The world is too loud. Read what matters.

Cautionary Tales with Tim Harford

The DeLorean Case: Industrial Policy Should Never Bet on a Single Firm

DeLorean built its factory on roughly $100 million in British government support, then slid into an $18 million fraud and an FBI cocaine sting. The real lesson of industrial policy is not whether to subsidize, but to keep several firms competing, use export markets as discipline, and never bet on one company.

Auto IndustryIndustrial PolicyGovernment SubsidiesEconomies of ScaleCorporate FraudDeLorean

The video won't play here. Listen to the audio instead:

The founder gossip is only the shell. The substance is how economies of scale raise the barrier to building cars, why subsidies get captured by a single firm, and why China's many-automakers-at-once approach did not repeat DeLorean's fate.

The argument · tap a timestamp to hear it

7:36

A big company's rulebook cannot stop innovation smuggled in as an option

General Motors had an explicit rule: no big engines in mid-size cars, because they would be too fast, too much like race cars, and would invite antitrust attention. The loophole DeLorean found was that the rule only applied to new models. Offer the big engine as an option on an existing model, and nobody had standing to veto it. So in 1964, you could pay $300 to fit a V8 racing engine into a Pontiac Tempest, a grandmother's car. He borrowed Ferrari's Gran Turismo Omologato to name the option package GTO, and it came to be called the first great American muscle car — while carrying his own career all the way up to vice president of General Motors. The story shows that however many rules a big company writes, it cannot stop innovation smuggled in as an option on a product that already exists.

17:50

A new carmaker can only start with a $100,000 sports car

Making cars has powerful economies of scale: design, testing and building the plant cost astronomical sums, so the first car is terrifyingly expensive and only then does each subsequent one get cheaper. That makes it extremely hard for new brands to enter. The United States had gone roughly fifty years without a successful new carmaker (the last one was founded by another former GM executive, Walter Chrysler). DeLorean's answer was to sell an expensive sports car, so the scale threshold could be cleared without spreading costs over too many units — and the distinctive gull-wing doors were the justification for the premium. This is exactly the same move Elon Musk later used: Tesla's first car was a $100,000 sports car, take deposits first, then build the cheap car for volume. Same strategy, different ending; the problem lay in what came after.

22:57

Governments subsidizing carmakers are buying supply chains, not jobs

Why are governments willing to throw money at car companies? On the surface it is jobs: manufacturing work is stable, does not require higher education, and supports a whole field of suppliers alongside it. One layer deeper it is indirect support: the story of Chinese electric vehicles is not only about cars but about the battery technology and rare earth supply chains riding on their back — though the host also admits he cannot explain exactly what rare earths are — and those same materials are the foundation of drones, missiles and high-end electronics. So supporting a car industry means indirectly maintaining an entire strategic supply chain. That was the premise on which DeLorean could ask for money, and it is why countries remain generous to carmakers to this day.

29:03

Once the money is in, the company is the one squeezing the government

Building the plant in Northern Ireland was far harder than imagined: barbed wire, body searches, schedule slippage. In 1980 DeLorean came back for more money, and the government's audit of the books found the American office had overspent by several million; internally, the recommendation was to cut. DeLorean's reaction: $2.5 million is peanuts, don't try to nickel-and-dime me; I don't run a penny-pinching company, and if the government is unhappy I'll shut the doors next Friday. The government gave him more money anyway, mostly in loan guarantees. Why? The money was already spent, and thousands of jobs had been promised to Northern Ireland — sunk costs left the government unable to walk away. When industrial policy bets on a single firm, the firm ends up holding the lever over the government.

36:22

What turned him from a business failure into a criminal case was not the cocaine

What pushed DeLorean from business failure into criminal scandal was not the bag of cocaine in a hotel room but GPD Services. The company was registered in Panama, yet the cheques went into post box number 33 in Geneva. DeLorean transferred roughly $18 million to it in total. Investigators found that GPD's founder was Lotus's boss, Colin Chapman — Lotus had genuinely done design work for DeLorean, but the design fees were paid separately, and this money never appeared on Lotus's books. The prosecution's version was that DeLorean pocketed about $8 million of it, using the money to buy a Utah company that made snowcats. Lotus's CFO pleaded guilty to fraud and was sentenced to three years, with the judge calling it fraud that was "bare-faced, outrageous and on a massive scale".

45:35

China's success lies in backing dozens of firms, not one

China is often held up as proof that industrial policy works: the government poured enormous sums into the car industry, and today it has powerful automotive, battery and solar industries, with Germany's solar industry essentially wiped out by China. But there is one crucial difference between China's approach and the DeLorean case: China backed dozens of carmakers at the same time rather than betting on one big company. Every local region had its own carmaker, and they fought it out domestically first. It was precisely that internal competition that preserved discipline and avoided the situation where one man squeezes the government. The host's inference is that the more sensible course for the British government back then would have been to build another, different carmaker outside Manchester, Birmingham or London and let them compete against each other, rather than building one factory and betting on one man.

46:35

The hard part of industrial policy is not picking right, it is daring to stop

The genuinely hard problem of industrial policy is one Tim sums up using a phrase from his former boss, the economist John Kay: "disciplined pluralism". Capitalism works not because of the textbook's price equilibrium and competitive equilibrium, but because it satisfies two things at once. Pluralism: you never know which idea will work — maybe gull-wing doors work, maybe electric cars work, maybe it isn't a car at all but an electric bicycle — so you leave room for all the possibilities. Discipline: when an idea is shown not to work, you have to be able to say out loud, "this is a bad idea, shut it down". In the DeLorean case the British government lost on both counts: it bet on only one project (no pluralism) and refused to stop because of sunk costs (no discipline). The approach many East Asian economies used during their growth phase was to rely on export markets as the discipline: the government can subsidize, but the product must sell abroad, to someone willing to buy it.

In their own words · checked verbatim

my concern was that there hadn't been an important product innovation in the industry since the automatic transmission and power steering in nineteen forty. Nine in the place of product, innovation the automobile industry went on a two decade marketing binge which genuinely offered up the same old product under the guise of some new and Useful

a two point five million dollars shortfall for THE us is. Peanuts i'm not going to be nickel and dimed by those. Bastards i'm not going to run a nickel shit. Company if they don't like, IT i will close it down Next. Friday

you built a time machine out of A. Dolorean

the thing that works about capitalism is not the kind of all the technical stuff that you read about in the economics textbooks about kind of price equilibration and competitive equilibrium and all that it's a much bigger picture. Thing it delivers something called disciplined. Pluralism

nothing can ever go wrong with a. Monorail

Figures

British government funding given to DeLoreanAbout $100 million (mostly grants, partly loans)23:57
Total the company wired to GPD ServicesAbout $18 million36:22
Amount DeLorean was accused of pocketingAbout $8 million (used to buy a Utah snowcat company)37:25
Price of the V8 engine option on a 1964 Pontiac Tempest$3007:36
Gap from America's last successful new car company to DeLoreanAbout 50 years (founded by Walter Chrysler)16:49
Price of Tesla's first sports car$100,00017:50
Weight of the cocaine in the FBI's room50 to 60 pounds3:28
John DeLorean's heightSix feet four inches2:28

Glossary

gull-wing doors
Doors that open upward rather than pulling outward, resembling a gull spreading its wings; DeLorean's signature design.
entrapment
When law enforcement sets up a scheme to induce a suspect into committing a crime; a common-law defense the accused can raise to claim innocence.
disciplined pluralism
John Kay's phrase: accommodate many business ideas at once, while cutting losses on bad ideas in time.
snowcat
A tracked snow vehicle used to clear and groom ski slopes.

How to listen

Who it's for

Founders and investors who follow industrial policy and manufacturing subsidies, and policy researchers who want to understand the trap governments fall into when they bargain with a single company.

Skip

You can skip the opening arrest scene and the roll call of celebrity investors. Start at the economies-of-scale section; the industrial policy analysis in the back half is where the value is.