Texas toll roads: the road can work while the investment fails

Consider the North Tarrant Express, a 13.3-mile managed-lane concession on I-820 and SH 121/183 in the Dallas–Fort Worth area. Asking “Who owns it?” produces seven answers.

1. Physical ownership. The State of Texas owns the highway and right-of-way. The concession company did not buy a strip of Texas real estate. It received contractual rights within a state highway corridor.

2. Statutory control. The Texas Transportation Commission and TxDOT retain public authority over the facility: enforcement of the concession, oversight of safety and standards, approval rights over specified actions, and the ability to exercise contractual remedies. A concession is broad, but it is not sovereignty.

3. Revenue rights. NTE Mobility Partners, LLC is entitled to the managed-lane toll economics through 2061, subject to operating costs, debt service, revenue sharing, contractual restrictions, and handback requirements. That right is the investable asset.

4. Equity ownership. As of Ferrovial’s May 2026 fact book, a Ferrovial/Cintra affiliate held 62.97% and a Meridiam affiliate held 37.03%. Those percentages refer to the concession business, not 62.97% and 37.03% shares of the highway itself.^S6

5. Debt control. At December 2025, NTE reported about $1.60 billion of project debt. Its bond documents restrict distributions if debt-service coverage falls below stated thresholds. Creditors hold liens on project revenues and contractual protections that can outweigh shareholder votes during distress. They can restrict equity distributions without owning the roadway.^S6

6. Operations. The project company is responsible for operating and maintaining the managed lanes. Toll billing is handled through the North Texas Tollway Authority’s collection platform; under the disclosed arrangement, TxDOT remits amounts to the concession project within two to three business days. That shifts much of the tag and collection risk away from the private operator even though traffic risk remains with the project.^S6

7. Residual ownership. When the concession ends, the contractual toll and operating rights expire and the asset remains with the state, subject to the agreement’s handback condition. The private toll rights are finite; continuing those receipts after 2061 requires support for continuing rights. Any settlement, reserve release or other residual claim needs its own contractual basis.

Following the payment separates those rights. The state holds the pavement; lenders, shareholders and operators claim different parts of its earnings. The billing agent processes the driver’s payment, while the agreement determines who can keep it.

Value depends on enforceable rights to charge tolls, set rates, operate, access the corridor, and claim contractual protection—plus the time remaining before those rights expire.

A toll-road model begins with:

Gross toll revenue = traffic volume × average toll paid

Define traffic consistently: transactions, trips, axles, and vehicle miles are different measures. Average tolls vary by trip, time, vehicle, occupancy, tag status, and congestion. Billed revenue also differs from collected cash.

The simplified cash waterfall is:

Billed tolls
  – HOV discounts, exemptions, leakage, and unpaid bills
  – collection and customer-service costs
= collected toll revenue
  – routine operations and maintenance
  – major maintenance and lifecycle capital spending
= cash available before financing
  – interest and scheduled principal
  – required debt-service and maintenance reserves
  – government revenue sharing
  – taxes, where applicable
= cash available for equity distributions

Each deduction has a different driver.

Traffic volume responds to population, employment, trip patterns, freight, fuel prices, remote work, and the quality of free alternatives. A growing metropolitan area can still produce weak toll demand if the corridor is in the wrong place, ramps do not match origins and destinations, or development arrives later than forecast.

Average toll reflects more than posted price. On a fixed schedule, it moves with trip length, vehicle mix, and periodic rate decisions. On a managed lane, software may reprice every few minutes to preserve speed. Drivers buy access and a higher probability of arriving on time.

Collection quality affects both cost and cash recovery. Electronic tags cost less to collect and are more likely to yield payment than mailed invoices. Administrative fees can improve recovery but create political resistance. The DFW concessions are unusual because NTTA handles billing and the project companies receive rapid public remittance. Their margins are not a benchmark for roads bearing their own violation and collection risk.

Operating costs may be modest in a mature electronic system, but pavement, bridges, software, gantries, drainage, and required lane additions create large, uneven capital obligations.

The North Tarrant Express illustrates the distinction. In 2025 it reported $323 million of revenue and $278 million of adjusted EBITDA, an 86.1% margin. LBJ Express reported $244 million and $202 million, an 82.8% margin. NTE 35W reported $368 million and $294 million, a 79.9% margin. Calculated revenue per transaction was about $8.73, $5.30, and $7.08, respectively. These sponsor-reported adjusted EBITDA figures benefit from the collection arrangement. Audited cash available for debt service is a different quantity.^S6

Debt changes what those margins mean. Reported 2025 net debt was about 5.3 times adjusted EBITDA at NTE, 10.1 times at LBJ, and 5.6 times at NTE 35W. LBJ can therefore have a high operating margin and still be the most financially sensitive of the three. Interest, principal, reserve requirements, and refinancing terms sit between EBITDA and the shareholder.^S6

Revenue sharing limits equity’s upside. The DFW agreements use progressive formulas that can transfer up to 75% of revenue above specified bands to TxDOT. During 2025, disclosed government-sharing amounts included $26.4 million at NTE 35W and $8.1 million at NTE, plus sharing of refinancing gains and the LBJ “Wishbone” extension. Strong performance does not flow one-for-one to equity.^S6

Mature roads still require capital. NTE’s mandatory capacity improvement was triggered in 2023. The project disclosed roughly $355 million of works, financed partly with a $414 million private activity bond issue; construction was 67% complete at year-end 2025. An investor valuing the road on current EBITDA without the contractual expansion would overstate equity value.^S6

Pricing power is conditional

The DFW managed lanes reprice as often as every five minutes. For 2026, the disclosed “soft cap” was $1.156 per mile, indexed to CPI-U. Rates can exceed it under mandatory congestion conditions, with the system designed around a 50 mph service target. Heavy vehicles face multipliers. Qualifying high-occupancy vehicles receive discounts that are reimbursed under public arrangements.^S6

Contractual pricing power remains subject to demand:

Revenue = toll price × paid traffic

Higher rates increase revenue only when lost volume is small enough. Managed lanes may raise peak prices to preserve speed; off-peak increases can empty the road. Political tolerance can constrain rates even when contracts permit them.

SH 130 uses a different framework. Its concession ties annual toll escalation to nominal Texas gross state product per capita and does not require reductions when that measure falls. The former SH 288 contract combined dynamic pricing with annual parameters: a minimum increase equal to the greater of 2% or CPI, and a cap equal to the greater of 3% or nominal Texas GSP per capita. These clauses can provide inflation protection, but only if drivers continue to pay.^{S13, S4}

A price-volume threshold makes the demand condition explicit. In an illustration, a 10% toll increase breaks even on gross revenue only if paid traffic is at least 1/1.10, or 90.9%, of its previous level. A 5% traffic decline raises revenue 4.5%; a 15% decline lowers it 6.5%. That holds mix and trip length fixed. Dynamic pricing can also change reliability, so the response is a corridor-and-time question rather than a universal elasticity.

Even the favorable revenue case is not an equal percentage increase in equity cash. Collection terms, government sharing, lifecycle work and financing tests still intervene. I would use the contractual waterfall with the new traffic/price case before crediting the residual owner with the full gross increase.

Leverage makes the equity behave like an option

Suppose a concession is worth 100 and carries 70 of debt. Equity is 30. A 10% decline in enterprise value reduces equity to 20, a 33% loss. A 20% increase raises equity to 50, a 67% gain. The road’s physical condition can be unchanged while the equity value moves sharply.

Greenfield projects must service debt before traffic matures. A miss in year three can exhaust liquidity even if demand improves by year fifteen. Public networks can spread ramp-up risk; a stand-alone concession may lack that capacity.

Concession life adds another constraint. A road with 35 years remaining is not equivalent to one with 70 years. Every year that passes removes a year of cash flow. A valuation can include a handback payment only if the contract provides one. It cannot append a perpetual terminal value to a right that legally expires.

Why institutions still want the exposure

Pension and infrastructure funds seek long duration, entry barriers, and indexed prices. Insurers and bond investors seek contractual debt service. Banks finance construction and refinancing; distressed funds buy impaired claims and may convert them into equity.

A seasoned corridor can remove much of the construction and traffic-ramp uncertainty while leaving a price-sensitive security. The buyer still inherits remaining capital works, debt, sharing and a finite term. A brownfield cash-flow record deserves a different valuation from a greenfield forecast; neither deserves a valuation independent of its contract.

Governments use concessions to add capacity and transfer risk. A private consortium can supply equity, construction execution, and balance-sheet capacity; it can accept cost, schedule, traffic, and lifecycle risks that a public agency does not want. The structure can advance a road earlier. SH 130’s federal project record says TxDOT estimated the P3 delivered Segments 5–6 about 20 years sooner than conventional funding would have allowed.^S11

NTE and LBJ combined private equity with federal TIFIA loans, tax-advantaged bonds, and public funds. Private equity costs more than public tax-exempt debt. Acceleration, transferred risk, and operating incentives must justify that premium and the contract’s constraints.

SH 288 followed a different path. Traffic and operating margins were strong enough to attract a major brownfield investor, but the government exercised an option that ended the private equity story.

TxDOT executed the comprehensive development agreement in March 2016 with Blueridge Transportation Group. The project added four managed lanes over 10.3 miles, rebuilt major interchange components, and placed operation and maintenance obligations on the concessionaire. Substantial completion occurred in November 2020. The contractual term ran to March 2068.^{S1, S3}

The reported project cost was $1.0635 billion. Financing included $298.6 million of private activity bonds, a $357.0 million TIFIA loan, $14.9 million of capitalized TIFIA interest, $17.1 million of public funds for the Texas Medical Center connection, and $375.3 million of developer and third-party equity. Original participants included ACS Infrastructure, InfraRed, Shikun & Binui, Northleaf, Clal, and Star America.^S3

By late 2023 the cap table had consolidated. Abertis acquired 56.76% for about $1.53 billion; ACS retained 43.24% through ACS SH288 Holdings, LLC. The acquisition presentation reported 2023 pro forma revenue of $90 million, adjusted EBITDA of $66 million, and gross debt of $654 million at December 31, 2022. The lanes used dynamic pricing, with contractual annual parameters tied to CPI and Texas nominal GSP per capita.^S4

Section 31 gave the state a termination-for-convenience right. In July 2024 the Texas Transportation Commission determined that exercising the right was in TxDOT’s interest. The concession ended on October 8. Control reverted to TxDOT, and the Texas Transportation Finance Corporation was authorized to develop and operate the project. The state funded the $1.7317 billion payment through a State Highway Fund loan, later refinanced with roughly $1.7 billion of TTFC toll-revenue and refunding bonds.^{S1, S2, S22}

The federal project record specifies the waterfall: termination proceeds first repaid all outstanding debt, including TIFIA, and the balance was available to shareholders. The TIFIA loan was repaid in full on the termination date. Abertis reported receiving $642.365 million and recording a €775.857 million pre-tax loss, or €581.892 million after tax.^{S3, S5}

Each valuation comparison has limits.

Implied equity value at Abertis entry: $1.53 billion divided by 56.76% equals approximately $2.70 billion for 100% of the equity. Adding the separately reported $654 million of gross debt produces an indicative enterprise value of $3.35 billion. This mixes a late-2023 transaction price with debt measured at year-end 2022, so it is a rough inference, not a closing balance sheet. Termination payment multiples: $1.7317 billion equals 19.2 times 2023 pro forma revenue and 26.2 times adjusted EBITDA. Those are gross transaction multiples because the payment covered debt and equity; they are not equity multiples. Estimated gross equity pool: subtracting approximately $650 million of debt from the payment leaves about $1.08 billion before fees, reserves, hedges, working-capital adjustments, and other claims. The actual shareholder distributions need the closing statement. Value per remaining concession year: the payment divided by approximately 43.4 remaining years is $39.9 million per year. This descriptive ratio ignores when the cash arrives; early cash flow is worth more than late cash flow. Payment versus original equity: the state’s gross payment was about 4.6 times* the original $375.3 million equity contribution. That does not mean the original sponsors earned 4.6 times. Debt was repaid first, ownership changed, capital may have been added or distributed, and the money was invested over years.

Abertis’s loss is a result for a particular buyer, entry price and holding period. The lender’s recovery is a result for a different claim. Texas acquired the ability to redirect decades of service and revenue, while financing the purchase itself. Collapsing those outcomes into “the private sector won” or “the state bought the road” loses the transaction’s most useful information. The closing waterfall and each investor’s dated cash flows are the right units of comparison.

For fiscal 2025, TxDOT reported $107.35 million of SH 288 operating revenue, $11.36 million of operating expense, and $2.73 million of maintenance expense, excluding the later bond refinancing treatment. TxDOT has said public control should permit lower tolls and slower escalation than under the former concession. HCTRA supports billing and account functions; TTFC holds the public operating and revenue mandate.^{S21, S22}

SH 288’s buyout cannot price NTE, LBJ, NTE 35W, or SH 130. Their termination language, toll rules, debt, remaining lives, and obligations differ. SH 130 also has a federal equity holder. Any comparison requires the relevant concession agreement and compensation provisions.

Public financing can use tax-exempt debt, pooled network pledges and an indefinite public operating life. A mature system can carry a new link through its ramp-up. Those advantages can lower financing cost while also making a weak individual project harder to see inside a strong network.

Private equity can absorb cost, schedule or demand losses and give one counterparty responsibility for lifecycle work. The original SH 130 equity loss is a concrete reason to take that allocation seriously. The public side still has to identify what it retained: federal credit exposure, public contributions, compensation promises, collection support and the cost of monitoring a long contract.

The useful comparison holds the delivered road, opening date and lifecycle standard constant. It then values the differences in construction risk, maintenance incentives, demand exposure, financing and flexibility. A lower public borrowing rate cannot by itself price earlier delivery or a genuine transfer of overruns. A high private EBITDA margin cannot by itself establish that transfer occurred.

There is a strong case for private participation when public constraints would otherwise delay useful capacity and the contract places manageable risks with a party able to control them. There is a weaker case when the public supplies most protection while paying an equity premium for risks it has effectively kept. The answer requires the project’s continuing public obligations to be identified. An ownership label cannot supply that account.

The analogy to a bond, utility or property helps only partway. Location persists, while the concession’s cash right ends. Traffic remains a choice, and late-life maintenance can consume cash before handback. The investor owns the contract’s residual after those claims. The public is purchasing a service and a risk allocation that should be compared with its available alternative.

Compare the public and private structures in the data appendix.

Texas has favorable population trends. The Census Bureau estimated 31.71 million residents on July 1, 2025, up 8.8% from the April 2020 base. Dallas–Fort Worth reached roughly 8.5 million, up about 11% over that period, with substantial growth in outer counties. Houston and DFW posted some of the country’s largest one-year metropolitan population gains in 2023–24.^S23

The state’s activity centers are dispersed. Jobs, homes, airports, warehouses, ports, and industrial sites are separated by long road trips. Transit is limited for many suburb-to-suburb routes. Texas also carries substantial freight: the state freight plan reported roughly 4 billion tons in 2019 and projects more than 8 billion by 2050. Congestion data show that most of Texas’s worst bottlenecks remain concentrated in the four largest metropolitan areas.^{S24, S25}

These conditions favor toll options but cannot rescue a badly placed road.

Value depends on trips: origins, destinations, bottlenecks, ramps, and minutes saved. A driver may pay $12 for reliability before a flight yet refuse $3 on a quiet weekend. Freight demand requires routes and toll costs that fit the shipper’s economics.

Free-road capacity is the clearest substitute. Widening a parallel highway can reduce toll demand, though construction disruption can temporarily increase it. Contractual protection against competing facilities is never absolute; governments preserve rights to improve their networks. SH 130’s agreement, for example, contained a competing-facilities zone but excluded I-35 and projects already in regional plans.^S12

Remote work changes commute frequency and peak timing. Fuel prices increase bypass costs. Development generates traffic only after occupancy and road connections; freight growth helps only where vehicle rules and logistics routes fit.

The valuation-input matrix lists the ten questions behind the investment decision.

Investment access, entry price, and the security’s rights matter as much as the corridor. A valuable road can offer no available equity or only a bond with limited upside.

The private concession table compares the contracts and capital structures of the five principal concessions.

NTE offers a seasoned corridor, strong reported cash flow, and about 35 years of term. Cumulative project distributions through 2025 were about $1.172 billion, 2.75 times original project equity before considering timing, additional capital, taxes, or residual value. The mandatory-capacity program and government sharing reduce free upside. A buyer would most plausibly enter through a sale by Meridiam, a sponsor-level transaction, or project bonds; Ferrovial’s listed shares provide diluted exposure across a much larger portfolio.^S6

LBJ has an attractive urban bottleneck and 2025 adjusted EBITDA of $202 million, but its roughly 10.1 times net-debt-to-EBITDA ratio leaves less room for error. Cumulative distributions of $952 million exceeded original equity in nominal dollars, yet that statistic says little about today’s entry price. APG- or Meridiam-managed interests could theoretically trade in the secondary market. At the cutoff, no reviewed evidence showed them offered for sale.^S6

NTE 35W produced the highest 2025 revenue of the three DFW concessions, $368 million, and has benefited from staged expansion. It also paid the largest disclosed 2025 revenue share. Its acquisition routes resemble LBJ’s: sponsor stakes, infrastructure-fund or pension secondaries, project bonds, or a recapitalization. The main valuation question is how much growth remains after sharing and future corridor obligations.^S6

SH 130 is the more specialized opportunity. It has become a freight and reliability asset after a failed greenfield ramp-up. SVP’s continuation-vehicle process shows that institutional secondary capital can access it, but the exact 2026 debt, distributions, and cap table are not public. The U.S. government’s minority interest and any TxDOT transfer consent narrow the buyer universe. Upside comes from traffic, freight, adjacent development, and a possible extension; downside comes from repeating the original mistake by capitalizing those possibilities before they are contractual or realized.

SH 288 no longer offers concession equity. A financial investor can consider TTFC toll-revenue bonds, service contracts, or indirect exposure through firms operating for the state. These provide debt or service income. A bondholder receives contracted debt service. The residual toll upside that Abertis expected belongs to the equity claim.

Public systems mainly offer municipal debt with different pledges, tax support, coverage, and maturities. TIFIA is a federal lending program. Distressed exposure requires a purchasable project claim and enforceable restructuring rights.

A proven concession with 35 years left can disappoint at a 4% equity yield and become attractive after deleveraging. Public efficiency does not give bondholders residual upside. Assess the available security’s price against traffic, debt, maintenance, contractual, and political risks.

Texas’s toll-road examples support a more precise judgment than either ownership slogan. Public title protects the public asset; it does not identify every financial obligation the public has accepted. Private equity can bear a real loss; that loss does not erase federal lending or contractual compensation. The contract determines the division.

SH 130’s original investors lost their equity while the public road remained open. SH 288’s lenders were repaid while a recent shareholder suffered an entry-price loss. A future investor can buy only an available security with its actual claims, at its actual price. The existence of an attractive corridor creates neither access to that residual nor a guaranteed return.

I would begin any purchase or policy comparison by tracing the enforceable cash right and its ending, then testing traffic, lifecycle work, senior claims and the public obligations left around it. That is where a toll dollar becomes an investment return—or a public commitment somebody else must finance.