriers’ prices move in line with one another
• prices in different regions are positively correlated (assets can move from one region to another if not)
• Contract prices eventually move in response to spot prices

​

• Is there an opportunity to take this “informal” market to more of a formal market?
• Large brokers are basically the market maker at this point

​

27

Contracting alternatives?

28

Potential contract improvements

• Binding service contract
• Carrier agrees to accept 100% of loads offered to them on a lane
• Reduce spot exposure for shipper, headaches associated with backup carriers/spot carriers
• Risky for carrier or broker – exposure to demand swings, potentially tie up assets at low prices when market swings in their favor

​

​

​

29

Potential contract improvements

• Binding service contract tied to market prices
• Reduces risk for carriers/brokers to price swings
• Reduces risk for the shipper of “default” or backing out by the carrier

​

• We piloted this with a large shipper and broker (2020 timeframe)
• Broker agreed to accept 100% of tenders on somewhat problematic lanes
• Tied initial contract prices to DAT spot price index

​

​

30

Futures market

• There could also be a centralized futures market
• provide a way for shippers/carriers/brokers to hedge against price volatility
• could simplify pricing/negotiations/procurement
• this was attempted by a company called Freightwaves in 2018-2019 timeframe
• do not think this is around anymore or had success

31

Important distinctions for TL

• Cannot inventory truck capacity (or, at least, is very expensive and difficult to do)

​

• The customer takes delivery of the “product”
• Unlike oil and gas markets, flour markets, etc.
• Must consider actual management of service

​

• Highly dispersed market with significant inertia

32

Similar issues in the ocean industry

• New York Shipping Exchange is trying to implement binding contracts in the ocean shipping industry, which has some similarities to trucking

​

• The Baltic Exchange/BIFFEX is an index of pricing in ocean freight and has had some success
• I don’t know if it’s used for pricing

33

34

�Questions?��

35

Sustainability

36

The problem

• Commercial freight transportation accounts for 16% of all corporate greenhouse gas (GHG) emissions in the US

​

• There is practically no visibility into the sustainability of the hundreds of thousands of carriers

​

​

​

​

37

Source: The Green Freight Handbook, Environmental Defense Fund https://supplychain.edf.org/resources/the-green-freight-handbook/

The problem

• The most important factor influencing emissions and under a carrier’s control is the equipment they use
• 2018 model-year trucks mandated to use 9% to 23% less fuel than 2010 model-year trucks
• Aerodynamics can affect fuel consumption from 10% to 30%

​

​

It’s the equipment that matters!

​

38

Sources: EPA Regulatory Announcement EPA-420-F-11-031; Trucking Efficiency Confidence Report, North American Council for Freight Efficiency

39

Source: EPA MOVES model documentation

​

40

Source: EPA MOVES model documentation

​

The problem

• There is a wide variety of carriers with different types of equipment – different truck types, model-years, and aerodynamics

​

• Many shippers want to use “green” carriers, but there is no way of knowing which carriers are green or not

​

There is no way for a shipper to select carriers based on their sustainability

​

41

The SmartWay Program

• The US Environmental Protection Agency (EPA) developed the SmartWay Program to close some of this gap

​

• Voluntary program for shippers, carriers, and OEMs

​

• Standardized, public reporting of emissions by participating carriers

​

• Participating shippers must ship at least 50% with SmartWay-approved carriers

42

The SmartWay Program

• Significant costs associated with the program

​

• Carriers must collect a lot of data to participate, put in the MOVES model; very complicated with lots of documentation
• Around 1% of carriers participate, perhaps 35% of trucks (2,922 carriers in 2020, ~720,000 trucks)

​

There is no visibility into the sustainability of 99% of carriers

​

43

The SmartWay Program

• SmartWay is a rough proxy for sustainability
• E.g., a non-SmartWay carrier with 5 brand new, clean trucks is not considered “green”

​

​

44

The SmartWay Program

45

Green Shipper

Clean trucks

Dirty trucks

Clean trucks

Dirty trucks

SmartWay carrier

Non-SmartWay carrier

How big is the problem?

• In 2020, 69% of SmartWay-approved trucks were operated by carriers NOT in SmartWay!

​

• Of carriers with fewer than 100 trucks, 99% of SmartWay-approved trucks were operated by non-SmartWay carriers

​

• By excluding non-SmartWay carriers, you exclude a majority of clean trucks

46

47

Clean trucks

Dirty trucks

SmartWay carrier

Non-SmartWay carrier

31%

Note: SmartWay approved = clean

69%

SmartWay carrier

Non-SmartWay carrier

19%

81%

A potential solution

• Fundamentally, we need insight into the equipment operated by all carriers

​

1. Can score all carriers based on the age of their trucks, whether they are SmartWay-approved, whether the trucks meet federal standards (e.g., >= 2018 model-year)

​

1. Can link emissions using SmartWay data to non-SmartWay carriers based on equipment

48

A potential solution

• Created a database of > 3 million unique class 8 trucks operated by practically all carriers on the road

​

• Includes characteristics of these trucks, including engine, model-year, SmartWay-approved, etc.

​

• This could be used to improve our understanding of carriers that are green, help shippers select green carriers/trucks

49

Thank you!��Questions?��ascott@utk.edu

50