If you haven’t heard about the American entrepreneur Bradley Jacobs, who built five-billion-dollar companies from scratch, then maybe this is the time you should take notice. Because his recent move in XPO might be a prelude to his biggest upcoming bet. Frustrated by the conglomerate discount ascribed to XPO by Mr. Market, Brad spun off the swanky logistics business (GXO) from XPO earlier this month. Thus, the stage is set for Brad to implement his old playbook of selling companies (United Rental, Waste Management) after scaling them to billion-dollar entities.
XPO is the third-largest non-unionized LTL (Less than Truckload) player operating in a high entry barrier industry with excellent oligopolistic economics. Less than a handful of players enjoy returns above the cost of capital. Unionized players plagued by pension costs continue to cede market share. No new asset-based LTL player has been able to enter in the last decade. This is despite the industry growing at 1.5x GDP buoyed by the e-commerce boom. Even players like UPS and FedEx, known for operational geniuses, haven’t been able to get their LTL strategies correct. Therefore, the sector is akin to Marathon AM’s “capital cycle approach,” where entry barriers and capital starvation has created superior returns for a few.
The Spin underscores Brad’s plan to showcase XPO as a benchmark to ODFL, the gold standard in LTL (like Canadian National in rail).
XPO piqued my interest because of its sheer undervaluation – trading at less than a half of ODFL’s EV ($13 bn vs. $31 bn) despite having an almost similar EBITDA! While this looked too appealing, a deep-dive into the company changed my opinion to the contrary.
But allow me to present both the Bull and Bear cases in this deep-dive.
Bear thesis – why I refrained from buying
- Poor earning quality is masking the true business economics
- XPO is eroding its long-term competitive advantages to boost short-term profits
- Home run enjoyed in brokerage to come under severe competitive pressure from tech-based players
- Value trap unless the sale of non-LTL business springs a surprise
- Large insider selling before the Spin
Bull thesis – upside risk
- LTL presents a multi-decadal profitable growth with reinvestment opportunity
- Run by an owner-operator (13% stake) with an excellent track record!
- XPO becomes a likely acquisition target (that’s how the Spin has been structured)
Table of Contents
- How does the new XPO stack up?
- LTL 101
- What makes XPO’s LTL strategy unique
- Why ODFL’s strategy is superior to XPO
- Quality of earnings
- Last Mile Industry – First-mover advantage in a high-growth fragmented industry
- Brokerage – Digital advantage in no entry barrier business?
- Valuations –Brad Jacobs premium or value trap?
1. How does the new XPO stack up?
Roll-ups have a terrible reputation of blowing up either from overpaying or cooking of books. But Brad has proved everyone wrong. He started XPO with an investment of $150 m ten years back to dominate the asset-light freight brokerage industry. Later he realized that his aspirations were too modest. So he went on acquiring 17 companies between 2011-2016. However, this made XPO a conglomerate comprising contract logistics (GXO) and transport (XPO).
I’m sure most of you might not have missed the GXO spin promotions on CNBC. GXO (38% of FY 20 sales) was separated into a new entity this month. GXO is one of the largest pure-play contract logistics providers with marquee clients like Apple, Disney, L’Oréal etc. GXO was mainly built up through three big acquisitions of France-based family-controlled company Norbert Dentrassangle, New Breed and Menlo.
The existing XPO is still not a pure-play LTL. It also houses freight brokerage, last mile and European businesses. Brad’s original plan was to sell the entire non-LTL business before the pandemic spoiled the plan. The non-LTL portion has a substantial brokerage commission business, which skews the contribution of operating profit.
2. LTL 101
LTL is a fascinating industry with a niche for freight larger than a Parcel (catered by FDX, UPS and AMZN) but too small to require an entire truckload. An ideal weight to qualify for LTL is between 150 to 15,000 lbs. An LTL operator collects freight from different shippers and then combines them into a full trailer load. Customers include a whole range of omnichannel retailers (Home Depot, Lowe’s), e-commerce (Amazon), Industrials (chemicals, grains) and grocers. The routes are pre-defined as it depends upon the network of the LTL operator. Unlike in Truckload, the freight in LTL is handled multiple times while loading and unloading. As a result, operators with better service levels (lower claim damages) command better pricing. In addition, driver turnover is very low because the driver can get home every night, unlike in Truckload, where they are on the road for weeks.
The top 5 players control 56% of the market. This has given them rock-solid pricing power, especially over the last decade, just like the Rails. Moreover, LTL players are classified as union and non-union. The non-unionized players have steadily increased their market share from 52% to 75% in the last two decades. To give you an idea, ODFL has outpaced the industry by growing at a CAGR of ~12%. On the contrary, the second-largest player, YRC, unionized, was twice on the brink of bankruptcy wherein it got saved by the union in 2009 and an equity injection by the Federal Government in 2020.
3. What makes XPO’s LTL strategy unique
XPO’s entry into LTL was led by the acquisition of Con-Way in 2015. The purchase came in as a surprise because it changed XPO’s strategy from an asset-light freight broker business into an asset-based player. Consequently, the stock was severely punished, and questions were raised on the success of this strategy. On the contrary, XPO proved everyone wrong by doubling the EBITDA within four years of acquisition. This was done by bringing in an owner-operator “Tony Brooks” to manage LTL. Tony had experience running the business’s shipping side with large companies (like Sears, Sysco, & Pepsi) and the LTL business at Roadway. Both Brad and Tony changed the culture at Con-Way by decentralizing and assigning P&L for each service centre. This meant that every service centre operator was paid depending upon the profitability at their hubs rather than an earlier system of organization-wide profitability. Second, XPO massively overhauled its service centres by closing down unprofitable ones. Third, it put an impressive investment into technology that ensured better pricing algorithm and route optimization.
…in an industry where players hardly make money
Despite a favourable industry dynamic, only two players make a double-digit operating profit margin. XPO’s operating ratio (a measure of revenues compared with expenses) is second only to ODFL but much better than other players, including FDX. Except for the top 3-4 players, none make reasonable returns above their cost of capital. This provides a vicious cycle for these top players to redeploy profits back into the business and create a formidable moat. UPS is a prime example, which exited LTL by selling to Montreal-based TFI (TSX: TFII). Based on a rough projection, UPS’s operating profit margin was a mere 3-4%. The problem with UPS was that managers tried to treat LTL similar to parcel and its union workforce. Consequently, LTL was axed after Carol Tome joined as CEO in 2020.
4. Why ODFL’s strategy is superior to XPO
No doubt XPO has made great progress in LTL. But I believe that ODFL’s superiority will continue to persist. ODFL’s secret sauce lies in – 1) Investing in strategic real estate and 2) Getting the network density better than anyone else.
ODFL’s Secret Sauce #1 – Investing in Real Estate
Going against the industry odds, ODFL has consistently expanded service centres over the last 10 years. For example, against XPO’s flattish growth of 1%, ODFL has grown by 13%. In addition, XPO’s underinvestment was due to the realignment of unprofitable centers. In this regard, ODFL’s real estate team has perfected the art better than anyone else. As a result, ODFL has created a significant competitive advantage, especially in urban areas. First, everyone wants same-day delivery, but no one wants a warehouse in their neighbourhood. Second, XPO has started the sale and leaseback of its properties recently. I believe it’s a trade-off between boosting your short-term profitability vs. control of your most strategic assets. Third, XPO has focussed on deploying pricing algorithms to improve profitability. On the other hand, ODFL gets a premium pricing from having the best network density from these real estate investments.
ODFL’s Secret Sauce #2 – Getting the correct Network Density
“We were somewhat like some of our competitors back in the day. We used a lot of purchase transportations. We were dependent upon those for our linehaul moves and what not. We eliminated that over the years. And I think, by and large, we’ve just gotten better. I don’t want to say that we did stupid things. Maybe that’s not exactly the right terminology, but I think we’ve just gotten smarter, and we’ve gotten better over the years.”
– ODFL CEO
An LTL carrier often purchases transportation from a competitor when a pickup/delivery destination is outside their network. Many times LTL still prefers purchasing over their own network if a competitor has a better density. Hence, you’ll see that purchase transportation is the second-highest cost after labor. The importance of perfecting the art of a profitable network density can be underpinned by FedEx’s recent decision to embargo freight not fitting its requirements due to capacity constraints. Hundreds of shippers, including big-box retailers like Home Depot and Lowe’s, were left scrambling for carriers. Not even premium pricing could help because FedEx didn’t have the network density.
ODFL has the lowest purchase transportation cost, which underscores its hugely profitable network density. On the other hand, XPO still has a long runway to improve its network density (despite having higher centers). Similarly, on pricing, weight, and mix, ODFL can be seen to have superior economics compared to XPO. This is despite Con-way starting being more regional before acquisition which ODFL developed later.
5. Quality of earnings
Adjustment to LTL Earnings
XPO makes a host of adjustments to arrive at an adjusted operating ratio for LTL. To give you an idea, the operating profit for LTL jumped from $487 m (GAAP) to $573 m (Non-GAAP; +18%) in 2020. I believe many of these expenses do not make economic sense to be added back. First, XPO adds back other income (read as GAAP pension expense). Since GAAP treats this as an expense, I presume it to be the service cost component of pension expense. Second, transaction and restructuring costs come from the roll-up strategy that should not be treated differently by adding them up. Third, XPO’s profit gets a boost from the sale and leaseback (16% of 2020 profit). Fourth, corporate costs are not allocated to segments, thereby increasing the reported profits. After making these adjustments (except amortization of intangibles), the operating profit margin falls from 16% to 10.6%!
Adjustments to the entire business
Adjustments to the entire business (pre spinoff) are also intriguing. Transaction and restructuring expenses worth $471 m have been added back over the last five years. If we combine sale & leaseback and pension income, a total adjustment of $945 m or ~66% of cumulative net profit has been made. XPO started reporting pension income as “other income” after adopting Accounting Standard Update (ASU) 2017-07 in 2018. Interestingly, XPO’s pension managers have made 15-20% actual return on US plan assets primarily consisting of fixed income portfolios over the last five years (quite impressive to have them as my manager!). I want to clarify that I do not doubt any irregularities but only highlight earnings’ accounting quality. The frequent resignation of CFO’s (the latest one just before the Spin) doesn’t help instill confidence.
6. Last Mile Industry – First-mover advantage in a high-growth fragmented industry
The last mile is the easy-to-understand last leg of movement of heavy goods from a transportation hub to your home. Imagine FedEx, UPS and Amazon, which are well-entrenched in the delivery of parcels at doorsteps. But last-mile of heavy goods (usually above 150 pounds) like a washing machine or a Peloton bike is an entirely different business model to crack since it requires a trucker to enter inside someone’s house for installation. In LTL, you have a single driver. But in the last mile, you have two people. This makes it a highly service-intensive business because the reputation of a big-box retailer like Home Depot will be at stake.
Last-mile is the most significant growth driver for XPO, having grown at a CAGR of 31% (FY 13-20). The industry is so fragmented that despite being the #1 player, XPO’s market share is a mere 7% in the US. XPO has the first-mover advantage by buying the biggest last-mile providers like 3PD between 2013 and 2015 to become the largest provider in North America. Companies such as FedEx, JB Hunt and Ryder have entered, sensing the opportunity. The strategic importance of the last mile was felt during the pandemic. But it is too late for someone to scale up quickly because there aren’t many assets available to buy. I believe XPO should not have a problem selling this business if it wants to unlock value.
XPO doesn’t disclose the profitability of last-mile separately. It should be a high-return business being as XPO uses independent contractors to perform deliveries. The investments are limited to creating 85 hubs that claim to position XPO within 125 miles of approximately 90% of the US population. The scale helps in making more stops per day and hence the flywheel for profitability. The other part that sets XPO apart is the technology piece – from finding the most efficient truck for delivery to providing real-time information to shippers and customers.
7. Brokerage – Digital advantage in no entry barrier business?
Truck brokerage is a straightforward and attractive high return business model which throws off free cash flow during all parts of the economic cycle. A broker sells the truck capacity to the shipper at a slightly higher price than the broker paid (usually for a 15-20% gross margin), but at a better price than the shipper could otherwise obtain on their own. Thus, the broker earns a profit on the spread. It is a highly fragmented market with millions of shippers and more than 10,000 licensed truck brokers, but only 25 of those have revenue greater than $200 m. Historically, there has been no entry barrier because all one needed was a small office set up anywhere with a set of phone/fax machines to match shippers with carriers.
But this has now changed.
C.H. Robinson (the largest), TQL, Echo etc. were long-established players before XPO made inroads in 2013. More than half of the revenues for C.H. Robinson stems from long-term contracts from its biggest 500 customers (out of a total customer base of >30,000). So XPO focussed on targeting small and medium-size shippers to gain scale. XPO was an early adopter of technology which helped it gain scale quickly to become the fourth-largest player. For example, its freight optimizer tool gave it a significant advantage in using data-driven algorithms for price discovery vs. human judgment based on small brokers. Consequently, XPO grew net revenues at a much faster rate than CHRW.
Competition to dent margins
“Broking is becoming more automated with more digital interactions. I think long-term, margins will come down; as costs come out of the way brokers do business, a good chunk of that cost savings will get passed along to customers, and that’s going to degrade margins. Having said that, lower margins on a much larger amount of business with lower SG&A can still be a beautiful thing, and you can still create a lot of value from that. And I think there will be a shrinking of the number of players. There’s not going to be 10,000 or 20,000 brokers 5, 10 years from now. I think it will be a smaller number of larger brokers who have very significant investments in technology and are tightly integrated electronically with both customers and carriers.”
– Brad Jacobs
Over the last few years, the incumbents have been challenged with an onslaught of tech-based disruptors (Uber Freight, Transfix, Convoy etc.) who have Uberised the industry. XPO was also not far behind in launching XPO Drive but was a little late to the party as it initially focussed on gaining scale with the truckers before approaching the shippers. As of December 2020, it exceeded 100,000 downloads of Drive XPO but is still below competitors. The biggest challenge for brokers will be to keep the trucker engaged, like getting them the backhaul load etc. XPO Connect (cloud-based digital freight marketplace) is a platform developed by XPO that connects freight optimizers with AI.
Discussion on the brokerage industry deserves a separate write-up for some other time. But it is widely accepted that the industry is becoming competitive, and it will be challenging for players like XPO.
“The difference between the digital players and the non-digital players has become almost non-existent. We’re as digital, more digital than the so-called digital ones. I think over time, with respect to margins, I think margins will come down. Having said that, I think profit will go up because I think volume will go up because I think the amount of transactions and the velocity of transactions will increase. So I think there’s — it’s going to evolve very much like you see commodity markets, particularly like the oil market, for example where it morphed from purely physical markets, then derivative markets and Wall Street got involved.”
– Brad Jacobs
8. Valuations –Brad Jacobs premium or a value trap?
XPO’s management has often emphasized EBITDA as a preferred valuation tool. But having discussed the perceived quality of earnings, cash flow would be a better metric. To show you what I mean, XPO has the lowest conversion rate of 58% from EBITDA to operating cash flow. So this means that XPO will look optically cheap on EV/EBITDA.
While on an EV/EBITDA basis, XPO looks dirt cheap to ODFL and SAIA, the same is not the case with P/FCFE. A ~40% discount to ODFL on EV/EBITDA falls to a 26% discount on P/FCFE. Or simply put, an 11x EBITDA translates into 20.5x Cash flows. Selecting a higher multiple is all about giving Brad Jacobs premium, in my opinion. Imagine if he was not a part of the company. Would you have given it a premium to the current multiple? While I agree that some excellent bull thesis exists and shorts have mainly been proven wrong, a good exercise for me was to check the growth and margins imbibed into the current price.
Implied valuation- reverse DCF implies 5% growth over the next decade
Reverse DCF is a process of checking what assumptions are implied in the current price. In the case of XPO, the market is factoring in a revenue CAGR of 5.1% over the next 10 years. I suppose is not cheap by any parameters. For context, XPO has had overall revenue growth of a mere 1.8% since 2016. LTL grew by 3.8% and Non-LTL by 1.2%.
Similarly, the market gives XPO full credit for margin expansion from 5.4% in 2021 to 8% in 2026. XPO expects $1 bn EBITDA in 2021 in LTL, which I assume will equate to an EBIT margin of 5.4-6% margins for the entire company. So for the market to assume a consistent 200 bps improvement will require margins to increase in brokerage (not possible) or new profitable avenues for LTL.
Being optimistic on efficient turnover and margin expansion, I have assumed ROIC will increase from 9.7% in 2019 to 17% in 2030. Of course, the user can tweak the other implied assumptions, such as perpetual rate and a discount factor. But the sensitivity should not be a breaking point. To illustrate, a 100 bps increase in WACC to 7.9% will reduce the intrinsic value by ~25% (on my assumption), thereby simply making the investment case unviable for me.
Finally, against the current P/FCF of 20x, the terminal P/FCF (after 10 years) comes to be at almost the same level of 20.5x. Thus, while I agree upon the potential growth opportunity for LTL to grow in the next 10 years, the growth and reinvestment potential would have certainly moderated by then. To conclude, XPO needs to come out with a show-me-story for me to revisit my assumptions. This could be from a better sale realization of non-LTL businesses and either redeploying the cash to LTL or returning it to the shareholders. Until then, it will continue looking cheap (optically) on EBITDA, especially when compared to ODFL, which has separate reasons to trade at a premium.
The Investing Principle (TIP)
Investing in Roll-ups should not be a straightaway criterion to disregard an investment. Serial acquirers like Danaher, Roper etc., have a robust DNA in replicating the culture onto the acquired companies. But here, investment in XPO needs to be seen from the lens of Brad Jacobs, who is a PE operator. The investment will thrive until the private-equity player is in the game. Once he exits, it will not be long to recognize the erosion of XPO’s LT competitive advantage. But the biggest upside risk is that the PE player finds an exit for all at a great price 🙂 !! Hence, I say this situation is akin – do not hate the player but hate the game.
Recomended Readings
- Orbis is a highly respected fundamental, long-term fund which takes a contrarian approach. One has to respect their conviction on being invested in XPO (top 3 holdings) throughout the drawdowns. BUY thesis HERE.
- Return on Capital has the best primer on the parcel industry (HERE).
- Vanck, who worked on sell-side, runs an excellent substack dedicated to the freight transport. ODFL thesis HERE