Who owns Texas’s private water?

One tariff, very different price changes

At Treetop Estates, a small water system in Parker County, the inherited residential tariff charged $28.45 for 5,000 gallons. At Council Creek, in Burnet County, the same amount of water cost $92. After the systems entered CSWR-Texas’s consolidated tariff, their base and usage charges converged: $75.89 under the 2024 schedule and $89.72 under the 2026 schedule. These figures exclude other assessed charges; they measure the same part of each tariff, at the same consumption and meter size. ^S1, ^S2, ^S3, ^S4

One customer group experienced a steep increase. Another initially experienced a decrease. Both came under the same owner and ultimately the same price schedule.

That is the structure underneath this story: the physical water system is the unit of service, but the rate pool is the unit that spreads costs. A sale can change both who owns the pipes and which customers pay for the owner’s wider collection of systems. The owner’s identity, the price paid, and the household bill are connected—but they are not interchangeable measurements.

This investigation assembled the Texas Public Utility Commission’s published investor-owned-utility registry, documented ownership chains for 11 platforms, and reconstructed 14 tariff groups covering 24 listed drinking-water systems. Within that panel, it verified three acquisition closings and tariff continuity, then joined four active physical systems to federal compliance records. It also examined the Quadvest transaction and the legislation governing acquisition valuation and rate consolidation.

The resulting evidence supports four findings:

• The 2019 law changed acquisition incentives, but later laws changed how those costs can reach customers. Fair-market-value valuation, the ability to request an existing tariff for newly acquired systems, and broader rate consolidation use different mechanisms.
• A customer can share acquisition costs from an FMV transaction even when their own system was purchased outside FMV. The three matched sales used the ordinary route, then entered a rate proceeding that also included seven FMV acquisitions. Consolidation makes the exposure extend beyond the system sold.
• Fewer recorded violations can mean better compliance administration without a demonstrated improvement in contaminant levels. The four-system case series produces sharply different results across systems and violation types.
• The largest current example is still a prospective rate story. H2O America says Quadvest’s $483.6 million regulated purchase price will become its ratemaking rate base. The sale-approval order explicitly reserved that determination. H2O expects revised rates in 2028. ^S5, ^S6, ^S7

Statewide ownership coverage remains incomplete. The registry contains 458 certificate or registration identifiers. Forty-nine were matched to the 11 documented platforms; 409 remain unresolved in this investigation. Those certificate counts require customer totals before they can describe market share. The workbook preserves every registry row and every unresolved match rather than filling gaps with guesses. ^S8

High bills predate the 2019 law

The basic concern did not begin in 2019. Texas Tribune reporting in 2012 and 2013 described large rural and small-town private-water rate increases. Those stories document a longstanding affordability problem; they cannot be evidence of an effect caused by a law enacted six or seven years later. ^S9

ABC13’s November 2025 report on a New Caney homebuyer is a different kind of evidence. The reported water and sewer base charges totaled $116.89 in October 2024 and $138.58 in September 2025: $21.69 more each month, or $260.28 annually, before using water. The utility was Texas Water Utilities, part of Nexus Water Group. It was not The Texas Water Company, H2O America’s subsidiary buying Quadvest. A high bill in a new subdivision cannot measure an acquisition effect without a before-and-after comparison. ^S10

Commercial analysts have already mapped much of this market. Bluefield Research marketed a Texas private-water ownership report in 2022 with 20 company profiles, county-level positioning, and transactions dating to 2015. It would be wrong to claim that institutional ownership of Texas water was just discovered, or that it first appeared after 2019. ^S11

Academic work has also described how fair-market-value laws can raise the price sellers receive and enlarge the asset value recovered through rates. What remains poorly resolved is the Texas incidence: which approved valuation increases entered which rate pools, what existing and newly acquired customers then paid, and what service changed in return. ^S12

The research gap is therefore narrower—and more useful—than another list of private-equity firms. It is a dated crosswalk from system → transaction → legal utility → capital owner → rate pool → customer outcome.

Who owns the companies

“Owner” needs a stopping rule. A listed holding company’s economic owners are its shareholders. Naming the chief executive or one asset manager leaves most of that ownership unexplained. A private fund manager is not identical to the pension funds, insurers, or other investors supplying the fund’s money. A cooperative’s members are a different ultimate constituency again.

The table reports the highest ownership or capital-control layer supported by the documents reviewed. It does not pretend to identify every fund investor or every intermediate subsidiary.

Nexus illustrates why stopping at a familiar financial name is misleading. BCI manages money for pension and other public-sector clients. The proposed merger chart put the other 50% under SWMAC Holdco, itself split 75% to IIF Subway and 25% to Bazos. The arithmetic would imply 37.5% and 12.5% of the combined company, respectively, under that proposed structure. The application identifies Bazos with Munich Re and IIF Subway with IIF US Holding 2. Calling the company simply “owned by JPMorgan” would erase those dated proportions and capital relationships. ^S13, ^S14, ^S15

There is a second trap: an asset sale does not necessarily transfer the whole named company. Crystal Springs’ own website says it sold to Texas Water Utilities while retaining Spring Branch Crossing and Pine Rock/White Rock accounts. Its registry name therefore cannot be assigned wholesale to Nexus. The dataset leaves those certificates unresolved pending service-area-level reconciliation. ^S25

And there is a third: neither government registry is a ready-made list of ultimate owners. The PUC file contains 460 rows, 384 utility-name strings, and 458 identifiers after duplicate certificates are reconciled. Its labels include cooperatives and associations. EPA’s “private” ownership category is also broader than for-profit investor ownership. GAO has specifically criticized inaccurate and outdated water-ownership information; its recommendation for better ownership definitions and verification remained open in February 2026. ^S8, ^S26

In the federal 2026 Q3 snapshot, 4,757 Texas systems are classified as active community water systems; 1,819 carry the broad private-owner code. That is not a count of 1,819 independently owned, for-profit companies available for acquisition. One company may operate many systems, and nonprofit or association ownership can appear within the private category. The proposition that “thousands of rural systems became an asset class” needs these units disentangled before it can be measured. ^S26, ^S27

An original data check exposed a practical join problem. The selected CSWR systems have Texas PWS identifiers but a Missouri value in the federal STATE_CODE field associated with their administrative address. Filtering that field for Texas would discard the very systems being investigated. The extraction defines the state through the Texas system identifier and primacy relationship. An owner’s mailing location would misclassify these systems. The federal organization-name field is not used as proof of ultimate ownership. ^S27

The sellers are not one category either. Treetop’s bill of sale names Tom and Barbara Crew alongside the utility corporation. Quiet Village’s seller was the estate of Donald Wilson, acting through its executor and beneficiary. Quadvest was a family-owned business building water and wastewater systems for new subdivisions; its lender’s case study identifies the Sequeira family and describes development-led growth. An estate exit, a small local utility sale, and a growing suburban platform are different economic events. Their statewide proportions remain unmeasured. ^S1, ^S38

How a sale becomes a claim on future bills

For a regulated utility, the annual revenue requirement can be represented as:

Operating expenses + depreciation + taxes + allowed return × rate base, with other authorized adjustments where applicable.

The rate base is the approved capital amount on which the utility may earn a return. Annual revenue, stock-market value and the purchase price each measure something else. Only approved capital enters the rate base. The allowed overall return combines debt and equity financing; it is not a shareholder profit margin applied to the whole bill. ^S28

Under Texas Water Code §13.305, created by HB 3542 in 2019, an eligible Class A or B buyer and seller can elect a valuation process. The PUC selects three experts. Each uses cost, market, and income approaches. The fair market value is the average of the three appraisals, and the acquisition’s ratemaking rate base is the lesser of that value and the negotiated purchase price. That amount is incorporated in the buyer’s next rate-base case. ^S29

This changes the bargaining structure. Under an original-cost approach, paying the seller substantially more than the remaining eligible asset value may leave a premium shareholders cannot recover from customers. An FMV pathway can make more of the acquisition payment recoverable. That can support a higher bid, provide an exit for a small owner, and attract buyers able to assemble systems into a larger portfolio.

But the seller receives the acquisition payment. The payment itself does not replace a pump, repair a well, or hire an operator. Repairs need their own budget. The customer’s economic question is whether the service and financing benefits justify both the purchase price and the rehabilitation.

Texas did not leave the appraisal completely unconstrained. The implementing rule bars appraisers from taking the negotiated price into account and excludes specified speculative-growth and goodwill adjustments. Independent appraisals are a real constraint; the mechanism is not “pay anything and send the bill to customers.” At the same time, the statute is not limited to distressed rural systems. The PUC rejected a proposed restriction to small selling utilities because the legislation was broader. ^S30

One especially consequential provision deserves more attention: the appraisal disregards the original funding source, including developer or customer contributions in aid of construction. That creates the possibility that infrastructure initially financed outside the utility’s investor capital can support an acquisition valuation later recovered through rates. Whether a household paid twice turns on the earlier financing and later rate treatment. Proving that requires the contributed-asset records, their treatment in the seller’s prior regulatory rate base, and the buyer’s approved valuation. A financial-accounting write-up alone is insufficient. ^S29, ^S30

The 2019 law is only part of the story:

The statutes and implementing rules treat these as different decisions. A study that codes every post-2019 sale as the same treatment would combine valuation changes, tariff adoption, later rate cases, and infrastructure surcharges. ^S29, ^S30, ^S31, ^S32

This also changes the appropriate comparison group. A system that was never sold can still experience acquisition-related costs if it already shares the buyer’s consolidated tariff. Such customers share the potential exposure. Using them as “untreated” controls would contaminate the comparison.

What the matched tariffs show

The three documented closings were Treetop Utilities on December 23, 2020; Council Creek on January 29, 2021; and the Quiet Village II utility on September 9, 2021. The associated tariffs list five PWS identifiers; four active systems enter the quality analysis. Closing filings document the transaction dates; acquisition and later approved tariffs supply the schedules. The broader 14-group panel is a convenience sample selected for extractable, comparable schedules. It leaves representativeness and statewide transaction coverage unmeasured. ^S1, ^S2, ^S3, ^S4

The comparison below holds consumption at 5,000 gallons a month and uses a standard 5/8- or 3/4-inch residential meter. The amounts are tariff calculations in nominal dollars. Paid invoices would also reflect the additional charges below. It isolates base and volume charges because local pass-through charges, rate-case surcharges, and the 1% regulatory assessment have their own charges.

Quiet Village requires a careful denominator. Its inherited schedule labeled $12.62 plus $3.84 per thousand gallons for water and sewer together. The original rate application’s billing comparison splits that combined charge equally between the two services. Treating it as a water-only tariff would substantially misstate the increase. ^S2, ^S3, ^S4

Consumption changes the result. Council Creek’s inherited water tariff was $70 plus $4 per thousand gallons for the first 4,000 gallons and $6 thereafter. The 2026 consolidated schedule is $61.60 plus $4.56 per thousand for the first 4,000, $9.88 for the next 10,000, and $19.76 above that. At 5,000 gallons, the base-plus-volume charge remains slightly below the inherited tariff. At 10,000 gallons, it is $139.12 versus $122—14.0% higher. ^S2, ^S4

This is why a single headline percentage cannot describe every customer. The sale’s closing date, the tariff’s effective date, the meter, consumption, the services purchased, and the customer’s entry into a common rate pool all matter.

The excluded charges are real. The 2024 tariff authorized a $1.60 monthly rate-case surcharge for each applicable service. The 2026 tariff has $1.86 for water and $2.14 for sewer, with specified recovery limits. Treetop and Quiet Village also have local pass-through provisions. For Quiet Village, the company’s published June 2026 schedule includes a sewer pass-through that was not found in the stamped 2026 sewer tariff examined. That source discrepancy is preserved in the workbook; a modeled total including it is not represented as a verified customer invoice. ^S3, ^S4

All three matched acquisitions were outside the §13.305 FMV process. A January 11, 2024 filing identifies the FMV acquisitions included in this rate case; Treetop, Council Creek, and Quiet Village are not among them. The final order specifically incorporates that classification. These are therefore examples of substantial post-acquisition tariff changes without FMV being the transaction’s valuation pathway. ^S28, ^S37

That does not mean their acquisition premiums were zero. The 2024 order approves positive acquisition adjustments for eligible systems acquired outside FMV, alongside different treatment for FMV systems. It also includes operating and capital costs. The precise inference is that acquisition followed by tariff consolidation can redistribute customer charges dramatically, while a bill difference alone cannot measure what the 2019 law added. The appropriate test compares FMV acquisitions with otherwise comparable ordinary acquisitions. A simple before-and-after-2019 comparison mixes those legal routes. ^S28

There is a further complication: outside FMV does not mean unexposed to FMV costs. The same proceeding brought seven FMV acquisitions into the consolidated water and sewer tariffs. CSWR’s filing identifies their previously approved acquisition rate bases as follows. These are reported regulatory asset values. Cash purchase prices and acquisition premiums need transaction-specific support. ^S28, ^S37

This is the most important consequence for measurement: the customers paying an acquisition-related revenue requirement need not be only the customers whose system was acquired under that law. The rate pool can move the burden. A comparison between “FMV customers” and “non-FMV customers” inside one consolidated tariff would miss that shared exposure.

The same filing shows a $1,193,311 adjustment between its plant-asset column and the total above. That number is tempting but cannot be called the premium over the sellers’ former regulatory rate bases: the plant column is not verified as the comparable pre-sale original-cost balance. The dataset preserves all three columns and labels the missing counterfactual explicitly.

Two prices can be recovered by combining documents. The North Orange and Leon Springs acquisition orders state that their purchase prices were below appraised FMV and therefore set the acquisition rate bases, while keeping the amounts in confidential attachments. The later public CSWR table supplies $1.75 million and $3.893 million. Read together, the documents support those purchase-price amounts by cross-document inference. They still do not supply the sellers’ comparable prior regulatory rate bases. ^S37, ^S40

Regulatory review also changes outcomes. In the 2024 CSWR case, the commission approved an overall return of 7.37%, compared with the requested 9.62%, and rejected an extra risk premium partly because the approved rates reduced the asserted business risk. That is evidence against assuming the regulator simply grants every request. ^S28

Did the water get better?

The analysis links the three sales to four active physical public-water-system identifiers. A fifth tariff-listed identifier, South Council Creek 1, is recorded as inactive since 1991 in the federal inventory even though it appears in the sale and tariff documents. The conflicting identifier is excluded. Its inactive label leaves physical closure and violation outcomes unknown. Federal records were deduplicated by PWSID and violation ID, because a violation can appear more than once with different enforcement actions. The counted date is the recorded beginning of the noncompliance period. “New violation records” is therefore the precise measure; these are not all violations active during an interval, unique contamination incidents, outages, or illnesses. ^S1, ^S27

The primary comparison uses the 36 months immediately before and after each closing. It counts health-based and other violations in their own columns. A second comparison delays the post-sale window by 12 months to allow for an operational transition. The workbook includes annual results from 2017 through 2025 and labels partial 2026 observations.

The mixed result survives the lagged window. Post-sale totals become 1, 3, 35, and 40, respectively, while the health-based counts remain 0, 0, 24, and 24. Each Council Creek system had health-based violation onsets in all 12 quarters in both primary windows. The 24 records represent repeated quarterly breaches of two standards. Treating each one as an independent accident would misdescribe their frequency. ^S27

The companies’ own 2025 consumer-confidence reports corroborate the unresolved contaminant problem. Council Creek Village’s report gives a highest combined-radium level of 44 pCi/L against a limit of 5 and a highest gross-alpha level of 131 pCi/L against a limit of 15. It says treatment options or a new source are being explored. South Council Creek 2 reports combined radium of 33.6 pCi/L against the same limit of 5 and describes work toward a new source. The measurements describe 2025. Today’s tap-water conditions require current testing. ^S33

The Council Creek problem was inherited. A 2010 draft feasibility study prepared for TCEQ already documented gross-alpha and combined-radium exceedances in measurements from 2001–2008 and examined treatment, alternative groundwater, and a connection to another supplier. The study describes proposed remedies and historical costs. Construction completion and current costs need later engineering records. It documents the age of the problem, while the later records test whether the new ownership resolved it within the observed period. ^S41

Conversely, the Treetop and Quiet Village records support a real improvement in compliance administration. It would be wrong to dismiss monitoring and reporting: those functions are necessary to know whether a system is safe. It would also be wrong to relabel a reduction in paperwork-related records as a measured reduction in contaminant exposure. Quiet Village’s 2025 report, for example, discusses a violation originating in 2024 even though no new violation onset appears in the 2025 count. ^S27, ^S33

The case series is small and deliberately selected for document matchability. It has no untreated comparison group and does not estimate a causal buyer effect. It measures three acquisitions, one buyer, and four systems. The price and quality windows also answer different questions: tariffs show later scheduled charges, while the primary compliance comparison covers the first three years of new ownership. They must not be presented as simultaneous experimental outcomes.

Research elsewhere gives the buyers’ argument a fair hearing. An August 2026 study of 15 Illinois municipal FMV sales to one buyer, compared with 34 municipal controls, reports large reductions in both health and non-health violations. The authors explicitly stop short of causal claims and do not estimate rate or affordability effects. It supports the possibility of service improvement; it doesn’t show that Texas customers must pay an acquisition premium to obtain it. ^S34

The counterfactual matters. The relevant question is not only whether a buyer improves on a failing predecessor. It is whether that buyer and financing structure outperform feasible alternatives—another buyer, a member-owned organization, a public consolidation, or professional operation without an asset sale—at a comparable total cost.

Quadvest: the largest lead is not yet an after-sale bill result

The Quadvest transaction has several dated milestones:

The latest status filing examined, dated October 19, reported the closing and said evidence addressing the transaction and customer deposits would be filed by November 2. The sale order also found that customer rates would not change because of the transaction. That immediate statement can coexist with a later rate case recognizing a larger asset value. Neither should be mistaken for the other. ^S5, ^S6, ^S7

Nor is Quadvest simply a collection of stagnant rural systems. H2O describes rapid connection growth, subdivision development, a wholesale business, and a large pipeline of contracted future connections. That makes it a growth acquisition as well as a regulated utility acquisition. The law’s relevance extends beyond rescuing distressed systems. ^S6, ^S7

Price per customer is especially easy to get wrong here. The March prospectus described about 38,000 regulated retail service connections at December 2025; the October earnings release described 62,200 total active connections at September 2026, including wholesale and retail service. Water and sewer connections can also refer to the same household. Dividing $483.6 million by those different denominators would produce superficially comparable numbers describing different things. Future contracted connections are not present bill-paying households either. ^S6, ^S7

The buyer is not receiving a free return on day one. H2O discloses expected earnings dilution while financing the acquisition before new rates take effect. That supports a more precise economic account: the seller is paid at closing; the buyer funds the interval and faces regulatory and operating risks; customers may subsequently finance approved acquisition capital and improvements through rates. The allocation depends on the final orders and tariffs. ^S7

This investigation does not report a realized Quadvest rate increase attributable to the acquisition, a finalized $483.6 million PUC rate-base award, or improved service after the sale. Those outcomes remain ahead of the reported sale.

The alternative capital stack already exists in Texas

South Plains Water Supply Corporation provides a concrete alternative to a binary choice between an undercapitalized small owner and an investor-owned consolidator. Customers from four systems around Shallowater organized a water supply corporation after prolonged operating and quality problems. Communities Unlimited, which participated in financing the transition, describes loans supporting the transaction and a prior $3 million forgivable state loan. ^S35

In February 2026, the Texas Water Development Board approved a further $3.3 million grant for improvements across all four sites, including wells, electrical equipment, meters, storage, pumps, and distribution. The grant provides a documented source of repair capital. ^S36

It is also not a demonstrated victory on every outcome. The participating organization’s August account says substantial quality and infrastructure needs remain. Grant approval is not completed construction, and member ownership is not proof of better water. ^S35

The example changes the financing question. A grant shifts some costs away from the utility’s future bills to the public funding program. A sale price shifts cash to a seller and can create a later claim on customer revenue. These choices can coexist with professional management and consolidation. Public funding eligibility is program-specific: Texas’s Drinking Water State Revolving Fund explicitly includes eligible privately owned community systems. It would be incorrect to treat access to all public financing as exclusive to government or cooperative ownership. ^S35, ^S36, ^S39

For public-versus-private comparisons, the honest denominator is the household’s total water-related burden: tariffs, relevant taxes or assessments, connection charges, and the incidence of public subsidies. A municipal bill and a private bill may not finance the same bundle of obligations. Conversely, the existence of municipal taxes does not excuse ignoring a private acquisition premium. Each must be measured.

A public district can finance the same infrastructure through taxes and developer reimbursement. The Austin-area MUD investigation follows bonds, unreimbursed construction and contract taxes. Comparing a district with a private-water tariff therefore requires the household’s tax bill alongside its water charge.

The measure that would answer the price question

A statewide series of announced acquisition prices divided by reported “customers” would not answer whether the 2019 law raised bills. It would mix water and sewer, asset and share purchases, wholesale and retail businesses, present and future connections, and growth systems with failing systems. It would also treat undisclosed prices as if they did not exist.

The more revealing measure is an acquisition-only bill increment. It can be constructed from regulatory records without pretending that every post-sale increase is acquisition profit:

1. Determine the seller’s remaining eligible original-cost rate base for the same assets and date, including the treatment of contributed infrastructure.
2. Record the acquisition rate base approved, alongside the announced cash price and the financial-accounting purchase-price allocation.
3. Identify the later rate case and the exact customer pool over which the approved acquisition amount is recovered.
4. Recalculate that approved rate model replacing only the acquisition valuation with the comparable original-cost amount, leaving repair capital, operating costs, financing assumptions, and rate design fixed.
5. Compare household charges at several consumption levels, for acquired customers and the buyer’s pre-existing customers in their own group.

The result would be an accounting counterfactual: the portion of the approved tariff associated with the valuation change, conditional on everything else in that rate case. It would not prove what would have happened if the sale had never occurred. A causal study with an appropriate comparison group would be needed for that claim.

An illustrative calculation shows the scale. If an additional $1 million entered rate base and were spread evenly over 1,000 accounts, an 8% allowed return plus 30-year straight-line depreciation would initially require about $9.44 per account per month, before taxes and other adjustments. Those inputs are hypothetical. Quadvest’s approved financing and any customer’s bill require their own rate-case inputs. Allocation would follow the tariff’s customer classes and usage charges.

For service, the companion measure should be quarters free of health-based noncompliance, supplemented by contaminant concentrations, monitoring completeness, boil-water days, and outage duration. That requires validated active noncompliance intervals and return-to-compliance dates; a quarter with no new violation onset cannot automatically be treated as a compliant quarter. Reporting improvements and physical water-quality improvements should be measured as different outcomes.

The dataset connects ownership, transactions, tariffs and compliance in a documented case series. A statewide trend in comparable purchase price per unique customer, an acquisition-premium burden for every transaction and a causal effect of Texas’s 2019 law remain unmeasured. Missing purchase prices, prior regulatory balances, historical household counts, and complete outage records remain substantive gaps—not zeroes.

What would change the conclusions

The most consequential unanswered question is now precise: how much extra customer revenue is generated by repricing already-existing assets, who receives the sale proceeds, and which customers carry that revenue requirement after consolidation? Texas’s public records contain pieces of that answer. They do not currently join them for the customer.

The rate pool carries the acquisition

The test of a water acquisition is the valuation customers must finance and the service they receive for it. The owner’s name is a starting point; the larger economic decision is which communities share the bill for buying and repairing the systems.