Ford is doing better than GM and Chrysler, appearing successful and widely sold in Germany/Europe.

GM cannot unilaterally terminate a franchise contract; by state law the franchisee is generally the only party who can end the contract unless the manufacturer can show cause (e.g., the dealer underperformed).

GM has only two profitable lines, Chevrolet and Cadillac, with the others ranging from a small loss to an enormous loss, yet contract provisions and state laws prevented closing the unprofitable lines.

Franchise contracts included regionally exclusive licenses (no competing same-brand dealer within a certain distance), which made the franchise more valuable and kept franchise fees higher than they otherwise would be.

The original explanation for selling automobiles through franchises—that imperfect capital markets in the early 1900s and 1930s forced manufacturers to raise capital via franchise fees from locally-collateralized dealers—is a plausible but false story, because the practice continued even after auto companies were well capitalized.

In 46 of the main U.S. states it is illegal for automobile companies to vertically integrate by owning their retail dealerships directly; dealers must be locally owned.

State laws prevent manufacturers from closing a dealership without the dealer's permission unless they pay back the entire franchise fee, which forces GM to keep manufacturing unprofitable lines (like Pontiac) because stopping production would trigger massive franchise-fee repayments to dealers nationwide.

The political power in the manufacturer-dealer relationship is the reverse of the intuitive picture: rather than the large corporation dominating small dealers, local franchise dealer associations hold disproportionate political clout (especially via contributions to House members), and bankruptcy is the only way for manufacturers to escape these constraints.

The U.S. auto industry of the 1940s-60s functioned as a cushy cartel: three large firms generating rents, with unions able to extract a share of those profits in the form of higher pay and especially long-term healthcare and pension benefits, while creativity was channeled into sheet-metal design rather than competition.

The 1970s oil price shock harmed U.S. automakers not just by reducing driving but by suddenly making small, high-mileage cars attractive—cars that American firms didn't make because cheap gasoline and abundant land had given Americans no incentive to demand them, while foreign makers (with high gas taxes) had been making them all along.

When an external environment is not competitive, firms develop a bureaucratic, non-nimble corporate culture that leaves them structurally unable to adapt when competition finally arrives—they don't merely need to 'try harder,' they don't know how, because their entire hierarchy, design teams, and structure were built for a non-competitive world.

Detroit executives suffered a psychological insularity in which they assumed everyone lived as they did (large station wagons for country picnics), were mystified that anyone would want a small car, and marketed by asking 'how do you like our cars' rather than 'what kind of car would you like.'

In any competitive market only the marginal buyer and marginal seller are indifferent at the market price; almost every buyer would have paid more and almost every seller would have accepted less, so money is 'left on the table'—but the gap is normally irrelevant because competition forces a single market price and buyers capture consumer surplus.

Haggling is a mechanism for price discrimination: the salesman's goal is to elicit the maximum price the buyer is willing to pay, which sellers can exploit when there is a gap between the buyer's maximum and the seller's minimum; this is why Priceline-style 'name your price' systems can actually disadvantage consumers who reveal their maximum.

Whether a market has a single 'no-haggle' price or negotiated prices depends on two possible explanations: either the transaction costs of negotiating individual prices are so high they swamp the profits from discrimination, or competition forces a single price everywhere except markets like houses and cars.

The persistence of haggling is a puzzle with two competing hypotheses: Munger believes haggling persists because it benefits manufacturers (allowing higher average prices via discrimination), while Roberts speculates it may persist because it produces a lower average price by revealing the day-to-day P-star that firms couldn't accurately set in advance given inventory uncertainty.

GM maintained multiple essentially identical car lines (Pontiac, Buick, Chevrolet, Cadillac) for price discrimination—the same underlying car with different sheet metal and trim sold at different price points—but became increasingly cynical, relying on brand loyalty as quality differences shrank, with the result that consumers defected to Honda rather than down-buying to Chevrolet.

The Reagan-era voluntary import restraints on Japanese cars created a corrupt bargain that benefited both sides: American makers were shielded because there weren't enough imports to compete away their advantage, and Honda's profits increased dramatically because the restraints (backed by Japanese government allocation among its three makers, effectively a non-compete that would have violated the Sherman Act) meant Japanese sellers weren't forced to drive prices down.

As dealers shifted from selling cars (flow) to servicing the large installed stock of existing cars, and as the auto business bundled three separable functions—sales, service, and financing—the only places profit was actually made were financing (GMAC) and service, while manufacturing and sales merely broke even, suggesting service and financing were the less competitive, tied-good segments.

GM faced an unsolvable principal-agent problem with its dealers: poor or predatory local service sours customers not just on the dealer but on the GM brand itself, yet GM could not enforce good service because it was bound by franchisee contracts and state laws—so by the time it recognized the problem (creating Mr. Goodwrench advertising, etc.) it was too late.

The Saturn experiment was GM's attempt to create a foreign-style, no-haggle, friendly car company from scratch in the hope that Saturn's culture would influence GM's broader corporate culture; instead the GM culture metastasized into Saturn, demonstrating that GM knew what it needed to do but was too constrained to change its existing lines.

Much of economics is ex-post storytelling, and there is nothing wrong with that because attempting to construct a coherent story about why an institution exists teaches a great deal about how the world works—even though many unknown facts could later render those stories untenable.

A Hayekian local-knowledge explanation may justify the franchise model: independent franchisees with a residual claim on profits have stronger incentives and better local-market knowledge than salaried employees, so a manufacturer may earn more by 'giving away' profit to a franchisee than by vertically integrating—supported by Richard Smith's 1982 Journal of Law and Economics analysis of profit rates and residual claimancy.

The coexistence of franchising (independent residual claimants) and direct-ownership models with internal incentives (like Walmart, whose local managers and even associates have high autonomy and bonus incentives) shows that each organizational form is the answer to some particular transaction-cost problem, and economic 'just-so stories' that explain one form must also account for why the alternative thrives.

The car rental industry illustrates residual-claimant logic: local branch managers can be among the highest-paid people in the chain because their bonuses are tightly tied to local office success, giving them strong incentives to monitor costs and prevent theft that a salaried employee would lack.

The dealership franchise laws were originally enacted, at least partly in good faith, to reduce the perceived power imbalance between large corporations and small local businesses—the supposed backbone of America—but produced the opposite effect by entrenching dealer power.

The average Chrysler and GM dealer sells a little over one car a day, whereas the average Toyota or Honda dealer sells three or four, indicating a large productivity gap in the dealer network.

It is necessary to close many unprofitable GM and Chrysler dealerships and there is no reason taxpayers should prop them up; bankruptcy is the important mechanism that finally allows this restructuring despite the human cost to dealership owners.