The North American International Auto Show is one of the biggest events in the car industry, but at this year’s gathering in January, reporters talked less about the new models on display and more about how gruesome the Detroit economy might become. The city’s foreclosure rate hovers at three times the national average, and upscale properties are being abandoned. (Some residents joke that they track the housing market by searching on Google for “Grosse Pointe foreclosures.”) Detroit has the highest crime rate and the highest poverty rate of any city in the U.S., and unemployment is soaring.

The mayor was recently indicted for lying under oath about an affair with an aide. Even at the auto show, events didn’t come off as planned. Chrysler paraded 120 longhorn cattle in front of the convention center to promote its new Dodge Ram pickup truck, and some of the confused animals started mounting each other. One onlooker dubbed the scene Brokeback Pickup. (View an interactive feature calling out underperforming makes and models.)

Both Ford and G.M. celebrate significant anniversaries this year (the centennial of the Model T and the founding of G.M., respectively), but it’s easy to envision executives at the two companies crying in their champagne. U.S. car sales peaked in 2000 with a record 17.4 million vehicles, and G.M., Ford, and Chrysler had two-thirds of the market. This year’s sales will total less than 15 million cars, the lowest number since 1994, says automotive research firm J.D. Power and Associates. Since 2000 alone, the three Detroit companies have shed 269,440 employees, nearly a third of their combined total, and they’re still issuing pink slips in bulk. In the past three years, they have racked up more
than $67 billion in losses, a number that would be larger if Chrysler were ­included for all of 2007. (Since Chrysler was bought by private equity powerhouse Cerberus last summer, its financials are no longer public.) The automotive press used to refer to the U.S. carmakers as the Big Three, but once Toyota surpassed Ford and Chrysler domestically, a new name was needed. The Smaller Three? The Shrinking Three? They’re now simply called the Detroit Three.

How much worse can it get? Try the Dwindling Duo, in which Chrysler is sold to a foreign company—yet again—and smaller versions of G.M. and Ford remain. Or, less likely, the Sole Survivor, in which those two merge out of desperation. The market share the Detroit Three have lost since 2000 is equivalent to a Ford-size company being wiped out. You don’t need higher math to see that such losses aren’t sustainable for long. Another decade like this and the American auto industry could disappear entirely, its factories razed and turned into a lot where people can park all their European- and Asian-made cars.

Yet there is a scenario in which the Detroit companies—at least Ford and G.M.—can emerge somewhat smaller but far stronger. Their size and costs would be based on current realities instead of on pining for the good old days. And their cars would actually be products that people want to buy instead of merely settle for. I’ve been covering the car business for 23 years and have seen corporate crises, ill-conceived acquisitions, boardroom revolts, C.E.O. sackings, and more. Through it all, Detroit’s cycles have been biblical: Prosper, go astray, repent, recover. It’s repentance time now, and there are concrete reasons to believe that the Detroit Three will recover.

One reason for optimism is history, which is filled with spectacular automotive comebacks. Italy’s Fiat posted record profits in 2007 and wiped out its debt after nearly going under a few years ago. Nissan did the same thing in 2001. The mother of all turnarounds was led by Lee Iacocca at Chrysler in the early 1980s. “Crisis is opportunity,” Ford C.E.O. Alan Mulally says, adding that tough times allow for measures that would otherwise be unthinkable.

For example, the new contracts all three U.S. manufacturers signed last fall with the United Auto Workers union allow new hourly employees to be hired for about half the wages and far fewer benefits than those paid to assembly-line veterans, a deal that would have been soundly rejected by the mighty U.A.W. of the past. (When the boat is taking on water, you stop bickering and start bailing.) Detroit also cut billions in retiree medical bills by funding a new, union-run health-care trust for retirees at some 60 cents on the dollar. So instead of selling cars at fire-sale prices just to pay retiree health costs, the Detroit companies can start making rational decisions about production and sales. And they can start making profits on small and midsize cars as well.

Another positive: Both G.M. and Ford have thriving international operations. Last year, Ford doubled its income in both South America and Europe and turned profitable in Asia. G.M.’s sales are surging in Russia and China. Those two companies also have substantial cash reserves. At the end of 2007, G.M. had more than $27 billion, and Ford had nearly $35 billion. Those stockpiles will shrink this year, but they’re still big enough to buy the companies time to execute their recovery plans and pray for the U.S. economy to turn around. (Chrysler has a smaller international operation and just $9 billion in cash.)

All three companies have, to varying degrees, some of the same basic problems: outdated designs, too many dealers, too much debt, and a late start on hybrids and other alternative-fuel models. But the road to survival, and the odds, are different for each firm. For the past five months, I’ve interviewed dozens of analysts, industry watchers, and executives at the Detroit Three, as well as some of their foreign competitors. Fixing any of the three automakers will be difficult and painful, but it is possible. Here’s how.

FORD: Needs Brand Rehab

The good news is that Ford has lots of cash, and its international operations are profitable and growing (60 percent of its car and truck sales now occur overseas). The bad news is that it hasn’t designed many exciting new cars for the U.S. market lately. That’s a significant issue for, you know, a car company. Because of its lackluster lineup, Ford has lost more market share in this decade—eight percentage points, or one-third of its total—than the other members of the Detroit Three, leaving it with just 16 percent of U.S. car sales.

The decline is partly due to extremely high management turnover in the past 10 years. Ford has had three C.E.O.’s since 2000. Of the 50 top executives listed in its 2002 annual report, just 11 remain today. The result has been zigzag decisions, like dropping certain model names in 2006 and replacing them with alphabetic codes like MKZ and MKX. Ford compounded the confusion by declaring that MKZ should be pronounced Mark Z, only to revert, Prince-like, to M-K-Z in what the New York Times termed “the first recall of car pronunciation.” “I wasn’t here then,” says C.E.O. Mulally. “It’s beyond me.”

Mulally, a former senior Boeing executive who arrived 22 months ago with no industry experience, did manage to save the Taurus name, which Ford management also wanted to scrap. And he nixed Ford’s plan to focus almost exclusively on trucks for the U.S. market, an approach that brought the company to the brink of disaster after gas prices soared three years ago. One thing he hasn’t changed is Ford’s penchant for management jargon. Company chiefs now hold weekly B.P.R.’s (or business-planning reviews), in which problems can be referred to an S.A.R. (special-attention review), where attendees discuss things like a car’s H.M.I. (human-machine interface, basically the vehicle’s knobs and buttons). All of this will be forgivable and not just silly if it results in cars that customers actually want to buy.

Step one in Mulally’s plan is to reduce the number of factories, workers, and dealers in North America. Ford has already laid off roughly a third of its employees in this decade and has closed seven plants in the past two years. The company has also taken steps to repair its reputation for producing low-quality cars and trucks. In November 2006, executives found last-minute problems with the Edge crossover vehicles they were about to release to dealers, and delivery was halted so the problems could be fixed. Contrast that to prior releases, most notoriously the Escape and Focus models. Both were sent to dealers even though they were riddled with flaws, and both were later panned in quality surveys.

An early sign of progress is the Fusion midsize sedan. The Fusion isn’t exciting, but it got a big credibility boost last year when Consumer Reports named it one of the most reliable cars on the market. The magazine’s annual reliability survey showed that 41 out of 44 Ford, Lincoln, and Mercury models scored average or better on quality. Also coming this year are the Ford Flex—a seven-passenger “crossover utility vehicle”—and a new Lincoln MKS (yes, MKS), which will be the company’s first luxury sedan in years. Finally, the company is learning to develop cars in one region and sell them around the globe, eliminating billions in redundant engineering costs. The Ford Fiesta was developed in Europe, but virtually identical versions will be released in Asia next year and in the U.S. in 2010.

“The key is going through this tough time and coming out with a cost structure and a model lineup that’s competitive,” Mulally says. After agreeing to sell Jaguar and Land Rover to India’s Tata Motors, the company retains a manageable stable of four brands: Ford, Lincoln, Mercury, and Volvo. You get the sense, talking to Ford executives, that they wouldn’t mind letting Mercury die with dignity, and they recently considered selling Volvo. But they’re committed to reviving the company’s core Ford and Lincoln lines.

In my view, Ford has the cleanest, most achievable recovery plan in Detroit. The firm isn’t saddled with the complications that plague G.M., which is why Ford’s market cap was 30 percent bigger in mid-April, even though G.M. sells far more cars. Ford even posted a surprise profit of $100 million for the first quarter of this year. That said, there are further steps the company could take. I’d bring back the Continental name and cooperate more closely with Mazda, which is one-third owned by Ford. Mazda has a solid lineup of cars, yet because of its weak dealer network, it has a tiny market share, just 1.8 percent of U.S. sales. Bringing the two companies closer together would give Mazda broader U.S. distribution and possibly inspire Ford to put a little zing into its models.

Mulally, who grew up in Kansas, has a tendency to punctuate his speech with expressions like “aaabsolutely” and “way cool,” but behind the hokeyness is someone who has convinced Wall Street that he can finally turn Ford around. He’s the only C.E.O. in Detroit to state publicly that his company will become consistently profitable again next year, and I think he’ll get there.

CHRYSLER: Needs More Foreign Revenue

When C.E.O. Bob Nardelli came to Chrysler last ­summer, he brought a reputation for executive arrogance. During his stint as head of Home Depot, he produced record profits but was disdainful of such niceties as investor relations. He famously told the board of directors to skip the company’s 2006 annual meeting, which he ended curtly after just 30 minutes. Seven months later, facing a shareholder revolt, he further infuriated investors by walking out with a diamond-studded, $210 million exit package.

At privately owned Chrysler, Nardelli no longer has to concern himself with annual meetings or public shareholders, and he can make hard decisions quickly. “As the first major automaker to be privately operated in more than 50 years, we have an ideal structure for the work we have to do,” he says via email while traveling in Asia. Before Cerberus bought 80.1 percent of Chrysler last summer, Chrysler had spent a disastrous decade as a division of Germany’s Daimler, which retains the other 19.9 percent. But Nardelli has moved fast—scaling back production, eliminating 10,000 jobs (on top of 13,000 layoffs before he showed up), and killing such slow-selling models as the Dodge Magnum and the Chrysler Crossfire, Pacifica, and PT Cruiser convertible. “We took 100,000 units out of our fourth-quarter production, and we did it in a seven-minute phone call,” says Jim Press, vice chairman and president.

Of course, Nardelli’s take-no-prisoners attitude will invariably alienate people, and Chrysler has already lost some management talent. Wolfgang Bernhard, a highly regarded executive from Germany who was C.O.O. of Chrysler during its brief revival four years ago, had been slated to return as chairman but walked out when he learned that he would be working with Nardelli. And in March, veteran Chrysler engineer Mike Donoughe also left, under even less pleasant circumstances. Donoughe ran a critical program to develop midsize sedans to replace the Chrysler Sebring and Dodge Avenger, both of which flopped with customers. At a meeting to discuss the program, says a person familiar with the events, Donoughe disagreed with a consultant that Nardelli had brought in. “If you listen to everything he says, you don’t need me,” Donoughe said to Nardelli.

"Fine," Nardelli shot back and then had Donoughe escorted from the building. (Donoughe declined to discuss the matter.) The company issued a statement saying, “Chrysler denies that the departure had anything to do with a clash with management.”

At the same time, Nardelli—who, like Mulally, did not have any prior automotive experience—has brought in other executives from G.M., Ford, Nissan, and Lexus. “It’s an auto-industry melting pot,” says Tom LaSorda, Chrysler president and vice chairman. An eight-year veteran who ran Chrysler under Daimler and who now oversees manufacturing, LaSorda is known around the hallways as Mr. Supply, while Jim Press, who came to Chrysler after 37 years at Toyota and is responsible for sales, marketing, and dealer relations, is known as Mr. Demand. And, of course, Nardelli has poached people from General Electric, where he made his reputation. All of these executives will make history, and millions, if they can lead Chrysler back to profitability.

Their plan calls for a complete overhaul of the company: shrinking it to a more realistic size, reinvigorating an out-of-date product lineup, and consolidating dealerships by putting all three brands—Chrysler, Dodge, and Jeep—under a single roof. Ford and G.M. face similar issues, but Chrysler is starting later and with less strength. Most significantly, it needs to boost overseas sales. The company sells just 10 percent of its vehicles outside of North America, the lowest of the Detroit Three by a wide margin. Building international operations the traditional way would require time and money that neither Chrysler nor Cerberus has, so LaSorda is striking deals with foreign companies. In one such arrangement, Nissan is producing a subcompact for Chrysler to sell in South America. And Chery Automobile, based in China, will make a small car for Chrysler to sell in international markets this year.

Chrysler’s deal with Chery will also let it bring to the U.S. the first Chinese-manufactured car, a fuel-efficient subcompact that looks a little like a Chevrolet Cobalt. Its name hasn’t been determined yet, but the model will address a deficiency in a lineup that is still—still!—centered around fuel-thirsty vehicles. Chrysler’s major in-house launches this year are the new Dodge Challenger retro muscle car (13 miles per gallon in city driving) and the Dodge Journey crossover vehicle, which gets just 19 m.p.g. in the city, 16 m.p.g. with the all-wheel-drive V6. Later this year, Chrysler will finally unveil its first hybrids: the Dodge Durango and Chrysler Aspen king-size S.U.V.’s.

Which brings up another issue: The Durango and the Aspen are virtually identical vehicles sold in separate dealerships, a costly system that the company wants to discontinue. Chrysler has about 3,500 dealers right now, a number Press and Nardelli would like to reduce—they won’t give a target figure, but 20 percent is probably realistic—by selling the Chrysler, Dodge, and Jeep brands through unified dealerships with no overlapping products. The premium cars would carry Chrysler badges, mid-market models and pickup trucks would be Dodges, and Jeep would be, well, Jeep, an iconic brand with global appeal and one of Chrysler’s most valuable assets. Of course, the dealers—who are akin to franchisees—are likely to push back and seek hefty buyouts from the company. But Press says there’s no room in the budget for such arrangements. “We aren’t cutting dealers, the market is,” he says. “We’re opening a dialogue so they’ll understand the same reality that we understand.” Rather than offer straight buyouts, Chrysler will encourage weaker dealers to sell to stronger ones.

At the New York International Auto Show in March, company officials said Chrysler is exceeding its financial targets so far. (Though again, because it’s a private company, who knows?) With about $9 billion in cash reserves, Chrysler can wait a year or two for Nardelli’s changes to take effect but not much longer. As he puts it, “We knew a correction in the economy was coming over the next few years, but no one expected it to happen in just six months. So the rate of change internally has to be faster, or we’re just backing up.” Chrysler needs so much work on so many fronts, but rejuvenating the dated product lineup has to be Nardelli’s top priority. To that end, it wouldn’t hurt to reach out to Mike Donoughe, the recently departed senior engineer, and ask him to return. Such a move might help Nardelli shake his arrogant image and boost the company’s efforts to come up with better cars. And he should kill off the Chrysler line, leaving just Dodge and Jeep. Nardelli wants Chrysler to be the company’s luxury line of cars, but no one has thought of the brand that way in a long, long time.

The bottom line? Within a few years, Chrysler will probably wind up right where it was when Cerberus came along, as a subsidiary of a foreign car company. (Cerberus’ other possible exit strategy, an initial public offering, seems far less likely, as Chrysler isn’t strong enough to survive independently.) The company’s partnerships with Chery and Nissan might be, in Detroit’s parlance, test drives, as either of those two manufacturers would be on the short list of likely buyers. For Chery, Chrysler could provide an instant U.S. presence, including well-known brands and a distribution network. As for Nissan, Chrysler’s strengths—Jeep, minivans, and pickup trucks—complement Nissan’s weaknesses. (Chrysler will soon start manufacturing Nissan’s trucks.) Carlos Ghosn, C.E.O. of both Nissan and Renault, has publicly stated that he wants a North American company to add to those two.

The good news for Nardelli is that he doesn’t have to be the second coming of Iacocca. He just has to be successful enough at cleaning house and streamlining Chrysler’s lineup for Cerberus to make money on its investment. Cerberus paid only $7.4 billion for the stake, compared with the $36 billion that Daimler paid a decade ago. In other words, Cerberus doesn’t need heroics from Nardelli. And it’s not likely to get them.

General Motors: Needs to Shed Brands

G.M. is headquartered in Detroit’s fortresslike Renaissance Center, an early 1970s structure that, ironically enough, was built by Henry Ford II. The goal was to reinvigorate the downtown area, but instead the building became a microcosm of the city’s problems. It was passed from one owner to the next—one of whom defaulted on its mortgage payments—before G.M. bought it at a fire-sale price in 1996.

There’s been no renaissance in G.M.’s stock price lately. Since C.E.O. Rick Wagoner took over in 2000, shares have dropped about 70 percent. A rally last fall was quashed when G.M. announced it was writing off $38 billion in tax credits, an implicit admission that the company didn’t expect to post a profit anytime soon. “I think it caught people by surprise,” Wagoner told the trade publication Ward’s AutoWorld, in perhaps the understatement of the year. After the announcement, G.M.’s stock slid from $43 to around $26. (Wagoner and other G.M. executives declined to be interviewed for this article.)

Even before the announcement, G.M.’s financial statements were loaded with a motley mix of special items. In the past three years, the company paid Fiat $2 billion to exit a disastrous partnership; incurred charges of $7.5 billion in the bankruptcy proceedings of Delphi, its former parts subsidiary; and took a succession of restructuring charges. And Delphi’s troubles aren’t over. Spun off from G.M. in 1999, it remains the maker’s largest supplier of auto components, manufacturing everything from sound systems to suspensions, but it’s still trying to extricate itself from a bankruptcy restructuring that has dragged on for more than two and a half years. The latest financing plan fell through in April, thanks to the credit crunch, and G.M. likely will have to put more money into the rescue effort.

The carmaker’s second lingering issue is GMAC, its financing arm. GMAC was once a major contributor to the company’s profits, but G.M. had to sell a 51 percent stake to Cerberus in 2006 to raise cash and prop up GMAC’s credit rating. Some executives argued against the move, though in hindsight it looks well-timed. GMAC has expanded into home mortgages in recent years to offset slumping results in the auto industry, a move that has since produced multibillion-dollar losses. Because of the sale to Cerberus, G.M. is on the hook for only about half of those losses ($2.3 billion last year), with Cerberus picking up the rest. In the auto business, that counts as good news.

G.M.’s biggest problem, however, is its unwieldy collection of eight brands—Cadillac, Chevrolet, Pontiac, Buick, Saturn, Saab, G.M.C., and Hummer. Each product line has its own dealers, its own development costs, and its own marketing budget, meaning that all those resources are diluted to the point of impotence. If you’re shopping for a midpriced sedan, for example, G.M. has six. Buick by itself has two. Toyota, by comparison, has just one—the Camry, which sells nearly as many vehicles each year as all six of G.M.’s offerings combined.

To reduce some of the overlap, Wagoner is trying to merge all the brands into four sales channels. Such consolidation is a tedious process, but cheaper than killing brands outright. When G.M. dropped Oldsmobile a few years ago, the bill totaled more than $500 million. The company still has about 6,700 dealers, a number that is down about 14 percent since 2005, and it’s now helping stronger dealers buy out weaker ones. Some executives privately concede that G.M. needs only three or four U.S. brands: Cadillac, Chevrolet, G.M.C., and perhaps Saturn.

One bright spot is alternative energies, an area in which G.M. is moving much more aggressively than its Detroit rivals. Exhibit A is the Chevy Volt, a battery-powered car now in development that will rely far more on electric power than current hybrids do. You plug it in at night like a cell phone, and it will run about 40 miles on a full charge before a gas-powered generator kicks in. Not scheduled to debut until 2010, the Volt has nonetheless become a public relations hit for a company often criticized for pushing big S.U.V.’s. At least it was until February, when vice chairman Bob Lutz, the cigar-chomping executive in charge of the Volt, told reporters that global warming was “a total crock of shit.”

The blogosphere erupted, and Lutz was forced to offer a clarification, which only muddied things further. “My thoughts on what has or hasn’t been the cause of climate change have nothing to do with the decisions I make to advance the cause of General Motors,” he wrote on G.M.’s Fastlane blog. “The Chevrolet Volt program is occurring under my personal watch because I—and others in senior management—believe in it.” Wagoner, for his part, told reporters that Lutz’s remarks “weren’t coming out of our company” but instead represented the vice chairman’s personal views—after which Automotive News suggested that Lutz wear a red tie when expressing his personal views and a blue one when speaking for the corporation, just so everyone would know.

The sad part is that these problems have overshadowed the real progress that G.M. has made. The company has cut billions from its overhead and promises to cut billions more. It has launched well-received new products like the Cadillac CTS sedan, which competes with the BMW 3 series, and the Chevy Malibu, G.M.’s first credible competitor to the Camry and Honda Accord in decades. And G.M.’s overseas operations are growing fast. Last year, it sold more cars internationally than in North America, a first for the firm. The Buick brand, stunningly unsexy in the U.S., actually has cachet in China. Book a limousine from the Shanghai airport into town, and you’ll likely get a Buick.

As with Chrysler and Ford, G.M. is now faced with an unusual business goal: not to grow but rather to shrink its U.S. operations intelligently until costs are in line with revenue. To that end, here are some prescriptions: Kill Saab and Pontiac today. Merge Saturn with Chevy (the cars aren’t that different anyway). Sell Hummer. Move Buick to China. Given that G.M. once sold one out of every two cars in the U.S., such steps might seem like radical surgery, but that’s what’s needed. The stock is at a 33-year low, and the company’s market share is 23 percent and dropping, a reality that industry insiders are slowly coming to accept. “Can G.M. be okay at 20 percent market share?” asks one former G.M. executive. “With the right size and structure, the answer is yes.”

Those aren’t the grand visions that propelled G.M. during its 77-year stint as the No. 1 seller of cars in the world. But given the other, scarier scenarios it now faces, a more modest definition of success might have to be good enough.