We are exceptionally sympathetic to the plight of General Motors. In researching 2,500 business failures over the past 25 years for our recent book, “Billion-Dollar Lessons,” we rarely came across an industry that faced as many challenges as the auto industry—and that was before the spike in gasoline prices turned car buyers away from GM’s profitable SUVs and the onset of current economic crisis dried up credit and forced potential customers to put their wallets away.

Given the onslaught from so many fronts, it’s hard to see what the right answer is for GM. It is, however, easier to identify wrong answers, and our research suggests strongly that acquiring Chrysler would be a disaster.

When an industry consolidates, as the auto industry is, there are four main red flags that can signal that problems are coming, and a GM purchase of Chrysler would raise all those flags:

–Acquirers can focus so much on the benefits of what they’re buying that they don’t pay enough attention to the problems. Ames Department Stores, for instance, bought three discount department store chains in the 1980s and 1990s, convinced that they would give Ames national scale and let it stand up to Wal-mart. Instead, problems at the stores being purchased overwhelmed Ames. It filed for bankruptcy protection twice, then liquidated in 2002.

GM, strapped for cash, seems to be mostly focused on getting its hands on the $11 billion of cash that Chrysler holds. That money would certainly help. But GM would be taking on Chrysler’s problems, too, and they are legion. Chrysler has excess manufacturing, too many dealers, a mediocre to poor product mix, too many brands and frightened customers—all the problems that GM itself has, and that GM management has been unable to solve. Cerberus, the hedge fund that took control of Chrysler last year, is full of plenty-smart folks, and they haven’t solved the problems, either, so the odds are that they will persist.

–While mergers focus on economies of scale, there can, in fact, be diseconomies of scale. When US Air purchased Piedmont Airlines in 1986, in an attempt at economies of scale, the airlines had operating profits six to seven percentage points better than the industry average. But the additional size overwhelmed US Air. Its information systems, for instance, often broke down, and armies of secretaries had to manually type thousands of checks on payday. Operating profits fell to 2.6 percentage points below the industry average, resulting in $3 billion of losses in the five years following the Piedmont takeover.

A takeover of Chrysler would bring enormous, new complexity, as GM would have to make sense of the bigger dealer network, additional manufacturing capacity, and a broadened product line. Management would have to spend so much time on integrating Chrysler that it would pretty much stop working on the pressing problems that currently occupy GM executives.

Now, GM says it expects $10 billion in cost savings from a Chrysler takeover, and that kind of money can cover the sins of a lot of diseconomies. But plans for such savings often failed to pan out, in our research—Daimler, for instance, never got the savings it expected after buying Chrysler in 1998. Even when cost savings are possible, they often require investments of cash in the short run and produce savings in the long run—and GM needs a quick hit, in this precarious environment; it can’t afford another drain on cash.

–Companies often assume that, without too much effort, they can hold on to all the customers that the companies had before the takeover. But that’s often not the case. Competitors use a takeover as a time to poach, and customers can be more willing to listen to a new pitch because of the trauma of a change in control. Even Alcatel and Lucent had to fight a price war to prevent defections following their 2006 merger, despite the fact that the complexity of their telecommunications equipment tends to lock in customers.

Toyota, which is already offering zero-percent financing as a way of putting the squeeze on its less-liquid competitors, would surely lead an assault on GM and Chrysler’s combined market share and would dislodge many customers.

–Companies often assume they should be the buyers when an industry consolidates, even though it can be better to be the seller. There are plenty of personal reasons for wanting to be the buyer. The executives at the buyer tend to keep their jobs and may even get a raise, because they’ll be running bigger operations. There’s also an emotional aspect. The seller’s identity may disappear, and that can be hard to stomach, as Yahoo CEO Jerry Yang showed when he passed up an enticing takeover offer from Microsoft early this year.

For investors, though, it often would have made more sense to sell. So, the instinct to sell at Cerberus—which is only about investment and doesn’t have a long history in the car business—may be the rational one.

It may be that there are considerations that would let a GM-Chrysler deal succeed, considerations that we, as outsiders, can’t see. If so, GM still needs to take a hard look at the red flags we raise, and consider whether they might apply, to counteract the tendency to smooth over any potential pitfalls once high-level agreement has been reached on a possible strategy. GM also would need to go through some exercises to make sure it is aware of all the assumptions that would have to hold true for a deal to work and that GM has investigated all possible reasons the deal might fail.

In the end, we’d bet that GM would find that buying Chrysler would be doubling down on a bad hand. In blackjack, you double your bet when you have a good two-card total, such as 10 or 11, not when you have a bad total like 16. Given the current environment, GM and Chrysler are the equivalent of 16.

What should GM management do instead? While we’re not privy to GM internal calculations, we can surmise that the consideration of the Chrysler acquisition means that the situation is desperate and that everything, even a “Hail Mary” pass, is on the table. We think management should take Chrysler off the table, given that buying it is a strategy doomed to fail and is clearly soaking up critical management attention. Instead, management should go back to work on several alternatives. One is to continue to shed assets and survive the recession, hoping the Chevy Volt will turn out to be the game changer that it could well be. Another is to make the case for a government rescue package. (There are even hints that GM is eyeing a rescue package as the way to fund the Chrysler acquisition, but that is politically naïve. How could either political party support a scheme that would entail the loss of thousands more jobs in Michigan and elsewhere?) Another option would be to assume that Chrysler is in a hopeless situation. If so, GM could prepare to poach disaffected Chrysler customers. Finally, there is the option that management is surely dead set against emotionally, which is to concede that they are out of options and should search instead for the best way out for the employees and investors.

If, instead of pursuing alternatives, GM acquires Chrysler, it could earn a prominent spot in our sequel.