By Gerry Donohue.

They called them "poof roll-ups," and for a few years in the mid-1990s they were the rage of Wall Street. Financiers combined or rolled up several private companies in an industry, such as flower shops or limousine services, into a single entity and ? poof ? created a $100-million public company. The Fortress Group was one of those companies. The sales pitch was simple: Through economies of scale and public money, these combined companies would dominate their industries. Initial results seemed to match the promises. Jonathan Ledecky rolled up four office-products companies in 1994 to create U.S. Office Products. Over the next three years, U.S. Office Products used its access to the public markets to acquire more than 250 competitors.

Investors flocked to the concept. According to Montgomery Securities, investors poured nearly $8 billion into roll-ups during the second half of the 1990s. Jeff Manning, of Legg Mason, estimates that banks put in an additional $10 billion to $12 billion. In one instance, investors put more than $500 million into a blind roll-up pool; they didn't know what industry the financier would roll up, but they blindly believed they would make money.

James Martell wanted a piece of the action. A housing industry veteran who had helped take NVR public in 1986, he saw the opportunity to create a roll-up in home building. He envisioned a company that would merge dozens of high-quality home builders in markets all across the country. These builders would have access to public money, national purchasing contracts, top-flight management expertise, and cutting-edge technology that would make them formidable competitors in their local markets.

Equally important, the individual builders would be able to liquefy their investments in their companies at a public multiple, which was several times higher than they could get by selling their companies in the private market.

"The concept behind these companies was compelling," says George Yeonas, who became CEO of The Fortress Group three years after its inception. "But I think there was a certain amount of naïveté about how bloody difficult they really are to create."

In the first three years of its existence, Fortress soared, acquiring companies at a dizzying pace and growing to more than 4,000 closings and nearly $700 million in revenues. Then, reality hit and the company started to unravel. Over the next four years, Fortress steadily sold off assets as management tried to service a crushing debt and build a viable company. In the end, so little of the company was left that the only choice was to sell off the remaining assets and close up shop.

Phase One

In 1995 and early 1996, Martell and his management team crisscrossed the country, talking to about three dozen potential builders. Eventually, they worked their way down to four: Buffington Homes in Austin, Texas; the Genesee Co. in Denver; Sunstar Homes in Raleigh, N.C.; and Christopher Homes in Las Vegas.

In April 1996, Fortress completed its IPO, selling three million shares and $100 million in bonds at 13.75 percent. In a simultaneous transaction, the company acquired the four builders for various combinations of cash and stock and paid off their outstanding debt. The investment banking firm Furman Selz managed the IPO.

Several signs at the outset should have alerted Martell's team that even the irrationally exuberant market had reservations about the group.

In its initial prospectus, the company used a stock price of $11 per share for its projections. At three million shares, the company would have raised about $33 million in equity, minus expenses. When Fortress went public, however, the price was $9 per share, bringing in $27 million, minus expenses, a drop of almost 20 percent.

More damaging were the senior bonds. In its initial prospectus, Fortress proposed that its long-term bonds would mature in 2006 and would be redeemable at any point after 2001. At the time, the company didn't specify the interest rate. When the company went public, however, the bonds were noncallable and matured in 2003. From day one, the bonds were like a ticking time bomb in the executive office. Even worse, the bonds were priced at a crippling 13.75 percent. One industry insider heard that Furman Selz changed the terms of the deal at the last moment because it couldn't get them to fly. Faced with the choice of accepting the lower stock price and high bond rate or walking, Martell went ahead.

Fortress Financials (in millions); Source: Thompson Financial Solutions "Those bonds are the whole Fortress story," says Ed Horne, who owned Wilshire Homes, which was acquired by Fortress in April 1997. Concurrent with Lennar Corp.'s deal to acquire Fortress' assets, announced in June of this year, Horne bought back Wilshire for $23 million. "The [bonds] were a deadweight. We made money everywhere, but we couldn't get out from under the debt," says Horne. After all the expenses of the start-up, Fortress had about $11 million in cash and long-term debt of $100 million. "That nine-to-one debt-to-equity ratio made it very tough to operate effectively," says Yeonas.

Race Against Debt Service

Fortress' debt load wasn't atypical; most roll-ups began their lives heavily leveraged. The accepted strategy was to use company shares as currency to acquire other companies. By rapidly growing its revenue stream through acquisitions, the company could stay ahead of the debt service. Additionally, the growth would produce a compelling story for investors, who would bid up the share prices, so the company could buy even more companies.

Fortress embraced the strategy. By mid-1998, the firm had acquired eight additional companies, including the largest builders in St. Louis and Jacksonville, Fla. For fiscal 1998, the company reported revenues of $693 million, net income of almost $13 million, and 4,016 closings. Just three years earlier, the four original builders had combined for about $100 million in sales, $6 million in net income, and fewer than 1,000 closings.

Investors, however, weren't smitten. The company's stock price declined steadily from the IPO, falling to about $4 at the end of 1997. Several factors contributed to this situation. First, in its first several quarters of existence, Fortress badly missed on its quarterly forecasts.

"Private builders don't think much about quarterly forecasts," Yeonas says. "If you miss a forecast, it's no big deal. But if a public company misses a quarter or two, the Street is very unforgiving."

Second, the company had such a small-market float that it couldn't attract the institutional investors who help create an active market for a company. Additionally, the company was too small to be covered by the brokerage houses, so no one in the market was talking it up.

The third factor was that although the company was growing rapidly, several others were growing more rapidly. Centex, Pulte, D.R. Horton, and Lennar had more compelling stories and better track records.

Fourth, and most important, Fortress wasn't realizing the operational benefits of a roll-up. The gross profit margin, which should have climbed because of economies of scale, fell, from about 17 percent at the IPO to between 14 and 15 percent for most of the company's existence. Net income, which should have climbed because of access to public financing, dropped from about 4 percent to less than 2 percent.

The Builders' Fate

Without investors pushing up the stock price, Fortress looked elsewhere to sustain its growth strategy. The investment bank of Lazard Freres invested $75 million in the company in mid-1997, but it was a stopgap solution. Facing a dead-end business strategy and the crushing bond debt, the board of directors in 1997 recruited Yeonas to be COO. Yeonas had been vice president and general manager of the Arvida Co.'s south Florida division. Then, in July 1998, the board forced out Martell, who went on to try roll-up companies in other industries, and replaced him with Yeonas as president. "It was almost a revolt by the original four builders," says Bob Short, who had owned Genesee. Martell and his team "were financial people," says Short. "They had little concept of how to run [the company] after the roll-up was done."

The long slide toward the end had begun.

Egos and Operations

The original concept of Fortress was for the individual building companies to have a large amount of operational autonomy. The central office would merely be a source of capital. Yeonas quickly decided that this concept was flawed. He determined to integrate all the building companies into one operational entity.

It was "an inordinate challenge," says Yeonas. "We didn't even have a single definition of what constituted a sale."

Short, the only original builder to stay with Fortress for its life-span, agrees: "We had a great deal of diversity in the operations. Each builder had his own tried and true method for doing everything. On top of that, you had some pretty diverse personalities and strong egos."

Nevertheless, Yeonas made significant progress. He consolidated human resources and information technology. For the first time, staff across all the companies could communicate through a central e-mail system. He standardized the reporting process, so everyone counted revenue, profits ? and sales ? the same way. He created councils across Fortress for many different functions. For example, purchasing managers from all the building companies worked together to standardize the purchasing process.

Yeonas' biggest hurdle was to forge a strong corporate culture. This is hard enough to accomplish in a single company; to build a common culture among 13 separate entities is Herculean. Employees at each entity had a natural loyalty to their own building company above all. In the end, the separate loyalties won out.

"They really didn't have a culture," says Lennar's Bob Strudler, who oversaw the acquisition of Fortress. "They had no common bond. They were trying to create it as they grew, but that is almost an impossible task."

The Sell-Off

And in the end, there were the long-term bonds. In the year before they merged to form The Fortress Group, the four builders paid $18 million in debt service, mostly for development and construction loans. Upon creating the company, the builders faced $13.75 million in annual debt service even before they went to the banks to acquire land or build houses. While the company was growing, it could absorb the hit, but once that growth stopped, the debt service became overwhelming.

Yeonas and the board determined that the only way the company could continue to operate was to retire the debt. Beginning in 1999, Fortress started selling off its building companies, using the proceeds to buy back bonds. In some cases, such as in Las Vegas, the operation was losing money. Several operations, however, such as Florida and Carolina, were profitable, and selling them cut deeply into the muscle of the organization. When Fortress sold its Carolina operations to Lennar less than a year ago, it had cut its senior debt to $33 million, but it had only two operating companies left and had lost almost $9 million for the year.

"By the time we sold the Carolina operations, the handwriting was probably on the wall that we were no longer large enough to really be a true public company," says Short.

Fabled Economy

The demise of The Fortress Group isn't a home building story as much as it is a fable about the Bubble Economy of the late 1990s. The company was a product of the belief that all that was necessary to succeed was public money and growth. In every industry, small privately held companies were rolled up together and sold in the public markets. These new companies were hailed as the model for the new millennium. In many cases, however, they failed because of old-millennium realities ? too much debt and a lack of management, integration, and corporate culture.

Fortress' fate is similar to that of many roll-ups. Over the past few years, dozens of 1990s roll-ups have broken up or been acquired. Legg Mason's Manning says he has been an adviser in the sale of 35 roll-ups in the past year and a half.

Among the ones that haven't been acquired, the track record is similarly dismal. Jeffrey Evans, an analyst with Advest, based in Hartford, Conn., completed a study of roll-ups last March. Of the 113 roll-ups that were still operating, 17 had filed for bankruptcy protection and 83 were trading below their offering prices. U.S. Office Products, which was the high-flying model that others emulated, filed for bankruptcy protection in early 2001.

"When it comes to roll-ups," says Manning, "you can practically count the successful ones on one hand."

The Fortress Group History
May 1, 1996 Begins operations with 4 divisions: Sunstar Homes (NC), Buffington Homes (TX), The Genesee Co., (CO, AZ), Christopher Homes (NV)
September 1996 Acquires Landmark Homes
January 1997 Acquires Brookstone Homes
March 1997 Acquires D.W. Hutson Construction
April 1997 Acquires Wilshire Homes
August 1997 Acquires Don Galloway Homes
December 1997 Acquires Iacobucci Organization
January - April 1998 Acquires WestBrook and Whitaker Homes, Quail Construction
July 1998 CEO James Martell resigns
March 1999 George Yeonas becomes CEO
April 1999 Steps up transition to integrating operations
July 2001 Sells St. Louis operation
August 2001 Announces sale of Las Vegas operation
September 2001 Completes restructuring with sale of Christopher Homes
December 2001 Sells Carolina operation
March 2002 Sells Philadelphia operation
June 2002 Announces Lennar-Fortress merger agreement
June 2002 Sells Texas operation
July 2002 Ceases operating
Source: Company Reports

?Gerry Donohue is based in Silver Spring, Md.

Published in BIG BUILDER Magazine, October 2002