Prices, Market Structure, and Sustainability in the Trucking Industry
Abstract:
Dr. Scott will discuss the market structure, pricing, and sustainability in the truck transportation industry. Dr. Scott will provide an overview of the trucking industry, including the participants, relationships, and most common contracting methods used by shippers, carriers, and brokers. Methods to estimate the factors that affect transportation prices will be reviewed as well as insights into market dynamics from hundreds of thousands of spot prices. Potential future states of contracting will be discussed, including an innovative industry-academia partnership that created and implemented a binding-price agreement between a shipper and carrier. An overview of the current state of sustainability in trucking will be provided, time permitting.
Bio:
Dr. Scott’s research focuses on transportation with an emphasis on market dynamics, public policy, and transportation sustainability. His articles have been published in Transportation Science, Transportation Research, Journal of Operations Management, Production and Operations Management, Journal of Business Logistics, and other journals. Dr. Scott’s research on public policy in transportation has been widely discussed in industry publications and was cited in testimony before a committee on transportation safety policy in the U.S. Congress. He also developed a method and comprehensive data source to measure the sustainability of transportation firms, which led to a partnership between the University of Tennessee and one of the largest supply chain visibility companies in the world. Prior to returning to academia, Dr. Scott worked in the transportation, logistics, and consulting industries for about a decade.
Summary:
Market size (annual):
Transportation markets: $1.5 trn
Trucking market: $800b
For-hire tricking market: $300b
For-hire trucking
Spillovers to other markets
Asset-based (own trucks, sell transport)
Brokers (markets that connect trucking companies to customers)
30k brokers (from huge to tiny)
Many carrier companies that own trucks (from huge to tiny)
Contracting methods
Dedicated fleets owned by shippers (e.g. coca cola, walmart)
Contract freight
80-90% of loads move this way
Temporary contracts
Non-binding
Carriers are free to reject loads when they want to
Shippers can choose to not actually offer loads under the contract
Spot freight:
Negotiated live on markets
Good measure of market value of transport: responsive to supply/demand events
Prices tend to be higher than in contract freight
Visibility into the spot market:
Can observe:
Lane (origin/destination)
Lead time (from price request to load pickup): stable after few days ahead
Time/day effects
Weather effects
Carrier effects
Use these to estimate market conditions
Spot price premium = operational characteristics (observed) + market status (estimated)
Bid premium = spot price / contract price
Data
2012-2015 loads from a large manufacturer with simple dry-van loads
3.4m invitations
425k bids from >100 carriers/brokers
>100k auctions
Insights:
Spot price index correlated with other truckload market
Morgan Stanley market sentiment index (1 month-lagged survey)
Rejections of contract prices correlated to spot prices (spot prices go up -> rejections go up)
There’s a cost for carriers to reject loads
But they do if the spot price is significantly higher than the contract
Wide range of bidding strategies for carriers
Asset-based carriers:
Few bids, price relatively low on market (apparently not aware of market price)
Their win percentage rises as market price increases
Brokers:
Many more bids, somewhat higher price (take advantage of the lower price charged by asset-based carriers)
Bid percentage is static since they adapt to market
Bids correlated with each other despite no central market (except small asset-based carriers)
Regional prices correlate with each other (e.g. Northeast vs Midwest)
So the market is national since trucks move easily
Further markets are less correlated due to
Distance of moving trucks across country
Many carriers operate regionally
Difference in weather patterns (e.g. slow-downs, agricultural production schedule)
While there is no central market, large brokers effectively act as a market
Process is largely automated
Shippers and carriers are matched and deal with each other
Potential contract improvements
Binding contract
Rejected option: Carrier agrees to accept 100% of offered loads: risky to carrier if price moves higher
Preferred Option: Price of contract tied to market price
Piloted this with a broker and a shipper
Broker agreed to accept 100% of shipments on problematic lanes
The agreed prices depended on the DAT spot price index
Futures market
There could be a centralized market
Provides hedge against price volatility
Freightwaves tried this in 2018-2019
Didn’t succeed but no information about why
Challenges:
Can’t inventory truck capacity
Customer takes delivery of “product”, so harder for intermediaries to trade smoothly
Highly dispersed market with significant inertia
Cost of providing service depends on load/unload characteristics, optimal shipping horse, appointment flexibility, etc. : highly idiosyncratic
Similar challenges:
New York Shipping Exchange: binding contracts in ocean shipping industry
Baltic Exchange: index of pricing in ocean freight
Sustainability
Commercial freight accounts for 16% of greenhouse gas emissions
The equipment matters this most: trucks have been getting cleaner over time in greenhouse gases and local pollution
E.g. Port of LA/Long Beach limits age of trucks used in port to reduce pollution in city
Challenge: shippers can’t choose carriers based on their sustainability
SmartWay program
Voluntary
Standardized reporting of emissions
Small carriers don’t participate in this because of complexity (99% don’t)
Many new, clean trucks are not considered clean because of lack of participation
Trucks are not in SmartWay program
69% of clean ones
81% of dirty ones
Alternative score: look at model year of trucks used by carriers
Convert SmartWay data about fleets in the program to fleets outside the program
Attributes: engine, model year, etc.