In the end, Pulte Homes came close to breaking even for the fourth quarter of 2009, losing $117 million, or $0.31 per share compared with the $0.19 per share analysts had expected. But consider the balancing act it took to come close to equilibrium.

It took an $800 million tax refund made possible by a recent tax law change to help cancel out $926 million worth of charges, including a $563 million decrease in goodwill triggered by the company's stock price fall between the Pulte's August merger with Centex and an 0ct. 31 testing date.

"We fully appreciate that you still have to count all those costs," CEO Richard Dugas told analysts during the company's Tuesday morning conference call. But, negating those two extraordinary, one-off events, "We have significantly moved the needle on profitability."

Dugas stopped short on promising profitability for 2010 but said, "We are much closer to being profitable than many expected."

An example is the company's new orders, which are up 32% combined for both Centex and Pulte together over the past year.

Gross margins, before the impact of impairments, land-related charges, and expenses related to the merger, were another bright spot at 14.2%.

Dugas said the merger is already on track to save the company $440 million by the end of 2010 and is expected to generate $350 million in annualized savings by the end of the first quarter of 2010. The company's SG&A expenses are at $188 million combined, down $17 million from what Pulte's alone were, Dugas said.

Dugas also announced that the company will soon begin to use its new post-merger name, Pulte Group, in honor of the 60th anniversary of Bill Pulte building the company's first home himself.

In two ways Pulte's strategy has diverged from that of the larger production home builder universe in the past couple of years. It has chosen to hold onto more of its land holdings than other builders, who have generally rushed to sell to generate cash and tax write-offs, and it has held off on building speculative homes to capitalize on the tax credit-fueled new-home buyer market.

While Pulte did sell off some "non-core assets" at the end of 2009, generating $90 million in revenue as well as a loss of $98 million on the sales, in general the company has held onto its land, declaring it is well located and would be difficult to buy back later at its carrying costs.

Dugas confirmed the company's land-hold strategy Tuesday, saying that the flood of cheap A-located lots that some predicted would be available to buy post-downturn have not appeared.

"A location lots remain scarce," he said. "The company's ample lot position puts Pulte Homes in a very strong position to manage through the downturn."

If the company needs liquidity, it has enough extra land to sell off to other builders, Dugas added. Though, with $1.9 million in cash on hand, that need doesn't seem to be likely in the near term.

Pulte has also resisted the temptation to build more speculative homes to have ready for buyers who have a deadline of June 30 to close on a new home to collect a tax rebate. Instead, the company has adopted a tenet from Centex of favoring selling homes before starting them, which typically generates broader margins.

"We are firm in our conviction that the large spec buildup only leads to large discounts in the future," Dugas said, adding that the company learns from its mistakes.

"We got caught with a lot of inventory in a bad time in the market, and it depressed our margins dramatically, and we are not going back there, God willing," he said.

He doesn't think the focus on pre-sales has hurt the company's sales at all.

"Thirty-two percent growth over pro forma numbers shows that we didn't give up a helluva lot in volume," he said.