For most of the twentieth century, electricity in Texas was sold the way it was sold almost everywhere: by a vertically integrated monopoly utility, at a rate set by a regulator, with the customer's only real choice being how much to use. That changed at the stroke of midnight on January 1, 2002, when retail electric competition began across most of the state. The decision to make that change was a long time coming, the law that authorized it ran two hundred pages, and the implementation has now had more than two decades to settle. This is how Texas got here.

The story is not really about a single law. Senate Bill 7, signed by Governor George W. Bush in June 1999, is the headline event, but it sat on top of three decades of policy work at both the state and federal level. To understand the market a Texas commercial buyer operates in today — what's competitive, what isn't, who the players are, and why the rules read the way they do — it helps to start a long way back.

The long road in

Texas built its grid as its own animal. Federal policy then spent twenty years prying open the markets that sat on top of it.

Texas is the only state in the contiguous forty-eight that operates its own electric grid, separated by design from the two larger interconnections that cover the rest of the country. The arrangement traces to the 1930s, when Texas utilities deliberately kept their lines from crossing state borders so they could avoid coming under the jurisdiction of the Federal Power Commission. The Electric Reliability Council of Texas, ERCOT, was formed in 1970 to coordinate operations across the in-state utilities. ERCOT covered then, and covers now, about ninety percent of the state's electric load.

Through the 1970s and 1980s, the structure inside ERCOT looked like the structure everywhere else. A handful of large investor-owned utilities — Texas Utilities, Houston Industries, Central Power and Light, West Texas Utilities, Texas-New Mexico Power — each owned the generation plants, the transmission lines, and the distribution wires inside their service territory. They billed customers directly. The Public Utility Commission of Texas, established in 1975, set rates by reviewing each utility's costs and approving a return on capital. Outside the investor-owned territories, electricity was supplied by municipal utilities (Austin, San Antonio, Garland, Bryan, and dozens of others) and by rural electric cooperatives that traced back to the Rural Electrification Act of 1936. The munis and the co-ops were, and remain, governed by their own boards rather than by the PUC.

Texas legislators followed the federal track. In 1995, Senate Bill 373 deregulated the Texas wholesale market, allowing independent generators and large industrial users to buy and sell power across the state's transmission system on equal terms with the incumbent utilities. The retail side — meaning the price you saw on your home or business electric bill — was still bundled with the wires and still set by the PUC. But the wholesale change was the first crack. Industrial users, who paid the largest bills and had the most to gain from market pricing, started lobbying immediately for the next step.

Two other forces were pushing in the same direction. Enron, then a fast-growing Houston-based energy trader, was pouring resources into deregulation advocacy across the country. And inside the PUC, by the late 1990s, the staff was acknowledging that the regulated utilities were earning more than the rates the commission had originally approved — not because anyone had cheated, but because the cost of generating power had been falling while customer demand was rising fast. Then-PUC chairman Pat Wood wrote in a 1998 report that the agency could see itself "facing a never-ending stream of rate cases in an attempt to harness utility over-earnings." Either the rate-setting model needed serious reform, or it needed to be replaced.

California's mistake, and a napkin doodle

The first state to deregulate did so badly enough to set the entire national project back a decade. Texas legislators flew out, watched, and learned.

California passed Assembly Bill 1890 in August 1996, becoming the first state to attempt comprehensive retail electric deregulation. The law uncapped wholesale prices but kept retail rates frozen, on the theory that wholesale competition would drive retail costs down and consumers should be protected during the transition. It was a fatal design choice. When wholesale prices spiked in 2000 — partly due to a hot summer, partly due to drought reducing hydroelectric supply, and substantially due to deliberate market manipulation by Enron and others — the state's utilities were forced to buy power at uncapped wholesale prices and resell it at capped retail prices. Pacific Gas and Electric and Southern California Edison were both pushed to the edge of insolvency. Rolling blackouts hit California in early 2001. PG&E filed for bankruptcy that April.

The damage was severe enough that several states with deregulation laws on the books delayed implementation indefinitely. Texas, with its own law already moving through the legislature, did neither. Instead, the bill's authors went to California to study what had gone wrong and came back with a list of design fixes intended to prevent the same outcome.

The Texas redesign that emerged from those flights had three meaningful differences from California. It capped wholesale prices rather than retail prices. It required new retail providers to enter into long-term supply contracts so they couldn't be wiped out by short-term spikes. And it gave the PUC explicit authority to investigate and penalize market manipulation. None of those choices were fully tested until the market actually opened, and the early years would reveal that some of the safeguards worked while others did not. But the worst-case California outcome — insolvent utilities, capped retail prices that bankrupted them, blackouts driven by financial collapse — was avoided.

Senate Bill 7

The actual law that opened the Texas retail market. About two hundred pages, four pillars, and a long list of things it deliberately left alone.

Senator David Sibley (R-Waco) and Representative Steve Wolens (D-Dallas) introduced parallel deregulation bills in January 1999, then merged them into a single proposal that ran approximately two hundred pages. The Senate adopted it unanimously in March. The House passed it in May. Governor George W. Bush signed it on June 18, declaring that "competition in the electric industry will benefit Texans by reducing monthly rates." The law took effect on January 1, 2002. The full enrolled text of Senate Bill 7 remains posted on the Texas Legislature's website, along with its complete legislative history.

What did it actually do? Four major things, plus a lot of plumbing.

And what did SB 7 leave alone? More than people sometimes realize. The law applied only to investor-owned utilities inside ERCOT. Municipal utilities — Austin Energy, CPS Energy in San Antonio, Garland Power & Light, and dozens of smaller city systems — were given the option to opt in and almost none did. Rural electric cooperatives were also exempted, and almost all chose to remain regulated. Areas of Texas outside ERCOT — El Paso to the west, the Beaumont/Port Arthur region to the southeast, and the upper Panhandle — weren't part of ERCOT's grid in 1999 and so weren't subject to the law. The result, more than two decades later, is that Texas has one of the most fragmented retail electric markets in the country: deregulated for most urban and suburban customers, regulated for the rest.

Who got choice, who didn't

Two reference maps from the Public Utility Commission of Texas show what the deregulated market actually covers. The first answers which grid am I on; the second answers which wires company delivers my power.

SB 7 only deregulated retail electricity inside ERCOT. The first map below shows where that boundary actually sits. The far western tip of Texas is in WECC and served by El Paso Electric. A slice of the southeast (Beaumont, Port Arthur, Orange, Jasper, Newton, and several adjacent counties) is in MISO and served by Entergy. The upper Panhandle and a corner of northeast Texas are in SPP, where customers are served by Xcel, AEP-SWEPCO, and others. None of those areas are subject to retail competition. Customers there pay the regulated tariff their utility files with the PUC or, in the case of Entergy, with FERC.

Figure 1 · Texas grid jurisdictions

Most of Texas is in ERCOT — but not all of it.

Source: Public Utility Commission of Texas. puc.texas.gov

The second map zooms in on what's inside the deregulated portion of ERCOT and shows the five Transmission and Distribution Utilities — the regulated wires companies that operate inside the competitive retail areas. Where you sit on this map determines two things: which TDU charges show up on your bill (each utility has its own approved tariff), and which set of REPs is competing to serve your meter (most REPs serve all five territories, but pricing varies meaningfully across them).

Figure 2 · Transmission & Distribution Utilities

The five TDUs serving Texas competitive retail areas.

Source: Public Utility Commission of Texas. puc.texas.gov

The white space on the second map is the part the headline can't capture. Those gaps are municipal utility territories (Austin Energy, CPS Energy in San Antonio, Garland, Denton, Bryan, College Station, Brownsville, and dozens of others) and rural electric cooperative territories (Pedernales, Bluebonnet, Bandera, Bartlett, and dozens more). Customers inside those gaps are still inside ERCOT — the lights still come from the same wholesale market — but they don't have retail choice. Their utility sets the rate.

Roughly eighty-five percent of Texas's commercial and industrial electric load sits inside the colored portions of the second map. The other fifteen percent is served by a muni, a co-op, or one of the out-of-ERCOT utilities. Whether your meter is in or out shapes everything about how you buy power: inside the colored areas you choose your supplier and negotiate your contract; outside them, your utility files a tariff with a regulator and you pay it.

The market opens, January 1, 2002

A new wholesale architecture, a new retail rulebook, and a transitional rate called the Price To Beat that did not, in the end, turn out the way its architects hoped.

By the time the market opened, the incumbent utilities had spun off their retail operations into branded REPs — TXU Energy, Reliant Energy, CPL Retail Energy, and the rest. Roughly a dozen new REPs had also been certified. Customers were free to switch immediately. Almost none did.

The reason was a transitional pricing rule called the Price To Beat. Under SB 7, each incumbent REP was required to charge its existing residential and small commercial customers a fixed rate set by the PUC, six percent below the bundled rate those customers had been paying on December 31, 2001. The rule had two purposes. It guaranteed every customer a modest rate cut at the moment of deregulation, regardless of whether they shopped. And it set a price ceiling that new entrants had to undercut to win business. The Price To Beat was supposed to remain in effect until either January 1, 2007 or the date when forty percent of an incumbent's customers had switched to a competitor — whichever came first.

The mechanism worked in some ways and not in others. The six percent cut was real. But the formula for adjusting the Price To Beat was tied to natural gas prices, and only to natural gas prices, and only in one direction. Incumbents could petition the PUC to raise the Price To Beat when gas costs went up. There was no symmetric mechanism to push it down when gas costs fell. Through 2002, 2003, and 2004, natural gas prices climbed steadily — and so did the Price To Beat, in some cases by close to ten percent in a single approval. Competitive REPs, rather than aggressively undercutting the moving target, generally clustered their pricing just below it. The market was open, but it didn't move.

The first full audit of how the market was actually operating came in early 2003, when the PUC reported on Enron-related abuses during the 2001 pilot phase. Several incumbent generators — TXU, Constellation, Mirant, Reliant, AEP, and Enron itself — were found to have collectively earned about $29 million in improper revenues by gaming a congestion-relief mechanism in ways the regulators had not anticipated. The penalties were modest. The political fallout was contained, partly because Enron had collapsed in late 2001 and was no longer in any position to defend itself, and partly because the gaming had occurred during a pilot, before retail competition formally began.

By 2005, the Price To Beat had risen well above the rate that prevailed at the moment of deregulation, and average residential bills in deregulated areas were running above the U.S. average. The PUC's own published reports acknowledged the gap. The political response was not to abandon the architecture but to keep building it out. Two specific decisions, both made during this period, would shape the next decade of the market.

Growing pains

A scandal at the grid operator. A market overhaul that ran ten times over budget. A leveraged buyout of the largest utility in the state. The years 2004 through 2010 were not quiet.

In December 2004, ERCOT's chief operating officer was indicted on charges of accepting kickbacks from a vendor selected to build the grid operator's customer registration system. The investigation expanded over the following months and led to the dismissal or resignation of several senior staff, a review by the State Auditor's Office, and a restructuring of ERCOT's governance. The episode was significant less for the dollar amounts involved than for what it revealed about oversight: ERCOT was a private nonprofit corporation handling tens of billions of dollars of energy transactions a year, with a board that did not have meaningful state-level scrutiny over its procurement. Legislative changes in 2005 placed ERCOT under direct PUC oversight and added independent directors to the board.

The 79th Texas Legislature passed Senate Bill 20, raising the state's renewable energy target to 5,880 megawatts by 2015 and creating Competitive Renewable Energy Zones — CREZ — in West Texas and the Panhandle. The CREZ initiative was, in retrospect, one of the most consequential single decisions in the history of the Texas grid. It directed the PUC to plan and approve roughly $7 billion in new transmission lines specifically to move wind power from where it was generated (windy, sparsely populated rural areas) to where it was consumed (the population centers along I-35 and the Gulf Coast). Construction took most of a decade. By the time the lines were finished in 2014, Texas had built more wind capacity than any other state, by a wide margin, and was on its way to leading the country in solar as well.

In February 2007, a private-equity consortium led by Kohlberg Kravis Roberts, TPG Capital, and Goldman Sachs announced a $45 billion leveraged buyout of TXU Corporation, then the largest electric utility in Texas and one of the largest in the country. It was, at the time, the biggest leveraged buyout in history. TXU was renamed Energy Future Holdings (EFH); its competitive generation arm became Luminant; the regulated wires business became Oncor; and its retail business kept the TXU Energy name. The deal was financed with about $40 billion in debt, most of it secured against the assumption that natural gas prices — and therefore wholesale power prices — would stay high. They did not. Within a few years EFH was carrying interest costs it could not service, and the company would file the seventh-largest corporate bankruptcy in U.S. history in April 2014.

From its 2002 opening through 2010, ERCOT operated under what was called a "zonal" wholesale market — the state was divided into a small number of large zones, and the wholesale price was the same everywhere within a zone. The design was simple, but it failed to send accurate price signals about where the grid was actually congested. Power generated in West Texas couldn't always reach the cities, and the zonal pricing didn't reflect that. The PUC authorized a switch to a "nodal" market in 2003, with separate prices at each of approximately 4,000 individual nodes on the grid. The original cost estimate was under $100 million. The nodal market finally went live on December 1, 2010, six years late and at a final cost of approximately $660 million. Once operational, however, the nodal design did what it was supposed to do: it sent generators clearer signals about where new capacity was needed and gave large customers a more accurate view of what their power actually cost to deliver.

By the close of 2010, the deregulated market had been open for nine years. Average residential rates in deregulated areas remained somewhat above the U.S. average and noticeably above the rates charged by Texas's exempt municipal utilities and co-ops. Customer complaints filed with the PUC were running at roughly five times the pre-deregulation level, though they had begun a steady decline that would continue for the next decade. The number of REPs operating in the state had grown into the dozens. Competition was, by any measure, real. Whether it had delivered on the original promise of lower prices was a question the data couldn't yet answer cleanly.

Coming of age

The 2010s were the decade in which the market the Texas legislature designed in 1999 actually started behaving like the market its designers had imagined.

The single most important development of the 2010s wasn't a piece of legislation. It was the collapse in natural gas prices. The shale revolution, driven by hydraulic fracturing in the Barnett, Eagle Ford, and Permian basins, pushed Henry Hub gas prices from a 2008 average above $8 per million BTU to a 2012 average below $3. Because natural gas plants set the marginal price of electricity in ERCOT during the majority of operating hours, wholesale power prices fell in lockstep. By 2013 and 2014, wholesale prices in ERCOT were running well below where they had been at the market's opening — and average residential rates in the deregulated zones, which had been above the U.S. average for the entire first decade of the market, finally crossed below it. They have remained below the U.S. average for most of the years since.

Three other developments mattered.

Energy Future Holdings filed for bankruptcy on April 29, 2014. Luminant, Oncor, and TXU Energy continued operating uninterrupted under bankruptcy protection. Oncor, the regulated wires business, had retained value through the price collapse because its earnings depended on rate-regulated returns, not wholesale energy prices. Luminant and TXU Energy were both restructured. The bankruptcy proceedings dragged on for nearly four years and produced a series of failed acquisition bids for Oncor — first by a Hunt Consortium proposing a complex tax structure that the PUC effectively blocked, then by NextEra Energy of Florida, also rejected, before Sempra Energy of California finally closed on Oncor in March 2018.

Wind and solar built out at a pace that surprised everyone. The CREZ transmission lines completed in 2014 unlocked roughly 18,500 megawatts of wind capacity. By the late 2010s, wind was routinely supplying 20 to 25 percent of ERCOT's annual energy — and on certain low-load days in shoulder seasons, as much as 60 percent in a given hour. Utility-scale solar followed, slower at first but accelerating sharply after 2018. The renewable buildout pushed wholesale prices lower during high-renewable hours and made the market more dependent on natural gas plants for reliability during low-renewable hours, particularly during summer evenings when solar production fell and load remained high.

REP competition got real. By 2015, more than a hundred REPs were certified to operate in ERCOT, offering several hundred distinct products: simple fixed rates, indexed plans, time-of-use rates, free-nights-and-weekends plans, prepaid plans, and an expanding range of products targeted specifically at commercial and industrial customers. The Power to Choose website, run by the PUC for residential shoppers, became simultaneously the most-criticized and most-used consumer tool in the market. On the commercial side, the rise of independent brokers and consultants (the role this firm plays) reflected the growing complexity of the products available and the difficulty of comparing them on price alone. By the late 2010s, well over half of all non-residential power purchasing in ERCOT was being conducted through third-party intermediaries, and the share continues to climb.

February 2021: Winter Storm Uri

The single most consequential event in the history of the Texas grid since deregulation began. It deserves its own treatment.

Where Texas sits today

The retail electric market in Texas is now in its third decade. The structure is mature; some of the rules are still being argued over. Here is what a commercial buyer is actually looking at.

Inside the deregulated portion of ERCOT, electricity service is delivered through three distinct businesses operating under three different sets of rules.

The competitive layers (generation and retail) are where the deregulation experiment plays out. The regulated layer in the middle is essentially unchanged from the pre-1999 model. That middle layer is also where most of the durable price increases of the last decade have actually come from: transmission and distribution charges have grown faster than inflation through the 2010s and 2020s, partly to fund the CREZ buildout, partly to fund ongoing reliability investments, and partly because the underlying cost of replacing aging infrastructure is high. None of that is negotiable on a customer's bill, regardless of which REP you choose.

What is negotiable is the energy charge, the contract structure, the term length, the indexing, the ancillary service treatment, the pass-through clauses, and a half-dozen other components most buyers never see in their bid documents. That is the part of the bill where the deregulated market gives a buyer leverage. It is also the part of the bill where most buyers, working alone and pricing infrequently, leave money on the table.

What this history means for a buyer

Three observations, drawn from the arc above, that translate directly into how to operate inside the market as it exists today.

The market is real, and it is more competitive than it used to be. The early Price-To-Beat years were sticky and the early REP failures were ugly, but those problems are mostly artifacts of the first decade. A commercial buyer in 2026 who runs a structured procurement process — multiple bids, a clean spec, attention to the pass-through clauses — reliably gets a better outcome than a buyer who renews on autopilot. That wasn't reliably true in 2005. It is reliably true now.

The market also isn't all of your bill. Roughly half of a typical commercial customer's monthly charge is the regulated transmission and distribution component, and that component is set by a tariff filed with the PUC, not by anything you negotiate. Understanding that the bid you're evaluating is a bid for the energy and the REP margin, not a bid for the wires, is the difference between thinking the market saved you ten percent and knowing the market saved you ten percent of the half of your bill that was actually competitive.

The architecture is durable, but the rules are still being written. The post-Uri reforms changed the wholesale price cap, the ancillary service products, and weatherization standards. Active debates over capacity payments, the Performance Credit Mechanism, and various reliability proposals continue. None of those changes are likely to undo retail competition, but they affect what your contract is exposed to and what your REP is hedging against. A multi-year contract signed today is a multi-year exposure to whatever the regulators do next. That isn't a reason not to sign one. It is a reason to know what you're signing.

The Texas legislature picked an architecture in 1999 and the market has spent twenty-five years working out what that architecture actually means in practice. It is not, as some of its early advocates promised, a free lunch. It is also not, as some of its early critics warned, a disaster. It is a set of rules. The buyers who do well inside those rules are the ones who understand them.