Texas Monthly writer-at-large Loren Steffy’s new book, George P. Mitchell: Fracking, Sustainability, and an Unorthodox Quest to Save the Planet (Texas A&M University Press), is the first comprehensive biography of Mitchell, a Galveston native who is best known for two things: developing the Woodlands, the master-planned community north of Houston, and helping create hydraulic fracturing, aka fracking, the method of gas well stimulation that has transformed the global energy industry.
In this exclusive excerpt, Mitchell’s company is in dire straits, a situation that puts tremendous pressure on him to sell his beloved Woodlands—and also leads to the revolutionary development of fracking.
In the late 1980s, Mitchell Energy and Development faced deteriorating financial prospects. The effects of the oil bust early in the decade—one of the most volatile periods in the industry’s history—were still being felt. Interest rates soared, increasing the cost of the company’s debt, and oil and natural gas prices had fallen, which meant it had less cash coming in to pay the higher financing costs. Maintaining gas production in Mitchell Energy’s biggest field, Boonsville Bend, which was located about forty miles northwest of Fort Worth, still required huge amounts of capital, as did the company’s other drilling prospects. George bristled at the growing constraints from the company’s primary lender, Chase Manhattan. When he decided to change banks, David Rockefeller, the bank’s chair and scion of the country’s first oil magnate, flew to Houston to change George’s mind, without success.
Thanks for reading Texas Monthly
Still, there weren’t a lot of options. By 1986, six of the state’s biggest banks were on the hook for $11 billion in outstanding energy loans. As oil prices slid below ten dollars a barrel, many banks refused to lend more money to the industry. Lenders questioned Mitchell Energy’s insatiable thirst for capital and its ability to maintain growth.
George managed to make it through the 1980s without losing money, and he was able to keep the banks at bay, but Mitchell Energy didn’t emerge unscathed. The company had about 3,500 employees in 1983—in both energy and real estate—but by 1988 it had cut its ranks to 2,200 through early retirements, attrition, and layoffs. “It’s a tough thing to cut back that much,” George said. “But we did; we had to. We had to make it work.” As difficult as it was, Mitchell Energy fared better than many other independents, who either were acquired or filed bankruptcy.
Still, there was a major drag on the company’s finances that other energy companies didn’t have to worry about: the continuing costs associated with the Woodlands, the planned community outside of Houston that Mitchell had created in 1974. Since Mitchell Energy had gone public in 1972, George had tried convincing investors, Wall Street analysts, and anyone who would listen that owning oil and gas and real estate made sense, but few could see the benefits. While some oil companies had ventured into real estate as a hedge against oil price fluctuations, Mitchell Energy had diversified in the same geographic region. Its biggest gas field was in North Texas, and most of its real estate was around Houston. Because of the prominence of the oil industry in the Texas economy, real estate values were closely linked to oil prices. That meant that when oil and gas fell, both sides of the business suffered. George often said that the Woodlands would take fifty years to pay off, far longer than most investors would wait for a return. Even within the company, it was becoming clear that for the good of the shareholders, the company needed to split in two.
Despite these long-term concerns, Mitchell Energy reported strong financial results from its oil and gas operations. In 1991, it earned more than it had during the previous five years of the oil slump. Iraq’s invasion of Kuwait drove up prices, pulling land values in Houston skyward along with them. Mitchell Energy had become one of the country’s biggest producers of natural gas liquids, thanks to its processing plant in North Texas. George personally was worth about $500 million, counting his 63 percent stake in Mitchell Energy and his private investments. But Mitchell Energy’s stock value depended almost entirely on the energy side of the business. Its real estate holdings had negative cash flow, and George didn’t expect that would reverse anytime soon. The company was taking profits from the gas business and using them to cover losses in real estate. Despite George’s efforts
to lure good-paying jobs to the Woodlands, the development still lacked enough office space. Houston was working through a commercial building glut from the 1980s, when developers overbuilt, and the city became known for its “see-through” buildings. In the Woodlands, offices rented for about fourteen dollars a square foot, but the company needed seventeen dollars to build a six-story building profitably.
Wall Street began pressuring the company to spin off its property business, so it could plow more profits into expanding its energy operations. George stubbornly refused to address the issue. “I haven’t decided what we should do, if anything,” he said, when asked about splitting the company. Because he controlled two-thirds of the stock, he didn’t worry too much about mollifying investors.
And there was another problem on the horizon that made the situation even more urgent. The company’s new COO, Bill Stevens, was a longtime Exxon executive who had the deep understanding of financial discipline that Mitchell Energy desperately needed. Stevens’s years at Exxon had taught him to look ahead and spot potential trouble. George, of course, thought frequently of the future, but not in the same way. He tended to look decades down the road, and his visions were grand and sweeping. He didn’t concern himself with how his ideas would be achieved. Stevens, by contrast, worried about where problems might crop up in two or three years. One area that concerned him in particular was the contract to supply gas to the Natural Gas Pipeline Co., which ran gas from the Texas Panhandle to Chicago.
Mitchell Energy’s agreement with NGPL had been extended several times, thanks to George’s negotiating skills. The last time he had convinced NGPL to renew its agreement, he had been in Austin for his granddaughter’s birthday party. George entered the house to find it filled with screaming five-year-olds. He asked his son-in-law whether there was quiet place he could make a call. He found an upstairs bathroom, closed the door, sat on the toilet, and renegotiated a deal that would keep his company operating for years to come. Then he calmly went downstairs and asked, “Is there any cake and ice cream left?”
Despite George’s skill as a negotiator, though, he and his executives knew that sooner or later NGPL would balk at paying above-market prices for gas, as it had for years. By the 1990s, NGPL was paying about five dollars per thousand cubic feet, though the market price was less than one dollar. When the contract expired, even if George renegotiated it, Stevens warned, Mitchell Energy could expect to receive less for its gas. And that would be a serious threat to the company’s financial health; the NGPL contract accounted for about one-quarter of Mitchell Energy’s total production.
Like many Wall Street analysts, Stevens could see what George chose to ignore: Real estate was a drag on the energy company. It didn’t matter if the Woodlands paid off in fifty years or just twenty. If the company wanted to survive the next ten, it needed to split in two. The challenge was convincing George to accept the idea.
As the decade progressed, both he and George grew increasingly concerned about the company’s long-term health. In the near term, Mitchell Energy was replacing the oil and gas it produced with new reserves. But Stevens saw that replacing reserves would become increasingly difficult. For publicly traded companies, reserve replacement is a financial treadmill. If the total reserves fall, so does the company’s stock price. Maintaining reserves requires additional capital for new drilling, and Mitchell Energy had no new plays like the Boonsville Bend to save the day. The gas fields in North Texas were still prolific, but the high costs to keep them producing meant that they weren’t the profit center they had once been. Because of its heavy debt burden, Mitchell Energy couldn’t easily borrow more money, and George refused to issue new shares to boost liquidity. How the company should resolve this conundrum created a rift in management’s top ranks that pitted Stevens’s resolve against George’s vision for the future. George saw the company’s salvation in an experiment that had been going on in a corner of the Mitchell Energy office for more than a decade. Stevens thought that idea was pure folly.
George had worried about the company’s crown jewel, Boonsville Bend, as far back as the late 1970s. For twenty-five years, Mitchell Energy had steadily produced gas from the three hundred thousand acre site. But the company’s analysts calculated that the field’s production would soon decline, and eventually it would cease to be profitable.
George didn’t know how to solve the problem, but he reminded his geologists and engineers that the company’s future depended on finding an answer. They already had an idea of where to look. Oil and gas fields aren’t defined just by square miles on the surface; they’re also defined by depth. The Boonsville Bend ran under the North Texas hills at a depth of about 3,500 feet. Below that, there were other geologic layers, some of which, it was thought, might also hold gas. Beginning at a depth of about 6,000 feet, there was a layer of dense, slatelike rock that geologists called the Barnett Shale.
In 1976, the Department of Energy funded the Eastern Gas Shales Project in Appalachia, which demonstrated that natural gas could be extracted from shale formations in the region. If gas could be recovered from those shales, Mitchell thought, maybe it could be drawn out of the Barnett, too. But there were a few problems. Not all shale formations are the same, and the ones in Appalachia differed from the Barnett. Besides, the DOE report concluded that while the gas was in the shale, drilling into it didn’t produce enough to be worth the expense.
Then, in 1981, a paper from one of George’s geologists showed up on his desk. Just as he reviewed the data on each well the company drilled and fretted over street signs and seemingly every tree cut in the Woodlands, he also insisted on reviewing every paper from his employees before they submitted them to industry trade journals. The paper, by a geologist named Jim Henry, indicated that the Barnett Shale had large natural fractures and that it might be possible to extract the gas by widening those fractures. The more George read, the more excited he became. If the Barnett Shale held more gas, it would be as if the company had discovered another field under the one it already had. That would be a lot cheaper than finding a new gas field somewhere else.
The complication was that shale is as solid as a tombstone; getting gas out of it wasn’t going to be easy, and would likely be expensive. But George didn’t want to hear about what wasn’t possible. At one point, he told his shale team: “If you don’t think you’re capable, tell me, and I’ll find people who are.” Dan Steward, the team’s geological coordinator, sensed that George wasn’t bluffing. He thought of his wife and four kids and assured George that the team would find a way to replace the depleting gas fields of the Boonsville Bend.
One of the first things Steward’s team did was review the data from another project Mitchell Energy had been involved in a few years earlier that used a technique known as hydraulic fracturing. Fracturing—or fracking as it would later become commonly known—dates to the 1860s, when liquid nitroglycerin was shot into the hard rock of wells in Pennsylvania, West Virginia, New York, and Kentucky. It was extremely dangerous, but it succeeded in breaking up the rock and releasing oil. In the 1930s, companies tried using nonexplosive liquids like acid under high pressure to break rocks apart. Over the years, napalm and dynamite were tried, too. The Atomic Energy Commission even tried a nuclear bomb. Working with the El Paso Natural Gas Company, it detonated at twenty-nine-kiloton nuclear warhead, dubbed the “Gasbuggy,” four thousand feet underground near Farmington, New Mexico, in late 1967. The explosion left a crater 80 feet wide and 335 feet deep, and the wells drilled into it produced gas from the sandstone underneath. But the gas was radioactive. The commission tried two more times in Colorado, with similar results, before the idea of “nuclear fracking” was abandoned.
Today, George is often erroneously called the “father of fracking.” If fracking has a father, it was an engineer named Floyd Farris, who worked for the Indiana-based Stanolind Oil and Gas Corp. In 1947, Farris used napalm and sand to fracture a well in southwestern Kansas. The process didn’t increase production appreciably, but Stanolind patented it anyway and later licensed the technology to Halliburton, which completed its first two commercial “fracks” in 1949, one in Archer County, Texas, and the other in Stephens County, Oklahoma. By the mid-1950s, more than three thousand wells had been fracked in some way. But no one had tried fracking shales.
As the Mitchell team studied the Barnett, an idea that the rest of the industry thought was ludicrous—that gas could be extracted from shales without natural fractures—began to seem feasible. Over the next few years, Mitchell Energy deepened several dozen Boonsville Bend wells into the Barnett formation, and later tried various fracking techniques. They found gas, but not in commercial quantities. Without the above-market price guaranteed under the NGPL contract, fracking the Barnett didn’t make economic sense.
By the late 1980s, the company was facing another problem: a series of lawsuits from North Texas residents claiming that Mitchell Energy’s work in the Boonsville Bend had contaminated their drinking water was draining the company through legal fees and a $200 million legal judgment (which was eventually overturned). As the stock price sank, analysts renewed their demands that George split the company and cast off the Woodlands. It was a tough choice for George. He loved the energy business, but it had always been, essentially, a means of making money. The Woodlands, on the other hand, was something special: an entire community he had created from his own imagination. Yet faced with the incessant financial pressure on his company, he didn’t have a choice. He could no longer ignore the reality that the company needed to be either a real estate firm or an energy concern. The Woodlands had to be sold.
The sale, to a partnership that included Crescent Real Estate Equities and Morgan Stanley, was completed in July 1997. After taxes, Mitchell Energy booked about $460 million from the sale, far less than it had invested in the property over the previous twenty-five years. In approving the wording of the press release, Mitchell wrote three words across the top of the draft: “OK but sad.”
Splitting the company turned out to be the right move. By the end of 1997, Mitchell Energy shares were trading at more than twenty-nine dollars, 45 percent higher than they had in early 1996. Mitchell Energy—which maintained its offices in the Woodlands—was now what Wall Street had always wanted it to be: a pure energy play. But the biggest challenges for George and the company still lay ahead.
By the time of the Woodlands sale, Mitchell Energy had been drilling into the Barnett Shale for more than fifteen years. It had settled the debate over whether gas could be extracted from shale by fracking—it could. But it was too expensive. Because of the special chemical gels used in cracking open the shale, the fracking process often cost more than the drilling.
The solution to that problem was largely the handiwork of Nick Steinsberger, a Mitchell Energy petroleum engineer who was working on the Barnett Shale wells. Steinberger had graduated from the University of Texas at Austin with a petroleum engineering degree in 1987—right in the teeth of a major oil bust. Eventually, he found one company that was hiring—Mitchell Energy—and went to work in North Texas, where in 1995 he was promoted to lead engineer for the wells being drilled into the Barnett Shale. The company had punched between fifty and seventy holes in the Barnett, but none was producing enough gas to justify the cost.
He drove to his first meeting with managers in the reservoir engineering group and was surprised by their assessment of the Barnett project. “I was told that we would probably drill another thirty or forty wells and give up because the Barnett wasn’t working. It was not economic,” Steinsberger said. “I had just gotten this promotion. It was a bit of a let-down.”
Steinsberger decided that all he could do was reduce drilling costs as much as possible. But the special gels used in fracking pushed the expense of a typical well to $750,000 or $850,000. About 40 percent of that was the frack job itself, and the rest went for a couple million pounds of sand and the high-viscosity gelled fluids that carried it into the fractures and held it in place. Steinsberger focused his savings program on the priciest of the ingredients—the gels.
When a well is fracked, the gels carry the sand into the fractures and then break down, leaving the grains of sand to keep the fractures open so the gas can flow through. Conventional wisdom said that the gel ensured the sand was carried all the way into the fractures. With less viscous fluids, the sand would simply settle in the bottom of the drill pipe, a problem known as “screening out.”
The oil-field service companies that Mitchell Energy hired to provide the gels sold them by the gallon—at markups as high as 1,000 percent—so Steinsberger decided to cut back on the amount of gel per well and see what happened. The service companies warned him that the high temperatures of the Barnett would cause the gels to break down and they wouldn’t distribute the sand to the fractures, but he didn’t notice any change in production from the cutbacks. Still, the change didn’t save enough money to make the Barnett wells profitable.
Then, one night in the fall of 1996, Steinsberger wound up at a Rangers game with a couple of engineers from the Fort Worth–based oil company Union Pacific Resources. Unlike in major energy hubs such as Houston or Midland, the petroleum engineering community in Fort Worth is small. Most of the engineers know each other, and it isn’t uncommon for them to share ideas or problems with colleagues at other companies.
One of the Union Pacific engineers mentioned that his company was doing something called a “slick-water frack” in an East Texas formation known as the Cotton Valley. As the name implies, Union Pacific was using water rather than gels in the fracking process. What they were doing wasn’t supposed to work, but the engineers claimed to have cut the cost of fracking in half while getting good production from the wells. No one knew why it worked, but it did—at least in limestone, which was far more porous than shale. Could the technique also work in the Barnett? Steinsberger was skeptical, but he decided he didn’t have much to lose.
At a meeting in the Woodlands in late 1996, he made his case to Mitchell Energy executives. He proposed a series of wells using the new slick-water technique. If it didn’t work, he argued, the company could always refrack the wells with gels. All they would lose was a little time and some water. The reaction in the room was mixed. “Some people thought I was full of shit to even try something like that and other people were on board,” Steinsberger said. “It wasn’t like ‘This is an a-ha moment and it’s a great thing to try.’ It was a risk.”
The management team allowed Steinsberger to drill three wells using slick-water fracks in the spring of 1997. Any optimism, however, was short-lived. Each of the three wells screened out. The water didn’t carry the sand far enough from the well bore, and pressure spiked. Steinsberger stopped the operation early to keep the production casing in the well from bursting.
He managed to complete the wells, but after monitoring their production for several months, it was about the same as if he’d used gel fractures. “They were all failures,” he said. After poring over the results, he realized his mistake. He’d basically taken Union Pacific’s recipe for the East Texas formation and applied it to the Barnett. “That was stupid in hindsight,” Steinsberger said. The East Texas limestone that Union Pacific fracked was more permeable than shale. “Pumping slick-water fracks with even a little bit of sand is hard to do in shale,” he said. “I didn’t know that at the time.”
The realization sent him back to the Woodlands in the winter of 1997 to plead for another chance. At the time, the company was in dire straits. Crude oil prices were sinking (and, as it turns out, would sink further; by the end of 1998 they had fallen by 58 percent in two years) and natural gas experienced a similar decline. Even the biggest companies were struggling. Within a few years, Exxon and Mobil, BP and Amoco, and Conoco and Phillips would merge. Smaller ones—many, in fact, the size of Mitchell Energy—would be swallowed up or liquidated.
Perhaps because they were desperate for a solution, senior management agreed to give Steinsberger three more wells to continue the experiment. For two of the wells, he tweaked the sand concentrations he had used on the other wells. But for the third, the S. H. Griffin #4, he tried a different technique. Rather than adding all the sand at once, he started with a smaller amount and gradually increased the concentration, hoping that it would prevent the screening-out problem.
Steinsberger kept charts that rated the wells, giving each one a letter grade, from “A” through “F.” based on a number of factors, most prominently how much gas it produced.
After a year, the results were clear. Though the wells with modestly tweaked sand concentrations showed little improvement, the Griffin well was off the charts. It averaged a staggering one million cubic feet of gas a day. “That was a huge success,” Steinsberger said. “The S. H. Griffin wound up being the best well that we’d ever completed at that time.”
If there was an “a-ha” moment in fracking’s evolution, this was it. Steinsberger had done the impossible—he’d figured out how to frack shale commercially. At the time, he had no idea of the magnitude of his discovery; all he knew was that he had found the solution to Mitchell Energy’s cost problems. The impact of Steinsberger’s discovery, combined with the newfound belief that there was more gas in the Barnett than anyone at Mitchell Energy had thought, was immediately apparent to George Mitchell. Staring at the information, a smile spread across his face. “This is huge,” he said.
But Mitchell Energy’s innovation ended up having repercussions that echoed well beyond one company’s fortunes. Today, more than two decades after Steinsberger’s discovery, his technique is still used at sites around the world. Without it, the fracking revolution that has transformed Texas and put the United States on the path to energy independence might never have happened.
As is often the case with such breakthroughs, it’s a story of ingenuity, and more than a little luck. In this case, it’s also a story of persistence—of George Mitchell’s steadfast belief that there must be some way to save the company he had spent decades building. It took seventeen years and cost him hundreds of millions of dollars to make it happen. But that investment is still paying off today for anyone who puts gas in their car or turns on a light switch.
This story is adapted with permission from George P. Mitchell: Fracking, Sustainability, and an Unorthodox Quest to Save the Planet by Loren C. Steffy, published by Texas A&M University Press.