|Shares Out. (in M):||88||P/E||0||0|
|Market Cap (in $M):||18,470||P/FCF||0||0|
|Net Debt (in $M):||2,578||EBIT||0||0|
Sign up for free guest access to view investment idea with a 45 days delay.
For those well versed in this space you may find your time best spent by starting at “Pushback”.
Note: Big 4 refers to ORLY, AAP, AZO, and GPC (NAPA).
DIY and DIFM
Auto part retailers compete in two distinct markets: DIY and DIFM.
DIY (do it yourself) customers are typically male and are below the national median household income. They work on their cars out of economic necessity and approximately 50% of DIY customers pay in cash. Most DIY projects value speed of service (i.e., ability to return life to “normal”) before price. Price is becoming increasingly transparent across retail thanks to e-commerce, which we will discuss in greater detail below. DIY sales maintain higher gross margins than DIFM as DIY customers are not only purchasing the part but the guidance of the store associate and the loan-a-tool program. These add-ons are included in the price of the product, driving margin regardless of the customer’s choice to utilize them.
DIFM (do it for me) is the faster-growing market segment and represents a smaller part of the Big 4’s existing business. DIFM customers maintain multiple relationships with different suppliers. Multiple suppliers are a necessity for a professional shop due to the massive pool of inventory they need to draw from to complete any one job. For comparison purposes, Monro, Inc.(MNRO), a professional auto shop maintains ~2,300 SKUs per service shop. Each player in the Big 4 has different inventory strategies, but all hold more than 100,000 SKUs system-wide.
Professionals refer to their suppliers as first call, second call, etc. AutoZone executives estimate first call suppliers get ~60% of that professional’s business. Second call and subsequent suppliers then act as a back up to meet demand when the first call supplier fails to have the part or to fulfill the order within the desired time frame. Professional customers expect inventory requests to be filled in 30 to 45 minutes as turnover is the key driver of profitability in the professionals’ business. Given the expectation on timing, pricing of the component is less important, especially when the costs are passed on to the customer.
The structure of the DIFM market makes the first call position extremely coveted and difficult to usurp unless first call fails to deliver on customer expectations, thereby creating an opportunity for competitors to step in.
Market Size and Share
It is well documented that store count for the industry has been relatively flat since 2000. During this same time period, the Big 4 have grown substantially, taking share from and acquiring smaller competitors. Due to the size of the Big 4 and store overlap amongst the top ten competitors, it is unlikely to see any large and accretive M&A transactions. Future growth opportunities will come from in-fills in existing locations and expansions to underpenetrated markets. However, same-store sales will be the main driver of revenue growth for the group over the next decade.
Uses of Capital
Capex has historically been allocated to growing store count and distribution centers to facilitate the growth of the supply chain. As the industry is maturing, the Big 4 will likely have a lower capital intensity going forward as maintenance capex is low. Store closures are often avoided as much as possible, because the location is important in maintaining a strong distribution network. We do expect increased investment in omnichannel capabilities, particularly the user interface which is currently a painstaking experience to navigate. O’Reilly appears to be the laggard in this race, however, we don’t believe that any website is a “winner” today. We expect this investment to be less substantial than the build-out of new stores and infrastructure as existing assets can be leveraged in an omnichannel strategy.
Going forward, we expect free cash flow to increase, in part due to reduced capital intensity of establishing new units, which fuels higher share repurchase programs. We believe this will be truer for O’Reilly and AutoZone than for Advance as Advance attempts to reinvent their supply chain.
AutoZone and O’Reilly are share buyback cannibals. They have steadily devoured themselves over the past decade with AutoZone pioneering the strategy and ORLY adopting the strategy in FY 2011. Since FY 2011 both O’Reilly and AutoZone have repurchased similar percentages of their common stock.
Managing inventory is the number one differentiator of these retailers, because the availability of parts is the key driver to sales in both the DIY and DIFM markets. Below, we highlight key aspects of AutoZone’s, Advance’s, and O’Reilly’s inventory management.
Historically, AutoZone delivered once-a-week to their stores. Once-a-week delivery is fine when the focus of the business is on DIY customers who require less breadth of inventory selection compared to professional customers. This allowed AutoZone’s supply chain to focus on depth (i.e., stocking levels) of products at their stores.
AutoZone began FY 2012 and FY 2013 with two new programs designed to improve inventory availability throughout its system. The first program increased the number of deliveries from distribution centers to stores from once a week to multiple times per week. The goal was to replenish inventory at a more rapid pace, thereby reducing lost sales due to out-of-stock inventory. On the FY2017 Q4 call, management announced they were reevaluating this program as the costs and strain on their supply chain were too high. Concurrently, they announced the program had been deployed to 2,300 stores or only 38% of the network. Management stated their intention going forward to lower frequencies of deliveries at some locations back to once or twice per week.
The second program was a decision to open larger hub stores. The typical AutoZone store is ~6,600 square feet. The new mega hubs are 20,000 to 30,000 square feet or 3 to 4.5x bigger than traditional stores. The extra space is used to add additional inventory which may not have been locally available before this expanded hub approach. Mega-hub stores act as a mini distribution center, increasing product availability to surrounding stores. Management is pleased with this program, but finding the proper real estate has created a bottleneck limiting expansion. Management plans to have 25 to 40 mega hubs.
There are two important takeaways from AutoZone’s inventory efforts. First, the frequency of delivery is important if the proper products are being held at the store. Increasing frequency at stores with lower product breadth is not a value-add if the breadth is not experiencing depth issues (i.e., stock-outs). Second, increasing products at the front lines through increased breadth at stores or improved delivery times from distribution centers to stores/hubs will generate increased sales as products are more accessible to customers.
Advance Auto Parts
Advance created a logistical nightmare with large amounts of complexity. Like AutoZone, Advance historically has catered to DIY customers, which allowed Advance’s supply chain to be built around once-a-week delivery to their stores. Prior management facilitated once-a-week deliveries through a network of hub stores acting as the backbone of the supply chain. Management changed directions in 2012 and opened its first daily delivery distribution center. While embarking on this change of direction in supply chain management, Advance made the largest acquisition in industry history by acquiring General Parts International Inc, (Carquest and Worldpac) in January 2014 for $2.08 billion. Complexity was then compounded by integrating a foreign operating system into a system which was transitioning from weekly to daily deliveries. These actions, along with the worst comps in the industry, led to a nearly complete turnover of the executive suite in 2016.
New management, led by Tom Greco, is focused on reinventing Advance’s supply chain while utilizing the assets on the ground. Interestingly, Advance is not pursuing the O’Reilly supply chain model (discussed below) but rather is rethinking from the ground up how to best integrate technology into their existing asset base. Early moves made by management include a reduction in delivery frequency with a focus on order accuracy and timeliness of delivery. When asked at the 2016 analyst day about a greenfield supply chain solution, the CEO responded hesitantly falling back to the large amount of capital already deployed in the current system. While not directly expressed, it did imply a lower ceiling of possibilities with the existing configuration of the asset base.
Today, Advance is operating five different sets of books, which won’t be fully merged before 2019, and four supply chains, which are in the process of being integrated and transformed. Resolving the supply chain issues would unlock tremendous value if the Company is successful in its efforts, but we have significant doubts about the “bones” of the infrastructure. The lesson we glean from Advance is the necessity of a clear strategy in constructing distribution centers and hubs and operating these assets in line with their designed capacity.
O’Reilly is the gold standard of inventory management. Approximately 50% of stores receive intraday deliveries 4-8x per day directly from a dedicated distribution center. Another 35% of stores get access to distribution centers up to twice per day from Hub stores. 10% of stores get intraday service from a hub but not directly from a distribution center due to distance constraints. The remaining 5% of stores do not receive intraday deliveries but receive deliveries five times per week. This system allows ORLY to maintain the best coverage of products/parts while maintaining the lowest inventory per store.
O’Reilly demonstrates its leadership in inventory management through its high degree of delivery frequencies and expanding operating margins. O’Reilly’s supply chain allows them the ability to utilize the distribution center as a storefront and reduce the amount of inventory held at individual stores creating a moat which both AutoZone and Advance have been unable to successfully copy to date. A strong supply chain will be critical in growing the DIFM business and enabling O’Reilly to earn and maintain the coveted first call status for their professional shoppers.
1. Weather – Weak Excuse
The excuses that rang through the conference calls in Q1 and Q2 of 2017 were weather and tax refund delays. While it is cringe-worthy to hear retail executives blame the weather, it is meaningfully different when your business is focused on maintenance and failure. We spoke with various individuals involved in the auto parts business, and all noted weather as a primary factor driving key replacement products such as batteries, wiper blades, air conditioning, etc... Mild winters can hide failures while cool summers reduce the wear and tear during summer driving.
We will find out over the coming months if this is the true reason for weakness or not. We think it is a primary factor of weakness along with the drop in new vehicle sales from 2008-2010 which will work through the system in due time.
2. Negative Working Capital
A primary concern of ours is the amount of leverage in the systems of these retailers. Currently, the major players have a significant number of payables, and in some cases, payables are financing more than working capital (i.e., capital expenditures, share repurchases).
Payables terms have been extended by two factors.
Scale – Big 4 comprise a significant portion of the market, greatly increasing their bargaining power with suppliers.
Strong Credit Profiles – AutoZone, Advance, and O’Reilly all maintain investment grade ratings with Moody’s and S&P. Notably, Advance maintains the weakest credit rating of the group.
We decided to take a historical look at the peer group and adjust debt to capitalize operating leases by including a multiple of rent expense (six times) and add payables in excess of 40% of inventory to debt. Secondarily, we adjusted the denominator (capital base) in our calculation of return on invested capital (ROIC) to include excess payables. As seen below, O’Reilly has the lowest debt loads and greatest capacity to expand its buyback, as well as being the only company with expanding returns on invested capital.
If working capital reverses, it will hurt our investment thesis as funding the deficit will hamper the O’Reilly’s ability to return capital to shareholders but it will not be a death blow to O’Reilly or AutoZone.
Notably O’Reilly had their debt capacity increased in Q1 when rent expense in adjusted debt was reduced from 6x to 5x increasing borrowing capacity by ~$300 million as a result. Figures below are based on a 6x multiple.
3. Dealers targeting Service and Repair
Growing complexity in automobiles is likely to shift DIY customers to DIFM professionals; however, this shift will probably take multiple years as the fleet turns over with new models and DIY customers upgrade. This probable shift is important as it generates threats to independent jobbers (i.e., repair shops) and their suppliers (i.e., the Big 4) as diagnostic and repair data related to these cars is currently held and maintained by the OEM’s (e.g., GM, Ford, Toyota…). Interestingly, the value of possessing this data is evidenced by the fact that O’Reilly, AutoZone, Advance, Genuine Parts, Bridgestone, and Firestone via CARE (Coalition for Automotive Repair Equality) are lobbying to move the data into possession of the customer rather than the OEM. By doing so, it would enable the aftermarket to gain access to this information.
The control of this data will further amplify an already current risk to the DIFM business. Dealerships, which are the primary beneficiaries of the data, are increasing their efforts to capture vehicle service and repair as these offerings bring in the highest gross margin dollars. Dealers are competing against the Big 4’s customer base, which could adversely affect the Big 4’s shares of DIFM.
4. Electric Vehicles (long section)
Electric Vehicles (EVs) will change the game when they arrive, as they have ~90% fewer moving parts than internal combustion engines (ICE). This gap in moving parts is driven by fundamental differences in the powertrain, which transmits power from the engine (power source) to the axle, between EVs and ICE. The reduction in moving parts will reduce wear and tear on vehicles and result in lower demand for replacement parts and maintenance. Notably, EVs do not require oil changes.
Limits to EV adoption?
The lack of infrastructure surrounding charging stations will act as a barrier to adoption. As of 2016, there were 363,337 charging stations around the world, which is up from 28,236 stations in 2011. In the US there are ~16,000 charging stations representing ~43,000 connectors. As of 2015, there were 127,588 convenience stores selling motor fuels in the United States according to 2015 NACS Retail Fuel Report. The key difference between a filling a car with gas and charging is the length of time. The average time to fill a tank of gas is ~4 minutes, and that allows for ~350 miles of travel. The fastest full charge for an electric car is 30 minutes; however, more typical chargers will only complete 30 miles of charge per hour. Assuming 3.2 trillion miles are traveled in the US, a fully electric fleet would require ~107 billion hours of charging per year or ~400 hours of charging per vehicle per year compared to ~2.3 billion hours for drivers filling gas in their vehicles. This problem could be resolved through charging at home. Approximately 60% of Norwegians cite the home as their daily charging facility (Norway has the largest penetration of electric vehicles of total vehicle population). However, not every car can be charged at home. For example, 43% of British car owners do not have access to off-street parking and are thus unable to charge their vehicles overnight. While we do not have a statistic for the United States, we do see significant constraints from a lack of overnight charging options.
North Americans are now buying more lightweight trucks (i.e., SUV’s, pickups, crossovers, vans) than cars. Currently, Ford, GM, and Ram sell 6,000 trucks per day or 2.2 million per year. Starting prices are around ~$27 thousand, and higher-end packages run into high $50’s. There are no EV equivalents for these trucks, today or in the near future. The Tesla Model X is the closest EV comparison to current lightweight trucks, and it has a purchase price around $80 thousand. A more economical EV SUV will be necessary for mass adoption of EVs
Distance & Cost
The Chevy Bolt and the Tesla Model 3 have the best price-per-mile vehicle range of any mass-market EVs. The top range for current mass-market EVs is 80-100 miles with the exception of the Chevy Bolt and Tesla Model 3 which boast 200 plus miles of range.
UBS estimates consumers in the U.S. will have total cost parity between ICE and EVs by 2025. Until then, EVs will rely on subsidies from OEMs or the government to obtain any meaningful market share. UBS further estimated OEM’s wouldn’t hit their target EBIT margin of 5% until 2028 in the United States.
Supply Chain (the largest bottleneck in our opinion)
In a world dominated by EVs, the demand for materials will fundamentally change. UBS completed a teardown of a Chevy Bolt and issued a report highlighting their estimates of profitability and changes in demand for raw materials and natural resources. The two materials, as determined by UBS, with the largest demand are Lithium and Cobalt. The majority of lithium supplies are found in Chile and Australia where rule of law and business practices are stable. Total reserves of lithium are unlikely to be the first bottleneck in production.
We believe the larger issue in the supply chain will come from cobalt. Over 50% of known cobalt reserves and 50% of annual production comes from the Democratic Republic of Congo (DRC). The country is highly unstable with portions controlled by warlords and armed groups. These groups use child laborers, who mine natural resources, to fund their operations. According to the CIA World Factbook, the DRC has ~8.3 million child laborers ages 5-14 or 42% of children ages 5-14 in the DRC.
The vast majority of cobalt production outside of the DRC is a byproduct from mining nickel, limiting its ability to scale to increased demand. How OEMs will solve this supply chain issue will be important, not only for EVs, but the citizens of the countries holding these resources. EV pundits point to development in battery technology to solve the cobalt conundrum. While there are some batteries that mitigate the use of cobalt, EVs being built today rely on the material as a key input for their batteries.
It is worth noting that incremental demand in a 100% EV world does not include power grid battery storage which will be required to supplement variable renewable energy generation. Thus, UBS is likely underestimating total demand for lithium and cobalt and therefore underestimating the total burden on the supply chain.
How far out is widespread adoption of EVs?
Adoption is currently behind pace of many expectations. However, announcements from regulators and manufacturers have been grabbing headlines. There are new initiatives starting up from various manufacturers, but to achieve these goals, the manufacturers will need to use hybrid vehicles (a combination of ICE and EV).
Norway – Set the goal of 100% EVs for all new car sales by 2025
United Kingdom – Ban the sale of ICE cars by 2040 and phase out all ICE cars from the fleet by 2050
France – Ban the sale of ICE cars by 2040
India – Considering 100% EVs for all new car sales by 2030
China – Discussing a ban on ICE car sales due to pollution. Set a rule forcing OEMs to increase production of EVs by 2019
China has a significant influence on international manufacturers. Why? China has an incredibly low number of cars per person. In the United States, the fleet is ~260 million vehicles with a population of ~320 million (or ~800 vehicles per thousand persons). In China, depending on your data source, that ratio is at most ~150 cars per thousand people. While China may not reach the same penetration as the United States, there is plenty of room to grow. This represents a large opportunity to increase auto sales for international manufactures who are willing to comply with Chinese regulations.
Recent developments have captured the news cycle, but there is significant room before EVs take the majority of new car sales. JPMorgan references the graph below when discussing the slow pace of EV adoption. It is worth noting that 2016 sales figures are still below 1% of total car sales.
But for argument’s sake, let’s say that everything we outlined above is resolved today. How long would it take for the entire U.S. fleet to turnover? Assuming ~17 million vehicles are sold per year; it would take 15.3 years to completely turnover the entire fleet in the United States. At the absolute minimum, we are a decade and a half away, and, realistically, multiple decades from electric vehicles dominating the fleet. Until then we will have ICE and Hybrid vehicles which will be more than enough to generate parts demand.
5. Autonomous Vehicles
Autonomous vehicles represent a threat to the Big 4 as they would theoretically reduce the number of cars on the road if car sharing took off. The declining trend in the number of high school seniors who have a driver’s license supports the thesis that there may be growing interest in car sharing involving autonomous vehicles. Despite teenagers being of legal age and ability to drive, it appears that teens are increasingly more likely to prefer someone else driving them around. These habits will help facilitate the leap to autonomous vehicles.