Description

Executive Summary

We built a position in the company in March when the share price was notably lower and IRR more attractive. We decided however to post Meigong because (1) our projections might prove conservative given how dismal most operators are in the industry, (2) IRR at these prices is still attactive (we own the company, so consider we bought it yesterday at these prices) and (3) we think it is good to share learnings about an industry not much followed, both for the sake of learning and because Meidong might sell-off at any point and provide a super attractrive entry point, so better to know it in advance.

Our Meidong’s ownership is predicated on the significant runway for growth provided by a large addressable market for luxury vehicles triggered by China’s emerging middle class and urbanization dynamics – 19% of the population lived in cities in 1980 vs. 61% in 2020 vs. projected 70%+ in 2030, a strategic focus on serving lower tier cities where competition is virtually absent and a strong data-driven culture led by Ye Tao, a terrific owner-operator.

There are however potentially good reasons to pass on Meidong. An only-China business operating in a low-margin and capital-intensive industry selling for 36x 2021 consensus after-tax earnings is not exactly the hallmark of an attractive investment opportunity. This report explains why we believe these reasons are wrong. 

Investment Highlights 

Large TAM – Dealerships currently serve only 345 cities in China for a total of 687 cities in the country, translating into 342 unserved cities – most in the Tier 3 and Tier 4 buckets. While not all of these citie ntly meet the minimum economic and population hurdles to justify a dealership operation. In addition there is still a lot of room for stores in cities with existing dealerships as car ownership in China remains at relatively low levels. Meidong focuses on Tier 2-4 cities and currently operates only 66 new-vehicle dealerships distributed amongst 45 cities. We believe there is a huge untapped opportunity and contemplate a base scenario of Meidong operating 146 stores by 2025. 

Attractive store economics – Most Meidong’s stores become profitable in year one. Three main factors drive Meidong’s superior store economics. 

1. Focus on luxury brands – Meidong’s stores consist of BMW, Lexus, Porsche, Audi, Toyota and Hyundai, with the first four brands (luxury) comprising 80%+ of the store count. Luxury stores command higher margins for both new vehicle sales and aftersales services and lower inventory days. 

2. Dominant competitive position – Meidong pursues a Single-City-Single-Store (“SCSS”) strategy, which essentially means that it opens most of its dealerships in cities where it becomes the first store of that brand, thus reducing competition. 

3. Capital and asset efficiency – The company sticks to the right metrics – cash flow generation, ROA and ROE, ranking at the top of the industry in this regard. Tao laser focuses on KPIs like fast returns on new stores, quick inventory turns, high services revenue, low account receivables, and low capital expenditures when possible. 

Store-base immaturity – Assuming acquired stores as immature (as they typically have significant room for volumes and margin improvement) 25%+ of the store base is less than three-years and 40%+ less than four years.  Most stores might take up to 8+ years to reach its full economic potential in volumes and margins, so Meidong is currently underearning on its current store base. 

Owner-operator – A superb executor, communicator, and out-of-the-box thinker, Ye Tao has impregnated Meidong with a simple, direct, and meritocratic culture. 

Valuation – While the company looks optically expensive, we believe valuation is attractive considering Meidong’s untapped growth opportunity, model superiority and operational excellence. 

Brief History

Tao’s younger brother, Ye Fan, started the first store in Dongguan in 1998, established Dadong Group in 2007 and started to invest in more dealerships to finally take together with Tao the operation public along in 2013 under the name of MeiDong. In 2013 Meidong operated 14 stores, did ¥3,480m sales, ¥185m EBIT, ¥110m net income and ¥0.14 EPS. At the end of 2020 it had 63 stores, did ¥20,207m sales, ¥1,146m EBIT, ¥771m net income and ¥0.61 EPS. Sales, EBIT, net income and EPS have grown at 7-year CAGR of 29%, 30%, 32% and 24% respectively. Since IPO the company has paid out ~40% of its profits in dividends, relied on mostly internally generated cash flow to fund its growth and its stock has compounded at 50%+ CAGR. 

Upon his brother’s request Tao joined Meidong in 2008. A bit of background on Tao himself is necessary as he has been instrumental to implement the data-driven results-oriented practices which have made it possible for the company to become what it is today. 

Engineer by education and entrepreneur by occupation, Tao majored in Mechanics from the Peking University in 1989, moving then to the US where he studied Material Science and Engineering at the University of California to finally obtain in 1996 Master’s degrees in mechanical engineering and Management at the MIT. To put things in context only a low single digit proportion of high-school students in the China of 1980s went to college, a few of those gained admission in Peking University - the best one in the country with Tsinghua - and even fewer had the opportunity to study abroad in the US, leave aside the MIT. 

Now to the business side. Tao first worked briefly for large corps - Kodak, Intel, and Teledyne - till he founded Objectiva, a software outsourcing business to enterprises, in 1999 alongside a classmate from the MIT. Tao quickly moved back to Beijing to manage the Chinese division of the company with 500+ people under him. The company sold to EMC in 2008 and precisely Tao joined Meidong after the sale. 

The combination in one individual of strong quantitative skills, entrepreneurial spirit, team-building capabilities, ability to communicate clearly and extraordinary execution is rare, even more so in the Chinese dealership industry. Chinese dealers typically are cash-burning unprofessionally managed businesses, making the Meidong-peers gap very pronounced. While we pride ourselves of the unparalleled taste of Spanish food, we find Tao’s letters to shareholders even more delicious to consume. 

Brief Industry Remarks 

Leaving aside the long-term trends of Chinese urbanization, middle class formation and rising consumption propensity, the car industry has been struggling with three consecutive years (2018-20) of declining unit sales amid a combination of factors resulting in the perfect storm. First, the government implemented purchase tax cuts from October 2015 to December 2017 (5% for Oct 2015 – Dec 2016, 7.5% for Jan 2017 – Dec 2017 and back to 10% afterwards) which pulled demand forward. Two, the P2P fundingcollapse triggered a consumption downgrade in lower-tier cities during 18-19. Three, the administration imposed new-vehicle registration caps for tier-1 cities. Four, the effects of COVID-19 in 2020. 

While this weakness in demand might be more structural for larger cities as the government incentivizes mass transportation through investments in infrastructure and subsidies to counter the high levels of air pollution and traffic congestion, it looks more like a transitory period or does not even will apply any time soon to smaller municipalities – the ones relevant to Meidong – due to a deficient transportation network in contrast to larger urban areas and the low penetration levels of car ownership compared to other countries including Beijing, which shows the highest penetration rate. 

Interestingly, luxury vehicles unit sales have been immune to this decline as a result of their inelastic nature – some brands and models are actually Veblen goods. 2017-19 luxury unit sales grew +9% CAGR vs. -6% CAGR for total new vehicles, thus taking share uninterruptedly - now ~17% of units sold in China vs. ~9% at the end of 2016.The booming demand for luxury mostly stems from the marginally lower prices as a result of the rising share of local production in order to bypass tariffs (currently still around 1 of every 4 luxury cars is imported), lower costs passed onto customers from the economies of scale amid higher factory utilization and entry-level luxury models gaining weight in the mix. 

Within luxury unit sales BMW, Mercedes-Benz (Meidong operates no Mercedes stores) and Audi command the lion large share with respective ~23%, 23% and 21% share. The other two luxury OEMs Meidong operates, Lexus and Porsche, show 7% and 3% share. 

There are 30,000+ stores in China, luxury ~15% of those so 4,500. 14% are Mercedes stores (660 stores), 14% BMW (650), 13% Audi (590), 5% Lexus (230) and 3% Porsche (130). Interestingly for Meidong, the market concentrates in lower density regions around its two main brands with BMW comprising 30-35% of the stores and Lexus 30-40%. Mercedes is a distant third with 20-25%. The virtual absence of other OEMs in smaller cities in aggregate makes intuitive sense. For one, less-known brands aim to reduce demand risk by positioning their stores in high traffic areas and second, high-end luxury brands are captive of large cities.  After all you do not expect many Porsches in a place with 150,000 people and $10,000 per capita income. These dynamics facilitate Meidong’s SCSS strategy, operating as the only player in town in many of its locations. 

Brands 

The Chinese car luxury is an oligopolistic market where the German brands BMW, Benz, and Audi dominate with over 67%+ market share, with the largest players taking share over time and the smaller players losing relevance. One outlier was Tesla, which started from a low base and has the lead in electric by not only offering a superior product but also benefiting from tax breaks and a new local factory which lowered prices. 

Meidong sticks to a strategy of working with best-in-class groups such as BMW, Toyota Group and Volkswagen. We differentiate though two phases in Meidong’s approach to brands. From its foundation to IPO the company was mainly about mid-to-high labels with Toyota and Hyundai at the forefront. When the company went public in 2013 it concentrated on luxury-only. In fact it has opened no Hyundai or Toyota stores for the last two years.

Luxury is less cyclical, less commoditized and bodes better for the single-city-single-store strategy. The increase in population required in a city to support a 2nd luxury store is higher than for lower-end categories, making therefore less likely that rational competition enters the space. Automakers also strengthen this monopolistic position Meidong benefits from in most cities as they want to avoid brand dilution at any cost so they are reluctant to quickly franchise a second dealership in the same city – e.g. a Porsche store closing would make it look like a declining brand. 

We also believe Meidong approach of working for multi-brand OEMs is a smart move. One advantage for dealers when working with multi-brand OEMs is their ability to translate their past successes into new brands added, facilitating store count growth and reducing brand-concentration risk. For example, when a dealer successfully operates a Porsche store it is more likely to get an offer from the OEM to open an Audi store.

For the reasons above it makes sense to us Meidong picked its spot in multi-brand and luxury. Now, why specifically BMW, Lexus and Audi? The short answer, at least for BMW and Audi, is because both automakers are top of mind of Chinese consumers. The long answer is more nuanced as each specific brand presents pros and cons for dealers. 

BMW is so structurally bullish on China that it will increase its stake on its local JV Brilliance from 50 to 75% in 2022 by investing ~$4.2B. BMW used to be a difficult automaker to work with for dealers. In 2014-15, years when luxury underperformed in China, BMW’s dealers were unable to meet their targets to get the floating rebate. We elaborate further on the rebate system in the “store operations” section, but long story short, automakers assign KPI-based targets to dealers to pay them a rebate (the floating rebate). There is also a rebate per vehicle unrelated to KPIs (the fixed rebate). Dealers complained of what they understood as an unfair rebate system and BMW reacted by implementing a model with higher fixed rebates instead, making now BMW a more dealer-friendly brand. 

While Audi’s dealers extract comparably lower margins on new sales than dealers operating other luxury brands, they benefit from the aftersales stream that the largest base of vehicle on the road confers. Knowledgeable of this fact Audi charges high prices to dealers on auto-parts. Managing a new Audi dealer is hard. Unit sales do not provide much money upfront and it takes a while to create a serviceable car base through new sales and by attracting other Audi vehicles to the area sold by other dealers. Meidong’s partnership with Audi stemmed as a hedge to its BMW dependency rather than a real interest to push hard the brand, and currently waits for more data to see whether or not is a good brand to expand. 

Lexus failed to catch on in China and has a small car base insofar because Toyota refused to manufacture Lexus cars in China, depriving the brand of an opportunity to catch its German rivals. Local production is critical to get massive scale as it removes import tariffs, reduces shipping costs, and limits geopolitical risk. It is thus unsurprising to us that, Audi, BMW, and Mercedes, with 92%,75% and 75% of its sales coming from units produced locally, command the lion share of the market whereas Lexus, with no local production, is not nearly as popular in China (less than 30%-unit sales vs each German peer in 2020). Lexus is nonetheless the fastest growing brand for the last years (ex-Tesla) because it is a terrific brand to work with for dealers as they extract great economics in the form of a favourable rebate system to get high margins on new sales. 

Mercedes benefitted from a strong cycle for the last few years by launching lots of new successful models and as a result operates at full capacity on its Chinese factories. Meidong does not partner with Mercedes partly because its inexistent track record with the brand and also because of the unwritten rule that states dealership groups working with BMW can’t work with Mercedes and the other way around. Anecdotal evidence of this is the fact that the largest Mercedes dealer, Zhong Sheng, started operating BMW stores only after 2017; the largest BMW dealer network in the country, Guanghui – formerly participated by Evergrande which in order to reduce its debt sold its 41% stake in 2020 to the government-owned electricity operator Shenergy – operates a single Mercedes store which added through an acquisition rather than organically; and last, another large BMW dealership group, Yongda, only runs three Mercedes stores. We believe Meidong will try to enter Mercedes-land in the following years and it will likely be through acquisitions. 

Porsche suffers theoretically from the same constraint as Lexus does. No local production, all imports. The implication here is however very different given Porsche’s super luxury status, isolating the brand from demand risk as customers behave far more inelastically at such price point. Porsche has made its path to riches from customer exclusion, making tons of money per vehicle (the store economics section goes into specifics). 

Competition 

OEMs typically prefer to work with large dealer groups – i.e. those exhibiting financial strength and a successful track record of operations – as they tend to outperform small ones. This is obvious. BMW is more reluctant to do business today with dealers which might go bankrupt tomorrow. There is still however a large proportion of regional inefficient dealers operating just one or a few stores benefiting from know-how of the local market and the guanxi (a Chinese term standing for relationships, meaning the store operator is usually a friend/relative of and OEM senior management or high ranks at the local government). 

Guanxi is still a relevant factor in terms of dealership selection, but as the market has matured merit-based criteria have become increasingly relevant as only the best performing stores can now make outsized returns. As they focus on the wrong metrics (margins), most dealerships struggle to make a profit. On the contrary Meidong is all about inventory turns, gross profit per vehicle and after sales conversion. The difference might seem subtle, but in a capital-intensive industry relying on 3P financing it means life or death. 

On top of picking the wrong KPIs most operators lack sufficient scale to survive for extended periods. For example, Yongda is the largest operator with 65 locations out of the 650 BMW store base in China and Zhong Seng is the elephant when it comes to Lexus and Audi with 45 and 30 locations for these brands vs. 230 and 590 nationwide. While larger groups together show some concentration, they leave a long-tail of cash-haemorrhaging mom and pops. Meidong’s strategy, execution and incentives represent a significant opportunity to grow both organically and by acquiring stores from the underperforming smaller competitors. 

Not only we believe that Meidong’s is well suited to capture this opportunity of market consolidation over time but also it is the best positioned. ZhonSeng, Yongda, ZhengTong, Harmony and Grand Baoxin are Meidong’s larger peers, with ZhongSeng the only one deserving some credit for its achievements. 

The largest independent luxury operation in China, Zhong’s store count is 6x Meidong’s. Unlike Meidong it focuses on Tier 1 and 2 cities, pursues a balanced portfolio between luxury and non-luxury (60% of its stores are luxury vs. 80% for Meidong) and has a more diversified brand catalogue. It does not deal with Porsche but operates the Mercedes brand (the second largest Mercedes operator in China after Lei Shing Hong, a private conglomerate operating 150 stores), Jaguar-Land Rover, Volvo, Nissan and Honda brands. Like Meidong, it operates BMW, Lexus, Audi and Toyota brands. 

We see Zhong’s management as data-driven, project-centric, and operationally thoughtful. In fact Meidong copied some of Zhong’s most valuable practices and implemented them in smaller cities. On the other hand, Zhong has also been historically more aggressive (we think too much) on its capital allocation initiatives. It heavily relies on growth through acquisitions (acquired more than half of its store base) and aims to profit from land appreciation by including land ownership in most of its purchases. We see this second practice as questionable, to say the least, in terms of prudency and resilience to the cycle. 

Yongda is 3x Meidong’s size and we believe it is mismanaged. Management does not see high inventory days as an issue as it prefers to have sufficient in-store inventory to service customers’ immediate needs, discounting heavily when inventory moves slowly. We recommend you read Tao 2020 letter to shareholders where he differentiates between fast moving inventory (Rabbits) and slow moving one (Turtles) and how detrimental for returns the latter is. Yongda’s management should read that letter in its own interest. Yongda also diworsified by entering the used car market, developing EVs and rolling-out independent after-sales stores, long-term car rental and auto financing. You know… any company, with time and effort, can do anything. But it cannot do everything!  So no matter Yongda is more than 3x the size of Meidong by all measures (units, store count and revenues) and has similar mix towards luxury, it is quite leveraged, has low margins and high inventory days. 

Zheng Tong is an almost dead 125-store operation (91 luxury with BMW, Mercedes, Porsche, and Audi at its core) paying a high-price for misguided strategic decisions resulting in excess leverage exacerbated by the COVID-19 crisis onset. Like Zhong, it includes land on its acquisitions but unlike Zhong it also focuses on consumer auto-finance (the only dealer in China with a banking license). In 2019 it spun-off and listed its finance vertical under the name of DonZheng (still owns 71%). Both ZhengTong and DonZheng have proven disastrous for shareholders. 

Harmony has exclusive focus on luxury through 10 premium brands (including BMW, Lexus and Audi) and 4 ultra-premium ones (Rolls Royce, Ferrari, Maserati, and Bentley). It is mainly positioned in Tier 1 cities with little success in reducing inventory days. The company also failed to scale other initiatives as its own independent aftermarket service (IAM), EVs and ride sharing. Now it tries to replicate Meidong’s strategy by deemphasizing IAM, concentrating on its main brands (BMW and Lexus will be the focus moving forward) and targeting smaller cities. While Harmony’s strategic shift is the right way to think about its long-term sustainability, we think the company will find it difficult to replicate Meidong’s success as smaller cities represent a quite different operating environment (guanxi, local rules and regulations, etc) and because the implementation of an inventory-rooted culture and its associated incentives company-wide is hard to achieve as Harmony needs to leave room for consumer upgrade on its brand portfolio by keeping alive its ultra-luxury vertical, which yields lower inventory turns. 

Grand Baoxin went bust and China Grand (State Owned Enterprise) bailed it out by acquiring a 54% stake in 2015. Baoxin was very aggressive expanding when the cycle was sunny, levering up and entering new businesses (used cars), but as the company had significant exposure to BMW and the brand set aggressive targets in 2014 Baoxin was unable to survive amid insufficient rebates. 

Sunfonda presents no store overlap with Meidong as most of its stores are in Northwest China (with a few in Jiangsu). We heard from some local befriended investors that there is nothing particularly impressive about management. We lack further insights on how Sunfonda runs its stores and brands but we track the company as its brand portfolio looks attractive, although we feel store locations are inferior as these regions may grow slower.

An additional potential threat for Meidong (and the other dealers) is OEMs going DTC. We think this OEM forward integration is unlikely to happen though. For one, the business of selling vehicles is radically different to that of manufacturing them. Different skills and mindset. While some companies like Apple are able to successfully integrate production and retail activities, we argue this is more the exception rather than the rule. Most manufacturers don't retail, and most retailers don't manufacture. 

Two, creating, scaling, and maintaining a retail network is absurdly expensive. The average BMW dealership set-up cost in China sits at ~¥17m ($2.7m) and a Porsche store might cost up to ~¥60m (more than $9m) excluding the land. By selling franchise rights to dealers, automakers increase their capital on hand rather than invest it into facilities. 

Three, OEMs love guaranteed cash. Sure they could technically go direct, but it would mean they would have to carry all the inventory themselves and bear the risks of seasonality and swings in customer demand. It is a little-known fact outside the auto industry, but automakers consider a vehicle to be "sold" when it hits a dealer's lot, not when it ends up in someone's driveway. This is because automakers sell their inventory to their dealers, which then in turn retail the inventory to the public. In fact most automakers mandate that their dealers take inventory on a regular basis as a prerequisite for their franchise, which means automakers can always sell cars to their dealers (even if the public is not buying!).

Four, dealers know their customer base better and excel at filling local preferences. This is because the concept of “nationwide market” is an illusion. A dealer in Hong Kong will carry a different mix of vehicles than a dealer in a remote Tier 3 city as the people in these areas have different wants and needs. Dealers can order the cars they believe their customers will want to buy, and automakers can build what is ordered rather than trying to guess about which package/feature will work best in any particular market. What is more, dealers can build community relationships that a corporate automaker might struggle to create.

Last, dealers provide local service. Cars are complicated machines, and once you have a few hundred thousand of them driving around, you need a network of service centres to make warranty repairs, conduct recalls, and provide maintenance services. In-N-Out is such a cult in the US that some customers might drive for hours to get a burger. We remain

sceptical car owners in China are as excited to drive that long to the nearest OEM factory to maintain their vehicles. 

On top of competition for new car sales, we also factor competition for the after-market. While a dealership sticks to the typical blade and razor model – i.e. selling a product which makes little or even lose money upfront in order to secure and installed base to make all the profits later by offering high margin goods/services – that has historically worked so well for elevator companies, printer operators and aircraft engine manufacturers, dealerships are theoretically unable to benefit from strong customer lock-in. In the end a BMW buyer can legitimately decide to run his car maintenance through a 3P player. 

We keep our eyes open on well-funded competitors, specially Tuhu. Tuhu is a 2011-founded online-to-offline maintenance and repair service operation which has raised over $600m – with Tencent, Sequoia and Carlyle participating – and plans to IPO in the US. It has considerable scale and reach, with the app connecting its 52m+ customers through WeChat and Alipay to over 2,400 car repair centres and 13,000 partner centres in 405 cities. The fact that Tuhu is more focused on the tire vertical and its value proposition more compelling to vehicle owners of mid and low tier brands mitigates the risk for Meidong.

We also believe luxury car owners have strong incentives to push back their vehicles to the official stores in the form of warranty periods, low price sensitivity and more importantly inventory availability, motivated by OEMs prioritizing delivery (or even serving exclusively) of auto parts to their branded stores, thus rendering Tuhu’s scale useless to under-price official dealers to get original components. Authenticity is a central aspect for most luxury vehicle owners and owners have no way to know whether Tuhu’s parts are originals or not. 

Store operations 

A Meidong store sells new cars and provides aftersales services. Every time Meidong sells a BMW is normally making 3-4% gross margin, 6-7% for Lexus, 7-8% for Porsche, and 3-4% for Toyota.  The “normally” nuance is central to the thesis. Let us take a step back here to explain why this is the case. 

When a dealer orders a vehicle, it typically funds 10% of it with its own capital while a bank or the manufacturer’s finance arm covers the remaining 90% through 30-to-90 days bank notes with a 30-day interest free period. With this funding structure and timing quick inventory turns are critical to survive. In Tao’s own words, in Meidong they “live or die by inventory turns”. Every dealer faces a dichotomy between offering some discount to the customer at the beginning to get rid of slow-moving inventory and attempting to get full margin at the risk of a much larger discount later if that inventory remains unmovable. While most dealers decide not to sacrifice margins Meidong is all about turning inventory quickly.

Some numbers to shed light on why this is the case. Imagine a worst-case scenario where Meidong decides to quickly sell a BMW it just received from the OEM at 1% gross margin and turns its inventory 12x a year. As that inventory is only 10% equity-funded and Meidong is not paying interest on the OEM loan because the car sold in just a month, it makes 10% return on its capital on every turn and 120% a year. Compare this scenario with another where a competitor is unwilling to discount hoping for a 4% gross margin. There are cases where the dealer makes full margin. However often times the dealer is unable to move that inventory at the end of the interest-free period, making it more likely to discount. Moreover as the time to pay back the principal to the lender gets closer the dealer must accept even a deeper discount to redeem the outstanding banknotes. In such scenario 20%+ discounts are usual practice, the store becoming an unsustainable cash burning machine. 

The actual OEM-dealer economic relationship is a bit more complicated than in the above example. In fact the actual gross margin for a dealer is the combination of a fixed rebate and a floating rebate. The fixed rebate is the gap between the sale price including accessories and the purchase price. The variable rebate depends on meeting specific targets typically based on unit sales and customer service. Each OEM has its own fixed and floating combination - BMW 10% fixed + 2% floating, Porsche 8% + 4.5% and Mercedes 8+4%. As dealers face uncertainty on future sales and razor-thin margins it is unsurprising they prefer a higher fixed component. In 2014 luxury underperformed in China and OEMs gave subsidies to many dealers to keep them afloat. BMW was especially aggressive on its targets and had to raise its fixed rebate. We do not factor on a perpetual mercifulness from OEMs when times get south, so sound inventory management is critical under the rebate system. We believe that putting inventory management as first priority is so ingrained in Meidong’s DNA that it confers the company a durable competitive advantage. 

Meidong commands 46%+ overall gross margin for aftersales services - repair and maintenance. We note however that each aftermarket transaction brings a different margin depending on (1) who funds the service, as OEMs and insurers paying below retail price (an OEM typically brings 30% margin for maintenance services during the warranty period and insurers 40%) and (2) the mix between labour (higher margin) and auto parts (lower margin). We assume 46%+ for modelling purposes. 

Meidong benefits from a third source of income in the form of additional services, finance referrals to 3Ps, and insurance commission for referrals (together accounted as “other income”). About 70% of customers buy a luxury car on credit and the typical LTV is about 60% of the retail price. The dealers get 2-3% of the amount financed as commission, so from 0.84% to 1.26% of the retail price (2 or 3% x 60% x 70%). The portion of insurance premium the dealer gets is about 1%. Add things as alarm systems and extended warranties and “other income” typically translates into 2-3% of the vehicle retail price.

Meidong’s stores typically get 8+ years to reach maturity. Interestingly, the oldest BMW stores (2010) still grow double digit for aftersales and single digit for new sales. The long ramp-up period is a function of extended vehicle replacement cycles (infrequent purchase), awareness (some people take some time to realize there is a new luxury store in town as word of mouth does not happen overnight), rising incomes (people who are not potential buyers now but will be in the future) and aftermarket gaining traction as sold vehicles go back to stores sequentially for repair and maintenance. Furthermore aftersales are not only accretive from a margin perspective but also because they expand the addressable market as a store can service same-brand vehicles bought from stores in other locations. In other words, a Meidong store services a higher number of vehicles than those it sells.

We illustrate below the store economics per brand down to gross profit. The two drivers of profit-growth are volumes and margin expansion as after-sales gains weight in the mix as the store ramps-up. While 400+ bps of margin improvement might look like a small feat in the SaaS-era, for the average BMW store means 80% incremental gross profit. Lexus is the most profitable brand from a margin perspective while a Porsche store commands the highest gross contribution despite delivering just ¾ of BMW/Lexus unit sales because its exclusivity translates into a 20-40% higher price per vehicle.

Nyala Investment Fund Estimates

Addressable market 

Dealerships serve 345 cities in China for a total of 687 cities in the country, translating into 342 unserved cities, but we believe not all these cities are addressable and the exact TAM is difficult to identify because it is dynamic. Simple math tells us that if a BMW/Lexus store needs to sell 1,000 units annually at scale, when a city sells around 10,000 vehicles per year is probably a good time to open a luxury store as luxury in Tier 2-4 cities commands high-single digit share, lower than the 30% share in Tier 1. Location also matters as cities see inflows and outflows of people over time. For example, setting up a dealer in the northeast sounds like a bad idea as many cities in the area have experienced ongoing population declines for years. 

How large is then the opportunity? We guess an educated conservative TAM of ~140 cities and 300+ stores, ~5x Meidong’s current store count, as Tier 2 cities can take in multiple stores and many Tier 3 and even some Tier 4 will be able to absorb two stores. 

In order to get conviction of Meidong capitalizing on this opportunity we track automaker’s appetite to grant new dealerships as part of their expansion plans (testing whether Meidong is the preferred option for them), Meidong’s inclination to M&A and store-level performance for locations with multiple stores. We are positive on all three. 

For one, it is not a secret China is central to BMW and Lexus’s strategy and that both brands have a strong preference to award dealerships to companies with a track record of profitability and operational excellence. A company which has made no losses for over a decade, operates with a net-cash position excluding inventory funding and takes only 9 days to turn its inventory (which we find mind-blowing) strikes us as the strongest contender to get most new openings. 

Concerning M&A, Meidong has historically played conservative in the space, especially after finding out that some of its initial targets included shady owners, undisclosed off-balance sheet transactions and blatant fraud. Meidong’s acquisition track record is limited to two operations. It first acquired 6 struggling BMW stores in 2018 for ¥48m. It was a terrific call considering Meidong typically spends ~¥17m to open a new BMW location and the acquired stores are now cash-generators. In January 2021 Meidong bought two stores – one BMW and one Lexus – for ¥310m together (undisclosed separately). Assuming ¥155m (half) goes for the BMW store, this is almost 20x what Meidong paid for each BMW store in 2018. You might think management lost its mind, but frankly it is quite the opposite. First, a buyer gets much less bargaining power when the seller is a strong player – i.e. Zhong Seng was the counterpart for the BMW store. More importantly, the 2021 operation includes the right to use the land for both stores (free rent). We estimate Meidong paid ¥60m for the BMW location excluding land, translating into 0.1x revenue and 0.7x gross profit at scale. If that is not an attractive valuation, we do not know what it is. While two operations might be too short of a track record to extract meaningful insights to judge management’s capital allocation capabilities, we find comfort in (1) the time management has spent in learning about deals (opportunities, structuring, complications) without incurring in FOMO… is not it true most successful people are patient and say “no” most of the times? and (2) Meidong’s M&A division comprises one woman with no headcount – the colloquial one-woman show – which brings accountability and responsibility to confer Meidong an edge. We feel confident that Meidong is now well positioned to step up its pace on acquisitions. 

Finally, we conducted checks in some of the cities in the list below and found no evidence to conclude that store-level performance is worse in cities with more than one store. Our finding is supportive of the idea that the main driver of performance is management execution rather than simply SCSS strategy. This idea is central to the thesis as it expands TAM without a linked deterioration in the underlying economics as Meidong gets larger. 

Valuation 

We contemplate three scenarios depending on store roll out – base, bull, and bear. Our assumptions are as follows. 

Operating line items – Applied to the 3 cases other income, distribution costs and admin expenses are 1.1%, 2.8% and 2.2% of revenues at scale. Total opex line at scale = 3.9% (2.8%+2.2%-1.1%). Administrative expenses get the most leverage as they comprise the largest proportion of fixed costs. We model opex at scale for the next five years. 

Initial investment outlay per store excluding land costs – For organic openings we assume ¥17.5 for BMW and Lexus stores, ¥60 for Porsche and ¥12 for Toyota and Hyundai. For acquisitions we project ¥60 for BMW and Lexus stores, ¥200 for Porsche and ¥42 for Toyota and Hyundai. This also applies to all scenarios. 

Total store additions – In 2020, Meidong raised capital with the main purpose of M&A. This is why we assume a higher number of M&A additions in the near term. BMW M&A adds are also slightly higher than other brands due to the higher addressable store base and the minor assumption of Mercedes store buyouts (for which we assume same store economics). 

Store economics down to gross margins – The store economics described in the store operations apply. We should ideally project scenarios with different store performance, but we believe Meidong would shut the underperforming stores (if any), which is somewhat earnings-equivalent to assume fewer stores but all of them operating well. 

Base case

Bull case

Bear Case

Risks

Electric vehicle – This risk applies from two angles. (1) Meidong brands being unable to adapt and losing market share and (2) shrinking aftermarket revenues as EVs contain fewer parts and break down less. We elaborate more on the brand risk below. On the aftermarket risk 50% of Meidong’s reparations come after an accident and thus are unaffected by EV penetration. On top of this, the most optimistic estimates we have read estate 25-30% EV penetration by 2030. We assume 30% EV cannibalization times 30% aftersales unit reduction by 2030, translating into 9% top line impact which we think will not be a big deal for Meidong to replace over a decade. 

Autonomous vehicles – While we might be bullish on the benefits and prospects of autonomous transportation over the very long term, we are a bit sceptical on autonomous vehicles dominating the road at the end of this decade and strongly believe self-driving car will not be the default transportation method in the next five years. We are not experts on it but believe there are still technical bottlenecks, regulatory burdens, human scepticism in terms of safety and ethical trade-offs that make it difficult quick implementation and adoption. We admit however this is one tail risk of our investment in Meidong. 

Brands – For established brands like BMW, Lexus and Audi, there is little brand risk associated with brand durability (these brands are not fads) in a non-electric world. Electrification however might bring a paradigm shift where old brands do not get traction as they happen to be technically inferior. With “technically inferior” we do not mean batteries/hardware components (which we think will get commoditized) but software (intelligent connectivity), more electrification-related. That might be more difficult to replicate. A mitigant for potential brand-shift in consumers’ minds is Meidong’s culture and efficiency, which might help to adapt its portfolio by rotating failing brands to upcoming ones. 

Getting comfortable with this risk is all about the vision of the vehicle for the next few years. If software will be the only thing that matters then Meidong would be in a challenging spot as German brands (or any of the combustion native players) have no apparent advantage vs. newcomers (even a disadvantage with their legacy production models and incentives). Thus there is no guarantee that the “BMW of the 2030s” is well… BMW. 

In our view however hardware will still take an important role for vehicles in the future, at least for the next years. This is because unlike smartphones, which are software oriented, cars have longer lifespans and require a higher level of comfort during their usage. The main mitigant to this “all-software” risk is human inertia, as we think it is still early for most people not caring about the design, comfortability, and ultimately ownership of their cars. We admit if the world in the next decade is all about software and ride hailing for everybody in all situations, Meidong will likely be a terrible investment. 

Models – There is potential concentration risk in terms of profit pool for some models. For example, in 2019, the X5 accounted for about 75% of BMW’s total new-car gross profit, when 3-Series, X1 and 7-Series were all loss-making. We think this is inherent to the new car business, but a portfolio comprising a variety of models as it is the case with Meidong’s brands reduce the risk. Automakers also exercise sometimes the option of life extensions for successful models (with few or no changes) to reduce this risk. 

Relationships with automakers – Underperformance and brand erosion might damage Meidong- automakers relationships, resulting in a reduction in opening, no new openings and even closures of existing stores in an extreme case. This was the case for Zheng Tong, which cut costs (impoverishing customer experience) and fire sold inventory (diluting brand value) amid its abysmal finances, angering OEMs which responded by revoking some of Zheng Tong’s dealer licenses. While Meidong is not exempt of this risk, we believe its best-in-class culture, strong financial position, track record and reputation reduce it. 

Execution – We see store execution at two levels. Organic and M&A. We think Meidong’s culture, strategy, and incentives counter execution risk at the organic level. We also believe Meidong can implement its best practices into the acquired stores as it successfully did with the 6 BMW stores. 

Key man – Tao might pass, get tired or decide to leave for whatever reason. On the first risk, we would love to be Indiana Jones to give Tao the Holy Grail so he can live forever. Unfortunately we are not. To our comfort, at least, Tao loves cars but happily he does not seem to drive them in Fast & Furious mode. 

Death aside, we think there is low risk of Tao leaving the company any time soon. Now 48, he has many years ahead to execute on his plans. On top of this Tao and his brother combinedly own ~57% of Meidong, which provides a strong financial inventive for both to remain in the company until it becomes a household name in most parts of the country. 

Not everything relevant in business are economic incentives though. In fact a mission-driven management that “does the right thing” is far more important determinant of corporate success. A single episode gives light on why this is the case. 

Wuhan and Huanggang, a nearby city, were under strict lockdowns in early 2020. Meidong has a Porsche and BMW store in each city. While the local government advised businesses to pay employees minimum stipends much smaller than their salaries, Meidong decided in a March 2020 meeting to pay those store employees their full salaries and bonuses regardless both stores had been closed for two months. While a difficult decision at the time, it is the type of decision that separates the average from the outstanding. To build loyalty results in low-employee turnover, to our understanding one of the most underappreciated competitive advantages of a business on these days. 

If you are sceptical, think management is promotional and want to run a sanity check to test whether management’s excellence permeates through the entire organization, we recommend you speak to Winny Yip, Head of IR. After years of rushing CEOs and IRs with hectic lives, you will get the benefit of someone who is eager to respond all of the questions she knows and look for an answer to those she does not.  Average time we spent on each call with her? Over three hours. No exaggeration. So whether you finally invest or not, if you want to get a free-MBA in Chinese dealerships Winny is the best school. We have yet to meet an IR-person with more enthusiasm when speaking about a business. We think we won’t.  

Catalyst

- Store roll-out (organically and M&A)

- Existing store base ramp-up, reaching scale volumes and expanding margins amid aftermarket gaining relevance in the mix