Why Entry into Making Electric Vehicles is Easy

Introduction

Entry

New entry into the auto industry is easy, with the Chinese cell phone maker Xiaomi, internet giant Baidu and another tech firm launching production this year in China. That’s on top of a hundred or so other Chinese firms¹. Then there are Tesla and battery maker BYD (both now with a decade-plus under their belt), Faraday, Lucid, Rivian and many others in the US and Europe, including Togg in Turkey and Rimac in Croatia. [Caution: A friend in Silicon Valley cautioned me that for a startup, $1 billion is a lot of money.]

Here I explore enablers of new entry, focusing on two elements. There’s the tech component, the seemingly daunting task of getting a reasonably competent vehicle into production. There’s the finance component, pulling together the necessary resources to get to Job One, or SOP (start of production).

To illustrate the process, I start by focusing on the US a century ago, before returning to the present. Venture capital is nothing new, and then and now suppliers do the heavy lifting.

Don’t Forget Exit! – Easy Come, Easy Go

Making money in the auto industry is a whole different matter. Most attempts to make vehicles fail, both in the early 1900s and this past decade. With a couple exceptions, recent entrants remain deeply in the red, whether in China, the US or Europe.

Overview

In the era leading up to the Great Depression, over 700 companies entered the passenger vehicle industry in the US, and at peak in 1910 about 270 were in operation at the same time. (Another 700 or so firms entered the industry in Europe.) And those are just the firms that managed to make and then sell a car. While the data are not comprehensive as for firms that made it into production, likely 3,000+ enterprises sought to enter the industry in the US. Most, obviously, never made beyond the initial fundraising stage. Out of those early entrants, only 3 remain in operation today. To reiterate, entry was and remains easy.²

The US Industry in the Early 1900s

The Formative Years

Autos were one of the turn-of-the-19th-century’s high-tech industries. Everything was uncertain. Should cars run on electricity, internal combustion engines, or steam? Who were they for? At the turn of the century, car races were immensely popular, and were Henry Ford’s entrée into the industry – he set a speed record in 1904. While that raised awareness of the industry, it didn’t answer the key questions, and making racecars wasn’t itself a viable business. Along the way entrants also had to figure out how to make cars efficiently, and how to sell and service them profitably.

First the high tech story. A count of articles in early issues of Scientific American illustrates the level of interest, and the debates over best methods in this emerging industry.³

Table 2.1 Number of Articles by Topic in Scientific American

Interest didn’t wane; Scientific American published an “Automobile Number” in January 16, 1909 (vol 100 issue 3), and continued to carry many articles.⁴ Topics included the relative cost of horse-drawn, electric and gasoline trucks (trucks were already cost competitive with horses); the physics of magnetos; the standardization of spark plugs; and lubrication.

That reflected copious entry, some 700 firms with documented production in the US. Depending on the source, the number of firms in the market peaked in 1910, while new entry peaked in 1915, perhaps stimulated by the success for Ford’s Model T and the wealth it created.

Number of Firms, US Auto Industry, from author’s firm count spreadsheet

Entry Example: Henry Ford and the Model T

For those passing through Detroit, the original Ford Model T manufacturing facility, the Ford Piquette Avenue Plant, still exists, unlike the later Highland Park plant, where the automotive assembly line was developed. Piquette is now a Model T museum, and a visit gives a flavor of the formative days of the industry. Visit!

Initially Ford made nothing; his workers simply assembled parts purchased from outside suppliers. Now “simply” is a stretch, because the parts came from one or another of the hundreds of local small metalworking and casting shops in Detroit, which used general-purpose equipment to be able to sell to an array of customers. The actual parts thus didn’t exactly match the specifications, given the variance in equipment and machinists’ skills. Workers were “fitters,” who came with their own tools: files, hammers, metal shears. The “factory,” such as it was, consisted of work bays on the second and third floors of the building, which backed onto a railway spur used to ship completed vehicles.

The Dodge Bros., for example, were a shop that made engines for Ford. At the start, in 1903, the two brothers each accepted 50 shares valued at $5,000 in Ford Motor Co in lieu of cash; in 1908, after some changes in shareholdings by other early investors, that represented a 10% stake in the company. A shop associated with the banker John S Gray, who had a hand in multiple ventures, supplied other parts. He too was an investor.

Now for years Ford proved a good customer, and soon was paying cash for parts, with payment due in 90 days. The dictates of mass production, however, led Ford to develop specialized machinery and move production of many components in-house. That eventually included a large casting operation at the Rouge in Dearborn, built when Ford outgrew Highland Park. As an aside, Ford was willing to employ black workers at the standard wage, but racial prejudice meant they were segregated to shops within the Rouge such as the hot and dirty casting operation, with black foremen and supervisors.

Anyway, by 1914 Henry and the Dodge Brothers disagreed about the direction of the firm, and following litigation, the Dodges sold their shareholding moved downstream into car production.⁵ Now within Ford Motor Co’s first 15 months, the firm paid out 100% in dividends. The brothers together earned $9.8 million in dividends during subsequent years, and when the lawsuit was settled in 1919, their estate was paid $25 million. So for their initial $10,000 investment they earned at least $34.8 million, not including what they earned as Ford suppliers. In the meantime, they continued to indirectly finance Ford Motor Co, as Ford was able to produce, sell, and receive cash from selling the Model T before it had to pay them as suppliers.⁶

The key point, though, is that to enter the industry, Ford didn’t have to make anything. He used rented facilities, he purchased parts, he could hire fitters as sales volume dictated.

Entry facilitators: automotive engine suppliers

One of my core claims is that a broad supplier base facilitated entry. Here I look at only one example, engine suppliers, but they provided the most expensive and complex component, so it’s key to document that there were many such suppliers.

Fortunately, an article in the most recent Antique Automobile traces early suppliers.⁷ I put together a table of 44 separate engine firms from Rossi’s article, rearranged his longer list. I don’t impose that on you as a reader, and instead highlight several that in their day were major players. Many existed only briefly, but others shifted to/from farm equipment, trucks, marine engines and aircraft, and successor operations survive today.

• Continental, formed in 1903, sold engines to Hudson, Studebaker, Checker, Durant, Willys and Kaiser-Fraser. They claimed 128 customers in the late 1920s. As carmakers turned to making their own engines, Continental shifted to trucks and aircraft. Its engine operation continues today as Continental Aerospace.

• Cummins was formed in 1919 by someone who headed a pit crew for the 1911 Indianapolis 500. They produced gas engines for Auburn and Packard, before turning to diesel engines, where they are today one of the biggest global players.

• Daimler entered US manufacturing in 1895, with a factory at Steinway in NYC, exiting when their factory burned down in 1907. Their engines powered many early car companies in Europe and the US. Daimler remains a global player in commercial vehicles and luxury cars.

• Northway, formed in 1903 by a former Dodge Brothers worker. They reorganized several times, and GM acquired successor operations. Those became the core of GM’s engine manufacturing. GM however did not preserve the Northway name, as GM chose to use Chevy and their other brand names on specific engines, even when the underlying engines were shared across brands.

• Rutenber, founded 1898, sold engines to about a dozen car companies. None of which survived, and nor did Rutenber.

• Sintz, started 1890 making marine engines, but also powered the Duryea, the first American car. They shifted back to marine engines as Michigan Yacht and Power, though a central figure co-founded Gray, another automotive engine producer. As noted, Gray himself was an early Ford investor.

• Premier operated during 1903-1914, and was then reorganized before finally failing in 1923. They had a wide array of customers, especially V-12s for premium makers. Typical of many of these engine makers, the founders were involved in an array of automotive ventures, in this case A.O. Smith. The latter firm was an innovator in first bicycle and then automotive frames, providing the first steel frame to Oldsmobile in 1903, and later supplied Ford. I have visited two different chassis plants (one now Tower, another Metalsa) that started as AO Smith facilities.

This sampling should help illustrate the availability of outside suppliers that enabled entry by automotive entrepreneurs, who often had their fingers in multiple automotive component ventures.

Financing entry

Obviously, a large number of firms were able to finance entry. The availability of numerous suppliers meant that they didn’t need much. Still, where did the money come from?

The answer was a combination of suppliers and the early 20th century equivalent of venture capitalists. As with several new technology booms over the past 25 years, the high tech aura and the wealth generated by early successes helped stoke the fever. Entrepreneurs could find individuals willing to bet on their horse, or rather horseless carriage.

In the case of Henry Ford, as noted above the Dodge Brothers accepted $5,000 each in shares, providing parts in lieu of cash up front. Additional money came from Alexander Malcomson, the coal magnate who supplied Detroit Edison, where Ford was the chief engineer. In the end, he raised $28,000 in cash, despite two previous failed attempts to enter the industry.

That actually left him better capitalized than most. In a later interview, Roy Chapin of Hudson Motors noted that they started with “considerably less” cash, though with the owner of the Hudson Dept Store as a prominent backer they presumably had access to additional resources. In any case, in the early years car companies didn’t need a lot of money to get started.

Others tapped capital markets, raising not tens of thousands but millions of dollars. As noted by Lawrence Seltzer, whose 1926 PhD was on the financing of auto companies, all of these de novo ventures failed.⁸ Another approach was to use the ongoing merger boom to put together an assemblage of auto-related companies. One example, pieces of which continued in operation, was the United States Motor Company. It was started in 1910 by East Coast financiers, who purchased 7 different car brands, and a large number of suppliers and retailers. However, the brands sold poorly and without a strong-selling car among its 28 models, the suppliers and retailers they also purchased weren’t profitable. The company failed in 1912, after raising $9 million in equity and bonds. After another infusion of $3 million, portions were reorganized as Maxwell Motors, with the cash used to recruit Walter Flanders as president by buying his then-profitable Flanders Motor. Maxwill too failed in 1921, and in reorganization ended up a core part of Chrysler Motors, itself a 1925 assemblage of several car firms.

As an aside, over subsequent decades Chrysler acquired other car companies, expanding into Europe with the purchase in stages starting in 1964 of the British firm Rootes and in 1970 the French firm Simca. Their operations were rationalized and a modern dealership network developed, but when Chrysler ran into financial problems in the latter 1970s they were sold for pennies on the dollar to Peugeot to become PSA. Ironically, Chrysler was acquired out of bankruptcy by Fiat in 2009, while Fiat in turn was sold to PSA to form Stellantis.⁹

Similarly, General Motors was put together in 1980 by William Durant, who used the (horse) carriage company he controlled to purchase Buick. Initially Henry Ford agreed to be acquired by the new venture, but when he demanded cash instead of shares, R. E. Olds of Oldsmobile did the same, and Durant was unable to raise enough to cover both. Durant nevertheless managed to put together an array of 20 auto-related companies, which immediately became the largest producer of cars in the US. However, many of the constituent pieces were available because they too were running out of cash, and (as Seltzer suggests) both needed capital to improve their facilities, and were acquired via expensive side deals. GM ran out of cash almost immediately and in 1910 was taken over by a banking consortium. The bankers focused on preserving cash, and in 1915 sold out when the share price rose. Ironically the biggest purchaser was to Durant, but he was no more successful the second time around, and by 1920 his main backers, the Du Pont family, gained control and forced him out amidst another reorganization.

But to return to the main point. Early entry didn’t require much cash, while ironically having ready cash available led to the purchase of marginal firms and failure. Lots of entry, exit, mergers, acquisitions and spinoffs ensued through 1925 and the formation of Chrysler, but the core pieces that survived depended on internally financed growth.

Changing Entry Requirements

By the 1920s entry requirements changed. In the 1910s two firms, Fisher and Briggs, developed the technology to make enclosed steel bodies, displacing the use of wooden frames. At the same time, Du Pont developed a new class of paints that could be baked on; until then the use of wooden frames meant that exteriors had to be air dried. (Hence the black Model T, because it was the color that dried the quickest.) That meant that a startup needed to invest in paint ovens. These are hundreds of feet long, and are preceded by an acid wash, an e-coat bath, and intervening rinses. The paint shop also has an expensive air handling system, to provide uniform temperature and humidity for the paint booths, and to control emissions.¹⁰ (“Painting the parking lot” just isn’t acceptable!) The paint shop remains the most expensive part of a modern assembly plant, costing more than the assembly line.

Furthermore, by 1925 Briggs and Fisher, the two biggest external suppliers of body stampings, had been acquired by their customers, so a new entrant could no longer turn to external suppliers and also had to invest in their own stamping facility.¹¹ A modern (that is, post-1920s) assembly plant thus is comprised of a “body-in-white” welding operation that puts together the chassis and exterior “Class A Surface” body panels, the paint shop, and then final assembly. Order of magnitude, the combination costs $1 billion.

Furthermore, except for diesels, today car companies don’t buy engines from independent firms, a trend that was cemented as engines became more complex (requiring more sophisticated casting operations) and particularly with the arrival of the fully automatic engine transfer machining line, which by the 1950s could require the then-stupendous investment of $1 billion.¹² There are independent transmission suppliers, but those build specific transmissions for specific customers.

Entry is thus not quite as easy today. There are mitigating factors.

First, EV makers don’t need a transmission plant or engine plant, though they do need reduction gears and a complicated housing for the electric drive motor, and enclosures for the battery pack. EV makers can buy drive motors from suppliers, but Bosch and others make winding equipment specialized for the armature wires of EV motors, and many EV companies make their own. The investment required is a fraction of that needed for the engine plant and transmission plant needed for an internal combustion powered vehicle.

Second, the 100 or so global Tier I suppliers such as Bosch, Denso, “Conti” and ZF all engineer the components that they supply, exemplified by the innovations recognized in the Automotive News PACE Awards. They not only fund the underlying R&D, but also the capital investment required for production, with plants in all major markets that allow them to supply firms on a global basis. With the growth of the EV market, new suppliers have risen to prominence, such as Infineon and Renasas, producers of the chips needed for inverters and battery power controls. Other suppliers provide the screens, software and computers needed for modern infotainment systems, and for driver assist and safety systems. Supplementing the engineering services of the Tier I suppliers are independent engineering consulting firms such as AVL, which with 12,000+ employees and a strong presence in China can undertake almost every part of the vehicle development process. New entrants can focus on designing a vehicle, and integrating components, but can rely on others for much of the work needed to bring a vehicle into production.

Third, car companies can turn to independent dealerships to fund the investment in distribution and service facilities. They pay an up-front franchise fee, supplementing other investment sources in the startup phase. Once a firm enters production, dealership buy vehicles as they roll off the assembly line, so improve cashflow. They undertake most marketing and finance inventory, not the car company. They also invest in service facilities, pay for an inventory of repair parts, and train their staff. Local reputation is important for a franchise’s sales, so providing repairs promptly is important. While some new ventures have chosen the direct sales route, Tesla for example has struggled to provide service, which is one important reason why the rental company Hertz has “defleeted” most of the Tesla vehicles it purchased. With sales no longer expanding, Tesla is having to fund unsold inventory.¹³

Fourth, and less reliably, new entrants can also use joint ventures and the purchase of idled factories to lessen the up-front needs for cash, or contract production of vehicles they have designed to other firms.¹⁴ In practice, joint ventures are unstable. When additional investment is needed, which is almost always the case, one or another party may be reluctant (or unable) to put more resources into the venture. Even if it becomes profitable, the partners may no longer agree on the direction the venture should take or the rate at which it should expand. That, in fact, was why Henry Ford’s first two ventures failed, and why he later bought out the Dodge Brothers and his other outside investors.

Finally, let’s not forget the expansion of existing car companies into new markets. In addition to the Detroit Three, we now have 3 German producers (VW, Mercedes and BMW), 4 Japanese firms (Toyota, Honda, Nissan and Subaru), 2 Korean firms (Hyundai and Kia). Being an existing maker is no guarantee of success. Hyundai and VW both failed in their initial forays, and Suzuki, Isuzu, Mitsubishi Motors, and Mazda have exited.¹⁵

Nevertheless, many developing markets required automotive entrants to pair with a local firm, though such strictures typically did not extend to suppliers. In China, joint ventures were mandated, starting in 1984 with the 50:50 tie-up of VW with SAIC in Shanghai. By 2010 the government talked about eliminating that requirement, and finally removed it in 2018. Nevertheless, until 2023Q3, these international JVs dominated China’s passenger car market, as the in-house brands of local producers such as FAW, Dongfeng, SAIC, BAIC and GAC, sold poorly. However, in the last decade private firms without joint venture partners have expanded their share, particularly BYD and the Geely group. From 70% of the market in 2020, joint ventures now only hold 40%. Meanwhile, Jeep, Suzuzi, and Mitsubishi Motors have exited their joint ventures, while Renault and Peugeot have minimal production. In the other direction, BMW now has a controlling interest in Brilliance, its joint venture partner, and VW has 100% ownership of its new EV facility in Anhui Province. Still, it remains an important means of lower the financing requirement and increasing the change of success; and firms are still forming joint ventures to enter the electric vehicle market, including in 2024 Sony and Sharp in Japan.¹⁶

Summary

A new entrant today requires not $28,000, as for Ford Motor Co, but at least $2 billion. Still, 4 new EV producers entered the market so far in 2024, including the cell phone maker Xiomi and (through a joint venture) Baidu, China’s dominant search engine provider.

Making money remains hard, and exit will be the fate of most new entrants, past and present. Of the 3,000+ would-be entrants in the US in the early 20th century, only 700 reached the production stage. Out of those, only 3 are still in business. Among EV entrants in China that were still in business in 2020 some 19 have zero sales, even if they have not formally exited, and another 9 had July-August sales under 1,000 vehicles.¹⁷

Note I will make a few changes.

On Sun 13 Oct 2024: I added a list of Chinese auto firms that produced at some point during 2020-2024 but sold no cars

1. Entry into truck manufacturing was even more prolific into the 1990s, as every prefecture and large city setting up a truck plant. A key enabler was the ready availability of Mitsubishi Fuso engines made under license in China, so these 300 or so small shops were able to purchase the most technically demanding component. (None of the sources I’ve read to date mention transmissions or axles.) In North America, there were and still are independent suppliers for all of these – Cummins for engines, Dana for axles, and Eaton for transmissions (to give only one example of each component).

2. Many more ventures never entered production, but are almost impossible to count as records disappeared with their collapse. Actually selling a car generated an enduring paper trail, and often documentation on founders and key employees survived. Wikipedia lists over 1,000 firms, but business historians stick to the narrower list of firms that survived to selling a vehicle.

The automotive case is a key part of the business history and management literature on population ecology, which looks at how an initial host of firms leads to a winnowing of the industry as successful business models and managers emerge from the fray. Steven Klepper was a key contributor to this literature, with Glenn Carroll and Michael Hannan key figures in the wider literature on other industries.

3. This table is from Chapter 2: “History 1: The Rise of Oligopoly”, pg. 9 in Michael Smitka and Peter Warrian (2017), A Profile of the Global Auto Industry: Technology and Dynamics. Business Expert Press.

4. The database I used to automate counts only covered articles through end-1908. I’ve spot-checked scattered subsequent years, and obviously read the 1909 special issue.

5. By that point Ford had already bought out a number of other shareholders, some of whom wanted Ford to focus on higher-margin vehicles. Disagreements among initial investors in the predecessor to FMC led Henry to quit almost immediately. FMC was Ford’s 3rd venture.

6. Unfortunately, the brothers died during the 1918 influenza epidemic, one directly of the flu and the other of illness-exacerbated cirrhosis. Their carmaking operation was combined with Maxwell to form Chrysler in 1925.

7. I thank Robert Barr, president of the Society of Automotive Historians, for directing me to this article.

8. See the published version, Lawrence H Seltzer, A Financial History of the American Automobile Industry, Houghton Mifflin, 1928. Pg 38f is on United States Motor. He also details the early investors in Ford, and notes in passing the many interconnections among prominent managers (mostly engineers), investors and firms in Michigan in the early 1900s.

9. I have many anecdotes from Bob Haddock, the then-head of European operations, a high school friend of my dad who by chance ended up a neighbor of my parents 5 decades later in the same small town in northern Michigan.

10. See Kachman, Nicholas. 2015. GM: Paint It Red. Buena Vista, VA: Mariner Publishing. Putting it crudely, painting the parking lot isn’t acceptable. Keeping the residual paint from doing so is expensive, and overspray has to be collected and conveyed to a landfill. Of course, for the last 50 years venting VOCs, the volatile organic compounds that are part of paint, into the air has also been unacceptable. Kachman documents how GM wasted at least $19 billion in the Roger Smith era in an unsuccessfully effort to use new painting methods to avoid EPA penalties. Those new technologies didn’t work, and GM was left with plants that never opened.

11. In summers 1972 and 1973 I worked in Chrysler’s Mack Avenue Stamping Plant, which started out as a Briggs factory. See my recollections in “Mack Stamping or a Tale of Detroit’s Decline” at a 2013 post on my old blog.

12. David Hounshell documents this evolution in the late 1940s and 1950s, and the inflexibilities it created. As a result, 80 years later it remains much more expensive to develop a new engine than to develop a new car model. As a result, engines – and particularly engine blocks – can remain in production for over 2 decades. See for example “Ford Automates: Technology and Organization in Theory and Practice.” Business and Economic History. Fall 1995, 24:1, 59-71.

13. That was the issue that forced the restructuring of Toyota in 1950, with banks taking control from the Toyoda family. As part of the deal, Toyota was forced to spin off its sales arm into a separate company, which thereafter increased its reliance on outside dealers, and reduced its direct inventory. When passenger car sales inside Japan expanded in the 1970s and 1980s, high real estate prices left the urban dealerships of Toyota and its competitors unable to fund expansion. In response, car companies invested in these dealers, particularly in Tokyo. I suspect the same challenge afflicts dealerships in China, but have not (yet!) tried to keep track of the news reports I come across.

14. Old factories generally need a new paint shop while the body shop and assembly line will have to be rebuilt. So while they have land, a building, utilities, and foundations appropriate for heavy equipment, they likely need $500 million in renovations. Examples include Tesla in Freemont CA, a GM plant that was sold to NUMMI, the short-lived GM-Toyota joint venture, and then became a wholly-owned Toyota plant. Rivian uses the old Mitsubishi Motors plant in Normal IL, which started as Diamond-Star, the short-lived joint venture between Chrysler and Mitsubishi Motors.

In China, NIO initially contracted out production to JAC in Anhui Province, which had excess capacity when sales of its own cars collapsed. However, Magna has never been able to expand its contract production business beyond a few low-production vehicles assembled at Magna-Steyr in Austria.

15. OK, the initial Hyundai venture and Suzuki were both in Canada, not the US. Mazda has re-entered with a plant in Mexico. Of course there are de novo entrants such as Tesla.

16. Neither venture is at the production stage.

17. Chinese exits and weak players. These include firms such as Kai and Karry that sell vans (CNHD, trucks) but have yet to succeed with passenger vehicles, and Polestar, a Swedish-based publicly traded subsidiary of Geely that is simply doing very poorly. 32 firms, 0 (no) sales in Sept 2024: Aiways, Borgward, Dorcen, DS, Fujian, Guojin, Haima, Hanteng, Jeep, Lifan, Lingtu, MMC, Qoros, Southeast, Suzuki, Weltmeister, Yantai, Yema, Young, Yujie, Yundu, Yunque, Zedriv, and Zotye. 7 firms, sales under 1,000 in Sept 2024: Jiangling, SWM, Polestar, CNHD, Letin, HiPhi, and Karry; 11 firms, sales 1,000 to 4,999 in Sept 2024: PSA, JAC, JLR, Renault, Avatr, Kai, Brilliance, Zhido, Dayun, Jimai, and Skyworth.