SOC와 금융의 이야기

#45 – Why Did the U.S. P3 Market Grow First in Roads and Bridges?

foweuniverse — Sun, 6 Sep 2026 12:30:25 +0900

Korean Original
SOCFI #45 – 미국 P3 시장은 왜 도로·교량에서 먼저 커졌을까?
https://socfi.tistory.com/49

#45 - 미국 P3 시장은 왜 도로·교량에서 먼저 커졌을까?

- 한국과 미국의 P3에서 왜 도로·교량의 투자비 비중이 높을까요? 한국 민자사업의 도로 투자비와 민자고속도로 현황, 미국 Highway P3·TIFIA 통계를 비교하고 Greenfield Toll Road, Express Lane, Toll Concession

socfi.tistory.com

- Why did roads and bridges become the early center of the U.S. P3 market? This article compares Korean and U.S. infrastructure data and explains how toll revenue, Express Lanes, DBFOM, Availability Payments, and federal financing tools such as TIFIA and PABs made highway P3s easier to structure. It also looks at why rail, ports, and airports can involve more complex interfaces and contractual boundaries. -

When people think about PPP projects in Korea, roads are one of the first sectors that come to mind.

The numbers support that impression.

According to the government’s 2024 Status of Public-Private Partnership Projects, Korea had promoted 872 PPP projects with a total value of KRW 153.5 trillion between 1994 and 2024.

Of these, 284 were revenue-type projects such as BTO, while 588 were lease-type projects such as BTL.

Looking at the cumulative sector breakdown available for 2022, there were 818 projects totaling KRW 125.7079 trillion.

Road projects accounted for only 66 projects, or about 8.1% of the total by number.

But in investment value, roads accounted for KRW 47.465 trillion, or about 37.8%.

Rail followed with 18 projects totaling KRW 28.5101 trillion.

So roads were not the most numerous PPP projects.

But in terms of the amount of money invested, they were overwhelmingly important.

The same pattern can be seen in Korea’s expressway network.

The Korea Transport Institute’s Private Road Management Support Center currently describes the total length of privately financed expressways as 977 kilometers.

Including routes under development and construction, privately financed expressways account for roughly 22% of Korea’s total expressway network.

The United States shows a similar pattern.

91 Express Lanes, Dulles Greenway, I-495 Express Lanes, North Tarrant Express, Elizabeth River Tunnels, I-4 Ultimate, SR 400 Express Lanes....

Many of the major P3 projects discussed so far have been roads, bridges, or tunnels.

That does not mean most U.S. roads are P3s.

FHWA’s review of the U.S. Highway P3 market found only 28 long-term Highway P3 Concessions that had reached Financial Close between 1992 and 2016.

Compared with the enormous size of the U.S. highway network, this was still a very selective market.

But once we look specifically at major P3 and infrastructure-finance transactions, the picture changes.

A 2016 USDOT TIFIA report showed that, among 56 projects that had received TIFIA credit assistance at the time:

37 projects, or 66.1%, were Highway projects

14 were Transit projects

5 were Intermodal projects

This should not be read as saying that highways represented 66.1% of all U.S. P3 projects.

It does show, however, how prominent roads were in the early large-scale U.S. infrastructure-finance market.

So the question in this article is:

Neither Korea nor the United States turned their entire road networks into P3s. Why, then, did roads and bridges account for such a large share of major P3 investment?

Roads Can Convert Usage Into a Cash Flow That Repays Debt

The most important question in P3 finance is actually quite simple.

“What will repay the money borrowed at the beginning over the next 20 or 30 years?”

Roads have a relatively intuitive answer.

Tolls.

When vehicles use the road, users pay tolls.

Past traffic volumes, population and employment growth, surrounding road networks, and expected toll levels can be used in a Traffic & Revenue Study to estimate future traffic and toll revenue.

Lenders then use that forecast cash flow to decide how much Debt the project can support.

This should not be interpreted as saying that road-demand forecasting is easy.

Early privately financed toll roads such as the Dulles Greenway experienced long-running issues because actual traffic and financial performance did not meet expectations.

Even in 2025, the operator sought a toll increase, but the Virginia State Corporation Commission rejected the request, and the Supreme Court of Virginia upheld that decision.

The real advantage of roads is more specific:

The link between usage and revenue is relatively clear.

Federal law also recognizes this structure.

23 U.S.C. §129 provides a legal framework for tolling certain Federal-aid highways, bridges, and tunnels.

23 U.S.C. §166 allows certain general-purpose vehicles to use HOV facilities by paying tolls under specified conditions.

Courts have also addressed the legal character of tolls.

In Elizabeth River Crossings OpCo, LLC v. Meeks (2013), residents argued that allowing a private operator to collect tolls amounted to delegating taxing power.

The Supreme Court of Virginia rejected that argument.

The court treated the project tolls as user fees, not taxes.

At the same time, it held that Virginia had not surrendered its broader regulatory authority simply by entering into a contract with a private concessionaire.

For P3 finance, that distinction matters.

Money paid by users of a specific road can be linked directly to that road’s long-term project cash flow.

Existing Toll Lanes Were Chosen More Often Than Entirely New Toll Roads

Looking more closely at early U.S. Toll P3s reveals something interesting.

Among the Highway P3s that had reached Financial Close by 2016, FHWA identified 14 Real Toll Concessions where the private sector actually bore Toll Revenue Risk.

Those 14 projects fell into three broad categories.

Of the 14 projects, 9 were Priced Managed Lanes.

That is about 64%.

Let’s unpack the terminology.

A Greenfield Toll Road is a completely new toll road built where no comparable facility previously existed.

In simple terms, the private sector builds a new highway from scratch and collects tolls across the whole facility.

Representative examples include:

Dulles Greenway in Virginia

SH 130 Segments 5–6 in Texas

FHWA listed the approximate project sizes at:

Dulles Greenway — $355 million

SH 130 Segments 5–6 — $1.336 billion

A Fixed Crossing refers to a bridge or tunnel crossing a river, bay, or other physical barrier.

A representative example is Virginia’s Elizabeth River Tunnels.

FHWA listed its project size at approximately $2.088 billion.

The third category, Priced Managed Lanes, became particularly important in the United States.

A Managed Lane is a lane within an existing highway that is operated under different rules from the general-purpose lanes.

A Priced Managed Lane allows users to pay for access.

In simple terms:

The free lanes remain, while drivers can choose to pay for a faster lane.

Projects such as:

91 Express Lanes

I-495 Express Lanes in Virginia

North Tarrant Express in Texas

fit this model.

FHWA’s 2016 data listed project sizes of approximately:

I-495 Capital Beltway HOT Lanes — $2.069 billion

North Tarrant Express Segments 1 & 2A — $2.122 billion

IH-635 Managed Lanes — $2.645 billion

Express Lanes Created a New Paid Service Without Eliminating the Existing Free Road

Why did this structure matter so much?

A completely new Greenfield Toll Road has no real operating history.

At opening, the project must estimate from scratch:

How many drivers will actually use it?

How much will they be willing to pay?

What happens if they continue using competing free roads?

An Express Lane added to an already congested Interstate Highway starts from a different position.

The government already has substantial data on:

Traffic volumes

Peak commuting periods

Travel patterns

Congestion levels

And the driver still has a choice.

They can stay in the congested general-purpose lanes for free.

Or they can pay a toll to use the faster lane.

This is also where the term HOT Lane appears frequently.

HOT stands for High-Occupancy Toll Lane.

It is easiest to understand as an HOV—High-Occupancy Vehicle—lane that also allows other vehicles to enter by paying a toll.

Current federal law under 23 U.S.C. §166 provides a framework for this type of operation.

More recent systems do not rely only on fixed toll prices either.

Dynamic Pricing can raise tolls when congestion is severe and lower them when capacity is available.

This allows the operator to manage traffic flow in the Express Lane itself.

FHWA permits electronic tolling and variable pricing as congestion-management tools in Managed Lanes.

So an Express Lane is not simply:

“A public highway handed over to a private operator.”

It is closer to this:

An additional paid service is created inside an existing public road system, while the free road remains available.

Roads and Bridges Make Long-Term Performance Easier to Measure

There is another reason roads fit P3s well.

Their long-term performance can be defined relatively clearly in a contract.

Let’s return to DBFOM.

Under a traditional construction contract, the contractor may finish the road, hand it over to the government, and largely complete its role.

DBFOM is different.

If the private partner must operate and maintain the road for another 30 or 40 years, it has an incentive to think about long-term maintenance from the design stage.

Road performance can also be measured using fairly concrete indicators:

Pavement condition

Lane availability

Facility failures

Incident-response time

Bridge condition

Cleaning and maintenance standards

FHWA also explains that Highway P3 contracts can combine not only design, construction, and financing, but also operations and maintenance.

Florida’s I-595 Corridor is a good example.

It is often described as the first U.S. transportation Availability Payment P3.

FDOT entered into a 35-year agreement with the private partner covering design, construction, finance, operations, and maintenance.

Under the contract, payments could be reduced if the facility was unavailable or if O&M obligations were not met.

So the government was not simply buying:

“One completed road.”

It was effectively buying:

A road service that would remain available at the promised performance level for decades.

If Demand Risk Is Hard to Transfer, the Contract Can Shift to Availability Payments

Another strength of road P3s is that tolling does not automatically require the private partner to bear all traffic-demand risk.

U.S. Highway P3s could broadly use two different models.

The first is a Toll Concession.

The private partner collects toll revenue.

If traffic is lower than expected, the private partner bears the financial consequences.

The second is an Availability Payment, or AP structure.

Instead of paying the private partner based on the number of cars using the road, the government pays based on whether the road is available and meets the contracted performance standards.

FHWA’s 2016 review of 28 Highway P3s identified 9 Availability Payment Concessions.

Florida’s I-4 Ultimate is a representative example.

The project includes Express Toll Lanes.

But the private partner does not keep Toll Revenue.

FDOT manages the toll revenue and makes long-term Availability Payments to the private partner.

So the road P3 structure can be simplified like this:

Private sector can efficiently bear Toll Revenue Risk → Toll Concession

Transferring demand risk would be inefficient → Availability Payment

That flexibility in Risk Allocation helped expand the range of projects that could use P3 delivery.

Bridges and Tunnels Also Fit P3s Because They Combine High Upfront Cost With Long-Term Maintenance

Bridges and tunnels fit P3s for a somewhat different reason.

They require large amounts of capital at the beginning.

And after construction, structural safety and maintenance remain important for decades.

So the key question becomes less:

“Who can build this at the lowest initial price?”

and more:

“Who can design, build, and maintain this asset most efficiently over its full lifecycle?”

A representative example is the Goethals Bridge Replacement.

The project replaced the bridge connecting New York and New Jersey.

USDOT described it as the first Surface Transportation P3 in the northeastern United States to combine Project Finance with long-term maintenance.

Its payment mechanism used Availability Payments.

This is important.

Even where no Toll Revenue is transferred to the private partner, a bridge P3 can still make sense because:

Large upfront investment + long-term maintenance responsibility

can be bundled into a single long-term contract.

TIFIA and PABs Made Highway P3 Finance Even Stronger

As discussed in #44, the United States developed federal financing tools that work particularly well with long-term infrastructure cash flows.

TIFIA provides long-term federal credit assistance.

PABs—Private Activity Bonds allow qualifying privately delivered infrastructure projects to use tax-advantaged bonds.

Long-term Debt ultimately raises the same question:

“What will repay this money 20 or 30 years from now?”

Highway P3s had relatively clear answers.

Toll Revenue

Availability Payments

That allowed long-term federal financing to match long-term project cash flow.

This helps explain why, in the 2016 TIFIA portfolio, 37 of 56 projects were Highway projects.

This does not mean the federal government supported only highways.

TIFIA also supports Transit and other transportation sectors.

The point is narrower:

At the time the U.S. P3 market was expanding, large Highway P3s with Toll or AP cash flows matched particularly well with TIFIA and PAB financing.

But Most U.S. Roads Are Still Not P3s

At this point it would be easy to think:

“So does the United States use P3s for most large road projects?”

No.

FHWA identified only 28 long-term Highway P3 Concessions that had reached Financial Close by 2016.

Even FHWA’s current list of P3 New-build projects remains limited compared with the overall U.S. road program.

So P3 is not the default method of road delivery in the United States.

It has been used selectively where several conditions line up:

The project is large

Private financing is useful

There is a long-term repayment source such as Toll Revenue or AP

Bundling design, construction, and long-term maintenance creates value

So it is more accurate to say:

Not that roads are inherently “P3 projects,” but that some large road projects make it relatively easy to create the conditions under which a P3 can work.

Why Is Rail More Complex?

Rail can absolutely be delivered through P3s.

A major U.S. example is Denver’s Eagle P3.

RTD—the Regional Transportation District—used a 34-year DBFOM arrangement covering design, construction, financing, operations, and maintenance.

But there is an important difference from a typical Toll Road Concession.

RTD retained ownership of the rail assets.

RTD also controlled fares and fare policy.

Fare revenue remained with RTD.

The private partner instead received monthly Availability Payments based on system availability and performance.

Why not simply transfer fare revenue to the private partner the way a Toll Road might transfer Toll Revenue?

Because rail systems involve many more interconnected systems.

Track

Rolling stock

Signals

Power supply

Stations

Train operations

Scheduling

Transfers

Safety management

If a train is delayed, the cause may be:

the track,

the train,

the signaling system,

another operator,

or a connection with another part of the network.

So the core difficulty of rail P3s is not that they cannot generate revenue.

The harder problem is defining the Interface—the boundary of responsibility between public and private parties.

That can make the contract more complex than a road project.

Ports Can Earn Revenue, but Cargo Demand and Operating Interfaces Are More Complicated

Ports also have many characteristics that can work well with private investment.

A container terminal can generate:

Cargo-handling fees

Lease revenue

Other operating income

One term that appears frequently in port projects is Berth.

A Berth is the location where a vessel docks to load or unload cargo.

So when a private company operates a specific Berth, it does not necessarily mean the entire port has been privatized.

It may mean that a single terminal or part of the port is given to a private operator under a long-term arrangement.

In that sense, port P3s can work quite well.

But port revenue can depend heavily on factors that are less local and predictable than commuter road traffic:

Global trade volumes

Shipping-line route decisions

Competing ports

Cargo owners

Changes in logistics networks

And multiple actors often operate at the same time:

Port Authority

Terminal Operator

Shipping lines

Railroads

Trucking companies

U.S. port infrastructure also operates under a separate federal legal framework.

46 U.S.C. §54301 establishes the Port Infrastructure Development Program and federal support for port and intermodal infrastructure.

So the challenge is not that ports cannot use P3s.

It is that:

Defining which revenues and which operating functions should be bundled into a single long-term contract can be more complicated than for one highway segment.

Airports Also Generate Strong Cash Flow, but the Contract Boundary Can Be Much Wider

Airports have many revenue sources.

Landing fees

Airline charges

Parking revenue

Terminal leases

Commercial concessions

From a financial perspective, airports can be attractive infrastructure assets.

But transferring the operation of an entire airport to a private party brings far more regulatory and contractual issues than a simple construction project.

The United States has a separate statutory framework under 49 U.S.C. §47134, the Airport Investment Partnership Program.

Even when a private party acquires or leases airport operating rights, separate requirements can apply, including commitments to maintain, improve, and modernize the facility.

This is a much broader contract boundary than an Express Lane.

Separating a few lanes on an Interstate Highway and operating them under a P3 is one thing.

Coordinating:

Airport operations

Safety regulation

Airline agreements

Federal grants

Terminal operations

is another.

So airports were not less common in U.S. P3s because P3 delivery was impossible.

The contractual scope and public-service interfaces were often simply broader and more complex.

Comparing the Sectors Makes the Advantage of Roads Easier to See

The point of this comparison is not:

“Roads are good, while rail, ports, and airports are bad.”

The important distinction is this:

In the early P3 market, roads made it relatively easier to define the project boundary, create a revenue stream, measure long-term performance, and specify which risks could be transferred to the private sector.

Roads Became the Center of P3s for More Than Just One Reason

Let’s return to the numbers.

In Korea, roads accounted for only 66 of 818 PPP projects in 2022, about 8.1%.

But they represented KRW 47.465 trillion, or about 37.8% of total investment value.

In the United States, Highway P3s did not dominate overall road construction.

But they remained central to large P3 transactions and federal P3 finance.

FHWA separately analyzed 28 Highway P3 Concessions that had reached Financial Close by 2016.

During the same period, 66.1% of TIFIA-supported projects were Highway projects.

Why?

I would summarize it in five points.

First, Toll Revenue created an independent project cash flow.

Second, Express Lanes created a paid service without removing the existing free road.

Third, roads made it relatively easy to bundle 30–40 years of operations and maintenance into a DBFOM contract.

Fourth, Risk Allocation could shift between Toll Concessions and Availability Payments depending on who should bear demand risk.

Fifth, long-term infrastructure financing such as TIFIA and PABs matched well with those long-term repayment sources.

So the U.S. P3 market did not first grow around roads and bridges simply because the United States is a car-oriented country.

Roads were relatively easy to turn into stand-alone projects with long-term cash flow, measurable performance, and contractually divisible risks.

Rail, ports, and airports were not less suitable because P3s were impossible.

Their challenge was often that:

The boundaries among revenue, public service, operators, and private-sector responsibility were more complicated to define.

This article brings us to the end of the basic history and institutional structure of U.S. P3s.

From the next article, I will move more deeply into the financing that actually drives these projects.

We now know that tools such as TIFIA and PABs exist.

The next questions are more financial:

Who actually provides the Equity?

What do banks and bond investors look at before lending?

How are Senior Debt and Subordinated Debt divided?

That is where the next part begins.

Key Sources

Government of Korea – 2024 Status and Performance of Public-Private Partnership Projects
Korea Transport Institute – Private Road Management Support Center
FHWA – Report on Highway P3 Concessions in the U.S.
USDOT – 2016 TIFIA Report to Congress
23 U.S.C. §§129, 166
Elizabeth River Crossings OpCo, LLC v. Meeks (Va. 2013)
Build America Bureau – Eagle P3 / Goethals Bridge
49 U.S.C. §47134
46 U.S.C. §54301

#44 – What Role Does the Federal Government Play in U.S. P3s?

foweuniverse — Sun, 6 Sep 2026 11:33:42 +0900

Korean Original

SOCFI #44 – 연방정부는 P3에서 무슨 역할을 할까?
https://socfi.tistory.com/48

#44 - 연방정부는 P3에서 무슨 역할을 할까?

- 미국 연방정부는 P3 사업을 직접 발주하지 않으면서 어떻게 금융을 지원할까요? INFRA·Mega 같은 보조금부터 TIFIA·RRIF 장기대출, PABs 세제혜택 채권까지 연방 금융지원 체계를 한눈에 정리하고, I-

socfi.tistory.com

- How does the U.S. federal government support P3 projects without directly procuring them? This article explains the federal financing framework—from INFRA and Mega grants to TIFIA and RRIF long-term credit and PABs tax-advantaged bonds—and looks at how federal support actually entered I-4 Ultimate, North Tarrant Express, and SR 400. It also considers what this structure could mean for Korea’s PPP financing policy. -

In #43, we saw that U.S. P3s can take very different forms depending on the laws and procuring agencies of each state.

In Korea, the central government is also an important player in PPPs.

The Ministry of Planning and Budget administers the Basic Plan for Public-Private Partnerships and the broader PPP framework, and in some projects the central government itself serves as the competent authority and enters into the concession agreement with the private party.

Financial and fiscal support tools such as construction subsidies and the Infrastructure Credit Guarantee Fund are also used.

In 2026, the Ministry of Planning and Budget is continuing to review revisions to the Basic Plan and measures to promote PPP investment through the Private Investment Project Review Committee.

The United States works somewhat differently.

As discussed earlier, the actual contracting party in a P3 is usually a State DOT, local government, or another public agency.

FHWA—the Federal Highway Administration—or USDOT—the U.S. Department of Transportation—generally does not sign the P3 contract itself.

Instead, the federal government becomes deeply involved through financing, federal funding, environmental and legal requirements, technical assistance, and credit review.

The Build America Bureau also describes U.S. P3s as contracts between public agencies and private parties, while the federal government helps state and local Project Sponsors use financing and P3 tools.

So when looking at a U.S. P3, I think it helps to separate two different questions.

Who decides to pursue the project as a P3 and signs the contract? → State and local governments

Who helps make the project financially viable? → Federal financing, grants, and tax-supported tools

This article focuses on the second question.

First, I will look at the main federal financing tools together.

Then I will follow actual projects to see how the process worked, who made the decisions, and how much federal financing was involved.

Federal P3 Support Is Not Just One Grant — It Is a Combination of Financing Tools

The main federal tools that can be used in U.S. transportation P3s can be summarized as follows.

The key organization connecting these programs is the Build America Bureau.

The Build America Bureau is USDOT’s specialized infrastructure finance office.

Today, the Bureau brings together TIFIA, RRIF, PABs, and technical assistance in one place and encourages state and local governments and Project Sponsors to engage with it as early as possible.

This is quite important.

The Build America Bureau is not simply a policy lender that waits until the private concessionaire has already been selected and then reviews a loan application.

It can become involved during the early project-development stage and ask questions such as:

“Which federal financing tools can this project use?”

“If this project is delivered as a P3, when should the financing process begin?”

“How should private capital and federal credit be combined?”

That early involvement is part of its role.

The Federal Government’s First Tool Is Money That Does Not Have to Be Repaid: Grants

The easiest tool to understand is a Grant, meaning direct federal grant funding.

Representative programs include INFRA and Mega.

INFRA stands for Infrastructure for Rebuilding America and is a competitive federal grant program.

It supports major transportation projects of national or regional significance, including highways, bridges, and freight infrastructure.

USDOT is continuing separate INFRA competitions in 2026.

Mega is the statutory National Infrastructure Project Assistance Program.

As its name suggests, it supports very large and complex nationally or regionally significant projects that may be difficult to deliver through existing funding programs alone.

The BUILD Grant program also competitively supports transportation infrastructure of local and regional significance.

In 2026, approximately $1.73 billion was selected for 127 projects.

The biggest difference between these Grants and TIFIA is simple.

They do not have to be repaid.

Suppose a project costs $1 billion and the federal government provides a $200 million Grant.

That $200 million becomes final Funding for project costs.

If the project has sufficient economic value but toll revenue or Availability Payments alone cannot cover the full cost, a Grant can fill part of the Funding Gap.

The Second Tool Is TIFIA — the Federal Government Becomes a Long-Term Lender

One of the financing tools that appears most often in U.S. P3s is TIFIA.

TIFIA stands for the Transportation Infrastructure Finance and Innovation Act.

Literally translated, it is roughly the Transportation Infrastructure Finance and Innovation Act, but in practice the term usually refers to the federal long-term transportation credit program created under that law.

The simplest way to understand it is this:

The U.S. government itself joins the P3 project-finance lending group as a lender.

TIFIA can generally provide credit assistance for up to 49% of eligible project costs.

It offers a fixed interest rate tied to the federal government’s Treasury borrowing cost.

Principal and interest payments can generally be deferred for up to five years after substantial completion, and the repayment period can generally extend up to 35 years, with longer periods available for some projects.

But TIFIA’s real advantage is not just a low interest rate.

Infrastructure projects do not necessarily generate stable cash flow immediately after construction.

A Toll Road, for example, may need a Ramp-up Period after opening before traffic reaches expected levels.

If commercial bank principal repayments begin immediately, the project can face heavy cash-flow pressure during its early years.

TIFIA is valuable because it can structure repayment timing and maturity around the long-term cash-flow characteristics of infrastructure projects.

But TIFIA Does Not Mean the Government Takes Over an Unviable Project

To receive TIFIA assistance, a project must have a Dedicated Revenue Source.

There must be a clearly identified source of money that will repay the TIFIA loan.

For a Toll Concession, that could be toll revenue.

For an Availability Payment project, it could be the long-term AP—Availability Payments made by the state government in return for keeping the facility available at the required performance level.

The project must also satisfy a certain level of Creditworthiness, meaning the ability to repay the loan.

So TIFIA should not be understood as:

“The project is not financially viable, so the government puts in money instead.”

It is closer to this:

“The project has long-term repayment capacity, but private financing alone is too expensive or difficult to structure, so the federal government enters as a long-term lender and improves the financing structure.”

The Third Tool Is RRIF — Similar to a Rail Version of TIFIA, but With Different Eligibility

When looking at road P3s, TIFIA appears frequently.

In rail projects, another acronym appears: RRIF.

RRIF stands for Railroad Rehabilitation and Improvement Financing.

In Korean terms, it can be understood roughly as a financing program for rail rehabilitation and improvement.

USDOT can currently provide up to $35 billion in direct loans or loan guarantees through RRIF.

At least $7 billion of that authority is reserved to some degree for railroads other than the largest Class I freight railroads.

RRIF can support tracks, bridges, rolling stock, rail facilities, new rail infrastructure, and certain TOD—Transit-Oriented Development projects.

For eligible rail projects, financing can cover up to 100% of project costs, with repayment terms of up to 35 years.

Very simply:

Roads, transit, and a broad range of transportation projects → TIFIA

Rail and certain rail-related development → RRIF

Actual eligibility still has to be checked project by project.

The Fourth Tool Is PABs — Instead of Lending Money, the Government Opens Access to a Lower-Cost Bond Market

Another important tool is PABs.

PABs stands for Private Activity Bonds.

A literal translation such as “private activity bonds” does not immediately explain why they matter.

The key idea is:

Eligible privately delivered public infrastructure projects are allowed to use bonds that receive federal tax advantages.

Under qualifying conditions, interest received by PAB investors is exempt from federal income tax.

Because investors pay less tax, they may accept lower interest rates than they would on otherwise comparable taxable bonds.

That reduces the private project company’s financing cost.

So the difference is:

TIFIA → the federal government directly lends money

PABs → the federal government allows access to a tax-advantaged bond market

PABs Are Not Bonds Issued by the Federal Government

PABs are generally issued into the capital markets through a state or local Conduit Issuer.

The federal government itself does not issue the bonds.

And the private P3 project company is responsible for repaying principal and interest.

For a Toll P3, repayment may come from toll revenues.

For an AP P3, repayment may ultimately rely on the Availability Payments received by the private project company.

USDOT also explicitly states that PABs are not direct federal Financial Assistance.

There is another interesting point: timing.

According to USDOT, in many P3 projects the State DOT applies for the PAB allocation on behalf of the future private concessionaire before the private party has been selected.

Why?

Because bidders need to know:

“How much lower-cost PAB financing can we assume is available for this project?”

Only then can they design their financing plan and submit a price.

PABs are therefore not simply a financing product added after the winning bidder is selected. They can be considered from the early stages of structuring the P3 procurement itself.

With PABs, “The Program Exists” and “A New Project Can Use It” Are Not Always the Same Thing

According to USDOT public information as of November 2025, the statutory cap for transportation infrastructure PABs was $30 billion.

Approximately $23.9 billion had already been issued.

The remaining approximately $6.1 billion had also been allocated to projects.

The amount shown as available for additional allocation was therefore $0.

So when evaluating a new P3, it would be a mistake to assume:

“PABs exist in the United States, so this project can use them.”

At the actual project stage, sponsors have to check whether Congress has increased the cap, whether previous allocations have been returned, or whether additional capacity has otherwise become available.

The Federal Government Can Ultimately Combine Grants, Credit, and Tax Benefits

Putting all of these tools together shows how federal financing can work as a system.

Suppose a major P3 requires $5 billion.

Part of the project cost could be supported with a Grant.

Part of the remaining financing could come from a long-term TIFIA loan.

Private Debt could use PABs.

Commercial Bank Debt and private Equity could then sit on top of those sources.

A U.S. P3 Capital Stack—the different layers of funding used to pay for the project—might look like this.

This is why it makes more sense to view the federal role as a financing platform rather than as a collection of separate programs.

The Build America Bureau connects TIFIA, RRIF, PABs, and technical assistance within one organization and encourages projects to engage with the Bureau as early as possible.

Who Inside the Federal Government Actually Decides Whether to Provide This Financing?

At this point another question comes up.

“Does a Build America Bureau staff member approve the loan?”

No.

Large federal credit assistance such as TIFIA goes through several stages of federal decision-making.

The Build America Bureau works with the Project Sponsor and reviews the financing structure.

A Credit Programs Team examines repayment capacity and the project’s financing plan.

A major decision-making body after that is the Council on Credit and Finance within USDOT.

The Council is chaired by the Deputy Secretary of Transportation.

It reviews applications for Bureau credit assistance and makes recommendations to the Secretary of Transportation.

The Council generally meets monthly.

The final decision on federal credit assistance is made by the Secretary of Transportation.

The sequence can be simplified as:

Project team review

→ Credit and finance review

→ Council on Credit and Finance

→ Secretary of Transportation decision

When Does TIFIA Enter the Project, and How Long Does the Process Take?

The TIFIA process begins with early consultation between the Project Sponsor and the Build America Bureau.

The next major step is the LOI—Letter of Interest.

The literal translation sounds like a simple expression of interest, but the LOI is much more substantial than a one-page letter.

It includes information such as:

Project cost
Expected financing structure
Requested TIFIA amount
Repayment source
Status of environmental review

Submitting the LOI begins substantive federal financing review.

Korean PPPs do not have an identical procedure.

Very roughly, it resembles a stage in which the government examines the potential availability of public-supported financing while the procurement plan and financing structure are still being developed.

USDOT then reviews Eligibility—whether the project qualifies—and Creditworthiness—whether the project can actually repay the loan.

After credit analysis and review of the financial model, the project moves to the formal Application stage.

The Council on Credit and Finance then reviews the transaction, the Secretary approves it, and the process ends with a Credit Agreement between the federal government and the Borrower.

The overall sequence looks like this.

The Law Says 30 Days + 60 Days, but the Full Process Is Much Longer

Federal law specifies deadlines once the formal TIFIA Application has been submitted.

Under 23 U.S.C. §602, USDOT must generally notify the applicant within 30 days whether the application is complete.

Once the application has been determined complete, USDOT generally has another 60 days to notify the applicant whether it has been approved or denied.

At first glance:

30 days + 60 days = about 90 days

It seems like the process should be over in three months.

But there is an important distinction.

The statutory clock does not begin on the day the LOI is submitted.

Before the formal Application, a project may still go through:

Eligibility review
Creditworthiness review
Financial-model revisions
Credit-rating review
NEPA environmental review
P3 procurement and negotiations

Those activities are not governed by the simple 90-day statutory clock.

So the actual process can be much longer.

Looking at Actual Projects, LOI to Credit Agreement Often Took Around a Year or More

Among Build America Bureau projects where dates are publicly available, the timing varies considerably.

For I-4 Ultimate:

July 16, 2013 — LOI
September 4, 2014 — Credit Agreement

About 14 months.

For the more recent SR 400 Express Lanes:

November 1, 2023 — LOI
August 5, 2025 — Credit Agreement

About 21 months.

Other recent examples include:

Midtown Bus Terminal — about 4 months

BART Rail Car Upgrade — about 7 months

Sacramento Airport Walkway — about 12 months

Complete 540 Phase 2 — almost 24 months

If I simply calculate the average across these six examples, the period from LOI to Credit Agreement is approximately 13.5 months.

This is not an official USDOT average processing time.

The projects differ in scale, environmental review, and procurement status, so it is only a rough reference point.

The statutory review period may be about 90 days, but the full financing process can take more than a year because much of the work happens during pre-application Credit Review and project preparation.

Now Let’s Look at an Actual Project — How Did This Work in I-4 Ultimate?

Now that we have looked at the programs, let’s see how the financing tools entered actual projects.

The first case is Florida’s I-4 Ultimate.

The project reconstructed approximately 21 miles, or 34 kilometers, of I-4 through central Orlando and added four Express Toll Lanes in the median.

Public information places the Project Cost at approximately $2.877 billion, or roughly KRW 4.03 trillion using the Korean original’s simple conversion of $1 = KRW 1,400.

The decision to deliver the project as a P3 was made not by the federal government but by FDOT—the Florida Department of Transportation.

But federal financing review began before the private concessionaire had been selected.

July 2013 — FDOT submitted the TIFIA LOI

USDOT then reviewed eligibility and creditworthiness.

April 23, 2014 — FDOT selected I-4 Mobility Partners as the Preferred Bidder

The term Preferred Bidder is closest to Korea’s preferred negotiating party.

The project agreement and financing conditions were then finalized.

September 4, 2014 — TIFIA Credit Agreement and Financial Close

Very roughly, in Korean PPP terms:

TIFIA early review → around the stage when the procurement framework and financing structure are being examined

Preferred Bidder → preferred negotiating party

P3 Agreement → concession agreement

Financial Close → execution of financing commitments and satisfaction of major financing conditions precedent

The legal procedures are, of course, not identical.

I-4 Combined Federal Credit, State Payments, Commercial Debt, and Private Equity

The financing structure of I-4 Ultimate looked like this.

Senior Bank Debt refers to senior loans provided by banks and other commercial lenders.

Private Equity is capital invested by private shareholders at their own risk.

A Milestone Payment is an amount paid by FDOT when specified major construction milestones are achieved.

So even though this was a privately financed P3, the private sector did not provide the entire project cost on its own.

State funding + federal long-term credit + commercial bank debt + private Equity were combined.

I-4 Split Its TIFIA Loan Into Two Tranches

The approximately $949.46 million TIFIA loan for I-4 was not structured as one single loan.

It was divided into:

Tranche A — approximately $127.29 million

Tranche B — approximately $822.17 million

A Tranche is simply one portion of a larger financing package separated by repayment conditions or repayment sources.

Tranche A was structured to be repaid from FDOT’s Final Acceptance Payment.

In simple terms, that is close to a lump-sum completion-stage payment made after the private party completes construction and FDOT confirms that contractual requirements have been satisfied.

Tranche B, by contrast, is repaid from the Availability Payments that FDOT makes to the private party over the operating period.

So I-4 matched:

money received around completion → shorter Tranche

AP cash flow received over more than 30 years → longer Tranche

The financing structure was matched to the project’s cash flow.

North Tarrant Express Used TIFIA and PABs Together in a Toll P3

The second case is Texas’s North Tarrant Express Segments 1 & 2A.

The project reconstructed approximately 13 miles of highway and added Managed Toll Lanes under a 52-year DBFOM Concession.

DBFOM stands for:

Design-Build-Finance-Operate-Maintain

It combines design, construction, financing, operation, and maintenance within one long-term contract.

There is a major difference from I-4.

The private concessionaire does not receive Availability Payments.

It bears Toll Revenue Risk.

On June 23, 2009, TxDOT and the private party entered into a CDA—Comprehensive Development Agreement.

CDA is the name used in Texas for this type of long-term P3 agreement and plays a role broadly comparable to a Korean concession agreement.

The TIFIA financing agreement followed in December 2009.

About $2 Billion Was Built Up in Four Layers of Capital

The North Tarrant Express financing structure is a useful example.

TIFIA was money directly lent by the federal government.

PABs were tax-advantaged bonds raised by the private project company in the capital markets.

Both ultimately had to be repaid from project revenues, including Toll Revenue.

The federal government did not take demand risk away from the private concessionaire.

The private party continued to bear Toll Risk.

TIFIA and PABs instead helped reduce financing cost and repayment pressure.

That is an important feature of the U.S. approach to P3 financial support.

In the Recent SR 400 Project, Procurement and Federal Financing Moved Almost in Parallel

The third case is Georgia’s SR 400 Express Lanes.

It is a major Revenue-risk DBFOM P3 adding Express Lanes along approximately 16 miles north of Atlanta.

This project is especially useful because it shows how the P3 procurement process and federal financing process can move side by side.

Let’s follow the terminology one step at a time.

NOIA Tells the Market: “We Plan to Advertise This Project Soon”

GDOT issued a NOIA in February 2022.

NOIA stands for Notice of Intent to Advertise.

A natural translation would be something like advance notice of an upcoming procurement.

In simple terms:

“We plan to bring this P3 project to market soon, so interested companies should prepare.”

There is no exact statutory equivalent in Korean PPP procedure.

Very roughly, it resembles an early market notice before the formal project announcement or third-party proposal competition.

RFQ Asks “Who Is Capable of Delivering This Project?” Before It Asks About Price

In March 2022, GDOT issued an RFQ—Request for Qualifications.

At this stage, the agency is not yet asking for a final price.

It first reviews:

Development experience

Construction capability

Financial capacity

Operating experience

In Korean PPP procurement, this is closest to PQ—Pre-Qualification.

Companies respond to the RFQ with an SOQ—Statement of Qualifications, presenting their qualifications, experience, and project-delivery capability.

The Shortlist Is Essentially the Group That Advances to the Main Competition

GDOT announced the Shortlist in September 2022.

The Shortlist consists of the qualified teams that passed the RFQ stage.

They are not yet the Preferred Bidder.

In Korean terms, they are closest to teams that have passed PQ and advanced to the main competition.

The Real Technical and Financial Competition Begins With the RFP

The next stage is the RFP—Request for Proposals.

This is the point where the agency effectively says:

“Now submit your actual design, construction, operating, and financing proposal.”

For a Korean government-solicited PPP, it is broadly comparable to submitting the project proposal under the Request for Proposals.

For an unsolicited PPP, it is closest to submitting a competing proposal under a third-party proposal announcement.

SR 400 went through several Draft RFPs and one-on-one discussions with bidders before issuing the Final RFP in November 2023.

The process also used ATCs—Alternative Technical Concepts.

An ATC allows private bidders to propose more efficient designs or construction methods while still satisfying the procuring agency’s fundamental requirements.

The TIFIA LOI Was Submitted at Almost the Same Time as the Final RFP

This is one of the most important parts of the SR 400 case.

The TIFIA LOI was received on November 1, 2023.

That was right around the Final RFP stage.

The final P3 concessionaire had not yet been selected.

So U.S. P3 procurement does not necessarily follow a simple sequence of:

select the private partner → then begin financing review

Instead:

P3 procurement and federal financing review can proceed in parallel.

That allows bidders to structure their private Equity and Debt with a clearer understanding of potential TIFIA and PAB availability.

Compared roughly with Korean PPP procedure, it is closer to reviewing potential government-supported financing to a significant degree before the preferred negotiating party is selected.

Best Value Proposer Is Closest to Korea’s Preferred Negotiating Party

SR 400 received final Proposals in May 2024 and evaluated technical factors, tolling, financing conditions, and overall project value.

On August 15, 2024, the Georgia State Transportation Board approved the selection of the Best Value Proposer.

This is not simply the team offering the lowest price.

It means the bidder judged most advantageous after considering technical quality, financing, pricing, and overall project value.

For Korean PPP readers, this is closest to the preferred negotiating party.

SR 400 Used $3.89 Billion of TIFIA and Approximately $3.4 Billion of PABs

The formal TIFIA Application for SR 400 was received on May 16, 2025.

The TIFIA Credit Agreement was executed on August 5, 2025.

That was approximately 81 days from the formal Application to the Credit Agreement.

Measured from the original LOI, however, the process took approximately 21 months.

This illustrates the difference between the statutory TIFIA review period and the actual overall financing process.

The financing amounts were substantial.

TIFIA loan — $3.89 billion
(approximately KRW 5.45 trillion)

PABs — approximately $3.4 billion
(approximately KRW 4.76 trillion)

The Build America Bureau lists a Project Cost of approximately $12.4 billion, or roughly KRW 17.4 trillion using the same simple conversion.

But this Project Cost reflects the cost scope of a long-term DBFOM Concession.

It should not be treated as directly equivalent to a simple Korean construction project cost.

Putting the Three Projects Side by Side Makes the Federal Role Much Easier to See

Looking at the three projects together also shows what the federal government does not do.

It does not automatically guarantee traffic-demand risk.

In both North Tarrant Express and SR 400, Toll Risk remains with the private party.

Instead, the federal government provides lower-cost long-term credit and tax-advantaged financing that helps create a financial structure capable of carrying that risk.

The Federal Government’s Role Does Not End at Financial Close

Financial Close is the point at which the necessary bank loans, bonds, Equity commitments, and other financing agreements have been executed and major financing conditions precedent have been satisfied.

Korean PPP practitioners are already familiar with a similar concept through financing commitments and financial closing.

But TIFIA does not end at Financial Close.

The federal government remains a Lender under a loan that may remain outstanding for decades.

It therefore continues monitoring the project’s credit status and cash flow until repayment.

The federal role can continue through the entire project lifecycle:

Early project development → financing-structure consultation

Procurement → review of TIFIA and PAB possibilities

Financial Close → provision of long-term federal credit

Construction and operations → monitoring of credit risk

That is why the Build America Bureau describes itself as more than a simple lending window.

The Federal Question Is Less “How Much Should We Support?” and More “What Kind of Money Should Go Where?”

Looking at the overall structure, federal support in the United States is quite interesting.

The government does not simply fill every project Funding Gap with direct budget spending.

Depending on the project, it asks:

Does this project need a Grant?

Would a long-term TIFIA loan work better?

Could PABs reduce the cost of private Debt?

If it is a rail project, would RRIF be more appropriate?

The Build America Bureau connects these tools through a single federal platform.

There is another interesting feature in how federal budget resources are used.

The Build America Bureau explains that $1 of federal budget authority can support up to $15 of TIFIA Credit Assistance, which can in turn help support up to approximately $50 of transportation infrastructure investment.

In other words, the government does not necessarily pay the full project cost directly.

By absorbing a limited amount of credit risk, it can help mobilize a much larger pool of public and private capital.

What Could Korea Learn From This?

There is no need to copy the U.S. system directly into Korea.

Korea already has the PPP Act, fiscal support, the Infrastructure Credit Guarantee Fund, policy finance, and infrastructure funds.

The Ministry of Planning and Budget’s 2026 PPP policy agenda also includes measures involving expanded credit guarantees and financial support.

Still, there are several aspects of the U.S. model worth thinking about.

First, Financial Review Could Begin Earlier Than Preferred Bidder Selection

In SR 400, the TIFIA LOI was submitted before the final private partner had been selected.

That does not mean the federal government approved the loan in advance.

It means the public financing that bidders might actually use was being structured during the procurement process.

For large Korean PPPs, one option would be to examine potential policy finance, credit guarantees, and fiscal support on a conditional basis while preparing the RFP or third-party proposal competition.

That could reduce the risk of selecting a preferred negotiating party, spending considerable time negotiating the concession agreement, and only later discovering:

“The project cannot actually be financed in the market under these conditions.”

Second, the Financing Functions of the Ministry of Planning and Budget, MOLIT, PIMAC, and Policy-Finance Institutions Could Be Connected More Like a Single Window

The most impressive feature of the Build America Bureau is not any one financial product.

It is that TIFIA, RRIF, PABs, technical assistance, and P3 structuring are considered together.

In Korea, PPP policy and fiscal support sit with the Ministry of Planning and Budget.

Transport policy and technical issues sit with the Ministry of Land, Infrastructure and Transport.

PPP feasibility and project evaluation are handled by PIMAC.

Credit guarantees and policy finance are again spread across separate institutions.

A completely new organization is not necessarily required.

But for major PPPs, a coordinated structure involving:

Ministry of Planning and Budget + MOLIT + PIMAC + credit-guarantee and policy-finance institutions

could operate from the early project stage as something like an Infrastructure Finance Desk.

Third, the Question Could Shift From “How Much Grant Should We Give?” to “Which Form of Government Support Is Cheapest?”

Suppose a project needs KRW 100 billion of government support.

For some projects, directly providing a KRW 100 billion construction subsidy may be the best option.

But for other projects, the same private investment might be mobilized with a smaller actual fiscal burden by combining:

Long-term policy loans

Principal-repayment deferral

Credit enhancement

Interest support

That is one of the main lessons from TIFIA.

The important question is not just the size of government support.

It is:

Which form of support can attract private capital most efficiently?

Fourth, If the Government Takes Long-Term Credit Risk, Credit Review Must Also Become Stronger

Long-term policy lending is not automatically better than a subsidy.

When the government provides a long-term loan, it takes on part of the Credit Risk that otherwise would have remained with private lenders.

So two questions need to be separated.

“Is this project necessary as a matter of public policy?”

“Can this project actually repay the loan?”

That is why TIFIA places separate and strong emphasis on Creditworthiness.

If a project has weak repayment capacity but still receives a large government loan simply because it is considered politically or socially necessary, the loan may become a disguised subsidy.

In such a case, explicit fiscal support may actually be more transparent.

The Federal Government Does Not Decide the P3 — but It Is Deeply Involved in Making the Project Financeable

The basic legal authority to pursue a U.S. P3 comes from state law and state or local public agencies.

The federal government did not tell Florida:

“Deliver I-4 as a P3.”

Nor did it tell Texas:

“Transfer North Tarrant’s demand risk to the private sector.”

But once those projects reached the stage where they needed to raise billions of dollars, the federal role became much larger.

Grants reduce the Funding Gap.

TIFIA provides long-term credit.

RRIF supports rail projects.

PABs reduce the financing cost of private Debt.

The Build America Bureau coordinates these tools from the early stages of project development.

I-4 Ultimate used approximately $949 million of TIFIA, divided into two Tranches aligned with its long-term cash flow.

North Tarrant Express used $650 million of TIFIA and $398 million of PABs to support a private concessionaire that continued to bear Toll Risk.

SR 400 began the TIFIA process before the private concessionaire had been selected and ultimately combined $3.89 billion of TIFIA with approximately $3.4 billion of PABs.

If I had to describe the federal government’s role in one sentence, I would put it this way:

“It is not the government that decides the P3 on behalf of the state. It is the government that structures the money and supplies the credit so that the project selected by the state can actually work in the financial market.”

For Korea, the more useful question may therefore be not simply:

“Should the government provide more support?”

but rather:

“At what stage, and in what form, should government financing enter so that the greatest amount of private capital can be mobilized with the smallest fiscal burden?”

That leads naturally to the next question.

The United States has all of these federal financing tools.

So why did its P3 market first grow so strongly around roads, bridges, Toll Roads, and Express Lanes?

In #45, I will look at why the U.S. P3 market developed first around transportation infrastructure—especially Toll Roads, Express Lanes, and Bridges—by examining both the revenue structures and the financing system.

Key Sources

U.S. DOT Build America Bureau – About the Bureau
U.S. DOT Build America Bureau – Financing / TIFIA Program Overview
23 U.S.C. §602
Build America Bureau – RRIF
Build America Bureau – Private Activity Bonds
Build America Bureau – I-4 Ultimate
Build America Bureau – North Tarrant Express Segments 1 & 2A
Georgia DOT / Build America Bureau – SR 400 Express Lanes
Ministry of Planning and Budget – 2026 PPP Promotion Measures

#43 – Why Do U.S. P3 Rules Differ by State? What Does This Mean for Korean Firms?

foweuniverse — Sun, 6 Sep 2026 05:44:39 +0900

Korean Original
SOCFI #43 – 왜 미국은 주마다 P3 사업방식이 다를까? 한국기업의 진출방안은?
https://socfi.tistory.com/47

#43 - 왜 미국은 주마다 P3 사업방식이 다를까? 한국기업의 진출방안은?

- 왜 미국 P3는 주마다 규칙이 다를까요? Virginia·Florida·Texas·Colorado의 P3 법률과 판례, Elizabeth River Tunnels·I-4 Ultimate·North Tarrant Express·US 36 사례를 비교합니다. 여기에 한국 건설사·금융사·엔지니

socfi.tistory.com

Meta Description

- Why do U.S. P3 rules differ from state to state? This article compares the P3 laws, institutions, court cases, and major projects of Virginia, Florida, Texas, and Colorado, including Elizabeth River Tunnels, I-4 Ultimate, North Tarrant Express, and U.S. 36. It also looks at practical market-entry strategies for Korean contractors, financial institutions, and engineering firms. -

When working on PPP projects in Korea, the governing PPP law does not change simply because the project is located in a different region.

Whether a rail PPP is developed in Seoul or a road PPP is developed in Busan, the project generally starts from the same national framework: Korea’s Act on Public-Private Partnerships in Infrastructure and the Basic Plan for Public-Private Partnerships.

The competent authority, project conditions, and concession agreement may differ.

But the basic legal framework is the same.

In the United States, things are different.

If you study a P3 in Virginia and then move to a project in Florida, you often have to start by looking up a different law.

Go to Texas, and another set of statutes and institutions appears.

In Colorado, you even find a specialized public organization focused on P3s and infrastructure finance.

So in the U.S. P3 market, there is one question that often comes before the delivery model itself:

“Which state is this project in?”

Why does that matter?

The United States does not have a single national P3 framework law comparable to Korea’s PPP Act. Each state uses its own laws to determine which public entities may pursue P3s, what types of projects are eligible, and what contractual authority those entities have.

And this is more than a legal technicality.

When I discussed overseas expansion with people in Korea’s infrastructure industry, there often seemed to be limited interest in developed markets such as the United States, perhaps because established global firms already had strong positions there or because Korean companies lacked local track records.

In this article, I want to look at the issue from another angle.

If Korean contractors, financial institutions, and engineering firms were to enter the U.S. P3 market:

Which agencies would they need to work with?
What risks would they need to review?
At what stage should they join a consortium?

I will compare Virginia, Florida, Texas, and Colorado from that practical perspective.

The First Question in a U.S. P3 Is: “Does the Law Allow This Contract?”

In #42, I introduced the term P3 enabling legislation.

“Enabling” simply means making something legally possible.

In practical terms, a state law gives a State DOT, Authority, or another public agency the authority to say:

“You may enter into this type of long-term P3 contract with a private partner.”

The four states considered here already differ in how that authority is structured.

Virginia has the Public-Private Transportation Act of 1995 (PPTA).

Today, Virginia Code Title 33.2, Chapter 18 provides a relatively integrated statutory structure covering the Responsible Public Entity, public-interest findings, Comprehensive Agreements, and project review.

Florida relies primarily on Florida Statutes §334.30, Public-private transportation facilities, which establishes FDOT’s authority to pursue transportation P3s.

FDOT may accept unsolicited proposals or issue its own solicitations and, subject to statutory procedures, enter into agreements with private parties.

Texas is more complicated.

TxDOT’s Alternative Delivery authority appears in provisions such as Transportation Code Chapter 223, while Regional Mobility Authorities have separate Comprehensive Development Agreement authority under provisions such as Chapter 370.

A Comprehensive Development Agreement (CDA) can include design and construction, and may also extend to financing, acquisition, maintenance, and operations.

Colorado has taken yet another approach.

The Colorado Transportation Investment Office (CTIO) — historically associated with the High-Performance Transportation Enterprise — concentrates expertise and authority involving P3s, tolling, bonds, and innovative finance.

CTIO states that it has authority to impose tolls and user fees, issue bonds, and enter into P3 agreements with public and private parties.

The differences are already obvious.

So in the United States, it is not enough to ask:

“Does this state allow P3s?”

You also need to ask:

Who has the contracting authority? Is that authority still valid today? And does the law actually authorize this particular project?

Virginia Builds Public-Interest Review and Risk Verification Directly into the P3 Framework

Among the four states, Virginia probably feels the most familiar to someone accustomed to Korea’s PPP framework.

The Virginia PPTA uses the concept of a Responsible Public Entity.

For Korean readers, this is probably the closest concept to a competent authority, although the two systems are not legally identical.

A private proposal does not automatically lead to a contract.

The Responsible Public Entity must determine whether using a P3 serves the public interest.

Virginia law requires consideration not only of public need, but also of the benefits of the private approach and the allocation of risks.

There is another fairly strong review mechanism.

Virginia Code §33.2-1803 requires independent review for projects with estimated construction costs exceeding $50 million.

Using the same simple conversion applied in the Korean original, at approximately KRW 1,400 per dollar, that is roughly KRW 70 billion.

The review is broad.

It can examine not only traffic forecasts and project costs submitted by the private party, but also potential additional public costs, taxpayer exposure, private-party default, and risks associated with project debt.

The cost of that review is borne by the private proposer.

Put simply, the law is effectively saying:

“For a major P3, do not simply accept the private party’s demand, cost, and financial assumptions. Have an independent third party test them again.”

For Korean companies, this matters.

Entering Virginia as a P3 Developer requires more than saying:

“We can build this cheaper and faster.”

Cost, schedule, traffic demand, and potential public exposure all need to be supported by information that can withstand independent review.

The Elizabeth River Tunnels Case Shows the Boundary Between Toll Rights and Government Authority in Virginia

Virginia also has actual P3 disputes that have reached the courts.

One important case is Elizabeth River Crossings OpCo, LLC v. Meeks, 286 Va. 286 (2013).

The dispute involved the large P3 project improving the Downtown and Midtown tunnels between Norfolk and Portsmouth.

VDOT entered into a Comprehensive Agreement with the private concessionaire, Elizabeth River Crossings.

The central issue was tolling.

Residents essentially argued:

“Has the government delegated its power to impose taxes to a private company?”

The Supreme Court of Virginia rejected that characterization.

It held that:

“The Project tolls are user fees.”

In other words, the tolls were payments for use of the facility rather than taxes.

The court therefore did not view Virginia’s statutory structure allowing tolls within the P3 arrangement as an unconstitutional delegation of taxing power.

It also concluded that the Comprehensive Agreement did not cause the Commonwealth to surrender its police power — the government’s underlying regulatory authority relating to public safety and order.

The message for companies looking at U.S. P3s is important.

A private operator may receive toll rights for 30 or 40 years.

That does not mean it also receives all of the government’s regulatory authority.

Contractual revenue rights and public powers remain different things.

Dulles Greenway Has Become Particularly Important for Financial Institutions

A more recent Virginia case is Toll Road Investors Partnership II v. State Corporation Commission (2025).

The private operator of the Dulles Greenway requested an increase in tolls.

Virginia’s State Corporation Commission, or SCC, did not approve the request.

Virginia law establishes standards for toll increases.

Among other things, tolls must be reasonable in relation to the benefit received by users, should not become so high that they materially discourage use of the road, and should not provide the operator with more than a reasonable return.

On July 17, 2025, the Supreme Court of Virginia upheld the SCC’s decision.

The court emphasized the requirement that tolls be:

“reasonable to the user in relation to the benefit obtained.”

For lenders and investors, this has a very practical implication.

A Toll P3 financial model cannot safely assume:

Traffic ↓ → Revenue ↓ → Raise Toll

Tolls are affected by politics, regulation, and law.

Toll Regulation Risk can become Credit Risk.

That distinction matters when evaluating long-term concession debt.

In Virginia, Korean Firms Need to Present a “Verifiable Project,” Not Just a “Good Project”

Looking at these rules and cases from the perspective of Korean companies changes the entry strategy.

For contractors, entering immediately as a stand-alone P3 Developer may not be the most realistic first step.

A more practical approach may be to join a consortium with an experienced local P3 Developer and initially participate as a Design-Build Contractor or major construction partner.

Virginia’s public-interest review and independent verification also mean that the ability to quantitatively support cost, schedule, design-change exposure, and construction risk matters at least as much as offering a low price. (SOC와 금융의 이야기)

For financial institutions, Virginia can be a useful market for understanding U.S. Toll P3 finance.

But a Traffic & Revenue Model is not enough.

They also need to understand where the following rights come from:

Toll-setting authority
SCC regulation
Lender Cure Rights
Step-in Rights
Termination Compensation

The Dulles Greenway case shows that financial distress does not automatically entitle a private operator to raise tolls.

Engineering firms may find opportunities even earlier.

Public Sector Analysis, traffic-demand studies, Independent Engineer work, and technical due diligence all create opportunities before final design or construction begins.

Virginia therefore has an important market not only for design after procurement, but also for verification before procurement. (SOC와 금융의 이야기)

Florida Is Led by FDOT, but the Legislature Guards the Door Through the Work Program

Florida places FDOT at the center of transportation P3s.

Florida Statutes §334.30 allows FDOT to accept unsolicited proposals, issue solicitations, and enter into P3 agreements.

But there is an important condition.

The statute refers to:

“with legislative approval as evidenced by approval of the project in the department’s work program”

In practical terms, the project needs to appear in the FDOT Work Program in a way that demonstrates legislative approval before the P3 contract can move forward. (SOC와 금융의 이야기)

The structure can be simplified like this:

FDOT leads project development and procurement
→ project enters the Work Program
→ the legislature exercises program and fiscal control
→ P3 contract

This arrangement gives the professional transportation agency strong project-development authority while preserving legislative control over long-term fiscal and policy commitments.

I-4 Ultimate Shows How Florida’s P3 Structure Works in Practice

A representative example is I-4 Ultimate in Orlando.

FDOT and I-4 Mobility Partners delivered the project using a DBFOM structure, reconstructing and expanding approximately 21 miles, or 34 kilometers, of I-4 through central Orlando.

The project includes Express Toll Lanes.

But it is not a Toll Concession in which the private partner takes traffic and toll-revenue risk.

FDOT retains that revenue risk and pays the private partner through Availability Payments.

Public financial information places the project cost at approximately $2.877 billion, or roughly KRW 4.03 trillion using the simple exchange-rate assumption in the Korean original.

The TIFIA loan was approximately $949 million, or about KRW 1.33 trillion. (SOC와 금융의 이야기)

The lesson is straightforward.

Florida law may authorize P3 delivery, but the statute does not by itself determine whether demand risk goes to the private sector, whether an Availability Payment is used, or how the financing package is structured.

Those decisions are designed again at the individual project and contract level.

In Florida, Korean Firms Should Watch the FDOT Pipeline Before Trying to Create Their Own P3

For a Korean contractor, the better approach in Florida may be to follow FDOT’s long-term pipeline and procurement schedule rather than starting with the idea:

“Let’s create our own P3 proposal.”

Building relationships with local Developers and Contractors around FDOT’s procurement pipeline is likely more important.

A DBFOM project such as I-4 Ultimate combines not only design and construction but also financing and long-term O&M.

The first strategic question therefore becomes:

Which risks are we prepared to take?

For financial institutions, Florida offers opportunities beyond pure Toll Revenue Risk.

In an Availability Payment project, the key credit issues shift toward the legal basis for government payments, performance-deduction mechanisms, the stability of long-term payment sources, and the way TIFIA interacts with the financing structure.

Engineering firms can look for roles involving FDOT Design Standards, federal NEPA review, Traffic Modeling, Owner’s Engineer services, and Developer-side technical support.

In Florida, understanding the FDOT system itself becomes a market-entry asset. (SOC와 금융의 이야기)

Texas Is a Market Where Looking Only at Past P3 Deals Can Be Misleading

Texas requires particular attention to the current legal framework.

In the past, Texas actively pursued major CDA projects such as:

North Tarrant Express
LBJ Express
SH 288

TxDOT describes a Comprehensive Development Agreement (CDA) as a structure that can combine design, construction, utility relocation, and maintenance in one agreement, and in some cases include private financing, long-term leasing, and operations.

But there has been a major change.

TxDOT now states:

“broad authority to enter into new CDAs expired in 2017.”

In other words, TxDOT’s broad authority to enter into new CDA projects expired in 2017. (SOC와 금융의 이야기)

This is extremely important.

A Korean company looking at the North Tarrant Express and concluding:

“Texas is still a market where TxDOT can continuously procure new large Toll Concessions in the same way.”

would be drawing the wrong conclusion.

Today, TxDOT’s Alternative Delivery activity is more heavily centered on Design-Build, while new CDA authority needs to be verified on a project-specific legal basis.

In Texas, Current Procurement Authority Matters More Than Historical P3 Experience

That changes the market-entry strategy.

For contractors, it makes more sense to examine TxDOT’s current Design-Build pipeline and Alternative Delivery Program than to build a strategy entirely around historical Toll Concessions such as North Tarrant.

TxDOT continues to procure Design-Build projects.

The Korean original, for example, notes that the SH 99 Grand Parkway Segment B-1 Design-Build contract was executed in July 2026. (SOC와 금융의 이야기)

Financial institutions need to be even more careful.

The first legal due-diligence question should not simply be:

“Was a CDA structure used before?”

It should be:

“Does the statutory authority to enter into this specific P3 or CDA actually exist today?”

In Texas, Statutory Authority Risk may need to be checked before designing an elegant Project Finance structure.

Engineering firms may find more direct opportunities in the current Design-Build market through Project Development, Design, Digital Delivery, QA/QC, Traffic, and ITS work.

Colorado’s Distinctive Feature Is That P3 and Finance Are Concentrated in a Specialized Organization

In Colorado, the key institution is the Colorado Transportation Investment Office (CTIO).

CTIO specializes in P3s, tolling, and innovative finance.

Its official materials describe authority involving Toll and User Fees, bond issuance, and P3 agreements with public and private parties. (SOC와 금융의 이야기)

That structure is quite different from Virginia or Florida.

Rather than managing P3 primarily as one part of a general State DOT road program, Colorado has developed a specialized organization focused on transportation investment and finance.

CTIO has used several types of structures, including revenue-risk, user-fee, and Availability Payment approaches.

U.S. 36 Is a Representative Colorado-Style P3

One of Colorado’s best-known examples is the U.S. 36 Managed Lanes/BRT Project connecting Denver and Boulder.

The combined cost of Phase 1 and Phase 2 was approximately $511 million, or around KRW 715 billion using the conversion applied in the Korean original.

The agreement with private partner Plenary Roads Denver used a 50-year Concession structure.

CDOT has described it as Colorado’s first P3 highway project, and the full Express Lanes corridor reached Substantial Completion in 2016.

Phase 2 also used approximately $60 million in TIFIA financing.

Build America Bureau materials place the Phase 2 project cost at $175 million. (SOC와 금융의 이야기)

In Colorado, CTIO’s toll authority + P3 expertise + federal financing + long-term private operations can be combined within a single structure.

In Colorado, Financial Institutions and Tolling/ITS Firms Stand Out

For contractors, Colorado should not necessarily be viewed as a simple EPC market.

It is more useful to see it as a specialized consortium market where Developers, financial institutions, and operators work together.

Even if project sizes are smaller than some of Texas’s mega Managed Lane projects, Tolling, Operations, and Lifecycle Maintenance can play significant roles.

For financial institutions and infrastructure funds, Colorado may actually be quite interesting.

There is a specialized Counterparty in CTIO, and it deals with Toll Revenue, Revenue Bonds, federal financing, and P3 structures.

Long-term Toll Concessions still require careful analysis of traffic, toll-setting, political, and termination risks.

Engineering firms can look for entry points in areas such as Managed Lanes, Tolling, ITS, Traffic Operations, and Lifecycle Asset Management. (SOC와 금융의 이야기)

The Differences Among the Four States Also Change How Korean Firms Should Enter the Market

When the legal frameworks and actual projects are placed together, the comparison becomes more practical.

The point is not simply to rank one state as “better” than another.

Contractors, financial institutions, and engineering firms are looking for different things, and they can bear different types of risk.

So I would change the question.

“If a Korean company is entering the U.S. market for the first time, which state offers a relatively predictable legal and procurement structure where it can build local experience without taking excessive risk?”

That makes comparison more useful. (SOC와 금융의 이야기)

Contractors Should Decide “How Much Risk to Take” Before Asking “Where to Build”

Imagine a large Korean contractor entering the U.S. P3 market.

Having substantial PPP experience in Korea does not mean becoming a full P3 Developer in the United States from day one is automatically the right strategy.

A U.S. Developer is not simply a construction contractor.

The role may include:

Long-term Equity investment
Financial Close
Completion Risk
O&M
Toll or Availability Payment performance risk
Relationships with lenders

That is a much broader responsibility.

A more realistic first step may be a consortium such as:

Local Developer + Korean contractor + local Contractor/Designer

The Korean company could initially take a role closer to DB (Design-Build) or EPC, learn U.S. Bonding, Claims, O&M, and contracting practices, and gradually expand toward Equity and long-term operations. (SOC와 금융의 이야기)

The approach also changes by state.

In Virginia, Cost, Schedule, and Risk data must withstand Independent Review.

In Florida, the ability to follow FDOT’s pipeline and assemble a local consortium for DBFOM procurement matters.

In Texas, the current Design-Build pipeline and local JV network may matter more than historical CDA experience.

In Colorado, the question is whether the company can join specialized consortia that integrate construction with Tolling, Managed Lanes, and long-term operations.

For contractors alone, I would view Florida and Texas as relatively realistic first-entry markets.

Florida offers exposure to P3 and DBFOM.

Texas offers a large Design-Build market where companies can first learn the U.S. construction and contracting system.

Financial Institutions Need to Ask “Who Protects Me on the Downside?” Before Looking at the Project Itself

The perspective of a financial institution is different.

Before asking whether the road is badly needed, I would first ask:

Who determines the toll?
Who regulates toll increases?
What is the legal basis for Availability Payments?
When may lenders step in after a private-party Default?
How much debt can be recovered through Termination Payment if the contract ends?

The Dulles Greenway example in Virginia makes the point clearly.

Financial difficulty did not give the private operator an unlimited right to increase tolls.

In a Toll P3, Toll Regulation Risk can become Credit Risk.

The risk profile of an Availability Payment project is different.

Traffic demand becomes less central, while the legal basis for long-term public payments, budgeting and payment mechanisms, performance deductions, and Termination Payments become more important.

So a Korean financial institution should not apply the same PF credit model to both a Toll P3 and an AP P3. (SOC와 금융의 이야기)

From a lender’s perspective, I would actually consider Virginia one of the first markets worth studying.

It has the PPTA framework and a meaningful body of case law dealing with toll regulation and the boundaries of private rights.

That makes Downside analysis relatively transparent.

Florida is also attractive.

Projects such as I-4 Ultimate combine Availability Payments, TIFIA, and public payment obligations without transferring full Toll Revenue Risk to the private party.

That can provide a useful entry point for Korean financial institutions learning U.S. P3 finance. (SOC와 금융의 이야기)

Engineering Firms Can Enter Before Construction Even Begins

Engineering firms have another set of opportunities.

If they view the P3 market only as a market for final design or construction design, they may miss a significant part of it.

Before procurement and construction, projects may require:

Traffic & Revenue Studies
Preliminary Engineering
Environmental Review
Cost Estimation
Independent Engineer services
Technical Due Diligence

Virginia is particularly relevant because independent review is institutionally required for certain large projects.

Florida rewards firms that understand both FDOT standards and federal NEPA requirements, and opportunities can arise as an Owner’s Engineer or Developer-side Technical Advisor.

In Texas, current Design-Build procurement creates opportunities in Design Integration, QA/QC, ITS, and Digital Delivery.

In Colorado, specialties such as Tolling, Managed Lanes, ITS, Traffic Operations, and Asset Management can be particularly relevant.

So an engineering firm does not have to become a P3 Developer to enter the U.S. P3 market.

Building a track record first in specialized professional services around P3 projects may be a more realistic strategy. (SOC와 금융의 이야기)

Which State Is the Most Realistic First U.S. Base for a Korean Company?

Let’s make the comparison more practical.

The discussion so far explains the legal and market characteristics of each state.

It does not mean one state is objectively superior to another.

But if the question is where a Korean company could first enter the U.S. market, build local experience, and then establish a longer-term position, I think some prioritization is possible.

I would look at five criteria:

How clear is the legal authority?
How sophisticated is the procuring agency?
Is there a real current procurement pipeline?
Can a Korean company enter without immediately taking excessive risk?
Is there room for contractors, financial institutions, and engineering firms?

Under those criteria:

I would rank Florida first. (SOC와 금융의 이야기)

Florida Looks Like the Most Stable First-Entry Market

Florida’s biggest advantage is that the answer to “Who do we deal with?” is relatively clear.

FDOT is a clearly identifiable procuring agency.

The legal authority for P3s and the legislative control exercised through the Work Program are also relatively visible.

Florida also has major experience with structures combining DBFOM, Availability Payment, and TIFIA, including I-4 Ultimate.

That creates multiple entry points for Korean firms.

Contractors can join local Developers in DBFOM or Design-Build consortia.

Financial institutions can learn U.S. PF structures through AP and TIFIA without immediately taking full Toll Revenue Risk.

Engineering firms can enter through FDOT standards, NEPA, Preliminary Engineering, and Owner’s Engineer work.

For a first U.S. base, Florida offers an entry point for construction, finance, and engineering at the same time.

That is why I would place it first. (SOC와 금융의 이야기)

Virginia Ranks Second, but It May Be Even Better for Finance and Technical Services

Virginia has strong legal predictability.

It has the PPTA, institutionalized public-interest review and independent verification, and a growing body of cases involving tolls and private rights.

For financial institutions and engineering firms, Virginia may actually be a better learning market than Florida.

But for a contractor entering the U.S. for the first time, the level of verification is demanding.

Cost, schedule, demand, and risk-allocation data need to withstand external review.

Virginia is therefore less a “market that is easy to enter” than a “market that is predictable but demanding.”

Overall, I would place it second. (SOC와 금융의 이야기)

Texas Is Highly Attractive for Contractors, but P3 Investors Need to Verify Authority Project by Project

Texas remains highly attractive from a contractor’s perspective.

The market is large.

There are many major highway projects.

Design-Build Alternative Delivery continues to be used.

For a large Korean contractor, it can be a good market to learn U.S. contracting, Bonding, Claims, and local workforce management.

But P3 finance and Equity investment require more caution.

The historical success of CDA projects such as North Tarrant Express does not mean TxDOT can freely launch new Toll P3s under the same broad authority today.

Its broad authority to enter into new CDAs expired in 2017.

In Texas, every project therefore requires an early legal due-diligence question: “Does valid CDA/P3 authority actually exist for this project today?”

For contractors, Texas may be nearly as attractive as Florida.

But as an all-around first base across construction, finance, and engineering, I would place it third. (SOC와 금융의 이야기)

Colorado Is a Good Market, but It Looks More Like a Specialized Expansion Market Than a First Entry Point

Colorado’s strength is CTIO.

Tolling, Revenue Bonds, P3s, and Innovative Finance are handled by a specialized Counterparty.

That makes Colorado quite interesting for infrastructure funds and companies specializing in Tolling, ITS, or Managed Lanes.

But as the first U.S. base for a Korean company, the overall market is smaller and more specialized than Florida or Texas.

Colorado is not necessarily a “high-entry-risk market.”
It is a market where you need a clearly defined specialty before entering.

It may therefore make more sense after a company has already accumulated some U.S. experience and wants to expand into Tolling, ITS, long-term Concessions, or Asset Management. (SOC와 금융의 이야기)

Overall, I Would Rank Them Florida → Virginia → Texas → Colorado

Putting these factors together, my ranking would be:

Florida → Virginia → Texas → Colorado

This is not a ranking of investment returns or total project volume.

It is based on a different question:

Where can a Korean company enter without immediately taking excessive risk, build experience with U.S. procurement, law, finance, and construction, and increase its chances of establishing a long-term presence?

The ranking also changes by industry.

Contractors, lenders, and engineering firms do not necessarily need the same first market.

If I Were Designing an Entry Strategy, I Would Start in Florida and Expand the Role Gradually

If I were designing a U.S. P3 entry strategy for a Korean company, I would probably not recommend becoming the Developer of a 50-year Toll Concession from day one.

I would start by forming a consortium with local Developers and Contractors in Florida.

The first roles could be:

DB
DBFOM participation
Technical Advisor
Financing

Then the company could expand according to its strengths.

A contractor could move toward Texas and its large Design-Build market.

A financial institution could expand toward Virginia’s Toll, regulation, and contract market.

An engineering company or investor with ITS, Tolling, or operations expertise could look toward Colorado’s specialized market.

Rather than entering the entire United States at once, build initial experience in Florida and then expand to different states according to sector-specific strengths.

The most important question when entering the U.S. P3 market may therefore not be:

“Where can we win the biggest project?”

A better question may be:

“In which market can we first take on risks we know how to manage, build a U.S. track record, and then expand our role?”

If P3 is fundamentally about allocating each risk to the party best able to manage it, overseas market entry should probably follow the same logic. (SOC와 금융의 이야기)

Korean Firms Should Not Think of U.S. P3 as One Single “U.S. Market”

That brings us back to the original question.

Why do U.S. P3 rules differ so much by state?

Because under the U.S. federal system, much of the authority to enter into P3 contracts comes from state law and individual public agencies.

Virginia has strongly institutionalized public-interest review and risk verification.

Florida combines FDOT’s professional expertise with legislative control through the Work Program.

Texas had a highly active CDA market in the past, but today Design-Build-centered Alternative Delivery is more important.

Colorado uses CTIO to specialize in Tolling, bonds, P3s, and innovative finance. (SOC와 금융의 이야기)

So before a Korean company enters a U.S. P3 market, the checklist becomes fairly clear.

1. Which state are we dealing with?

2. Which law authorizes this contract?

3. Is that contracting authority actually valid today?

4. Who procures the project, and who controls tolling and other public powers?

5. Where do we fit — Developer, Contractor, Financial Institution, or Engineering Firm?

6. In that role, what level of risk can we actually manage?

The last question may be the most important.

Entering the United States does not mean taking responsibility for the entire project from day one.

Take the risks you can manage best first, build local experience, and expand your role over time.

That approach closely resembles the basic principle of risk allocation in P3 itself.

And at this point, another question naturally appears.

If state governments decide and contract the projects, why do names such as USDOT, FHWA, Build America Bureau, TIFIA, and PABs keep appearing in U.S. P3 materials?

In #44, I will look at how the federal government influences state-led P3s through grants, financing, tax-supported tools, environmental review, and regulation. (SOC와 금융의 이야기)

Key Sources

Virginia — Virginia Code Title 33.2, Chapter 18, Public-Private Transportation Act of 1995, Responsible Public Entity
Elizabeth River Crossings OpCo, LLC v. Meeks, 286 Va. 286 (2013)
Toll Road Investors Partnership II v. State Corporation Commission (Va. 2025)
Florida Statutes §334.30, Public-private transportation facilities
Texas Department of Transportation — Alternative Delivery Program
North Tarrant Express
Colorado Transportation Investment Office (CTIO)
U.S. 36 Managed Lanes/BRT

#42 - Who Decides and Who Is Responsible in a U.S. P3?

foweuniverse — Sun, 6 Sep 2026 05:08:28 +0900

Korean Original
SOCFI #42 – 미국 P3는 누가 결정하고 책임질까?
https://socfi.tistory.com/45

#42 - 미국 P3는 누가 결정하고 책임질까?

- 미국 P3는 누가 결정하고 누가 책임질까요? Virginia와 Florida P3법을 통해 State DOT·Toll Authority·Transit Agency·연방정부의 권한을 나누어 보고, Charles River Bridge·Monongahela·Kelo·Seven County 등 미국 연방대

socfi.tistory.com

- Who decides a U.S. P3, and who is actually responsible for it? This article explains how State DOTs, Toll Authorities, Transit Agencies, and the federal government divide authority in U.S. P3s, using Virginia and Florida P3 laws and major U.S. Supreme Court cases on toll rights, eminent domain, public authority, and NEPA. -

If you spend some time reading U.S. P3 materials, the first thing that gets confusing is usually the institutions.

USDOT, FHWA, State DOT, Toll Authority, Transit Agency, Build America Bureau....

At first glance, they all look like similar government bodies.

So the natural question is: who is actually the competent public authority here?

In Korea, the structure is relatively clear. The competent authority usually promotes the project and enters into the concession agreement, while the central government manages the overall system through the law and the Basic Plan for PPPs.

That makes it easy to assume the U.S. must work in a similar way. Maybe USDOT approves the project and the State DOT carries it out.

But if you look at it that way, U.S. P3s actually become harder to understand.

In the United States, the federal government, state governments, local governments, and various transportation public agencies each hold different legal powers, and the actual P3 contract is signed by the public agency that has the legal authority to do so.

The federal government may lend money or conduct environmental review, but that does not make it the contracting party to the P3.

And just because a private partner designs, builds, finances, operates, and maintains a road does not mean all public powers — such as eminent domain, toll approval, or public safety authority — are handed over to the private sector.

In the end, to understand a U.S. P3, you have to separate three questions.

Who procures the project?
Who exercises public authority?
Who bears the contractual risks?

Sometimes those three are the same entity.

Sometimes they are not.

U.S. P3s First Ask: “Is This Something the Agency Is Legally Allowed to Do?”

As discussed in #40, the United States does not have a single nationwide framework law that governs all P3s in the way Korea’s Act on Public-Private Partnerships in Infrastructure does.

Instead, each state gives particular public agencies the legal authority to pursue P3s through state legislation.

This legal basis is often called enabling legislation.

“Enabling” is a useful word here. It simply means the law opens the legal door and says, in effect, “this agency is allowed to enter into this kind of long-term agreement with a private partner.”

In Virginia, the representative example is the PPTA, the Public-Private Transportation Act of 1995.

That law lays out the main legal structure of a P3, including the Responsible Public Entity, private entities, Comprehensive Agreements, user fees, eminent domain, contract default, and other core elements.

In Florida, FDOT’s P3 authority is set out in Florida Statutes §334.30, Public-private transportation facilities.

The current statute addresses unsolicited proposals, Comprehensive Agreements, tolls, traffic and revenue analysis, financing plans, legal compliance, and eminent domain.

Put simply, state law answers this question first.

In Korea, the first question is often, “Should this project use a BTO or BTL structure?”

In the United States, before that, there is often an earlier question:

“Does this agency even have the legal authority to enter into this kind of contract in the first place?”

In Many Cases, the Real Center of a U.S. Road P3 Is the State DOT, Not the Federal Government

The agency that appears most often in road P3s is the State DOT, the State Department of Transportation.

In Florida, that is FDOT, the Florida Department of Transportation.

In Virginia, that is VDOT, the Virginia Department of Transportation.

If I were to compare them to Korea, they are much broader than a local road management agency. They are better understood as state-level transportation departments responsible for road policy, construction, and operations across the state.

The important point is that a State DOT is not an agency under the federal government.

FHWA is a federal agency. FDOT and VDOT are state agencies.

So USDOT → FHWA → FDOT should not be understood as a simple command hierarchy like a Korean central ministry and its regional offices.

For example, the DBFOM agreement for I-4 Ultimate discussed in #40 was signed not by USDOT or FHWA, but by FDOT.

Florida §334.30 authorizes FDOT to enter into a Comprehensive Agreement with a private partner and even to grant the right to impose tolls or other charges.

At the same time, the project must comply with federal, state, and local law, as well as FDOT standards, and FDOT’s eminent domain powers may also be used for the P3 project.

That is why a State DOT often feels like the U.S. institution closest to Korea’s competent authority in a PPP.

Still, that comparison only goes so far.

Depending on the project, the contracting public entity may instead be a Toll Authority or a Transit Agency.

If You Find the “Responsible Public Entity,” You Can See Who the Public-Sector Lead Really Is

Virginia law uses a particularly helpful term: Responsible Public Entity.

Taken literally, it means the public entity that holds responsibility.

Under the Virginia PPTA, a Public Entity can include not only the state government but also state agencies, authorities, counties, cities, and towns.

And in an actual project, the agency that takes responsibility becomes the Responsible Public Entity.

If I had to compare it to Korea, it is the concept closest to the competent authority.

But it is not exactly the same institution.

A Responsible Public Entity does not receive a private proposal and simply sign a contract right away. It must determine whether the project serves public need and public benefit.

The Virginia legal structure separately requires a Finding of Public Interest and a Public Sector Analysis.

That means the law is ultimately asking a simple question:

Not “Can this be handed over to the private sector?”

But “Is it in the public interest to hand this over to the private sector?”

Even When the Private Sector Receives Operating Rights, Government Authority Does Not All Move With It

A P3 private partner may take on a great deal.

It may handle design, construction, financing, operations, and maintenance.

In a Toll Concession, it may even bear traffic and toll-revenue risk.

That can make it seem as if the road has simply been handed over to the private sector.

Legally, though, the structure is more limited than that.

A Virginia Comprehensive Agreement may include terms on user fees, compensation to the private entity, project term, public support, lender rights, default, and cure rights.

User fees themselves can also be set through the agreement between the parties.

Florida can also grant a private partner the right to impose tolls or fares, but only within the limits allowed by state law and the Comprehensive Agreement.

So the private partner does not become an independent government.

It is closer to receiving, for a long period of time, certain rights that a public agency is legally allowed to grant by contract.

And interestingly, the U.S. Supreme Court has been thinking about the boundaries of these public franchises for a very long time.

The Supreme Court in 1837 Did Not Read Private Rights Too Broadly

Long before the term P3 ever appeared, similar questions were already reaching the U.S. Supreme Court.

A classic example is Charles River Bridge v. Warren Bridge, 36 U.S. 420 (1837).

Massachusetts had granted a private bridge company, the Charles River Bridge company, the right to operate a bridge and collect tolls.

Later, it authorized another nearby bridge, the Warren Bridge.

The original company was essentially arguing this:

“The government granted us the bridge franchise, so it should not authorize a competing bridge.”

But the Supreme Court was very cautious about reading into the original grant an exclusive right that had not been expressly given.

The basic direction of the case was this: when interpreting a franchise granted by the government to a private party, courts should not lightly presume that the government surrendered public powers or granted an exclusive right unless it clearly said so.

Later Supreme Court cases continued to cite Charles River Bridge for the principle that exemptions from government regulation are not lightly presumed.

The idea can be stated briefly like this:

“Grants of immunity from legitimate governmental control are never to be presumed.”

That line strongly recalls the non-compete issue discussed in #41 in connection with the 91 Express Lanes.

If demand risk is transferred to the private sector, some revenue protection may be needed.

But if that protection becomes too strong, the government may later find it difficult to widen general-purpose lanes or pursue new transportation policies.

In the nineteenth century and in modern P3s, the underlying question is surprisingly similar.

How far can private investment be protected without tying up the government’s future policy authority?

A Toll Concession Is Not Just a Business Opportunity. It Is a Right With Economic Value

There is another side to the story.

The government cannot simply wipe away private rights whenever it wants, either.

A representative case is Monongahela Navigation Co. v. United States, 148 U.S. 312 (1893).

Monongahela Navigation Company had been authorized by Pennsylvania to build and operate locks and dams on inland waterways and collect tolls.

Later, the U.S. government took the facility through condemnation.

The issue was whether compensation only had to cover the physical facility, or whether it also had to cover the value of the franchise to collect tolls.

The Supreme Court held that the toll-collection franchise also had economic value.

One of the key phrases in the case was “the franchise of this company to collect tolls.”

In other words, the company’s right to collect tolls was itself a compensable property interest.

The Court also emphasized that Just Compensation is not simply whatever the government chooses to pay. It is ultimately a matter for judicial judgment.

Later cases citing Monongahela reaffirmed the principle that compensation must fully reflect the value of the property that the government takes.

From a P3 perspective, the point is clear.

Granting a private party the right to collect tolls for 30 or 40 years under a Toll Concession is not just a matter of saying, “we are outsourcing road management.”

That right itself carries economic value.

So if the government buys out the project, terminates it early, or takes the rights back during the concession period, issues such as government takeover, termination, compensation, and lender protections become central.

That is very similar to why termination payments are so important in Korean PPP practice.

Eminent Domain Is One of the Powers That Typically Remains With the Public Sector

Road and rail projects ultimately need land.

In the United States, the government’s compulsory acquisition power is called eminent domain.

It is the power to take private property for a public purpose, subject to just compensation.

A P3 does not mean the private partner can freely exercise eminent domain.

Florida law provides that FDOT may use the eminent domain authority legally granted to it for purposes of a P3 project.

In Virginia as well, condemnations for transportation facilities operate within the legal authority of public agencies under state law.

Virginia separately states principles on property rights and just compensation in its laws.

So the basic structure looks like this:

The private partner needs land
→ the public agency determines public purpose and legal requirements
→ the public agency exercises eminent domain under the law
→ just compensation is paid

The Kelo Case Shows That Private Participation Does Not Automatically Eliminate Public Purpose

When eminent domain is discussed in the United States, the most famous case is Kelo v. City of New London, 545 U.S. 469 (2005).

The City of New London in Connecticut condemned private property as part of a broader economic development plan, and some of that land was expected to be used later for private development.

The landowners challenged the taking in very direct terms:

“How can taking my land and giving it to another private party count as public use?”

The Fifth Amendment to the U.S. Constitution requires just compensation when private property is taken for “public use.”

The Supreme Court held that the project satisfied the public use requirement because it was not simply designed to benefit a particular private party. It was part of a broader economic development plan serving a public purpose.

But Kelo did not mean that any transfer to a private party is automatically allowed.

One limit repeated in the Court’s cases is that the government cannot justify a taking merely by taking property from one private party and handing it to another.

The key concept here is public purpose.

That matters in P3s as well.

The fact that a private partner builds the facility and earns revenue does not by itself erase the project’s public character.

At the same time, the participation of a private partner does not automatically justify eminent domain.

The public agency must independently determine the public purpose and the legal requirements.

That is why eminent domain in a P3 is not simply a power handed over to the private sector.

The U.S. Counterpart to Korea’s Concession Agreement Is Usually a Comprehensive Agreement or a Concession Agreement

If I had to name one of the most important documents in a Korean PPP, it would be the concession agreement.

It is where the project cost, construction period, operating period, charges, risk allocation, termination, and lender rights are usually set.

There is a similar contract in the United States.

In Virginia and Florida, the common term is Comprehensive Agreement.

In other P3 contexts, Concession Agreement is also widely used.

Under the Virginia PPTA, a Comprehensive Agreement may include the private entity’s compensation, the contract term, user fees, government grants or loans, default and cure rights, lender protections, treatment of excess revenues, and contract changes.

So for Korean readers trying to understand the structure, it is useful to think of Comprehensive Agreement / Concession Agreement as roughly corresponding to 실시협약.

But that does not mean they are legally identical.

Because state laws and institutional authorities differ, the contents of the contract can also vary substantially from project to project.

And if you place the two Supreme Court cases discussed above side by side, the importance of the contract becomes even clearer.

Charles River Bridge refused to expand into an exclusive right something that had not clearly been granted.

Monongahela, by contrast, recognized the economic value of a toll franchise that had in fact been granted.

In other words, in a P3, it matters enormously how clearly the contract states what rights the private partner receives.

Private Financial Trouble Does Not Make the Government’s Public-Service Responsibility Disappear

There is another important question in P3s.

“If the private partner goes bankrupt, does the road also shut down?”

Not necessarily.

Virginia’s PPTA allows the Comprehensive Agreement to address default by the private entity and cure rights for lenders. The Responsible Public Entity can structure lender cure rights and continuity mechanisms through the contract.

This is one place where the difference between a P3 and full privatization becomes very clear.

Even if design, construction, financing, and operating risks are transferred to the private sector, the government’s public responsibility to keep an essential service such as a road or rail line functioning does not disappear.

At the same time, if the private partner runs into financial distress, that does not mean the state government automatically repays all of the private party’s debts.

[Image]
Default, Cure Rights, Step-In Rights, and Termination Compensation

That is why terms such as Default, Cure Period, Step-in Rights, and Termination Compensation are so important in actual P3 finance documents.

This is a topic worth examining more deeply later in the finance section.

The Federal Government Is Often Less the Contracting Party Than the Institution That Holds the Money and the Rules

So where do USDOT and FHWA fit into all this?

USDOT is the U.S. Department of Transportation.

FHWA, the Federal Highway Administration, is the federal agency responsible for highways and bridges.

In most state or local P3s, they do not directly sign the Concession Agreement with the private partner.

Instead, the federal government holds very powerful tools.

The first is funding.

Through the Federal-Aid Highway Program, it can provide federal funds to state governments, and through the Build America Bureau it can connect projects to financial tools such as TIFIA and PABs.

The second is federal law.

If federal funding or federal approval is involved, federal legal requirements on matters such as environment, labor, and safety also come with it.

The third is approval and oversight.

In major federally supported projects, federal agencies can play an important role in environmental review and financial planning.

So a U.S. P3 can end up looking like this:

The state government signs the contract, while the federal government holds part of the money and the rules.

That is why it is important to distinguish the contracting party from the regulator.

NEPA Is Not a Law That Says “Choose the Environmentally Best Project”

There is one acronym you really need to know in U.S. infrastructure.

NEPA, the National Environmental Policy Act.

When a major road or rail project is tied to federal funding or federal approval, NEPA review becomes a significant part of the process.

For roads, FHWA may be involved as the federal agency.

For transit, it may be FTA.

For rail, FRA.

But NEPA should not be treated as exactly the same thing as Korea’s environmental impact assessment system.

And in 2025, the Supreme Court once again clarified the character of the statute.

In 2025, the Supreme Court Reemphasized That NEPA Is a Procedural Statute

The case was Seven County Infrastructure Coalition v. Eagle County, 605 U.S. 168 (2025).

It involved environmental review for a project to build roughly 88 miles of rail line in Utah’s Uinta Basin to connect crude oil production to the national rail network.

On May 29, 2025, the U.S. Supreme Court reversed the lower court.

The key point emphasized by the Court was that NEPA is “a purely procedural statute.”

Put simply, NEPA is less a law that commands the government to choose the environmentally best alternative and more a law that requires the government to examine significant environmental effects and reasonable alternatives before making a decision, and to disclose that decision-making process.

The decision also emphasized that agencies deserve substantial deference in deciding which environmental effects fall within the scope of the analysis.

That distinction matters in P3s too.

Selecting a private partner first does not mean the private partner can then decide the route and the technical method entirely on its own.

If the project requires federal approval, the public agency still has its own separate legal responsibilities under NEPA.

In Rail and Transit, the Main Public Actor Is Often Not the State DOT

If you look only at roads, State DOTs can appear to be the center of U.S. P3s.

But rail and transit are different.

In urban rail or bus projects, a Transit Agency can become the Project Sponsor.

At the federal level, the relevant agency is the FTA, the Federal Transit Administration.

If I had to compare it loosely to Korea, it is closer to a federal institution combining elements of transit policy, urban rail oversight, national funding support, and supervision.

In the rail sector, FRA, the Federal Railroad Administration, comes in.

In Virginia, the Virginia Passenger Rail Authority is also expressly identified in the statute as an entity that may serve as a Responsible Public Entity under the PPTA.

So depending on the sector, the institution closest to the competent authority can change.

A Toll Authority Is a Specialized Public Agency for Toll Roads

Another institution that appears frequently in U.S. road projects is the Toll Authority.

In Korean, a rough description would be something like a toll-road authority or toll-road public corporation.

It is a public entity created separately from the State DOT to develop and operate particular toll roads, bridges, or tunnels.

If I compare it loosely to Korea, it looks something like a hybrid of some Korea Expressway Corporation functions and some local public-enterprise functions.

What distinguishes many U.S. Toll Authorities is that they often have their own toll revenue, bond-issuance authority, and debt-repayment structure.

Florida law also allows separate authorities to enter into P3 agreements, authorize tolls or fares, and use their own eminent domain powers where authorized.

So when you come across a U.S. P3 and think,

“Why is this agency — not the State DOT — signing with the private sector?”

the first place to look is the statute that created that authority and defined its powers.

Responsibility in a U.S. P3 Is Not Concentrated in a Single Institution

Let’s pause and pull the structure together.

That is why, when someone asks, “Who is responsible in a U.S. P3?”, it is difficult to answer with just one institution.

You have to ask: responsible for what?

Responsibility for procuring the project?
Responsibility for environmental review?
Responsibility for eminent domain?
Responsibility for cost overruns?
Responsibility for demand risk?
Responsibility for operating performance?

The answer changes depending on the question.

If You Put the Four Supreme Court Cases Together, the Boundaries of U.S. P3s Become Clearer

The four cases discussed here are from different periods, and not all of them are direct P3 cases.

But when placed side by side, they show surprisingly clearly the legal boundaries within which U.S. P3s operate.

What is interesting is that the first two cases can look like opposites, but in fact they point to the same underlying principle.

A private party cannot claim rights the government did not clearly grant.

But the government also cannot casually ignore rights it actually did grant.

That is exactly why contracts matter so much in P3s.

The contract has to spell out clearly what is being granted to the private partner, what remains with the government, and under what circumstances the government may step back in later.

In the End, U.S. P3s Decide Authority in Law and Risk in Contract

At first glance, the public structure of U.S. P3s looks complicated.

But once you compress it down to the core, it is simpler than it first seems.

The law determines who is allowed to do what, and the P3 contract divides the actual risks and rewards.

A State DOT cannot create a P3 just because it wants to.

It first needs authority under state law.

And even if the private sector is allowed to operate a road, that does not mean every public power moves with it.

Eminent domain, environmental review, and public safety remain, in important ways, on the government side.

At the same time, once the government grants a long-term toll right or operating right to a private partner, it cannot simply ignore the economic value of that grant without consequence.

That is why, when looking at a U.S. P3, there are at least three things that should be checked first.

What law authorized the project?
Which public entity is the contracting party?
What rights and risks were transferred to the private sector through the contract?

I would add one more.

What public powers still remain with the government anyway?

Once you look at those four points, the legal structure of a U.S. P3 starts to become much easier to read.

And then another question naturally follows.

Why do Florida and Virginia, or Texas and Colorado, have different laws and procedures even though they are all in the same country?

In #43, I will compare the actual P3 enabling laws of Virginia, Florida, Texas, and Colorado, and look at why the rules of U.S. P3s differ from state to state.

Key Sources

Virginia Code, Public-Private Transportation Act of 1995
Florida Statutes §334.30, Public-private transportation facilities
Charles River Bridge v. Warren Bridge, 36 U.S. 420 (1837)
Monongahela Navigation Co. v. United States, 148 U.S. 312 (1893)
Kelo v. City of New London, 545 U.S. 469 (2005)
Seven County Infrastructure Coalition v. Eagle County, 605 U.S. 168 (2025)

#41 – How Did U.S. P3s Begin?

foweuniverse — Sun, 6 Sep 2026 04:41:33 +0900

Korean Original
SOCFI #41 – 미국 P3는 어떻게 시작되었을까?
https://socfi.tistory.com/44

#41 - 미국 P3는 어떻게 시작되었을까?

- 미국 P3는 어떻게 시작됐을까요? 18세기 Turnpike부터 1956년 Interstate Highway System, 1991년 ISTEA, 91 Express Lanes·Dulles Greenway·Teodoro Moscoso Bridge까지 미국 교통 P3의 발전과정을 살펴봅니다. 민간자본보다

socfi.tistory.com

- How did U.S. P3s begin? This article traces the development of U.S. transportation P3s from early turnpikes and the 1956 Interstate Highway System to ISTEA and early modern projects such as the 91 Express Lanes, Dulles Greenway, and Teodoro Moscoso Bridge. It also examines how demand risk, tolls, non-compete clauses, and public policy authority shaped today’s P3 structures. -

When studying U.S. P3s, something can feel a little strange.

Some sources describe P3s as a relatively new way to deliver infrastructure. Yet other materials show that the history of toll roads in the United States goes back more than 200 years.

Private companies were already investing in roads and collecting tolls in the eighteenth century through what were known as turnpikes.

That raises an obvious question:

“Did U.S. P3s already exist 200 years ago?”

Not really.

The combination we now associate with modern P3s — private financing, long-term contracts, integrated design, construction and operations, and structured risk allocation — emerged much later.

For U.S. transportation P3s, the late 1980s and the 1990s were especially important.

Projects such as California’s 91 Express Lanes, Virginia’s Dulles Greenway, and Puerto Rico’s Teodoro Moscoso Bridge emerged around the same period.

At the federal level, the Intermodal Surface Transportation Efficiency Act of 1991, or ISTEA, also began opening more room for toll facilities and private participation within federal transportation policy.

But these early projects did not provide only success stories.

They exposed practical problems involving traffic forecasts, tolls, non-compete clauses, and the government’s ability to expand competing roads.

That is why the history of modern U.S. P3s is better understood not simply as a history of bringing private money into infrastructure, but as a process of learning — often through trial and error — which rights and risks should be transferred to the private sector, and how far that transfer should go.

Toll Roads in the United States Are Much Older Than Modern P3s

The history of toll roads in the United States goes back surprisingly far.

According to FHWA historical materials, demand for roads grew after American independence as westward expansion and trade between regions increased.

Government finances alone were often insufficient to meet that demand, which led to the emergence of private turnpike companies.

One of the best-known early examples was the Philadelphia and Lancaster Turnpike.

Pennsylvania chartered the company for the road in 1792, and it became one of the major early examples of toll-road development in the United States.

FHWA describes it as an important starting point in the early history of American turnpikes.

The word turnpike itself has an interesting origin.

It comes from the rotating barrier — the “turning pike” — that blocked the road until the traveler paid a toll.

The term survives today in road names such as the Pennsylvania Turnpike and Florida’s Turnpike.

Even today, turnpikes tend to be relatively expensive toll roads. When possible, I often find myself looking for routes that avoid them.

In broad terms, those early roads do resemble modern Toll Concessions.

Private companies raised capital, built roads, and recovered their investment by collecting tolls from users.

But they should not be treated as equivalent to today’s DBFOM P3s.

Modern P3s are procurement structures built around long-term contracts between public and private entities, detailed performance requirements, financing arrangements, and explicit risk allocation.

Eighteenth- and nineteenth-century turnpikes can reasonably be described as historical roots of modern P3s.

But that does not mean today’s P3 system already existed in the same form.

In the Twentieth Century, Government Became the Center of Highway Development

The United States did not continue developing primarily through private toll roads.

By the early twentieth century, the direction had changed.

FHWA historical materials show that many nineteenth-century turnpikes experienced financial difficulties or were eventually taken over by state or local governments and converted into public roads.

As automobiles became widespread and highways became essential to the national economy and national defense, the role of the public sector grew substantially.

A decisive turning point came with the Federal-Aid Highway Act of 1956.

That law launched the large-scale development of the Interstate Highway System.

For new Interstate construction, the standard funding structure was generally:

Federal government: 90%
State government: 10%

There is one point here that can be confusing.

“If a highway is part of the Interstate system, doesn’t that mean it must be free?”

Not necessarily.

Interstate refers to a highway’s role within the national Interstate Highway System. It does not mean “free road.”

When the Interstate network began expanding rapidly after 1956, toll highways such as the Pennsylvania Turnpike already existed.

From the government’s perspective, there was little reason to build an identical free highway directly alongside an existing toll road.

Some existing toll roads were therefore allowed to become part of the Interstate network.

The Pennsylvania Turnpike, for example, opened in 1940, before the Interstate system was created.

Portions of it were later designated as I-70 and I-76.

They remain tolled today.

A simple way to separate the concepts is:

Interstate = the road’s network classification and function
Toll road = whether users pay a toll

The larger trend, however, was clear.

From the 1950s onward, highway development in the United States became centered much more heavily on federal and state public funding than on privately financed toll roads.

Modern P3s emerged later, after this public-centered highway system had already become firmly established.

By the Late 1980s, States Began Looking Again to Private Capital

Why, then, did private capital return?

By the late 1980s, the transportation environment in the United States was changing.

A substantial highway network had already been built, but congestion in major metropolitan areas was worsening.

At the same time, the cost of building new roads and expanding existing facilities continued to rise.

Existing taxes and public funding were increasingly unable to cover every transportation investment need.

Some states began experimenting again with private financing and private operation.

Two important examples were California and Virginia.

California enacted Assembly Bill 680, or AB 680, in 1989.

The law authorized up to four pilot transportation infrastructure projects financed entirely with private funds.

According to FHWA materials, one of the purposes was to test whether private participation could help close the growing gap between transportation investment needs and limited public resources.

Virginia moved in a similar direction.

In 1988, the Virginia legislature enacted the Virginia Highway Corporation Act, allowing private companies to construct and operate toll roads for profit.

The Dulles Greenway was later developed under that authority.

This is an important point in the history of U.S. P3s.

The modern market was not created all at once by a nationwide federal P3 law.

Individual states facing fiscal constraints and growing transportation demand changed their own laws first and began experimenting with privately financed roads.

The decentralized structure discussed in #40 was therefore present from the beginning of the modern U.S. P3 era.

The Teodoro Moscoso Bridge Shows What an Early Long-Term Concession Looked Like

One major early-1990s example was the Teodoro Moscoso Bridge in Puerto Rico.

Puerto Rico is not a U.S. state, but it is a U.S. territory under U.S. jurisdiction.

According to the Puerto Rico Highways and Transportation Authority, or PRHTA, PRHTA and the private concessionaire Autopistas de Puerto Rico entered into a concession agreement on December 20, 1991.

The agreement covered the bridge’s design, construction, operation, and maintenance.

The bridge began operations on February 23, 1994.

Puerto Rico government materials place the investment at approximately $126 million.

Using a simple current conversion of KRW 1,400 per U.S. dollar, that would be roughly KRW 176 billion.

As in the Korean original, this KRW figure is only a simple conversion to help readers understand the scale. It is not the historical exchange-rate value of the project.

The financing structure is also interesting.

Part of the bridge construction cost was financed through Special Facilities Revenue Bonds issued by PRHTA.

The private operator was expected to repay the financing primarily through toll revenue.

The original concession term was 35 years, although the agreement was later amended and extended.

Some Puerto Rico government materials have described the project as the “first P3 in a U.S. jurisdiction.”

I would be careful about describing it unconditionally as the first P3 in the entire United States.

The answer depends on how P3 is defined.

If eighteenth- and nineteenth-century private turnpikes are included, the historical starting point changes.

If the definition is limited to privately operated toll roads or to modern long-term concessions, the answer changes again.

The safest way to understand the Teodoro Moscoso Bridge is as one of the representative early projects in the formation of modern U.S. transportation P3s.

Dulles Greenway Showed What It Really Means for the Private Sector to Bear Risk

Around the same time, Virginia was developing the Dulles Greenway.

The Dulles Greenway extends Route 267 from the area near Washington Dulles International Airport toward Leesburg.

The road was developed under the Virginia Highway Corporation Act of 1988 and opened in 1995.

At the time, Virginia government materials sometimes described it as the first private toll road in the United States in more than 150 years.

Again, that should not be read as meaning that it was literally the first privately developed road in U.S. history.

The eighteenth- and nineteenth-century turnpikes came much earlier.

A more accurate interpretation is that the Dulles Greenway represented the return of a modern privately financed toll road after a long period in which U.S. highway development had been dominated by the public sector.

What makes the Dulles Greenway important is not simply that a private company built a road.

It became an early example of the difficulty of a Toll Concession in which the private sector bears demand risk.

Building a road does not guarantee that the expected number of vehicles will use it.

If traffic volumes are lower than forecast, toll revenue falls.

That immediately affects the project’s ability to repay debt and generate returns for investors.

This is fundamentally the same demand-risk issue that Korea has dealt with for years in BTO projects.

The experience of these early privately financed toll roads eventually led to another question in U.S. P3 practice:

Is transferring demand risk to the private sector always the most efficient form of risk allocation?

The rise of the Availability Payment model discussed in #40 is connected to this history.

The 91 Express Lanes Showed Both the Potential and the Risks of P3 Contracts

One of the most interesting early modern U.S. P3 projects was California’s 91 Express Lanes.

The project was one of the four private transportation infrastructure pilot projects authorized under AB 680 in 1989.

The project added approximately 10 miles, or about 16 kilometers, of Express Lanes in the median of SR-91 in Orange County.

There were four tolled express lanes in total, two in each direction.

The private consortium California Private Transportation Company, or CPTC, financed and built the project.

After construction, ownership of the facility transferred to the State of California through Caltrans, while CPTC retained the right to operate it over the long term.

The contractual private operating period could extend for up to 35 years.

At Financial Close in July 1993, three banks provided approximately $65 million in project loans.

Kiewit provided another approximately $35 million in project debt.

CPTC itself invested approximately $19 million in Equity.

The total private investment at the time was approximately $130 million.

Using the same simple conversion of KRW 1,400 per dollar, that would be roughly KRW 182 billion.

The road opened on December 20, 1995.

It became one of the first major U.S. examples of what we would now call a priced managed lane.

In other words, toll prices could vary depending on factors such as traffic congestion and time of day to manage demand.

Up to this point, the project looks like a highly successful privately financed highway project.

And in terms of usage, the lanes were indeed popular.

The major problem emerged somewhere else.

A Contract Designed to Protect Private Revenue Ended Up Restricting Public Road Expansion

The 91 Express Lanes concession included a non-compete clause.

The logic behind it was understandable.

If the private concessionaire was going to assume demand risk, it needed some protection for future toll revenue.

The problem was how that protection was structured.

The contract restricted certain capacity expansions on the general-purpose lanes of SR-91.

The basic logic was straightforward.

If the private sector invested its own money to build tolled express lanes and the government then kept expanding free lanes immediately next to them, traffic could shift away from the tolled facility.

From the private investor’s perspective, some form of protection made sense.

Over time, however, congestion on the general-purpose lanes became increasingly severe.

Public authorities wanted to expand road capacity, but the contractual non-compete provision became an obstacle.

A provision designed to protect the private concessionaire from revenue risk was now limiting the government’s ability to make future transportation-policy decisions.

Eventually, the Orange County Transportation Authority, or OCTA, purchased CPTC’s operating rights in January 2003 for approximately $207.5 million.

Using the same simple current conversion, that would be about KRW 290 billion.

The problematic non-compete clause was then removed.

The lesson this project left for later U.S. P3s was significant.

When risk is transferred to the private sector, the private sector must also receive rights that correspond to that risk.

But if those rights are protected too strongly, they can restrict the government’s ability to make future policy decisions.

The key question in a P3 contract is therefore not simply how much risk is transferred.

It is which risk is transferred, and what rights the private sector receives in return for taking that risk.

ISTEA Began Changing the Federal Government’s Role in 1991

While these state-level experiments were developing, federal transportation policy was also changing.

The key legislation was ISTEA.

ISTEA stands for the Intermodal Surface Transportation Efficiency Act of 1991.

A literal Korean translation would be something like the “1991 Intermodal Transportation Efficiency Act.”

Enacted in 1991, ISTEA became one of the major turning points in U.S. transportation policy.

From a P3 perspective, Section 1012 was particularly important.

According to FHWA, the section expanded the ability of federal highway programs to work with toll facilities and public-private cost sharing.

It created more flexibility to combine federal support with toll facilities and widened the potential use of innovative finance and private investment.

FHWA materials indicate that, following ISTEA, 15 states enacted supplementary legislation to authorize P3s, while other states substantially modified their existing laws.

But the distinction is important.

ISTEA did not create a single nationwide U.S. P3 law.

Its role was closer to expanding the space within federal transportation policy in which these kinds of projects could be used.

States then developed their own laws and expanded their own P3 markets.

The structure described in #40 — federal policy + state law + individual contracts — began taking much clearer shape during this period.

Modern U.S. P3s Were Not Created All at Once

When the early projects of the 1990s are viewed together, a pattern becomes visible.

What is particularly interesting is that the law was not fully completed first and the projects launched afterward.

Instead, the process was iterative.

States changed their laws.

They launched actual projects.

Problems emerged.

Contracts and policies were then revised in response.

California confronted the non-compete issue through the 91 Express Lanes.

Virginia continued to deal with questions involving traffic demand and toll levels on privately financed roads.

At the same time, these projects demonstrated the potential of private financing, electronic tolling, congestion pricing, and long-term concessions.

As these experiences accumulated, U.S. P3s became more sophisticated during the 2000s.

They evolved into the more structured DBFOM, Toll Concession, and Availability Payment models seen today.

The Starting Point of U.S. P3s Was Ultimately About Risk, Not Just Money

Looking at the history of U.S. P3s, private capital is the first thing that tends to stand out.

When public funding is insufficient, private investors provide money and build infrastructure.

That is certainly part of the story.

But the deeper questions raised by these early projects were about risk.

The 91 Express Lanes raised this question:

If demand risk is transferred to the private sector, how far should the government go in protecting private revenue?

The Dulles Greenway raised another:

If traffic is lower than expected, can the private sector really bear that risk all the way through the concession?

The Teodoro Moscoso Bridge raised yet another:

How should private operation, toll revenue, and bond financing be combined within a long-term concession?

These questions are not unfamiliar in Korean PPP practice.

In the end, Korea and the United States arrive at many of the same issues:

Who bears construction risk?
Who bears demand risk?
How far should private revenue be protected?
How much policy flexibility should remain with the public sector?

The United States did not answer these questions through a single nationwide PPP statute.

Instead, the answers developed gradually through state laws, individual contracts, and federal transportation and financing policies.

That is why U.S. P3s still look different from state to state today.

In the next article, I will go one step deeper into this structure.

Who actually decides and takes responsibility for a U.S. P3?

In #42, I will look at how authority is divided among USDOT, FHWA, State DOTs, local governments, Toll Authorities, and Transit Agencies.

Key Sources

Federal Highway Administration (FHWA) – Toll Facilities in the United States: History and Policy
Federal Highway Administration (FHWA) – Why Does the Interstate System Include Toll Facilities?
Federal Highway Administration (FHWA) / Build America Bureau – 91 Express Lanes Project Profile
Virginia Department of Transportation – materials on the Virginia Highway Corporation Act of 1988
Puerto Rico Highways and Transportation Authority (PRHTA) and Puerto Rico government financing materials

#40 – How Are U.S. P3s Different from Korea’s PPP Projects?

foweuniverse — Sun, 6 Sep 2026 03:50:40 +0900

Korean Original:
SOCFI #40 – 미국 P3는 한국 민자사업과 무엇이 다를까?
https://socfi.tistory.com/43

#40 - 미국 P3는 한국 민자사업과 무엇이 다를까?

- 미국 P3는 한국 BTO·BTL과 무엇이 다를까요? 민자철도 업무에서 접했던 Availability Payment(AP)에 대한 의문에서 출발해 DBFOM, Toll Concession, AP의 차이와 USDOT·FHWA·State DOT·Build America Bureau의 역할을 설명

socfi.tistory.com

- How are U.S. P3s different from Korea’s BTO and BTL projects? This article compares DBFOM, Toll Concessions, Availability Payments, risk allocation, and the roles of USDOT, FHWA, State DOTs, and the Build America Bureau, using Florida’s I-4 Ultimate, I-595, and PortMiami Tunnel as examples. -

One of the questions I dealt with repeatedly while working on privately financed rail projects in Korea was how demand risk should be shared between the private sector and the government.

If traffic demand turns out to be lower than expected, who should bear that risk? If demand is higher than expected, who should benefit from the additional revenue? And if the government takes on part of the risk, how far should that support go?

These questions come up again and again in infrastructure PPPs.

At one point, while reviewing proposals for improving Korea’s PPP system, I heard quite a few suggestions that Korea should adopt the U.S. Availability Payment, or AP, model.

My first reaction was:

“How is the AP model really that different from Korea’s BTL?”

At a high level, they do look quite similar. The private sector builds the facility and arranges financing, and the government makes payments over a long period.

But after coming to the United States, studying here, and reading more U.S. infrastructure and P3 reports, I began to see the issue differently.

The difference between an AP and a BTL is not simply a matter of using a different name for the payment mechanism.

Behind the AP model is a broader system that differs considerably from Korea’s: the laws authorizing P3s, the government structure that decides which projects move forward, contract design, financing arrangements, and the way toll and demand risks are allocated.

So, starting with this article, I want to take a closer look at U.S. Public-Private Partnerships, generally referred to as P3s.

And I will start with the most basic question.

How are U.S. P3s different from Korea’s PPP projects?

Thinking of U.S. P3s as “American Versions of BTO or BTL” Can Actually Make Them Harder to Understand

Before going further, it helps to introduce several U.S. institutions that will appear repeatedly throughout this series.

The U.S. Department of Transportation (USDOT) is the federal department responsible for transportation policy. In Korean institutional terms, it is closest to the transportation functions of Korea’s Ministry of Land, Infrastructure and Transport.

Under USDOT are several modal agencies.

One that will appear frequently in this series is the Federal Highway Administration (FHWA), which focuses on highways and bridges.

FHWA administers federal highway funding, develops highway policies and technical standards, and provides guidance and analytical tools on issues such as P3 risk allocation, delivery methods, and financing.

But FHWA should not be thought of as the equivalent of a PPP division within Korea’s Ministry of Land, Infrastructure and Transport.

In the United States, actual highway P3 projects are often procured by State Departments of Transportation, or State DOTs.

Florida, for example, has the Florida Department of Transportation (FDOT).

For convenience, the relationship can be pictured roughly as:

Federal level: USDOT / FHWA
State level: State DOTs

But this does not mean there is an organizational chain of command between them.

FHWA is a federal agency. State DOTs are agencies of their respective state governments. A State DOT is not a regional office subordinate to FHWA.

Another important institution is the Build America Bureau.

Located within USDOT, the Build America Bureau focuses on infrastructure finance programs such as TIFIA, RRIF, and PABs, while also providing technical assistance for P3 projects.

So when looking at U.S. highway policy and P3 delivery, FHWA appears frequently. When looking more specifically at P3 financing, the Build America Bureau often becomes the more directly relevant institution.

USDOT describes P3s as contractual arrangements between public and private entities in which the private sector may assume additional project risks beyond design and construction, including financing, long-term operations, and in some cases transportation revenue risk.

FHWA similarly describes transportation P3s as long-term contractual arrangements that transfer additional responsibilities and risks to the private sector.

So far, this may sound quite similar to Korea’s PPP system.

Under a Korean BTO (Build-Transfer-Operate) project, the private sector finances and builds the facility and then operates it over a long concession period.

Under a BTL (Build-Transfer-Lease) project, the private sector also builds and finances the facility and remains involved in long-term management.

The difference begins to emerge in the way the system itself is structured.

For the U.S. side, this series will focus primarily on transportation infrastructure P3s, where USDOT and FHWA have accumulated relatively extensive guidance and project documentation.

The same framework should not automatically be generalized to P3s in U.S. water, energy, or public-building sectors.

Korea Defines PPP Delivery Methods in Law, While in the U.S. the Structure Often Becomes Clearer Through the Contract

Korea has a nationwide statute governing PPP projects: the Act on Public-Private Partnerships in Infrastructure.

Article 4 of the Act directly specifies the methods through which PPP projects may be implemented.

The structure commonly referred to as BTO, or Build-Transfer-Operate, involves the private sector building the facility, transferring ownership to the national or local government upon completion, and then receiving the right to operate and manage the facility for a specified period.

In other words:

Build → Transfer → Operate

BTL stands for Build-Transfer-Lease.

The private sector builds the facility and transfers ownership to the government, after which the government or another public entity leases the facility during the agreement period.

The Korean government also establishes and announces a national Basic Plan for Public-Private Partnership Projects, under which individual competent authorities develop specific projects.

As a result, Korea’s PPP system can generally be understood through a relatively common sequence:

PPP Act
→ Enforcement Decree and Basic Plan for PPP Projects
→ Project designation and announcement
→ Selection of the private concessionaire
→ Concession Agreement

The United States does not have a single nationwide P3 framework law that corresponds exactly to Korea’s PPP Act.

And when reading U.S. P3 documents, another acronym appears very frequently: DBFOM.

DBFOM stands for:

Design-Build-Finance-Operate-Maintain

The structure becomes clear when the acronym is broken down:

Design + Build + Finance + Operate + Maintain

These responsibilities are bundled into a single long-term contract.

There is an important distinction here.

BTO and BTL are project delivery structures defined within Korea’s PPP legal framework.

DBFOM, by contrast, primarily describes which responsibilities are bundled together and assigned to the private sector under a contract.

Acronym	Full Name	Scope Assigned to the Private Sector
DB	Design-Build	Design + Construction
DBF	Design-Build-Finance	Design + Construction + Financing
DBOM	Design-Build-Operate-Maintain	Design + Construction + Operations + Maintenance
DBFOM	Design-Build-Finance-Operate-Maintain	Design + Construction + Financing + Operations + Maintenance

DBFOM itself does not tell us how the private concessionaire earns its revenue.

Two DBFOM projects can use very different payment structures.

In one project, the private concessionaire may collect toll revenue to recover its investment.

In another, the public agency may make long-term Availability Payments.

Put very simply:

DBFOM tells us what the private sector is responsible for.
Toll or AP tells us how the private sector gets paid.

Once this distinction is clear, U.S. P3 terminology becomes much easier to follow.

Even Under the Same DBFOM Structure, Toll Concessions and AP Projects Can Have Very Different Revenue Models

DBFOM does not tell us where the private concessionaire’s revenue comes from.

Two common structures can exist under the same DBFOM framework.

The first is a Toll Concession.

A toll is a user charge. A concession can be understood as a long-term agreement under which a public authority grants a private concessionaire the right and responsibility to operate an infrastructure asset for a specified period.

In a typical Toll Concession, toll revenue becomes the private concessionaire’s main source of income, and the private party assumes Traffic and Revenue Risk — the risk associated with traffic demand and toll revenue.

But it would be wrong to assume that every Toll Concession transfers 100 percent of demand risk to the private sector.

The actual allocation of risk depends on the individual contract. In theory, a government could provide minimum-revenue support or share part of the risk.

So the important question is not simply what the project is called.

What matters is who actually bears the risk of lower traffic volumes and insufficient toll revenue under the concession agreement.

The second major structure is the Availability Payment Concession, or AP model.

“Availability” refers, quite literally, to whether the facility is available for use.

If the private concessionaire keeps the infrastructure available at the agreed performance level, the public agency makes long-term payments.

If the facility is unavailable or fails to meet agreed operating and performance standards, those payments may be reduced.

FHWA identifies factors such as completion timing, incident response, and facility condition as examples of performance criteria that can be tied to Availability Payments.

A very simplified comparison with Korea looks like this:

Category	Similar Structure That May Come to Mind in Korea	U.S. P3
Scope of Private-Sector Responsibilities	Determined within BTO, BTL, etc.	DB, DBF, DBFOM, etc.
User-Fee-Based Revenue	BTO	Toll Concession
Long-Term Government Payment	BTL	Availability Payment
Core Project Contract	Concession Agreement under Korea’s PPP framework	Concession Agreement, Project Agreement, etc.

There is one point that requires particular caution.

BTO ≠ Toll Concession, and BTL ≠ Availability Payment.

There are economic similarities, but ownership arrangements, toll-setting authority, government payment mechanisms, termination compensation, financial support, and risk allocation can all differ depending on the applicable law and individual contract.

AP Looks Similar to BTL, but They Start from Different Institutional Concepts

This was exactly what I wondered about when I first encountered the AP model.

“If the government pays a private concessionaire over a long period, isn’t that basically the same as BTL?”

On the surface, the similarities are obvious.

Under Korea’s BTL structure, the private sector builds the facility and arranges financing.

The government makes payments over a long period, and those payments may be reduced when operating performance falls below specified standards.

In a U.S. AP P3, the private sector may design, build, finance, operate, and maintain the facility, while the public agency pays long-term Availability Payments.

So AP and BTL clearly share a number of characteristics.

But the starting point of the two concepts is somewhat different.

In Korea, BTL is a method of implementing PPP projects expressly provided for under Article 4 of the PPP Act.

In the United States, an AP is more accurately understood as a payment mechanism that determines how a private concessionaire is compensated under a long-term P3 contract such as a DBFOM agreement.

To simplify the distinction:

BTL is closer to answering, “What type of PPP project is this?”
AP is closer to answering, “Under what conditions will the private concessionaire be paid?”

That does not mean BTL payments are unrelated to performance while AP payments are performance-based.

Korean BTL projects also have performance requirements and operating-performance evaluations, and government payments can be reduced depending on the results.

The difference between AP and BTL therefore needs to be understood through the broader legal framework, risk allocation, and contract design rather than through the existence or absence of performance assessment alone.

Florida’s I-4 Ultimate Shows That Availability Payments and Tolls Can Exist in the Same Project

The distinction becomes clearer when we look at an actual project.

One of the roads I use when traveling to places such as Walt Disney World or Universal Studios is I-4 Ultimate, one of Florida’s major P3 projects.

The project was procured by the Florida Department of Transportation (FDOT).

I-4 Ultimate involved the full reconstruction and expansion of approximately 21 miles, or about 34 kilometers, of Interstate 4 through central Orlando.

The project reconstructed the existing general-purpose lanes, added four Express Toll Lanes in the median, and constructed, rebuilt, or expanded 15 major interchanges and approximately 140 bridges.

The private concessionaire is I-4 Mobility Partners.

The project uses a DBFOM structure, bundling design, construction, financing, operations, and maintenance into a long-term P3 agreement.

The contract was executed in 2014, major construction began in 2015, and the Express Lanes opened in February 2022.

Here is the interesting part.

The road has Express Lanes that actually charge tolls.

At first glance, one might expect a Toll Concession structure in which the private concessionaire collects the tolls and uses that revenue to recover its investment.

That is not how this project works.

FDOT sets the toll rates and receives the toll revenue.

The private concessionaire instead receives Availability Payments from FDOT.

So a tolled road is not automatically a Toll Concession.

In the I-4 Ultimate project, long-term traffic demand and toll-revenue risk remain with the public sector, while the private sector primarily assumes design, construction, financing, operations, maintenance, and performance risks.

This does not mean every AP project allocates risk exactly the same way as I-4 Ultimate.

The specific allocation of demand, toll-revenue, and operating risks cannot be determined simply because a project is called an AP project. The contract for each individual project must be examined.

This is why, rather than translating AP simply as an “American version of BTL,” it is more useful to look directly at how risks are allocated.

I-4 Ultimate Used Federal TIFIA Financing Alongside Private Capital

According to publicly available project information, the Project Cost of I-4 Ultimate was approximately $2.877 billion (about KRW 4.03 trillion).

For the KRW figures in this article, I use a simple conversion rate of approximately KRW 1,400 per U.S. dollar to make the scale of the projects easier for Korean readers to understand.

These figures do not represent the actual exchange rate at the time of the project or any official KRW-denominated project cost.

The approximately $2.8 billion project was not financed entirely with private equity and commercial bank loans.

A major component was TIFIA, another term that appears frequently in U.S. P3 finance.

TIFIA stands for the Transportation Infrastructure Finance and Innovation Act.

Technically, TIFIA is the name of a federal statute, but in practice the term is widely used to refer to the federal transportation infrastructure credit program authorized under that law.

TIFIA provides federal credit assistance, including direct loans and loan guarantees, for large transportation infrastructure projects.

Put simply:

It allows the federal government to provide long-term financing on relatively favorable terms for major transportation infrastructure projects.

The TIFIA financing for I-4 Ultimate was approximately $949 million (about KRW 1.33 trillion).

The project also included approximately $484 million (about KRW 680 billion) in Senior Bank Debt, and approximately $103.5 million (about KRW 145 billion) in private Equity.

FDOT also provided approximately $1.035 billion (about KRW 1.45 trillion) in Milestone Payments during the construction period.

A simplified version of the financing structure looks like this:

Private Equity + Bank Debt + TIFIA + FDOT Funding + Long-Term Availability Payments

All of these elements were combined within a single P3 project.

So when we use the term “private investment” in the context of U.S. P3s, we should not assume that the entire project cost is financed through purely private capital.

The way public and private financing are combined is one of the important features of U.S. P3 finance.

PABs Are Another Important Part of U.S. P3 Finance

Another acronym that appears repeatedly in U.S. P3 finance is PABs.

PABs stands for Private Activity Bonds.

A direct translation into Korean would sound something like “private activity bonds,” but understanding the financing structure is more useful than relying on a literal translation.

In the United States, when a private entity develops, builds, or operates qualifying transportation infrastructure that serves a public purpose, it may under certain conditions use tax-exempt debt to finance the project.

Because the interest received by bond investors may receive favorable tax treatment, these bonds can potentially be issued at lower interest rates than ordinary taxable bonds.

This can reduce the financing cost of a P3 project.

PABs should not be confused with a government grant.

The federal government does not directly pay the private concessionaire’s project cost through PABs.

Rather, qualifying public infrastructure projects can access bonds with tax advantages. Since investors receive favorable tax treatment on the interest, financing may be available at lower rates than through comparable taxable debt.

That can reduce overall project financing costs.

TIFIA and PABs will be examined separately in greater detail later in this series on U.S. P3 finance.

Even Within Florida, P3 Contracts Do Not Follow a Single Model

I-4 Ultimate is not Florida’s only Availability Payment P3.

Earlier projects included the I-595 Improvements Project and the PortMiami Tunnel.

The I-595 Improvements Project entered into a 35-year DBFOM concession agreement in 2009.

The facility opened in 2014, and the private concessionaire receives performance-based Availability Payments during the operations and maintenance period.

FDOT reports a Capital Cost of approximately $1.2 billion (about KRW 1.68 trillion).

The PortMiami Tunnel followed a similar structure.

A 35-year DBFOM concession agreement was executed in 2009, and the tunnel opened in 2014.

The private concessionaire also receives performance-based Availability Payments during the long-term operations and maintenance period.

FDOT reports a Capital Cost of approximately $667 million (about KRW 930 billion).

Even within Florida, FDOT has used a range of structures, including DBF, BF, and DBFOM.

U.S. P3 development therefore cannot be understood as the adoption of a single project model called “AP.”

It is closer to a project-by-project combination of decisions:

Which responsibilities among design, construction, financing, operations, and maintenance should be bundled into one contract?

Which risks should be transferred to the private sector?

And how should the private concessionaire be paid?

The United States Does Not Have a Single Nationwide P3 Law Equivalent to Korea’s PPP Act

If I had to identify one of the most important institutional differences between Korean and U.S. P3s, this would be it.

Whether a PPP project is developed in Jeju or Seoul, Korea generally starts from the same Act on Public-Private Partnerships in Infrastructure.

The PPP Act defines project implementation methods, and the national government establishes the Basic Plan for PPP Projects.

There is no nationwide U.S. P3 law that corresponds exactly to this Korean framework.

The federal government influences P3 projects through transportation policy, federal funding, tax policy, environmental review, and financing programs.

But whether a particular P3 can actually be undertaken depends heavily on state law and the legal authority of the relevant public agency.

That changes the order in which questions are asked.

In Korea, a common question is:

“Should this project be implemented as a BTO, a BTL, or another PPP structure?”

In the United States, an earlier question may need to be:

“Does this state and this public agency have the legal authority to enter into this type of long-term P3 contract?”

The answer may be different in Florida, Virginia, Texas, and Colorado.

This does not mean every U.S. state must have a separate standalone P3 statute.

State legal structures and authorization vary. Some P3s may rely on existing transportation statutes or specific legal powers granted to particular agencies.

So when examining U.S. P3 authority, it is not enough simply to ask whether a state “has a P3 law.”

We also need to ask what types of infrastructure are authorized, which public entities may enter into these contracts, and what approval procedures are required.

I will look more closely at why P3 rules differ from state to state in #43.

Understanding the Different Roles of USDOT, FHWA, State DOTs, and the Build America Bureau Helps Explain U.S. P3s

Here is a summary of the institutions discussed so far./

Institution	Full Name	Meaning Role in U.S. P3s
USDOT	U.S. Department of Transportation	Overall federal transportation policy
FHWA	Federal Highway Administration	Highway and bridge policy, federal funding, technical assistance for highway P3s
State DOT	Department of Transportation of each state	Key procurement and contracting authority for state highway projects
Build America Bureau	USDOT infrastructure finance and project-support office	TIFIA, RRIF, PABs, P3 finance and technical assistance
FTA	Federal Transit Administration	Public transportation, including urban rail and bus systems
FRA	Federal Railroad Administration	Railroad policy, safety, and rail project support

The key is to distinguish the roles of the federal government and state governments.

FHWA may administer federal highway funding and policy, but the agency that actually entered into the I-4 Ultimate contract was FDOT, not FHWA.

At the same time, the federal government was deeply involved through financing support such as TIFIA.

In U.S. P3s, the agency that signs the contract and the government that supports the project through funding, financing, or policy are often not the same entity.

I will look at this division of governmental responsibilities in more detail in #42.

Private Money Alone Does Not Make a Project a P3

A project does not become a P3 simply because a private construction company builds a highway.

Nor does it become a P3 simply because a private financial institution buys project bonds.

Under the traditional DBB (Design-Bid-Build) model, the public agency completes the design and then selects a construction contractor through bidding.

Under DB (Design-Build), design and construction are combined into one contract.

Under DBF (Design-Build-Finance), financing is added.

Under DBFOM, long-term operations and maintenance are added as well.

As more functions are bundled, the private sector may assume broader responsibilities and greater lifecycle obligations.

But bundling more responsibilities does not automatically make a project a “better P3.”

Efficiency can improve when risks are transferred to the party that is better able to manage them.

If risks that the private sector cannot realistically control are transferred too aggressively, financing costs or overall project costs may instead increase.

So the real question is not simply whether private capital is involved.

The more useful questions are: What responsibilities and risks have been transferred to the private sector under a long-term contract, and is that allocation reasonable?

When looking at U.S. P3 projects, I find it more useful to start with questions like these:

Question	What Does It Tell Us?
Who finances the construction cost?	Public and private financing responsibilities
Who bears demand risk?	A central distinction between Toll and AP structures
How does the private sector recover its investment?	Toll revenue, Availability Payments, or other revenue structures
How does operating performance affect revenue?	Performance risk and long-term management responsibility
Who has authority to set toll rates?	Revenue risk and the degree of public control

Viewed this way, it becomes easier not to get too caught up in labels such as BTO, BTL, DBFOM, Toll Concession, and AP.

It also becomes easier to understand a project like I-4 Ultimate, where tolls and Availability Payments exist at the same time.

What matters more than what a model is called is who actually bears each risk.

That brings me back to the same question I kept dealing with while working on privately financed rail projects in Korea.

Korea and the United States Are Trying to Solve Similar Problems Through Different Institutional Systems

When I first encountered Availability Payments, I wondered how different they really were from Korea’s BTL structure.

I now see the issue somewhat differently.

Both systems use private capital and private-sector expertise to build infrastructure and manage it over a long period.

But they construct that relationship differently.

In Korea, the starting point is a nationwide institutional framework: the PPP Act and the Basic Plan for PPP Projects.

Delivery structures such as BTO and BTL also originate from the statutory framework.

In the United States, understanding a P3 requires looking at several pieces together: state-level legal authority, the actual procuring agency, the contractual scope defined through structures such as DBFOM, Toll or AP revenue mechanisms, and federal financing tools such as TIFIA and PABs.

So instead of immediately translating a U.S. P3 into Korean PPP terminology, it is more accurate to ask, in order:

Who decides to pursue the project?
Who signs the contract?
Which risks are transferred to the private sector?
How does the private concessionaire recover its investment?
And where does the federal government provide support?

The comparisons in this article between BTO and Toll Concessions, and between BTL and Availability Payments, are intended to make the concepts easier to understand.

They do not mean that the underlying legal systems are equivalent.

DBFOM is also not a term at the same conceptual level as BTO or BTL.

DBFOM describes the scope of responsibilities assigned to the private sector.

Toll and AP mainly describe the private concessionaire’s revenue or payment structure.

BTO and BTL are project implementation methods established under Korea’s PPP legal framework.

Once this distinction is clear, U.S. P3 documents become much easier to read.

And this will be the basic framework I use throughout the rest of this U.S. P3 series.

So, did the United States have this kind of system from the beginning?

No.

The United States also went through an era of older toll roads and predominantly public highway construction.

Over time, changes in federal transportation policy and various state-level experiments with privately financed roads contributed to the development of today’s long-term concession-based P3 market.

In the next article, I will trace that development through early projects such as the 91 Express Lanes, Dulles Greenway, and Teodoro Moscoso Bridge, along with changes in federal transportation policy, to see how the U.S. P3 market evolved into its current form.

Key Sources

U.S. Department of Transportation, Build America Bureau – Public-Private Partnerships (P3)
Federal Highway Administration (FHWA) – P3 Fact Sheet / Availability Payments
Build America Bureau – I-4 Ultimate Project
Florida Department of Transportation – Public-Private Partnership Summary
Build America Bureau – TIFIA / Private Activity Bonds
Korean Law Information Center – Act on Public-Private Partnerships in Infrastructure and the Basic Plan for Public-Private Partnership Projects

#70 - 미국 P3에서 한국 민자사업이 실제로 가져올 것은 무엇일까?

foweuniverse — Tue, 1 Sep 2026 13:46:49 +0900

- 한국 민자사업에서 민간은 사업기간 단축, 위험배분 개선, 자기자본 부담 완화, 자금재조달 기준 개선, 부대사업 확대와 금융조달 다양화를 꾸준히 요구해왔습니다. 미국의 Market Sounding, TIFIA, PABs와 정부 지원체계를 놓고 어떤 요구는 현행 제도로 풀 수 있는지, 어떤 것은 법·세제·예산·조직 개편이 필요한지 실무적으로 살펴봅니다. -

※ 미국 금액은 비교 편의를 위해 $1=1,400원으로 단순 환산했습니다.

* 이 글은 필자의 개인적인 의견으로 정부의 정책방향 및 의도와는 전혀 무관합니다.

미국 P3를 거의 30편 가까이 살펴보고 나니 마지막에는 자연스럽게 이런 질문이 남았습니다.

그래서 한국 민자사업에 실제로 무엇을 가져오면 좋을까요?

처음에는 TIFIA나 PABs처럼 눈에 띄는 금융제도가 먼저 보였습니다. 그런데 미국 제도를 가져오기 전에 한국 민자시장에서 오래 반복된 문제부터 다시 볼 필요가 있습니다.

민간이 그동안 정부에 요구해온 내용은 꽤 일관됩니다. 사업을 너무 오래 끌지 말아달라, 민간이 통제하기 어려운 위험을 일률적으로 넘기지 말아달라, 자기자본과 자금재조달 기준을 현실화해달라, 부대사업을 넓혀달라, 주무관청의 전문성도 높여달라는 요구입니다.

대한건설협회가 2022년 정부에 제출한 건의에도 정부고시사업 확대, 신유형 사업 활성화, 부대사업 확대, 민간제안사업 신속처리, 최소 자기자본비율 완화, 자금재조달 기준 개선 등이 들어 있었습니다. (뉴시스)

그런데 2026년 현재 다시 보면 일부는 이미 움직였습니다.

정부는 AI 데이터센터·전력망·철도복합시설 등으로 민자 대상을 넓히고, 단순 운영형 민자도 도입하기로 했습니다. 국민참여 인프라펀드를 만들고 적격성조사와 민투심 절차도 줄이기로 했습니다. 물가반영과 전력비 정산 같은 현장의 요구도 일부 손봤습니다. 정부 발표 기준 지난 30년간 추진된 민자사업은 872건, 약 154조원입니다. (대한민국 정책브리핑)

그러니 지금 필요한 것은 오래된 건의사항을 다시 나열하는 일이 아닙니다.

왜 일부 요구는 여전히 남아 있는지, 그리고 미국에서 그 문제를 다른 방식으로 풀고 있다면 무엇을 가져올 수 있는지를 보는 게 더 중요합니다.

협상기간을 줄이려면 먼저 한국과 미국이 어디에서 시간을 쓰는지 봐야 합니다

민간이 가장 오래 요구해온 것 가운데 하나가 사업추진기간 단축입니다.

과거 KDI 자료에서는 도로 민자사업의 전체 추진기간이 약 10년 수준까지 길어지는 문제가 지적되기도 했습니다. 다만 이 숫자를 미국 P3의 RFQ부터 Financial Close까지 걸리는 기간과 바로 비교해서는 안 됩니다. 사업발굴·검토·승인·협상 등 포함범위가 다르기 때문입니다.

미국이라고 P3가 금방 끝나는 것도 아닙니다.

FHWA 자료를 보면 민간금융 P3는 RFQ, RFP, 제안평가, 협상, Commercial Close, Financial Close까지 상당한 시간이 필요합니다. TIFIA나 PABs가 들어가면 연방 심사와 금융절차가 추가됩니다.

그런데 절차를 따라가다 보면 숫자보다 더 중요한 차이가 보입니다.

미국은 Preferred Bidder를 선정하기 전에 시장과 계약조건을 상당 부분 검증합니다.

RFI(Request for Information·시장정보요청), Industry Forum, Draft RFP, Shortlisted Bidder와의 1:1 미팅, Alternative Technical Concept(ATC·대안기술제안) 등을 이용해 Final RFP를 만들기 전에 기술·계약·금융조건을 반복해서 확인합니다. (교통부)

이 차이가 생각보다 큽니다.

한국에서 우선협상대상자를 선정하면 이후에는 사실상 한 팀과 협상합니다. 그때 금융조건이나 중요한 위험배분 문제가 새로 드러나면 협상이 길어질 수밖에 없습니다.

미국식 접근은 경쟁자가 여러 명 남아 있을 때 더 많은 조건을 확정하는 쪽에 가깝습니다.

그래서 한국의 협상기간을 줄이는 방법도 단순히 “6개월 안에 끝내자”는 식으로 가면 안 됩니다.

필자가 보기에는 적격성조사 → 제3자공고 → 우선협상대상자 → 장기간 협상의 일부를

적격성조사 → Market Sounding → Draft 공고조건 → Industry Review → Final 공고 → 우선협상대상자 → Closing Negotiation으로 앞당기는 것이 더 효과적이라고 생각합니다.

민간 입장에서도 좋을겁니다.

입찰하기 전에 금융기관이 Bankability(금융조달 가능성)를 검토할 수 있기 때문입니다. “이 협약조건으로는 대출이 안 됩니다”라는 이야기가 협상 막판에 나오는 것을 줄일 수 있습니다.

다만, 왜 한국에서 이런 방식이 쉽게 확산되지 않았는지도 들여다 봐야합니다.

특정 사업자와 공고 전에 깊게 이야기를 하면 특혜나 정보비대칭 논란이 생길 수 있습니다. 그래서 미국도 Industry Review를 하면서 비밀정보의 범위, 이해충돌, 동일정보 제공과 기록관리 같은 절차를 같이 둡니다.

한국에 도입하려면 Market Sounding 한 줄을 넣는 것으로는 부족합니다.

참여자 등록, 표준질문, 제공자료의 범위, 1:1 미팅 기록, 영업비밀 처리, Final RFP 변경사유 관리까지 같이 만들어야 합니다.

이 부분은 우선 법 개정보다 민간투자사업기본계획과 PIMAC 세부지침에서 시작할 수 있지 않을까 생각합니다.

위험을 민간에 넘긴다고 계약서에 쓰면 그 위험이 없어질까요?

민간투자사업에서 거의 모든 쟁점은 결국 돈으로 돌아옵니다.

물가, 보상, 인허가, 지장물, 금리, 전력비, 법령변경, 불가항력, 수요 등이 쟁점의 유형입니다.

민간 입장에서는 자신이 통제할 수 없는 위험까지 부담하면 그 위험을 입찰가격에 넣을 수밖에 없습니다.

정부 입장도 이해됩니다.

민간 비용이 늘 때마다 정부가 보전해주면 Risk Transfer(위험이전)의 의미가 사라집니다.

이 문제는 국내 학계에서도 오래 지적됐습니다. 이동훈은 「고려법학」에서 민자사업에서 합리적인 위험배분이 중요하다는 점은 널리 인정되고 있지만, 실제 법령과 실시협약에서 위험이 어떻게 배분되고 이를 어떻게 보완해야 하는지에 대한 법리적 논의는 충분하지 않았다고 지적했습니다. (KCI)

미국 FHWA가 강조하는 원칙도 비슷합니다.

위험을 가장 많이 넘기는 것이 아니라 가장 잘 통제할 수 있는 주체에게 두는 것입니다. 민간이 통제하기 어려운 위험을 과도하게 넘기면 Risk Premium이 올라가고 결국 공사비·통행료·Availability Payment에 반영됩니다.

반대로 정부가 모든 위험을 떠안으면 민간의 Capital at Risk가 줄어 성과유인도 약해집니다.

그래서 필자는 한국에서도 Risk Register를 단순 참고자료보다 한 단계 높여 사업방식과 정부지원 규모를 결정하기 전 공식 산출물로 만드는 게 어떨까하고 제안해봅니다.

위험별로 발생확률 × 비용영향 × 공기영향 × 통제주체 × 금융비용을 붙여보는 겁니다.

이 결과가 BTO냐 BTL이냐보다 먼저 나와야 합니다.

법률을 고쳐야 할 내용은 아닙니다. 기본계획과 적격성조사 세부요령에서 상당부분 할 수 있습니다.

자기자본비율도 ‘몇 % 넣었느냐’만 보면 조금 부족합니다

최소 자기자본비율 완화 역시 민간에서 꾸준히 요구해온 부분입니다.

정부가 조심스러운 이유는 분명합니다. Equity는 사업에 문제가 생겼을 때 가장 먼저 손실을 보는 돈이기 때문에 민간의 자기 돈이 충분히 들어가 있어야 사업을 끝까지 관리할 유인이 생깁니다.

그런데 미국 P3를 보다 보면 자기자본비율 하나만으로 그 유인을 평가하지 않는다는 점이 눈에 들어옵니다. Revenue-risk Toll P3와 Availability Payment P3가 같은 Equity 비율을 가질 이유가 없습니다. 수요위험의 크기가 다르고 Debt Capacity도 다르기 때문입니다.

그래서 장기적으로는 단순한 최소비율 대신 Equity + 후순위성 + Lock-up + Completion Support + Sponsor Guarantee + 해지 시 실제 손실을 같이 보는 Capital at Risk 개념을 더 활용해볼 수 있습니다.

자기자본비율이 낮아도 Sponsor가 공사완료 위험을 강하게 부담하고 배당이 제한돼 있으면 실제 위험노출은 클 수 있습니다.

반대로 표면상 Equity 비율이 높더라도 정부지원과 해지보상으로 상당부분 보호된다면 실제 위험자본은 예상보다 작을 수 있습니다.

그래서 “15%냐 10%냐”를 놓고 싸우기보다 어느 정도의 민간자본을 실제로 위험에 묶어야
원하는 성과유인이 생기는지를 보는 게 금융적으로 더 정확합니다.

자금재조달도 ‘민간이 돈을 더 벌었다’에서 끝낼 문제는 아닙니다

자금재조달은 민간이 계속 제도개선을 요구한 영역입니다.

사업이 안정되고 위험이 낮아지면 기존 고금리 Debt를 낮은 금리로 바꾸는 것은 자연스럽습니다.

정부 입장에서는 정부지원과 공공의 신용 때문에 위험이 줄었다면 그 이익을 국민과 나누는 것이 맞다고 볼 수 있습니다.

민간 입장에서는 운영을 잘해 신용도를 높인 결과까지 정부가 과도하게 가져가면 Refinancing을 적극적으로 할 이유가 줄어듭니다.

국내 실증연구도 이 문제를 비교적 잘 보여줍니다.

조혁종·남윤길·김형진은 「국토연구」에서 BTO 사업의 자금재조달과 사업재구조화 사례를 분석해 정부와 이용자에게 생긴 공익적 효과뿐 아니라 사업자가 부담한 위험과 투자수익률도 같이 살펴봤습니다. (KCI)

여기서 제가 가져오고 싶은 방향은 공유비율보다 Source of Gain, 즉 이익이 어디에서 생겼는지를 더 세밀하게 보는 것입니다.

예를 들어, 정부의 보증이나 새로운 정책금융 때문에 생긴 금리절감, 시장금리 하락으로 생긴 Windfall, 민간의 운영성과와 신용도 상승으로 생긴 이익, 당초 Financial Model에서 이미 예상한 Refinancing을 같은 방식으로 보지 않는 겁니다.

정부지원으로 생긴 이익은 정부 몫을 더 크게 인정하고, 민간이 실제 운영성과를 개선해 만들어낸 부분은 Incentive를 더 남길 수 있습니다.

이건 민투법 개정보다 기본계획의 자금재조달 기준과 표준실시협약의 문구를 조정하는 것이 먼저라고 봅니다.

그런데 금융에서는 여기서 한 단계 더 가야 합니다. TIFIA가 핵심입니다

한국에도 민자금융 수단은 많습니다.

건설보조금을 지급하고 있고, 장기대부 근거가 있으며, 산업기반신용보증기금과 사회기반시설채권도 있습니다.

그런데 미국의 Capital Stack을 보면 우리보다 두드러지는 층이 하나 있습니다.

정부가 Financing Gap에 맞춰 만기·상환시점·위험순위를 설계하는 공공 Credit입니다.

TIFIA가 그 역할을 합니다.

현재 TIFIA는 적격 사업비의 최대 "49%"까지 지원할 수 있습니다. 금리는 미국 국채금리 수준의 고정금리이고, 통상 최대 35년 상환, 일부는 75년까지 가능하며 실질적 완공 후 최대 5년까지 상환을 늦출 수도 있습니다. (교통부)

그런데 필자는 낮은 금리보다 다른 표현이 더 중요하다고 봅니다.

USDOT는 TIFIA를 Supplemental and Subordinate Investment, 즉 민간자금을 대체하기보다 빈틈을 메우는 보완적·후순위성 투자로 설명하는데(교통부), 이게 핵심이라고 생각합니다.

사업 초기에는 현금흐름이 약해서 일반은행이 30년 이상 Debt를 넣기 어려울 수 있습니다. 그렇다고 정부가 건설보조금을 늘리면 처음부터 재정을 투입해야 합니다. TIFIA는 그 사이에서 장기 Credit을 제공합니다. 공공이 일부 위험을 받아줌으로써 Senior Debt가 더 들어올 수 있게 만드는 겁니다.

학계에서도 TIFIA를 단순 저금리대출보다 민간·지방자본을 동원해 공공편익이 큰 고위험 교통사업을 추진하는 연방 직접신용수단으로 분석합니다. Lee와 Gifford의 Transportation Research Record 연구도 TIFIA의 시장자본 유인기능에 초점을 맞춥니다. (Sage Journals)

물론 장점만 있는 건 아닙니다.

TIFIA는 Creditworthiness(신용도), 전용 상환재원, 투자등급 등 엄격한 요건을 봅니다. LOI 이후에도 독립 Financial Advisor 분석과 신용심사를 거치고, 신청 초청 뒤에도 계약조건을 다시 협상합니다. (교통부)

그러니 한국에 가져올 때 단순하게 “정부 저리대출 하나 만들면 된다”고 생각하면 안 됩니다.

한국형 TIFIA라면 국토교통부가 중심이 되는 게 맞습니다

미국 TIFIA는 U.S. Department of Transportation, 즉 미국 교통부의 프로그램입니다.

대상도 도로·교량·터널, 대중교통, 철도, 항만 접근시설, 공항 등 교통인프라입니다. (교통부)

따라서 한국에서 먼저 교통형 TIFIA를 만든다면 국토교통부가 주관하는 것이 자연스럽습니다.

가칭 K-TIFIA Transportation Credit Program을 생각해볼 수 있습니다.

국토교통부는 교통정책 적합성, 사업구조, 수요·건설·운영위험과 Financing Plan을 보고, 재정당국은 신용지원 총한도, 예상손실과 국가재정 위험을 관리합니다. 여기에 더해 정책금융기관, 신용보증기금은 Credit Underwriting과 자금관리를 지원할 수 있습니다.

여기서 PIMAC의 역할은 구분하는 편이 좋습니다.

PIMAC은 한국 민자제도에서 사업을 함께 만들어가는 Financial Advisor라기보다 적격성·VfM·제도기준을 검증하는 성격이 강합니다. 따라서 K-TIFIA의 대출심사·구조화를 PIMAC가 직접 맡기보다는 기존 독립검증 기능을 유지하는 쪽이 낫습니다.

처음부터 법을 만들 필요도 없습니다.

3~5개 도로·철도 사업의 Financial Model에 가상의 K-TIFIA를 넣어보면 됩니다.

기존 Bank Debt 일부를 장기 고정금리, 상환유예, 장기만기, 제한적 후순위 조건의 공공 Credit으로 바꿨을 때

WACC는 얼마나 떨어지는지, 정부보조금은 얼마나 줄어드는지, DSCR은 어떻게 바뀌는지, Senior Debt가 얼마나 더 들어오는지,

정부가 감수하는 Expected Credit Loss는 얼마인지 먼저 계산해보는 겁니다.

효과가 반복적으로 확인되면 그때 상설화합니다.

상설 K-TIFIA까지 가려면 민간투자법상의 장기대부 근거를 넘어 융자한도, 상환유예, 후순위성, 손실충당, 회수금 재융자 등을 규율할 별도 계정이나 기금 근거를 검토해야 할 가능성이 큽니다.

PABs는 TIFIA와 또 다른 방법으로 금융비용을 낮춥니다

PABs(Private Activity Bonds·민간활동채권)는 정부대출이 아닙니다.

민간 P3가 Tax-exempt Bond Market(면세채권시장)에 접근할 수 있도록 허용하는 제도입니다.

미국에서는 적격 PAB의 이자가 연방소득세에서 면제됩니다. 투자자는 과세채권보다 낮은 금리도 받아들일 수 있고, 결과적으로 프로젝트의 조달비용이 낮아집니다. (교통부) 현재 교통 PAB의 한도는 $30B, 약 42조원입니다.

2025년 11월 기준 발행액은 $23.9B, 추가 배정액은 $6.1B로 전체 한도가 모두 배정됐습니다. Transform 66에는 $737M, Purple Line에는 초기 $313.035M이 발행됐습니다. (교통부)

TIFIA와 PAB가 같이 쓰이는 이유도 여기 있습니다.

TIFIA는 Cash Flow와 위험의 빈틈을 메우고, PAB는 Senior Debt의 가격을 낮춥니다.

학계에서도 P3와 Tax-exempt Bond Financing을 별도의 중요한 금융구조로 분석해왔습니다. Tsilas와 Brashares는 Municipal Finance Journal에서 P3와 면세채권 결합이 어떻게 작동하는지, 그리고 그 과정에서 생기는 정책·행정적 과제를 다뤘습니다. (IDEAS/RePEc)

미국 P3가 ‘민간 돈으로 정부가 손을 떼는 방식’이라고 이해하면 이 부분이 잘 설명되지 않습니다.

미국 역시 연방 Credit과 조세정책을 이용해 민간자본의 가격과 위험을 적극적으로 조정합니다.

한국형 PAB는 오히려 TIFIA보다 제도개편이 더 클 수 있습니다

한국에도 사회기반시설채권이 있습니다. 산업기반신용보증과 결합할 수도 있습니다.

2026년 정부는 국민참여 공모 인프라펀드에 선순위채와 산업기반신용보증기금 보증을 결합하고, 개인투자자에게 1억원 한도 15.4% 분리과세를 적용하기로 했습니다. (대한민국 정책브리핑)

방향만 놓고 보면 PAB와 닿는 부분이 있습니다.

보증 + 세제 + 자본시장을 이용해 민자금융 비용을 낮추는 접근입니다. 하지만 미국식 PAB까지 가려면 민간투자법만 고쳐서는 부족합니다. 핵심이 세제이기 때문입니다. 적격 인프라채권에 비과세 또는 추가적인 저율과세를 적용한다면 조세특례제한법 등 세법 검토가 필요합니다.

그리고 왜 민간사업자의 채권에 세제혜택을 줘야 하느냐는 질문에도 답해야 합니다.

세금감면은 현금예산으로 지급하지 않을 뿐 결국 Tax Expenditure, 즉 조세지출입니다.

그래서 저는 처음부터 전면도입보다 한도를 정한 K-Infrastructure Bond Pilot이 낫다고 봅니다.

예를 들어 세제혜택으로 Debt Coupon이 40bp 낮아졌다면 그 절감액 중 상당 부분이 사용료·AP·건설보조금 감소로 돌아오도록 Financial Model에 반영합니다.

이래야 PAB형 제도가 민간 특혜가 아닌 정부가 세제를 이용해 전체 인프라 조달비용을 낮추는 정책이 됩니다.

부대사업 확대 요구도 왜 쉽게 받아들이지 못했는지 봐야 합니다

민간은 부대사업 확대도 오래 요구했습니다.

철도라면 역세권, 상업시설, 주차, 광고, 에너지·통신 같은 수익을 본사업과 연결할 수 있습니다.

2026년 민간투자 관련 세미나에서도 부대사업을 Positive 방식에서 Negative 방식으로 전환하고, 민간투자 전문기업과 산업 생태계를 육성해야 한다는 의견이 다시 나왔습니다. (철도산업정보센터)

그런데 정부가 그동안 부대사업을 무한정 열지 않은 데도 이유가 있습니다.

공공부지의 개발이익은 누구 것인지, 그 개발권을 왜 해당 민자사업자에게 주는지, 토지가격은 어떻게 산정하는지, 부대사업의 수익 중 얼마를 본사업에 돌려야 하는지 같은 문제가 같이 따라옵니다.

21년 대장동 개발사업 이후 실제가 어떠한지와는 별개로 민간사업자+부동산 개발의 조합은 문제가 많다는 이미지가 된것도 정부의 담당자들이 쉽게 적극적으로 덤비지 못하는 이유이기도 합니다.

그래서 규제를 풀기 전에 개발이익 환원원칙부터 명확하게 만드는 게 중요합니다.

철도역세권·Mobility Hub·에너지·디지털 인프라처럼 본시설과 연계성이 비교적 분명한 분야부터 Pilot으로 열고, 실제 사업에서 현재 법의 열거방식이 반복적으로 장애물이 된다면 그때 민간투자법 제21조 개정을 검토하는 편이 더 안전합니다.

미국식 유연성을 가져오려면 정부 담당자의 전문성부터 같이 높여야 합니다

여기까지 제도를 유연하게 만들면 다른 문제가 생깁니다.

누가 이 계약을 협상할 것인가입니다.

민간 컨소시엄에는 건설사뿐 아니라 금융기관, 회계법인, 로펌, 기술자문사와 보험 전문가가 붙습니다.

정부 담당자가 부임한 지 몇 달 만에 수백 페이지의 실시협약과 PF Financial Model을 보고 이들과 협상해야 한다면 정보의 격차가 생길 수밖에 없습니다.

해외 연구에서도 P3의 성공조건을 단순한 계약형식이나 조달방식으로만 보지 않습니다. Martin, Lawther, Hodge, Greve는 국제 P3 경험을 검토하면서 정책·절차·조직역량을 포함한 여러 Best Practice를 함께 다뤘습니다. (CCSE)

한국은 다행히 완전히 새 조직부터 만들 필요는 없습니다.

도로에는 민자도로 관리지원센터, 철도에는 민자철도 관리지원센터가 있습니다. 철도사업법 제25조의5도 국토교통부장관이 민자철도에 대한 전문성을 고려해 관리지원센터를 지정할 수 있도록 하고 있습니다. (법률정보센터)

이 조직들을 Transaction Support까지 할 수 있도록 조직과 역량을 더 키우는 게 좋겠습니다.

예로, Financial Model 검토, 협상지원, 계약변경, Refinancing, Claim, 보험, 법률·회계 같은 개별 사업의 실무를 지원합니다.

대신 범정부 교육은 PIMAC이 맡는 게 맞습니다.

PIMAC은 이미 2026년 주무부처·지자체·공공기관 담당자를 대상으로 기초·기본과정뿐 아니라 재무심화와 재무모델 실습까지 운영하고 있습니다. 도로 BTO와 BTL 재무모델 실습자료도 별도로 제공하고 있습니다. (PIMAC)

여기에 조금 더 실무성을 넣었으면 합니다.

초급은 민간투자법·기본계획·절차, 중급은 Risk Allocation·Financial Model·해지시지급금, 고급은 DSCR·IRR·Debt Sculpting·Refinancing·TIFIA·PABs·협상 Simulation 정도로 나눌 수 있습니다.

공무원의 보직도 같이 봐야 합니다.

민자사업의 핵심 직위는 가능하면 2년 이상 장기근무하는 전문직위 또는 전문직위군으로 운영하고, 전문교육·수당·경력평정·관련 보직 연계를 붙여야 합니다. 30~40년짜리 계약을 관리하면서 담당자가 계속 바뀌면 정부 쪽 Institutional Memory가 쌓이기 어렵습니다.

변호사와 회계사에게 도움을 받는 데도 예산이 필요합니다

전문성 이야기를 하면서 교육만 강조하면 현실성이 떨어집니다.

담당 공무원이 PF 교육을 몇 번 들었다고 프로젝트금융 전문 변호사나 회계사를 대신할 수는 없습니다.

필요할 때 그 사람들을 쓸 수 있어야 합니다.

분야별 관리지원센터에 법률, 회계·세무, Project Finance, 보험, 건설 Claim, 수요, 기술 전문가 Pool을 만들어두고 국토부나 주무관청이 필요할 때 바로 자문을 요청할 수 있도록 하는 겁니다.

이를 위해 민자철도, 도로 관리지원센터에 전문가를 영입할 수 있도록 인건비 반영이 되어야 합니다. 재정당국 입장에서도 이렇게 들어가는 인건비가 사업 한두개에서 터지는 수십억~수천억원의 비용을 줄일 수 있다면 그것이 더 재정절감에 도움이 될 것입니다.

중요한 것은 돈만 편성하는 게 아닙니다.

자문사가 필요할 때 다시 몇 달짜리 일반용역 절차부터 밟으면 협상에는 쓸 수 없습니다. 전문가 Pool을 사전에 계약해두고 필요할 때 Task Order 방식으로 개별 업무를 맡기는 식의 신속한 전문자문 조달방식도 같이 설계할 필요가 있습니다.

그렇다면 실제 제도개선은 어떤 순서로 하면 될까요?

한꺼번에 법을 바꾸는 것은 추천하고 싶지 않습니다.

먼저 해보고, 그다음 바꾸는 편이 낫습니다.

단계	제안	필요한 조치
바로 가능	Market Sounding·Industry Review	기본계획·PIMAC 지침
바로 가능	Risk Register 선행	적격성조사 세부요령
바로 가능	PIMAC 전문교육 강화	교육체계
바로 가능	2년 이상 전문보직·외부전문가 활용	인사·국토부 예산
Pilot	Risk-based Equity	기본계획·표준협약
Pilot	Source of Gain 기반 자금재조달	기본계획·표준협약
Pilot	국토부 K-TIFIA	시범 금융구조·예산
Pilot	PAB형 인프라채권	금융·세제효과 검증
법제화 검토	상설 K-TIFIA Credit Account	민간투자법·재정제도
법·세제 검토	PAB형 세제지원	조세특례 등
법 개정 검토	부대사업 근본 확대	민간투자법 제21조 등

민간의 요구를 모두 들어주는 것이 민자 활성화는 아닙니다

여기까지 쓰다 보면 민간이 요구한 규제를 하나씩 풀어주자는 글처럼 보일 수도 있습니다.

하지만 그건 제가 이야기하고 싶은 방향과 조금 다릅니다.

정부의 자기자본 요구, 자금재조달 이익 공유 주장도 이유가 있습니다.

부대사업을 제한한 것은 공공부지의 개발이익과 특혜 문제 때문입니다.

사전협상을 조심스러워한 것은 조달의 공정성 때문입니다.

그리고 TIFIA나 PAB도 공짜 돈이 아닙니다.

TIFIA는 정부가 Credit Risk를 부담하고, PAB는 조세지출이 발생합니다.

그러니 “미국은 해주는데 우리는 왜 안 해주느냐”로 가면 답이 잘 안 나옵니다.

오히려 미국에서 가져올 만한 것은 같은 공익목표를 좀 더 세밀한 가격과 계약으로 관리하는 방식입니다.

자기자본 15%가 중요한 게 아니라 실제 얼마가 위험에 노출되는지 봅니다.

재조달 이익이 얼마인지보다 왜 생겼는지 봅니다.

모든 위험을 민간에 넘겼는지보다 누가 그 위험을 더 싸게 통제할 수 있는지 봅니다.

정부보조금을 얼마 줄지 결정하기 전에 Financing Gap이 무엇인지 봅니다.

그리고 그 Gap이 장기 Credit이면 TIFIA를, Debt Pricing이면 PAB형 제도를 검토합니다.

이렇게 보면 금융지원과 재정지원도 조금 다르게 보입니다.

정부가 부족한 사업비를 채워주는 것이 재정지원에 가깝다면, 금융지원은 민간자본이 들어오지 못하게 만드는 위험과 만기·금리 구조를 바꾸는 일에 더 가깝습니다.

미국 P3를 30편 가까이 살펴본 끝에 제가 한국 민자사업에 가장 가져오고 싶은 것도 결국 이 사고순서입니다.

사업방식보다 위험을 먼저 보고, 정부지원액보다 Capital Stack을 먼저 보는 것.

한국은 이미 많은 도구를 가지고 있습니다. 민간투자법도 있고, PIMAC도 있습니다. 건설보조금과 장기대부가 있고, 산업기반신용보증과 사회기반시설채권도 있습니다. 최근에는 국민참여 인프라펀드까지 들어왔습니다.

먼저 이 도구들을 한 화면에 올려놓고 빈칸을 찾아보면 됩니다.

그 빈칸이 장기·유연한 공공 Credit이라면 국토교통부가 K-TIFIA를 시험해볼 수 있습니다. 그 빈칸이 장기 Debt의 가격이라면 PAB형 세제를 검토할 수 있습니다. 그리고 실제 사업을 몇 개 해본 뒤 법이 막는 부분이 보이면 그때 법을 고치는 겁니다.

미국 제도를 복사하는 것보다 이 방식이 한국 민자사업에는 훨씬 현실적인 답이라고 생각합니다.

핵심 참고자료

기획예산처, 2026년 민간투자 활성화 방안
https://www.korea.kr/news/policyNewsView.do?newsId=148959421&pWise=main&pWiseMain=L4

대한건설협회, 민간투자사업 활성화 제도개선 건의 관련 보도
https://www.newsis.com/view/NISX20220330_0001813365

USDOT Build America Bureau — TIFIA Program
https://www.transportation.gov/buildamerica/financing/tifia

USDOT Build America Bureau — TIFIA Eligibility
https://www.transportation.gov/buildamerica/financing/tifia/eligibility

USDOT Build America Bureau — Private Activity Bonds
https://www.transportation.gov/buildamerica/financing/private-activity-bonds

USDOT — PAB Allocations
https://www.transportation.gov/buildamerica/financing/private-activity-bonds-pabs/private-activity-bonds-allocations

KDI PIMAC — 2026년 민간투자사업 교육
https://pimac.kdi.re.kr/cooperation/edu_view.jsp?pageNo=1&seq_no=22443&showListSize=30

이동훈, 「실시협약을 통한 민간투자사업에서의 합리적인 위험배분」, 고려법학 제80호, 2016
https://www.kci.go.kr/kciportal/landing/article.kci?arti_id=ART002091892

조혁종·남윤길·김형진, 「수익형 민간투자사업(BTO)의 자금재조달 및 사업재구조화 특성과 효과분석」, 국토연구 제89권, 2016
https://www.kci.go.kr/kciportal/landing/article.kci?arti_id=ART002120544

Narae Lee & Jonathan L. Gifford, “MAP-21 to FAST Act: Did the TIFIA Program Better Support High-Risk Transportation Infrastructure Projects?”, Transportation Research Record, 2021
https://journals.sagepub.com/doi/10.1177/03611981211011644

Vicky Tsilas & Edith Brashares, “P3 Projects and Tax-Exempt Bond Financing: How Does the Puzzle Work?”, Municipal Finance Journal, 2016
https://ideas.repec.org/a/ucp/munifj/doi10.1086-mfj37010051.html

Lawrence Martin·Wendell Lawther·Graeme Hodge·Carsten Greve, “Internationally Recommended Best Practices in Transportation Financing Public-Private Partnerships (P3s)”, 2013
https://www.ccsenet.org/journal/index.php/par/article/view/30857

철도산업정보센터, 민간투자 교통SOC의 성과와 발전방향 세미나 보도, 2026
https://www.kric.go.kr/jsp/board/portal/sub01/railNewsDetail.jsp?p_id1=M01060101&p_id2=725681&p_id3=2026.08.27

#69 - 미국 P3 29편을 보고 나니 한국과 무엇이 가장 달랐을까?

foweuniverse — Tue, 1 Sep 2026 00:48:01 +0900

- 미국 P3 29편을 따라가고 나니 한국 민자사업과 미국 P3의 차이는 ‘민간을 얼마나 많이 활용하느냐’보다 불확실성을 다루는 방식에 있었습니다. I-4 Ultimate, Purple Line, Port of Miami Tunnel과 TIFIA·PABs 사례를 통해 한국의 제도·표준화와 미국의 위험배분·시장검증·자본구조 설계가 어떻게 다른지 금융과 계약의 관점에서 비교합니다. -

※ 미국 사례의 원화는 비교 편의를 위해 $1=1,400원으로 단순 환산했습니다. 실제 사업 당시 환율과는 다릅니다.

미국 P3에 대해 공부하면서 처음에는 제도 차이가 가장 먼저 눈에 들어왔습니다.

한국에는 「사회기반시설에 대한 민간투자법」이 있습니다. 민간투자사업기본계획이 있고, BTO·BTL 같은 사업방식도 익숙합니다. 민간제안이면 적격성조사를 하고, 제3자 공고를 거쳐 우선협상대상자를 선정하고, 협상 끝에 실시협약을 체결합니다.

미국은 달랐습니다.

Virginia를 보다 Florida로 넘어가면 P3 권한부터 다시 확인해야 했습니다. State DOT(주 교통부), Toll Authority(유료도로기관), Transit Agency(대중교통기관), 지방정부가 각각 다른 법적 권한과 조달절차를 가지고 있었습니다.

같은 DBFOM(Design-Build-Finance-Operate-Maintain·설계·건설·금융·운영·유지관리)이라는 이름을 붙여도 어떤 사업은 민간이 Toll Revenue(통행료수입)를 가져가고, 어떤 사업은 정부가 Availability Payment(AP·가용성지급금)를 지급합니다.

처음에는 그저 “미국은 주마다 달라서 복잡하구나” 정도로 생각했습니다. 그런데 절차와 계약을 보고, TIFIA와 PABs를 보고, Financial Close(금융종결)를 보고, 마지막에는 데이터센터와 UAM까지 따라가다 보니 조금 다른 차이가 보였습니다.

한국과 미국은 민간투자사업의 불확실성을 줄이는 방식 자체가 다릅니다.

한국은 법, 기본계획, 전문기관 심사와 표준화된 계약을 통해 불확실성을 줄이는 쪽에 가깝습니다.

미국은 사업마다 위험을 더 잘게 나눈 뒤 계약과 금융을 다시 조합하고, 그 구조가 실제 시장에서 가격을 받을 수 있는지 확인하면서 불확실성을 처리합니다.

표현을 조금 단순화하면 이렇습니다.

한국은 제도가 신뢰를 먼저 만들어주고, 미국은 계약과 금융구조가 신뢰를 만들어가는 경향이 강합니다.

한국은 ‘민자사업이라는 제도’에서 시작하고, 미국은 ‘이 위험을 누가 맡을까’에서 시작합니다

한국 민자사업의 장점은 출발점이 명확하다는 데 있습니다.

2026년 현재 KDI 공공투자관리센터가 정리한 민간투자법·시행령·기본계획 자료만 해도 348쪽입니다. 「사회기반시설에 대한 민간투자법」, 시행령, 2026년 민간투자사업기본계획이 하나의 제도체계로 묶여 있습니다. (PIMAC)

민간제안사업의 경우 시행령 제7조는 제안서에 타당성조사, 사업계획, 총사업비와 자금조달계획, 관리운영계획, 사용료와 수입·지출계획 등을 포함하도록 합니다. 총사업비 2,000억원 이상 제안사업에는 경제성, 정책적 필요성, 재정사업 대비 민간투자방식의 적정성을 포함한 적격성조사를 하도록 하고 있습니다. (법제처)

중앙 심의도 강합니다.

현재 시행령 제6조는 일반적으로 총사업비 2,000억원 이상, 법 제4조제2호 방식은 1,000억원 이상인 대상사업을 민간투자사업심의위원회 심의대상으로 둡니다. 시설사업기본계획도 총사업비 2,000억원 이상이면 심의대상이 됩니다. (법제처)

실무자 입장에서는 이게 꽤나 큰 장점입니다. 왜냐하면 담당자가 바뀌어도 제도는 유지 되기에 업무가 이어집니다.

어느 지역 사업이든 어느 분야 사업이든 같은 법률과 기본계획이라는 공통언어가 있습니다.

민간사업자와 금융기관도 협상이 어디서 시작되고 어떤 관문을 거칠지 어느 정도 예상이 가능합니다.

미국은 처음부터 질문이 조금 다릅니다.

USDOT가 설명하는 P3도 하나의 고정된 법적 사업유형이라기보다 민간이 설계·건설·금융·장기운영과 수입위험까지 추가로 부담할 수 있는 계약의 스펙트럼에 가깝습니다.

그래서 처음부터 이런 질문이 나옵니다.

누가 부지를 확보할지, 누가 설계오류를 책임질지, 지장물 이설이 늦어지면 누가 책임질지, 교통 수요 예측이 틀리면 누구 책임인지, 물가 급등 시 O&M 비용은 어떻게 조정할지 , 30년 뒤 Handback(시설인계) 시점에 시설 상태는 어떨지 등 입니다.

이 위험을 하나씩 나눈 뒤 DBFOM인지 Toll Concession(통행료 양허)인지, Availability Payment 방식인지가 정해집니다.

FHWA도 이 부분을 아주 명확하게 설명합니다.

민간이 실제로 관리할 수 있는 위험을 넘겨야 효율성이 생기고, 통제할 수 없는 위험까지 넘기면 민간은 이를 Risk Premium(위험프리미엄)에 넣어 가격을 올립니다. (Federal Highway Administration)

제가 미국편을 쓰면서 느낀 차이는 바로 여기였습니다.

한국에서는 사업방식이 어느 정도 먼저 보이고 그 안에서 위험을 조정하는 느낌이 강합니다.

미국에서는 위험을 어떻게 배분할지가 먼저이고 그 결과로 사업방식과 금융구조가 만들어지는 경우가 더 많았습니다.

완전히 다른 제도라는 뜻은 아닙니다.

다만 사고를 시작하는 순서가 조금 다릅니다.

I-4 Ultimate를 보면 ‘유료도로인데 왜 민간이 교통량 위험을 안 가져갔나’라는 질문부터 생깁니다

Central Florida를 통과하는 I-4 Ultimate를 다시 보겠습니다.

Orlando 도심을 중심으로 I-4 약 21마일, 약 34km를 재건하고 중앙부에 4개 Express Toll Lanes(유료 급행차로)를 추가한 사업입니다. FHWA 기준 총사업비는 $2.877B, 약 4조278억원입니다. 40년 DBFOM Availability Payment Concession으로 체결됐고 Financial Close는 2014년 9월 4일이었습니다. (Federal Highway Administration)

그런데 이 사업은 유료도로인데 민간사업자가 통행료수입을 가져가지 않습니다. FDOT가 Toll Rate(통행료율)를 정하고 수입도 가져갑니다. 민간사업자는 시설이 계약상 요구되는 성능을 충족하면 Availability Payment를 받습니다. 통행료수입과 AP는 직접 연동되지 않습니다. (Federal Highway Administration)

왜 그렇게 만들었을까요?

정부가 Traffic Demand Risk(교통수요위험)를 남겨두고, 민간에는 설계·건설·금융·운영·유지관리와 성능위험을 넘긴 겁니다.

금융구조를 보면 이 선택이 더 잘 보입니다.

Senior Bank Debt(선순위 은행대출): $484M, 약 6,776억원
TIFIA Tranche A: $127.3M, 약 1,782억원

TIFIA Tranche B: $822.2M, 약 1조1,511억원

Equity(자기자본): $103M, 약 1,442억원

FDOT Milestone Payments: $1.035B, 약 1조4,490억원

TIFIA 자본화이자 및 이자수익: $306M, 약 4,284억원

여기서 Milestone Payment를 특별한 미국식 제도로 볼 필요는 없습니다.

경제적 기능만 보면 한국 민자사업의 건설보조금과 상당히 유사합니다.

건설기간 중 공공이 자본비 일부를 부담해서 민간이 조달해야 할 Debt와 Equity를 줄입니다.

그래서 미국과 한국의 차이를 “한국에는 건설보조금이 있고 미국에는 Milestone Payment가 있다”라고 설명하면 별 의미가 없습니다.

더 중요한 질문은 따로 있습니다.

공공자금을 얼마만큼 먼저 지급하고, 얼마만큼의 민간자본을 계속 위험에 노출시켜 둘 것인가.

이게 금융구조의 핵심입니다.

공공자금을 많이 넣으면 항상 좋은 걸까요? 여기서 P3 금융이 조금 복잡해집니다

이건 한국의 건설보조금을 볼 때도 같은 질문입니다.

공공이 공사비를 많이 부담하면 민간의 조달비용은 줄어듭니다.

예를 들어 총사업비 100을 민간이 전부 Debt와 Equity로 조달하도록 하면 정부는 초기재정 부담을 늦출 수 있습니다.

그런데 나중에는 높은 민간자본비용까지 포함해 상환해야 합니다.

반대로 정부가 90을 건설기간에 지급하고 민간이 10만 조달하면 금융비용은 확 내려갑니다.

문제는 민간이 실패했을 때 잃을 자본도 10밖에 남지 않는다는 것입니다.

P3에서 중요한 Capital at Risk(위험에 실제 노출된 민간자본)가 줄어드는 겁니다.

그러면 “민간이 자기 돈을 걸었으니 공기와 비용을 더 철저히 관리할 것”이라는 P3의 중요한 유인이 약해질 수 있습니다.

CBO도 Private Financing 자체가 돈을 싸게 만드는 것은 아니라고 봅니다.

민간금융을 사용하면 정부의 초기현금지출을 줄일 수 있지만, 결국 민간은 미래 세금이나 이용료로 상환받고 위험을 부담한 만큼 수익률을 요구합니다. CBO는 이자보조, 위험비용과 Transaction Cost까지 고려하면 민간금융 자체의 전체비용이 공공금융보다 본질적으로 낮다고 보기 어렵다고 평가합니다. (의회예산국)

그럼 굳이 왜 민간돈을 넣을까요?

돈이 필요해서만은 아닙니다. 그 돈을 위험에 노출시키기 위해서입니다. 민간 Equity와 Debt가 실제 성과에 따라 손실을 볼 수 있어야 Design-Build-Operate-Maintain을 한 사업자에게 묶는 의미가 생깁니다.

따라서 이상적인 구조는 공공재정을 최소화하는 구조도 아니고, 민간금융을 최대화하는 구조도 아닙니다.

공공자금을 너무 많이 넣으면 성과유인이 줄고, 너무 적게 넣으면 필요 이상으로 비싼 민간자본을 사용합니다.

중간에 최적점이 있습니다.

이에 따라, 정부 담당자의 질문도 그래서 이렇게 바뀌어야 합니다.

“건설보조금을 몇 % 줄까?”보다

“민간의 성과유인은 충분히 남기면서 전체 WACC를 가장 낮출 수 있는 공공자금 투입 수준은 얼마인가?”

이 질문이 금융적으로는 더 정확합니다.

I-4의 TIFIA는 $949M이라는 금액보다 ‘둘로 나눈 방식’이 더 흥미롭습니다

TIFIA를 단순히 미국판 저리 정책대출이라고 보면 중요한 부분을 놓치게 됩니다.

I-4 Ultimate의 TIFIA는 총 $949M, 약 1조3,286억원입니다.

그런데 하나의 대출이 아닙니다.

Tranche A $127.3M은 FDOT가 지급하는 두 번째 Final Acceptance Payment를 통해 전액 상환하도록 했습니다.

반면 Tranche B $822.2M은 운영기간 Availability Payment에서 상환하고 최종 만기는 2052년입니다. (Federal Highway Administration)

같은 TIFIA인데 Cash Flow가 다릅니다. A는 건설 완료 전후를 연결하는 비교적 짧은 자금에 가깝고, B는 장기 Infrastructure Debt 역할을 합니다.

현재 TIFIA는 일반적으로 적격사업비의 최대 "49%"까지 지원할 수 있고 금리는 U.S. Treasury Rate(미 국채금리)에 연동됩니다. 개별 만기와 상환조건은 사업의 경제성, 비용과 수입 패턴 등을 바탕으로 협상합니다. (교통부)

USDOT가 TIFIA를 설명하면서 사용하는 표현도 중요합니다. Supplemental and Subordinate Capital, 즉 민간금융을 대신하는 돈이라기보다 시장에서 부족한 부분을 채우는 보완적·후순위성 자본입니다. (교통부)

이게 미국 금융에서 가장 눈여겨본 부분입니다.

공공금융을 “싸게 빌려주는 돈”으로만 보지 않습니다.

상환시점, 상환원천, 순위와 만기를 바꿔 다른 돈이 들어올 수 있는 구조를 만듭니다.

Transform 66을 보면 이 Capital Stack이 더 극단적으로 보입니다

Virginia의 Transform 66 – Outside the Beltway는 약 22.5마일, 약 36km의 I-66을 확장하고 Managed Toll Lanes(관리형 유료차로)를 건설한 50년 DBFOM P3입니다. 총사업비는 $3.724B, 약 5조2,136억원입니다.

자금구조는 다음과 같습니다.

Equity: $1.523B, 약 2조1,322억원
TIFIA: $1.229B, 약 1조7,206억원
TIFIA Capitalized Interest: $301M, 약 4,214억원
PABs: $737M, 약 1조318억원
State Infrastructure Bank Loan: $33M, 약 462억원

TIFIA Credit Agreement는 2017년 11월 7일 체결됐습니다. (교통부)

여기서는 민간 Equity만 해도 $1.5B가 넘습니다. I-4 Ultimate와 성격이 완전히 다릅니다.

왜냐하면 Transform 66에서는 TIFIA가 Net Toll Revenues(순통행료수입)를 상환재원으로 잡고 있기 때문입니다. (교통부)

I-4에서는 FDOT가 교통량 위험을 가지고 AP를 지급했지만, Transform 66에서는 민간 Capital이 Toll Revenue에 훨씬 직접적으로 노출됩니다. 같은 도로 P3인데 금융구조가 다른 이유입니다.

Financial Structure는 조달기법이 아니라 Risk Allocation의 결과물입니다.

PABs까지 들어오면 미국 정부는 ‘보조금 지급자’보다 금융시장 설계자에 가까워집니다

PABs(Private Activity Bonds·민간활동채권)는 또 다른 방식입니다. 민간 개발사업자가 공공성이 인정되는 교통시설을 추진할 때 Tax-Exempt Interest Rate(면세채권 금리)에 접근하도록 해 자본비용을 낮춥니다.

2026년 현재 법정 총한도는 $30B, 약 42조원입니다. (교통부) 정부가 사업비를 직접 지급하는 게 아니라, 세제와 채권시장의 구조를 이용해 민간 Debt의 가격을 낮춥니다.

이 점에서 미국 공공금융의 역할이 조금 다르게 보입니다. Grant는 사업비를 직접 줄이고, Milestone Payment도 민간이 금융해야 할 공사비를 줄입니다. TIFIA는 장기·유연한 공공 Credit을 넣고, PABs는 민간 Debt의 금리를 낮춥니다. 그 위에 Bank Debt와 Equity가 들어갑니다.

하나의 정책수단으로 사업을 지원하는 게 아니라 Capital Stack의 서로 다른 층을 건드립니다.

미국의 금융이 복잡해지는 이유이기도 합니다.

그런데 금융수단이 많다고 위험이 사라지는 것은 아닙니다. Purple Line이 그걸 잘 보여줬습니다

Maryland Purple Line을 보면 미국 금융을 너무 낙관적으로 볼 수 없습니다.

처음 2016년 TIFIA 대출은 $874.6M, 약 1조2,244억원이었습니다. 연방정부의 Capital Investment Grant도 $900M, 약 1조2,600억원이 예정돼 있었습니다. (교통부) 그런데 사업은 환경소송, 부지, Utility Relocation, 설계변경과 공사지연을 둘러싼 분쟁에 휘말렸습니다. 2020년 기존 Design-Build Contractor가 사업에서 철수했습니다.

결국 Maryland는 2022년 새로운 시공팀과 계약을 체결했습니다.

Washington Post 보도에 따르면 새 시공계약으로 총 건설비는 $3.4B, 약 4조7,600억원까지 늘었습니다. 최초 약 $1.97B에서 약 75% 증가한 수준이었습니다. 장기 P3 협약의 전체 가치도 약 $5.6B에서 $9.3B, 약 7조8,400억원에서 13조200억원으로 커졌습니다. (The Washington Post)

금융도 다시 만들었습니다.

2022년 USDOT는 기존 $874.6M TIFIA를 대체하는 최대 $1.76B, 약 2조4,640억원의 새 TIFIA Loan을 제공했습니다. 당시 TIFIA 적격사업비는 $5.9B, 약 8조2,600억원이었습니다. (교통부)

이 사례에서 중요한 건 “P3가 실패했다”는 단순평가가 아닙니다. 민간 시공 측은 부지와 환경승인, 정부 요구사항 변경 등 공공이 통제해야 할 영역 때문에 비용이 늘었다고 주장했고, 정부는 상당 부분이 계약상 민간 책임이라고 봤습니다.

결국 사업은 재협상과 금융재구조화를 거쳤습니다.

이게 Risk Allocation의 현실입니다.

계약서에 “민간 부담”이라고 적는 것과 민간이 실제로 그 위험을 통제할 수 있는 것은 다릅니다.

Port of Miami Tunnel은 오히려 ‘모든 위험을 민간에 넘기지 않는 것’이 더 좋은 P3일 수 있다는 사례입니다

Florida의 Port of Miami Tunnel은 위험배분을 조금 다른 방식으로 했습니다.

Biscayne Bay 아래를 뚫는 터널이다 보니 예상하지 못한 지반조건이 큰 문제였습니다. 민간에 전부 넘길 수도 있었을 겁니다.

그런데 FDOT와 민간은 그렇게 하지 않았습니다.

FHWA가 정리한 구조는 다음과 같습니다.

예측하지 못한 지반조건으로 추가비용이 발생하면 첫 $10M, 약 140억원은 민간, 다음 $150M, 약 2,100억원은 FDOT, 다음 $20M, 약 280억원은 다시 민간, 그 이상이면 양측이 계약종료를 선택할 수 있는 구조였습니다. (Federal Highway Administration)

처음 보면 이상합니다.

민간 P3인데 왜 정부가 $150M까지 부담할까요?

이유는 간단합니다.

민간도 통제할 수 없는 지반위험을 무제한 넘기면 입찰자는 최악의 상황을 가정해 가격을 크게 올립니다. 그리고 실제로 문제가 발생하지 않아도 정부는 그 Risk Premium을 지불하게 됩니다.

따라서 Risk Transfer의 목표는 최대화가 아닙니다.

위험비용 최소화가 목표입니다.

정부 담당자에게는 꽤 중요한 차이입니다.

“민간에게 얼마나 많이 넘겼나?”보다 “이 위험을 누가 가장 싸게 관리할 수 있는가?”를 봐야 합니다.

금융기관은 단순히 돈을 빌려주는 사람이 아니라 정부의 위험배분을 다시 심사합니다

Project Finance에서 은행의 역할을 생각하면 미국식 구조가 조금 더 이해됩니다.

Financial Close 전에 Lender는 단순히 이자율만 정하지 않습니다.

EPC 계약을 보고, Schedule을 보고, Liquidated Damages(지체상금)가 충분한지 봅니다.

그리고 보험조건을 확인하고, 정부의 지급의무와 Termination Compensation(해지시지급금)을 봅니다.

DSCR(Debt Service Coverage Ratio·부채상환커버리지)을 돌려보고 Downside Scenario(하방 시나리오)를 만듭니다.

Construction Cost가 10% 오르면 어떻게 되는지, 개통이 1년 늦으면 어떻게 되는지, 교통량이 20% 적으면 Debt Service가 가능한지, 그걸 확인하고 돈을 넣습니다.

운영 중에도 마찬가지입니다.

Reserve Account, Distribution Lock-up, 추가차입 제한, Debt Covenant를 통해 계속 사업을 감시합니다.

이러한 맥락에서 Private Debt는 단순 조달금이 아니라 Governance Mechanism(사업감시장치)이기도 합니다.

정부가 Risk Matrix에서 “민간 부담”이라고 적었다고 금융시장이 그대로 인정해주는 것은 아닙니다. 은행이 그 위험을 감당할 수 없다고 판단하면 Debt Margin이 올라갑니다. Loan Tenor가 짧아지고, DSCR 요구가 높아지며, Equity가 더 많이 필요합니다.

심하면 Financing이 안 됩니다.

결국 Bankability(금융가능성)는 정부의 위험배분안에 대해 금융시장이 내리는
두 번째 심사라고 볼 수 있습니다.

그래서 미국의 Market Sounding은 설명회라기보다 금융구조 설계과정에 더 가깝습니다

한국도 민간사업자와 금융기관의 의견을 듣습니다.

하지만 미국 자료를 계속 보다 보면 Market Sounding의 위치가 조금 다르게 보입니다.

사업구조가 완전히 결정되기 전에 묻습니다.

이 Traffic Risk를 가져갈 수 있습니까.
이 계약기간으로 Debt가 가능합니까.
이 Termination Payment라면 Lender가 들어옵니까.
이 Inflation Indexation(물가연동)으로 O&M Risk를 받을 수 있습니까.
이 정도 공공 Contribution이면 Equity Return이 나옵니까.

그리고 시장 반응이 좋지 않으면 구조를 수정합니다.

Transform 66 같은 사업에서 시장과 금융의 반응이 계약구조 자체를 결정하는 이유입니다.

이 점은 다음 #70에서 한국이 가져올 만한 부분으로 다시 다루겠습니다.

그렇다고 미국 방식이 더 효율적이라고 바로 결론 내리기도 어렵습니다

유연한 구조에는 대가가 있습니다.

사업마다 법률자문이 필요하고, Financial Advisor(금융자문사)가 붙어야 하며, Technical Advisor(기술자문사), Traffic Consultant(교통수요 자문사), Insurance Advisor, Bond Counsel까지 들어옵니다.

공공 쪽도 많은 돈을 쓰고 민간 입찰자도 상당한 Bid Cost를 부담합니다.

이 비용이 바로 Transaction Cost(거래비용)입니다.

CBO도 P3의 비용을 평가할 때 민간금융 비용만이 아니라 위험비용과 거래비용까지 같이 봐야 한다고 설명합니다. (의회예산국)

GAO의 경고는 더 직접적입니다. P3에는 “free money”, 공짜 돈이 없습니다.

민간이 빌려온 돈도 결국 장래의 Toll이나 Tax로 갚아야 합니다. GAO는 P3가 위험공유와 운영효율 같은 장점을 가질 수 있지만 Toll, 민간수익, 장기 계약통제, 거래비용과 공공성 문제까지 함께 분석해야 한다고 지적했습니다. (미국 정부 회계 감사국)

그래서 미국의 유연성은 공짜가 아닙니다.

잘 설계하면 사업 특성에 매우 잘 맞는 계약을 만들 수 있지만, 그 구조를 만들기 위해 상당한 비용과 전문성을 사용합니다.

한국의 표준화는 그 반대입니다.

모든 사업에서 계약언어를 처음부터 만들지 않습니다. 법과 기본계획, PIMAC, 축적된 실시협약과 금융관행이 이미 있습니다.

따라서 더 정확한 비교는 이렇습니다.

한국은 표준화로 거래비용과 제도적 불확실성을 낮추는 대신 사업별 설계폭이 좁아질 수 있습니다.
미국은 사업별 최적화 폭이 큰 대신 그 최적점을 찾는 비용도 큽니다.

정부 담당자, 민간사업자, 금융기관은 같은 P3를 보면서 전혀 다른 것을 봅니다

정부 담당자는 이런 질문을 합니다.

이번에 재정을 얼마나 써야 할까요.
공사를 앞당길 수 있을까요.
민간에 어느 위험까지 넘기는 게 합리적일까요.
향후 30년 AP 지급의무는 재정에 감당 가능한가요.
민간이 부도나면 서비스를 어떻게 계속할까요.

민간 Sponsor는 다른 것을 봅니다.

Equity IRR은 얼마인가.
EPC Contractor에게 Construction Risk를 Back-to-Back으로 넘길 수 있는가.
정부의 지급의무는 신용도가 충분한가.
Change in Law(법령변경)는 누가 부담하는가.
Refinancing Gain은 얼마나 공유해야 하는가.
해지되면 Equity는 무엇을 회수하는가.

은행은 또 다릅니다.

“Debt Service가 끊기지 않는가?”

그리고 시민은 더 단순한 질문을 합니다.

“결국 내가 얼마를 내는가?”
“왜 또 공사가 늦어졌는가?”
“민간이 위험을 가져갔다면서 왜 정부가 추가로 돈을 지급하는가?”

이 질문이 틀린 것도 아닙니다.

실제 P3는 이 네 가지 관점을 모두 만족해야 합니다.

어느 하나만 최적화하면 문제가 생깁니다.

I-4 Ultimate도 정부 입장과 지역사회 입장이 완전히 같지는 않았습니다

FDOT 입장에서 I-4 Ultimate는 매우 큰 교통병목을 한 번에 재건하고 장기 성능관리까지 묶은 프로젝트였습니다.

그리고 2022년에는 Express Lanes까지 개통했습니다. 하지만 공사기간 지역사회가 본 모습은 달랐습니다. 공사는 수년 동안 이어졌고, 차선변경과 진출입로 변화, 야간공사, 사고와 주민불편이 계속됐습니다.

즉, P3가 Financial Close를 했다고 시민에게 사업이 성공한 것은 아닙니다. Financial Close는 금융의 시작일 뿐이고, 실제 이용자에게는 준공과 서비스 수준이 훨씬 중요합니다.

이건 한국 민자사업도 같습니다.

우리는 실시협약 체결과 금융약정을 매우 중요한 성과로 봅니다. 오죽하면 실시협약문을 만드는 협상과정을 민자사업의 꽃이라고 지칭할까요. 하지만 이용자에게는 개통날의 통행료와 소요시간이 훨씬 중요합니다.

P3의 최종 성과평가는 결국 그쪽으로 돌아와야 합니다.

Purple Line은 정부와 민간이 모두 자신에게 유리한 Risk Matrix를 읽을 수 있다는 것도 보여줬습니다

Purple Line 분쟁이 흥미로운 이유도 여기에 있습니다.

민간 입장에서는 정부가 제때 해결해야 할 승인·부지·외부 인터페이스 때문에 지연됐다고 볼 수 있습니다. 정부는 민간이 계약한 공기와 비용을 제대로 관리하지 못했다고 볼 수 있습니다.

둘 다 계약서를 근거로 말합니다. 그래서 Risk Allocation에서 정말 어려운 건 표를 만드는 일이 아닙니다.

위험과 위험 사이의 경계를 정하는 일입니다.

예를 들어 Utility Relocation을 정부위험으로 두었다고 해도 민간의 설계변경 때문에 Utility 작업이 추가됐다면 누구 책임일까요.

환경소송은 정부위험인데 민간 설계가 환경검토를 다시 하게 만들었다면 어떨까요.

이런 Interface Risk(경계위험)가 실제 분쟁을 만듭니다. 그래서 고도로 정교한 P3 계약일수록 Risk Matrix보다 그 뒤의 Relief Event, Compensation Event, Change Order, Dispute Resolution 조항이 중요해집니다.

이 부분은 미국식 계약유연성의 장점이면서 동시에 복잡성을 키우는 이유입니다.

한국 제도를 오래 봐온 입장에서는 오히려 미국편을 끝내고 한국의 장점도 더 많이 보였습니다

미국 자료를 처음 볼 때는 TIFIA와 PAB가 굉장히 눈에 들어왔습니다. 주마다 완전히 다른 구조를 만드는 것도 흥미로웠구요.

그런데 29편을 다 보고 나니 한국이 이미 갖고 있는 것들이 생각보다 큽니다.

같은 법체계가 있고, 민간투자사업기본계획이라는 공통 기준이 있습니다. 민간투자사업심의위원회가 중요한 절차를 밀도있게 들여다보고 PIMAC이라는 전문기관이 철저하게 검증합니다.적격성조사가 제도화되어 있고, 표준적인 협약구조와 수십 년간 쌓인 금융관행이 자리를 잡았습니다. 이건 단순한 행정절차가 아닙니다.

30년짜리 계약에서 상당히 중요한 Institutional Memory(제도적 기억)입니다.

정부 담당자는 바뀝니다. 민간 Sponsor도 지분을 팔 수 있습니다. Lender도 Refinancing으로 바뀝니다.

그런데 제도는 남습니다. 장기 P3에서는 이런 지속성이 매우 중요합니다.

미국에서는 이 기능을 강한 계약과 전문 자문시장으로 보완하는 경우가 많습니다.

한국에서는 중앙제도와 전문기관이 상당 부분 담당합니다.

서로 다른 방식입니다.

반대로 한국이 앞으로 어려워질 부분은 ‘표준 밖의 사업’일 가능성이 큽니다

도로와 철도에서는 기존 체계가 상당히 잘 작동합니다.

문제는 #66~#68에서 본 새로운 분야입니다.

데이터센터, 송전망, 디지털 인프라, UAM, 자율주행 인프라 이런 사업에서 처음부터 “BTO인가 BTL인가?”라고 묻는 순간 질문이 너무 빨리 좁아질 수 있습니다.

예를 들어 데이터센터는 정부가 수요를 보장할 필요가 없습니다. 이미 민간수요가 있습니다. 문제는 부지와 송전망, 전력공급, 인허가일 수 있습니다.

UAM 역시 30년 승객수요를 정부가 보장하기보다 여러 운영자가 쓸 수 있는 Vertiport, 전력, 환승시설이 더 중요할 수 있습니다.

이런 사업에서는 미국식 사고순서가 도움이 됩니다.

먼저 Risk Register를 만들어서 사업을 막는 위험이 무엇인지 봅니다. 그 다음 누가 그 위험을 실제로 통제할 수 있는지 봅니다.

이와 동시에 그 위험 때문에 Cash Flow가 어떻게 달라지는지 봅니다. 그다음에 자본구조를 만들고 계약방식을 선택합니다.

이 순서를 한국 제도 안에 조금 더 적극적으로 넣을 수 있지 않을까 하는 생각이 듭니다.

29편을 보고 나니 가장 큰 차이는 ‘누가 더 민간친화적인가’가 아니었습니다

처음 미국편을 시작할 때는 미국이 훨씬 민간친화적이고 유연한 제도를 가지고 있다고 생각했습니다.

지금도 어느 정도는 맞다고 생각합니다. 그런데 그 표현만으로는 부족합니다.

한국은 제도와 표준으로 불확실성을 사전에 줄입니다.

미국은 위험을 더 세분화하고 그것을 계약과 금융시장에 가격으로 반영하면서 불확실성을 처리합니다.

한국에서는 제도가 공통언어입니다. 미국에서는 Risk Allocation과 Financial Model이 공통언어가 되는 경우가 많습니다.

한국의 가장 큰 장점은 예측가능성입니다. 미국의 가장 큰 장점은 맞춤형 구조를 만들 수 있다는 겁니다.

한국의 약점은 표준 밖의 새로운 사업에서 대응이 느릴 수 있다는 것이고, 미국의 약점은 복잡성과 높은 거래비용, 발주기관별 역량격차입니다.

정리하면 이렇습니다.

그런데 이 표보다 저는 한 문장이 더 중요하다고 생각합니다.

한국은 제도를 통해 신뢰를 만들고, 미국은 계약과 금융을 통해 신뢰를 만드는 경향이 강합니다.

미국식 P3로 바꿀 필요는 없습니다. 그런데 미국에서 배울 것은 분명히 있습니다

미국 사례를 오래 보다 보면 제도 몇가지가 벤치마킹 하면 좋겠다고 생각이 듭니다.

TIFIA 같은 장기공공금융, PABs 같은 세제지원채권, 보다 적극적인 Market Sounding, 자유로운 Risk Allocation 등이 좋아보이고 흥미로운 제도입니다.

그런데 TIFIA만 떼어다가 한국에 만든다고 미국 TIFIA가 되는 건 아닙니다. 미국에는 국채시장, PAB 시장, 장기 Institutional Capital, Credit Rating, State DOT의 강한 사업권한과 금융자문시장이 함께 있습니다.

계약유연성도 마찬가지입니다. 미국처럼 주무관청에 선택권만 많이 주고 법률·금융·기술 전문역량을 충분히 지원하지 않으면 오히려 민간과의 협상력이 약해질 수 있습니다.

그래서 미국편을 다 쓰고 난 지금은 오히려 이렇게 생각합니다.

한국의 민자제도 뼈대를 미국식으로 바꿀 필요는 없습니다. PIMAC의 전문성, 중앙 심사체계, 적격성조사, 표준화는 상당히 큰 자산입니다. 대신 그 위에 몇 가지를 얹어볼 수 있습니다.

예를 들면, Risk Register에서 사고를 시작하는 것, Market Sounding을 조금 더 일찍 하는 것, 공공금융을 단순한 저리자금이 아니라 Capital Stack 속 위험완충수단으로 보는 것, 그리고 데이터센터·전력망·UAM처럼 기존 BTO·BTL의 언어만으로 설명하기 어려운 사업에는 계약을 조금 더 유연하게 설계하는 것입니다.

그런데 여기서 다음 질문이 생깁니다.

미국에서 실제로 가져올 만한 것은 무엇이고, 반대로 미국에서는 가능하지만 한국에서는 가져오면 안 되는 것은 무엇일까요?

#70은 그 문제를 좀 더 구체적으로 보겠습니다.

미국 P3 30편 가까이를 따라온 이유도 결국 거기에 있을 것 같습니다.

미국 제도를 소개하는 데서 끝내는 것이 아니라, 한국 민자사업을 다시 보기 위해서 말입니다.

핵심 참고자료

KDI 공공투자관리센터 — 2026년 사회기반시설에 대한 민간투자법·시행령·민간투자사업기본계획 원문
국가법령정보센터 — 사회기반시설에 대한 민간투자법 시행령 제6조 원문
국가법령정보센터 — 사회기반시설에 대한 민간투자법 시행령 제7조 원문
FHWA — I-4 Ultimate Project Profile 원문
USDOT Build America Bureau — TIFIA Program 원문
USDOT Build America Bureau — Financing / Private Activity Bonds 원문
USDOT Build America Bureau — Transform 66 Outside the Beltway 원문
USDOT — Purple Line TIFIA Restructuring, 2022 원문
Washington Post — Purple Line New Construction Contract, 2022 원문
FHWA — Port of Miami Tunnel Risk Allocation 원문
Congressional Budget Office — Public-Private Partnerships for Transportation and Water Infrastructure 원문
Congressional Budget Office — Using Public-Private Partnerships to Carry Out Highway Projects 원문
U.S. Government Accountability Office — Highway Public-Private Partnerships 원문

#68 - UAM·자율주행·하이퍼루프에도 P3를 적용할 수 있을까?

foweuniverse — Mon, 31 Aug 2026 13:17:01 +0900

- UAM·자율주행·하이퍼루프 같은 미래교통에도 P3를 적용할 수 있을까요? Florida SunTrax와 Eve, Orlando·Lake Nona의 Lilium, Utah Smart Roadways, Great Lakes Hyperloop 사례를 통해 전통 도로 P3와 다른 기술·수요·규제·금융위험을 살펴보고, 어떤 자산에 어떤 자본과 위험배분이 필요한지 정리합니다. -

※ 본문의 원화는 사례 비교를 위해 $1=1,400원으로 단순 환산했습니다. 실제 사업 당시 환율과는 다릅니다.

도로 P3를 오래 보다 보면 어느 정도 익숙한 공식이 생깁니다.

예상 교통량을 계산하고 계약을 한 후 도로를 건설합니다. 통행료를 받거나 정부가 Availability Payment(AP·가용성지급금)를 지급하여 수입이 발생합니다. 민간은 앞으로 20년, 30년 동안 들어올 Cash Flow(현금흐름)를 보고 Debt(부채)와 Equity(자기자본)를 넣습니다.

#66과 #67에서는 이 공식을 조금씩 넓혀봤습니다.

통행료가 없는 대학과 공공청사도 장기 지급계약이 있으면 금융이 가능했고, 데이터센터처럼 민간이 스스로 수익을 만드는 산업에서는 정부가 돈을 주기보다 부지·전력·인허가 같은 병목을 풀어주는 것이 더 중요했습니다.

그럼 마지막으로 한 걸음 더 가보겠습니다. UAM(Urban Air Mobility·도심항공교통), Autonomous Vehicle(자율주행차), Hyperloop(하이퍼루프)처럼 아직 기술과 시장 자체가 만들어지는 중인 교통수단입니다.

얼핏 보면 P3에 아주 잘 어울립니다. 정부에는 새로운 교통망을 만들 부지와 규제권한이 있고, 민간에는 기술과 자본이 있습니다.

둘이 역할을 나누면 훨씬 빨리 갈 수 있을 것 같습니다.

그런데 금융계약까지 생각해보면 이야기가 달라집니다.

도로는 30년 뒤에도 도로일 겁니다. 철도의 신호시스템과 차량은 바뀌어도 노선과 역사, 궤도는 남습니다.

미래교통은 30년짜리 P3 계약보다 기술의 수명이 먼저 끝날 가능성이 있습니다.

항공기 제조사가 사라질 수 있고, 충전규격이 바뀔 수도 있습니다. 자율주행차가 도로와 통신하는 방식 자체가 달라질 수 있고, Hyperloop는 아직 상업노선 하나도 운영하지 않았습니다.

저는 이 차이가 미래교통 P3를 이해하는 출발점이라고 봅니다.

P3를 적용할 수 있느냐보다 무엇을 P3로 묶어야 하느냐가 더 중요한 문제입니다.

도로 P3와 같은 계약을 쓰기 어려운 이유는 위험의 종류부터 다르기 때문입니다

전통적인 도로·철도 P3에서도 위험은 많습니다.

Construction Risk(건설위험), Demand Risk(수요위험), Operation & Maintenance Risk(운영·유지관리위험), 금리와 물가변동까지 봅니다.

그런데 도로와 철도는 시설 자체의 기술적 성립 여부를 다시 검증하는 경우는 많지 않습니다.

하지만 미래교통은 그 앞에 위험이 몇 개 더 붙습니다.

구분	전통 도로·철도 P3	미래교통 P3
기술	상당부분 검증	계속 변화·인증 진행
수요자료	장기간 실적 존재	실제 상업수요 부족
규제	비교적 안정	새 규칙이 만들어지는 중
주요 위험	건설·수요·운영	기술·인증·표준·기업생존 + 건설·수요
자산수명	장기	구성요소별 편차 큼
금융	PF Debt 조기 투입 가능	초기에는 Equity·Grant 중심
실패 시 자산	도로·철도 대부분 잔존	기술에 따라 가치 급락 가능

은행의 입장에서 생각하면 차이가 더 선명합니다.

유료도로에 돈을 빌려줄 때는 “차가 예상보다 적게 다니면 어떻게 할까?”를 봅니다.

Vertiport(버티포트·수직이착륙장)에 돈을 빌려줄 때는 그 전에 질문이 생깁니다.

“여기에 착륙할 항공기가 인증은 받을 수 있는가?”

Hyperloop라면 한 단계 더 갑니다.

“상업규모로 완성한 사람이 아직 없는데 누가 Fixed-Price EPC(확정가격 설계·조달·시공)를 보증할 것인가?”

B/C(편익비용비)가 높다고 은행이 대출해주는 것이 아닌 이유입니다.

Florida를 보면 정부가 왜 먼저 움직여야 하는지 꽤 잘 보입니다

Florida가 지금 재미있는 실험을 하고 있습니다. Polk County의 Auburndale, I-4 옆에 SunTrax라는 시설이 있습니다.

Florida Department of Transportation, FDOT(플로리다 교통부)가 개발·운영하는 775 acres, 약 313만㎡ 규모의 미래교통 연구·시험시설입니다. 처음에는 Connected and Autonomous Vehicle, CAV(커넥티드·자율주행차) 시험이 중심이었지만, 지금은 항공까지 붙었습니다.

SunTrax Air만 212 acres, 약 86만㎡입니다. Vertiport, 20,000ft²(약 1,858㎡) 격납고, 차량 크기와 전력수요 변화에 대응할 수 있는 고속 충전시설을 갖추고 있습니다. FDOT는 이를 미국 최초의 전용 AAM Aerial Test Track(첨단항공교통 공중 시험트랙)을 갖춘 시설로 설명합니다. (SunTrax)

여기서 눈여겨볼 부분이 있습니다. FDOT는 eVTOL 항공기를 직접 개발하지 않습니다. 어느 회사가 승자가 될지 미리 선택하지도 않았습니다. 대신 여러 기업이 들어와 시험할 수 있는 공용 인프라를 먼저 만들었습니다.

이 접근이 미래교통 P3에 상당히 잘 맞는다고 생각합니다.

특정 회사가 실패해도 Test Track(시험트랙), Vertiport, 격납고, 충전·통신시설과 정부 전문인력은 남는데다가

Technology Sponsor Risk(기술사업자 위험)를 공공자산과 분리할 수 있기 때문입니다.

Florida는 이걸 연구사업으로만 두지 않고 아예 법에 넣었습니다

정책의 강도도 한 단계 올라갔습니다.

Florida는 2025년 Florida Statutes §332.15를 신설했습니다.

이 조항은 FDOT가 Vertiport와 AAM을 Statewide Aviation System Plan(주 항공시스템계획)에 반영하도록 합니다. AAM 담당 전문가를 지정하고, Airport Hazard Zone(공항 위험구역) 규정을 검토하며, Florida Department of Commerce(플로리다 상무부)와 함께 주 전체 AAM 시스템계획과 Corridor(회랑)를 마련해야 합니다. 법은 2025년 Chapter 2025-155를 통해 만들어졌고 2026년에도 그대로 시행되고 있습니다. (Florida Senate)

여기서 정부 담당자의 시각이 보입니다.

회사마다 각기 다른 Vertiport를 만들고 지방정부마다 다른 규칙을 적용하면 시장이 쪼개집니다. 어느 도시에서는 허용되고 바로 옆 카운티에서는 안 된다면 민간은 네트워크를 만들기 어렵습니다.

그래서 정부가 먼저 하는 일은 돈을 주는 게 아닙니다.

공통된 규칙과 인프라 지도를 만드는 것입니다.

이런 Regulatory Certainty(규제의 예측가능성)는 눈에 보이는 보조금은 아니지만 금융에는 직접 영향을 줍니다.

상업운항 시점을 2028년으로 잡았는데 인허가가 3년 늦어지면 Equity IRR(자기자본 내부수익률)이 크게 떨어집니다. 건설을 먼저 시작했다면 IDC(Interest During Construction·건설기간이자)까지 늘어납니다.

정부가 인허가 불확실성을 낮춰주는 것 자체가 일종의 De-risking(위험완화)입니다.

연방정부도 2026년에는 아예 ‘정부+민간’ 방식으로 실증하도록 만들었습니다

이런 방향은 Florida만의 선택도 아닙니다.

미국 연방정부는 2025년 Executive Order(대통령 행정명령)를 통해 eVTOL Integration Pilot Program, eIPP(전기 수직이착륙기 통합 실증프로그램)를 만들었습니다.

중요한 조건이 있습니다.

주·지방·부족·영토정부가 단독으로 신청할 수 없습니다. eVTOL 개발·제조·운영 경험을 가진 민간사업자와 파트너십을 구성해야 합니다.

법정 프로그램은 아니지만 대통령 행정명령과 FAA의 Federal Register 절차를 통해 운영되고 있으며, 최소 5개 사업을 고르도록 했습니다. 정부와 기업이 Operational Concept(운영개념), 규제수요, 일정, 데이터공유, 책임을 협약으로 정하도록 설계됐습니다. (The White House)

실제로 2026년 3월 FAA는 30개 이상의 제안 가운데 8개 사업, 26개 주를 선정했습니다. Florida도 포함됐습니다.

Florida 사업은 Archer, BETA, Electra, Joby 등을 포함한 여러 산업 파트너와 함께 화물·승객운송·자동화·의료대응을 3단계로 시험하는 주 전체 프로그램입니다. FAA는 상당한 Public and Private Investment(공공·민간 투자)가 함께 들어간다고 설명했습니다. (연방항공청)

미국 정부가 P3를 요구하는 이유도 꽤 노골적입니다.

Sean Duffy U.S. Transportation Secretary(미 교통부 장관)는 AAM이 지역연결, 도시혼잡 완화, 응급·의료수송과 함께 미국 항공산업의 일자리와 경쟁력을 만들 수 있다고 설명했고, FAA Administrator Bryan Bedford(FAA 청장)는 실증에서 얻은 경험을 전국적으로 확장 가능한 안전·운영규칙을 만드는 데 활용하겠다고 밝혔습니다. (연방항공청)

여기에는 교통정책과 산업정책이 같이 들어 있습니다.

그리고 바로 그 때문에 정부 담당자가 더 조심해야 합니다.

산업을 키워야 한다는 정책목표 때문에 금융적으로 성립하지 않는 사업까지 P3로 밀어서는 안 됩니다.

2026년 Eve–FDOT 협력은 정부와 기업이 무엇을 서로 원하는지 잘 보여줍니다

2026년 7월 FDOT와 Eve Air Mobility(이브 에어 모빌리티)는 SunTrax Air에서 AAM 공동프로그램을 추진한다고 발표했습니다.

정부는 시설을 제공하고, Eve는 eVTOL 항공기·운항·서비스 전문성을 제공합니다. 같이 검증하는 대상은 항공기 성능 하나가 아닙니다. Infrastructure Readiness(인프라 준비도), 운영절차, 공역운항, 승객경험까지 봅니다. (FDOT)

Eve CEO Johann Bordais도 이 협력을 설명하면서 항공기만으로는 AAM이 성립하지 않고 인프라·운영·지역사회까지 하나의 생태계로 맞아야 한다는 점을 강조했습니다.

FDOT Secretary Jared Perdue(플로리다 교통부 장관)가 보는 방향은 다릅니다.

그는 SunTrax를 eVTOL을 Florida 교통망에 안전하게 통합하기 위한 시설로 설명합니다. 2026년 8월에는 SunTrax가 연말까지 실제 eVTOL 시험을 시작하고 자율·커넥티드 차량의 Florida 전역 도입도 지원할 것이라고 밝혔습니다. (FDOT)

같은 사업을 보고 있지만 목적은 다릅니다. 기업은 시장진입의 확실성을 원하고, 정부는 안전한 통합과 공공교통체계의 준비를 원합니다.

P3가 필요한 이유도 여기에 있습니다. 양쪽 목적을 하나의 실증·계약구조 안에서 맞춰야 하기 때문입니다.

지역 언론의 시각은 더 현실적입니다.

WFTV는 2026년 7월 협력을 보도하면서 “어떤 멋진 항공기가 만들어지는가”보다 Florida가 실제 전기 Air Taxi(에어택시)를 운영하려면 어떤 인프라와 절차가 필요한지를 시험한다는 점을 중심에 놓았습니다. Kissimmee Gateway Airport(키시미 게이트웨이 공항)의 Vertiport 계획도 같이 언급했습니다. (WFTV)

WESH 역시 8월 SunTrax 확장을 소개하면서 Florida가 공중 eVTOL과 지상 자율주행을 같은 공공 시험인프라에서 검증하고 있다는 점에 초점을 맞췄습니다. (Wesh)

이 정도가 되면 미래교통 P3는 더 이상 먼 미래 이야기는 아닙니다.

Orlando가 먼저 하는 것도 Vertiport 건설이 아니라 규칙 만들기입니다

Central Florida 안에서도 Orlando는 흥미로운 위치에 있습니다.

City of Orlando(올랜도시)는 NASA의 AAM Community Integration Partnership(첨단항공교통 지역사회 통합 파트너십)에 선정된 5개 정부기관 가운데 유일한 도시정부였습니다. (Orlando)

2025년 4월 Orlando City Council Workshop(올랜도 시의회 워크숍) 내용을 보면 현장 담당자가 실제 무엇을 고민하는지 더 잘 보입니다.

Jacques Coulon Mobility Innovation Manager(모빌리티혁신담당자)는 시가 잠재 Vertiport 위치, 지역 주요 Activity Center(거점) 연결, 현행 도시코드 Chapter 58의 Heliport 규정, GEICO Garage의 Vertiport 활용 가능성 등을 검토하고 있다고 설명했습니다.

당장 시설을 짓기보다 Phase 1에서는 규제·인프라·경제효과를 보고, Phase 2에서는 Zoning(토지이용규제)과 실제 상황을 가정한 Tabletop Exercise(모의계획훈련)를 했습니다. Phase 3에서는 Travel Demand Modeling(교통수요모형)을 통해 실제 연결축과 Vertiport 후보를 좁혀갈 계획입니다.

당시 GOAA(Greater Orlando Aviation Authority·광역올랜도항공청)는 Orlando International Airport에 2개 Vertiport 후보를 위한 RFP(Request for Proposals·제안요청서)도 진행하고 있었습니다. (Pub Orlando)

필는 이 순서가 꽤 중요하다고 생각합니다.

도로 P3라면 기본설계와 수요예측을 갖고 민간투자자를 찾는 것이 자연스럽습니다.

Orlando는 아직 도시 안에서 이 시장을 어디에 넣을 수 있는지부터 찾고 있습니다.

기술이 아직 움직이고 있기 때문입니다.

그런데 Orlando에는 이미 한번 꽤 큰 기대를 걸었다가 멈춘 경험도 있습니다

2020년에는 분위기가 지금보다 훨씬 낙관적이었습니다.

독일 eVTOL 기업 Lilium(릴리움)이 Tavistock Development Company(태비스톡 개발회사)와 Lake Nona에 미국 최초급 Vertiport를 만들겠다고 발표했습니다.

계획된 시설은 56,000ft², 약 5,200㎡였습니다.

5년 동안 143개 일자리를 만들고 임금은 Orange County(오렌지카운티) 중위임금의 150% 이상으로 제시했습니다.

City of Orlando는 Tax Rebate(세금환급)를 제공하기로 했고, Orlando Utilities Commission, OUC(올랜도 공공전력회사)와 GOAA도 협력했습니다. Lilium COO Remo Gerber는 Orlando가 혁신에 개방적이라는 점이 Florida를 선택한 이유라고 설명했습니다. (News)

지역언론 WESH도 당시 시의회가 Lilium과의 Incentive Agreement(인센티브 협약)를 심의한다며 “Orlando에 Flying Taxi가 오는가”라는 기대를 크게 보도했습니다. (Wesh)

2025년이 목표였습니다. 지금 보면 꽤 과감한 계획이었습니다.

문제는 시설보다 Lilium이라는 기술기업의 생존이었습니다.

Reuters는 2024년 Lilium의 붕괴를 보도하면서 eVTOL 업계가 2019년 이후 약 $13B(약 18조2,000억원)을 조달했지만 인증·생산시설·상업화까지 필요한 자금이 워낙 커서 상당수 기업이 Cash Crunch(현금부족)에 직면했다고 분석했습니다. (Reuters)

Lilium만의 문제도 아니었습니다.

같은 해 12월 독일 Volocopter도 자금조달에 실패해 Insolvency(지급불능) 절차에 들어갔습니다. (Reuters)

이 사례를 실패했다고 조롱할 일은 아닙니다. 오히려 P3 계약을 설계하는 사람에게 굉장히 좋은 자료입니다.

도로 P3에서 사업자가 부도나도 도로는 남습니다. 다른 운영자를 데려올 수 있습니다.

하지만, 특정 eVTOL 회사의 기종·충전방식·운영망에 맞춘 Vertiport라면 이야기가 다릅니다.

회사와 기술이 같이 사라질 수 있습니다.

이것이 미래교통에서 처음 등장하는 Technology Sponsor Risk(기술기업 존속위험)입니다.

민간기업도 이제 항공기보다 ‘네트워크’를 같이 확보하기 시작했습니다

그렇다고 eVTOL 민간기업이 공공에만 의존하고 있는 것도 아닙니다.

오히려 대규모 자본을 직접 투입하면서 인프라 확보에 나서고 있습니다.

Archer Aviation(아처 에비에이션)은 2025년 Hawthorne Airport(호손공항)를 $126M(약 1,764억원)에 인수하기로 했습니다.

80 acres, 약 32만㎡의 부지와 약 190,000ft²(약 1만7,650㎡)의 터미널·사무·격납시설을 가진 공항입니다. Archer는 이를 Los Angeles 에어택시 네트워크의 운영거점이자 AI 항공기술 Testbed로 활용할 계획입니다. (Archer Aviation)

Joby Aviation(조비 에비에이션)은 또 다른 방식을 택했습니다.

2025년 Metropolis와 미국 전역 25개 Vertiport를 개발하기로 했습니다. Metropolis가 보유한 4,200개 이상의 Parking Location(주차시설)과 350개 이상의 항공서비스 지점을 활용하는 전략입니다. (Joby Aero, Inc.)

2026년 Joby는 Los Angeles Century City의 기존 Helipad(헬리패드), Santa Monica Airport, John Wayne Airport 등을 연결하는 인프라 네트워크를 이미 확보 중이라고 투자자에게 설명했습니다. (Joby Aero, Inc.)

민간기업도 학습하고 있는 셈입니다. 좋은 eVTOL을 만드는 것만으로 사업은 끝나지 않습니다.

항공기와 동시에 부동산·공항·전력·운항권을 확보해야 합니다.

기업 입장에서도 인프라가 사업가치의 일부가 됐습니다.

그런데 금융시장은 아직 UAM을 ‘도로 같은 인프라’로 보지 않습니다

여기서 중요한 구분을 해야 합니다.

Archer나 Joby에 돈이 많이 들어갔다고 Project Finance가 가능하다는 뜻은 아닙니다.

2025년 6월 Archer는 미국 정부가 eIPP를 추진하는 Executive Order를 발표한 직후 $850M(약 1조1,900억원)을 추가 조달했습니다. 앞서 같은 해 $300M도 조달했고, $850M 조달 이후 회사는 약 $2B(약 2조8,000억원)의 유동성을 확보했다고 밝혔습니다. (Reuters)

이 돈은 우리가 민자도로에서 보는 Senior Debt(선순위 대출)가 아닙니다. 대부분 Corporate Equity(기업자본)에 가깝습니다.

왜일까요?

은행이 Infrastructure Debt를 넣으려면 아래 사항들이 확인돼야 합니다.

기술이 작동해야 합니다.
FAA 인증이 있어야 합니다.
안정적인 운영규칙이 있어야 합니다.
반복 가능한 수요가 있어야 합니다.
운영사업자가 부도났을 때 다른 사업자가 시설을 대신 사용할 수 있어야 합니다.

초기 eVTOL 시장은 아직 이 조건들이 동시에 충족되지 않다보니 금융순서도 기존 P3와 반대에 가깝습니다.

전통 도로 P3는 비교적 빨리 Equity + Debt → Construction → Operation으로 갑니다.

반면에 미래교통은 R&D Grant(연구개발 보조금) → Venture/Corporate Equity(벤처·기업자본)

→ Pilot·Demonstration(시험·실증) → 인증·표준 확립 → 초기 상업운영 → 실제 수요자료 축적

→ Infrastructure Project Finance 순서가 더 자연스럽습니다.

Project Finance가 출발점이 아니라 마지막에 가까워집니다.

그래서 한 사업 안에서도 돈의 가격이 달라야 합니다

UAM 사업 하나를 SPC(Special Purpose Company·특수목적법인)에 전부 넣는다고 생각해보겠습니다.

토지도 있고, Vertiport 건물도 있습니다. 변전소와 충전기도 있고, Software도 있고 항공기도 있습니다.

전부 같은 WACC(Weighted Average Cost of Capital·가중평균자본비용)를 적용하는 게 맞을까요?

그렇지 않다고 봅니다.

토지와 장기간 활용할 건축물은 Infrastructure Fund(인프라펀드)가 좋아할 만한 자산입니다. 전력 인입과 변전시설도 장기 Debt를 붙일 여지가 있습니다. 반면 eVTOL 항공기와 운항 Software는 아직 Technology Equity(기술기업 자기자본)가 감당해야 할 위험이 큽니다. 충전장비와 통신설비는 중간쯤입니다. 자산 자체는 있지만 기술교체주기가 짧습니다.

그래서 미래교통 P3에서는 Asset Layer(자산층)별로 금융을 분리하는 것이 더 합리적일 가능성이 큽니다.

예를 들면 이런 식입니다.

자산	예상 특성	더 어울리는 자본
토지·건축물	장기·저기술위험	장기 Debt + Infra Equity
변전·기본전력	장기·규제형	Utility/Infrastructure Capital
충전기	기술교체 필요	중기 Debt/Lease
통신·센서	짧은 기술수명	Service Contract
eVTOL 항공기	인증·기술위험 큼	Corporate/Strategic Equity
운영 Software	높은 기술변화	기업자본·서비스계약

여기서 P3 계약의 모습도 바뀝니다.

30년 동안 모든 설비를 같은 사업자가 유지하는 DBFOM보다 장기 부동산계약 + 중기 설비계약 + 짧은 기술서비스 계약을 한 사업 안에 계층화하는 편이 더 나을 수 있습니다.

자율주행에서는 자동차보다 ‘도로의 디지털 계층’이 P3에 더 가깝습니다

자율주행도 비슷합니다.

정부가 자율주행차를 직접 만드는 것은 이상합니다. 특정 자동차회사의 Software에 투자하는 것도 정부 역할과 거리가 있습니다.

하지만 Vehicle-to-Everything, V2X(차량·도로·다른 차량·보행자 간 통신)에 필요한 공공 디지털 인프라는 다릅니다.

2024년 USDOT는 전국 V2X Deployment Plan(배치계획)을 발표하면서 차량과 신호기·Roadside Unit(도로변 통신장치)·보행자 장비가 정보를 주고받는 상호운용 가능한 체계를 구축하겠다고 밝혔습니다. 정책목표에는 안전뿐 아니라 민간투자에 필요한 시장의 확실성을 높이는 것도 포함됩니다. (교통부)

구체적인 P3가 Utah에 있습니다.

Utah Department of Transportation, UDOT(유타 교통부)과 Panasonic은 $50M(약 700억원) 규모의 Smart Roadways Transportation Data Network를 추진했습니다. FHWA(연방도로청)는 공식적으로 이 사업의 계약방식을 Public-Private Partnership, UDOT/Panasonic Joint Venture로 분류합니다.

1단계는 $8M(약 112억원)입니다.

40개 설치지점과 주정부 차량 30대로 시작해 향후 220개 지점과 2,000대 차량까지 확대하도록 설계했습니다. 전체는 5개 단계, 5년에 걸친 방식입니다. (Federal Highway Administration)

재미있는 건 금융입니다. Funding Source(재원)는 UDOT와 Panasonic이고, Panasonic은 Equity Provider(자기자본 제공자)로 표시됩니다.

Lender는 없습니다.

민자도로처럼 금융기관이 수십 년 Debt를 제공하고 통행료로 갚는 사업이 아니라 정부와 기술회사가 초기시장을 같이 만드는 단계입니다.

이게 미래교통 P3의 또 다른 모습입니다.

연방정부도 처음부터 크게 투자하지 말고 ‘검증한 뒤 키우라’는 구조를 사용했습니다

V2X에서도 같은 방향이 보입니다.

USDOT는 2024년 Arizona, Texas, Utah의 대규모 V2X 구축사업에 약 $60M(약 840억원)의 연방자금을 지급했습니다.

Arizona는 $19.6M(약 274억원)을 받아 750개의 물리·가상 Roadside Unit과 약 400개 차량장치를 구축하는 사업을 추진했고, Texas에는 $19.2M(약 269억원) 등이 배정됐습니다. (교통부)

SMART Grants Program도 처음부터 단계형으로 만들었습니다.

2026년 현재 신규공모는 종료됐지만 기존 122개 Stage 1 사업과 7개 Stage 2 사업은 계속 진행되고 있습니다. 첫 Stage 2에는 총 $85M(약 1,190억원)이 투입됐습니다. (교통부)

정책의 메시지는 꽤 분명합니다.

기술이 검증되지 않았는데 처음부터 30년짜리 민자협약부터 맺지 말라는 것입니다.

먼저 공공자금과 민간자본으로 작은 실증을 진행하여 성과와 수요를 확인합니다. 그 다음 확장합니다.

마지막에야 장기금융이 들어옵니다.

저는 미래교통 P3에서는 이 Stage-Gate(단계별 관문) 구조가 거의 필수라고 생각합니다.

Hyperloop는 P3가 가능하다는 사례라기보다 ‘어디까지 가야 금융이 가능한가’를 보여줍니다

Hyperloop로 가면 이야기가 더 어렵습니다.

2018년 Northeast Ohio Areawide Coordinating Agency, NOACA(북동오하이오 광역계획기구)와 Hyperloop Transportation Technologies, HyperloopTT가 Great Lakes Hyperloop 사업을 위해 P3를 체결했습니다.

여기서 정확히 봐야 합니다.

$25B짜리 Hyperloop 건설 P3를 체결한 것이 아닙니다. P3 대상은 Feasibility Study(타당성조사)였습니다.

전체 조사비는 $1.2M(약 16억8,000만원)이었고 공공과 HyperloopTT가 50%씩, 최대 $600,000씩 부담하는 구조였습니다. 실제 NOACA 측 비용에는 Ohio DOT, Ohio Turnpike and Infrastructure Commission, Cleveland Foundation 등이 참여했습니다.

2019년 Pittsburgh까지 분석범위를 넓히면서 $100,000(약 1억4,000만원)이 추가됐습니다. (HyperloopTT)

당시 기대는 컸습니다.

NOACA Executive Director Grace Gallucci는 이 기술이 지역을 변화시킬 가능성이 있으며 이번 연구를 통해 정말 지역에 맞는 교통수단인지 확인하겠다고 설명했습니다. 당시 Ohio의 정치권도 초당적으로 관심을 보였습니다. (노스이스트 오하이오 지역 조정 기관)

2019년 결과도 상당히 낙관적이었습니다.

Benefit-Cost Ratio는 2.20으로 제시됐습니다. 전체 Chicago–Cleveland–Pittsburgh 건설비는 약 $25B~30B(약 35조~42조원)으로 추정됐습니다. (ITS KRS) 경제적 편익만 보면 상당합니다.

그런데 2026년 HyperloopTT의 공식 프로젝트 페이지를 보면 여전히 Great Lakes 사업은 Feasibility Study 단계로 표시돼 있습니다. 다음 단계는 환경검토, 역사계획, 설계, 금융이라고 적혀 있습니다. (HyperloopTT)

7년이 흘렀습니다.

왜 Financial Close가 안 나왔을까요?

이 질문이 핵심입니다.

B/C 2.20과 Bankability는 완전히 다른 숫자입니다

정부의 타당성조사는 사회 전체를 봅니다. 여행시간이 줄고 지역경제가 성장하고 환경비용이 줄어들면 Benefit(편익)에 들어갑니다.

은행은 그렇게 돈을 빌려주지 않습니다. 은행은 Project Cash Flow를 봅니다.

$30B(약 42조원)를 빌려준다고 생각해보겠습니다.

누가 EPC(설계·조달·시공)를 Fixed Price로 맡습니까?
튜브와 추진시스템이 목표성능을 못 내면 누가 보상합니까?
안전인증을 받을 수 있습니까?
운영사업자가 파산하면 다른 사업자가 시설을 이어받을 수 있습니까?
실제 Fare Revenue(운임수입)로 Debt Service(원리금)를 낼 수 있습니까?

현재 Hyperloop는 이 질문 가운데 상당수에 확정적인 답을 하기 어렵습니다.

그래서 B/C 2.20은 경제적 타당성을 말할 뿐 Bankability(금융가능성)를 보장하지 않습니다.

이 구분은 미래교통에서 특히 중요합니다.

P3를 쓰면 기술위험을 민간에 넘길 수 있다는 생각도 위험합니다

정부 입장에서는 이런 유혹이 생길 수 있습니다.

“기술이 불확실하니까 민간이 모든 Technology Risk를 부담하도록 하면 되지 않을까?”

실시협약 Risk Matrix(위험배분표)에 그렇게 적는 것은 쉽습니다.

하지만 민간이 그 위험을 실제로 통제할 수 없다면 위험이전이 아닙니다.

가격이 올라가고, 더 높은 Equity IRR을 요구하며, Contingency(예비비)가 커집니다.

은행이 대출을 거부할 수도 있습니다.

결국 경쟁력 있는 사업자가 입찰하지 않고 위험을 과소평가한 사업자가 들어올 가능성도 있습니다.

KentuckyWired나 Purple Line에서 봤던 문제와 본질적으로 같습니다.

계약서에 위험을 넘긴다고 써놓는다고 위험이 사라지는 것은 아닙니다.

미래교통에서는 이 원칙이 더 중요합니다.

그래서 계약기간도 한 번에 30년을 주기보다 자산수명에 맞춰 잘라야 합니다

UAM을 예로 들어보겠습니다.

Vertiport의 토지는 30년 뒤에도 가치가 있습니다. 건물도 쓸 수 있죠. 변전시설도 상당기간 남습니다. 그런데 충전기와 통신장비, Software는 그렇지 않습니다.

FAA의 현재 Engineering Brief 105A는 Vertiport를 Heliport(헬리포트)의 한 유형으로 다루고, 조종사가 탑승하고 VMC(시계비행기상조건)에서 운항하는 최대이륙중량 12,500lb 이하 eVTOL을 기준으로 설계지침을 제시합니다. FAA는 2024년 12월 최종 EB 105A를 공개했고 신규 Vertiport에는 14 CFR Part 157의 시설통지와 Part 77의 공역안전검토도 적용합니다. (연방항공청)

이 기준도 미래에는 바뀔 수 있습니다.

따라서 필자는 한 계약에서 이렇게 구분하는 것이 더 낫다고 봅니다.

Land & Building(토지·건물) : 20~30년.
Power Infrastructure(기본 전력설비) : 15~25년.
Charging Equipment(충전장비) : 5~10년 뒤 재투자.
Communication & Sensor(통신·센서) : 3~7년 Technology Refresh.
Operating Software(운영 소프트웨어) : 성과기준 계약 + 주기적 경쟁.

쉽게 말하면 오래가는 자산만 오래 묶는 것입니다.

미래교통에서는 Handback보다 Technology Refresh가 더 중요해질 수도 있습니다

전통적인 도로 P3에서는 계약 종료 시 Handback Requirement(시설인계조건)가 중요합니다.

30년 뒤 포장상태가 어떠해야 하는지, 구조물 잔존수명은 몇 년이어야 하는지를 계약에 넣습니다.

미래교통에서는 다른 조항이 하나 더 필요합니다.

바로 Technology Refresh Clause(기술갱신조항)입니다. 예를 들어 5년마다 충전·통신·센서 기술을 검토합니다. 업계 표준이 바뀌면 일정 범위까지 사업자가 교체합니다. 큰 변화라면 공공과 비용을 나눕니다. 여기에 Open Standard(개방형 표준)도 필요합니다.

특정 eVTOL 제작사만 착륙할 수 있는 Vertiport를 30년 동안 운영하게 하면 공공이 그 기업에 Lock-in(종속)됩니다. 특정 통신기업의 V2X Protocol에 한 주의 도로를 전부 묶는 것도 마찬가지입니다. 이건 기존 도로 P3에서는 상대적으로 덜 중요했던 계약조건입니다.

실패했을 때 남는 자산을 처음부터 계산해야 합니다

저는 미래교통 P3에서 Residual Value(잔존가치)를 기존 사업보다 훨씬 중요하게 봐야 한다고 생각합니다.

UAM이 예상보다 늦어졌습니다. 그런데 Vertiport가 공항·헬리포트·물류·주차·상업시설로 활용될 수 있다면 손실을 줄일 수 있습니다.

자율주행차가 V2X 없이 차량 자체센서 중심으로 발전했습니다. 그래도 통신망을 버스우선신호, 교통관리, 응급차량 Priority Service에 쓸 수 있다면 가치가 남습니다.

Hyperloop 기술이 상업화되지 않았다고 해도 확보한 Right-of-Way가 철도·전력·통신 Corridor로 전환될 수 있다면 역시 가치가 있습니다.

그래서 사업타당성조사에서 저는 이런 질문을 하나 더 넣고 싶습니다.

“Base Case가 아니라 Technology Failure Case에서 공공이 돌려받는 자산의 가치는 얼마인가?”

미래교통의 VfM(Value for Money·재정효율성)은 이 Downside Value(하방가치)까지 봐야 합니다.

정부와 민간의 역할도 훨씬 잘게 나눠야 합니다

정부가 잘하는 영역은 꽤 분명합니다.

공공부지, 인허가, 안전규제, 공역, 도로·교통망 연결, 데이터표준과 지역사회 협의입니다.

민간이 더 잘할 수 있는 영역도 있습니다.

항공기와 차량기술, 생산, Software, 고객서비스, 운영효율, 기술갱신입니다.

초기수요는 애매합니다.

여기서 정부가 Minimum Revenue Guarantee(최소수입보장)를 바로 제공할 필요는 없다고 봅니다.

대신 의료수송, 공공기관 이동, 긴급물류 같은 서비스를 실제로 구매하면서 Anchor Customer(초기 핵심고객)가 되는 방식이 있을 수 있습니다.

#67에서 송전망의 Transmission Facilitation Program을 보며 봤던 구조입니다.

수요를 보장하는 것과 공공이 실제 필요한 서비스를 사는 것은 다릅니다.

한국에서 시작한다면 저는 ‘UAM 전용시설’보다 Mobility Hub부터 볼 것 같습니다

한국에서도 UAM은 정부와 민간이 같이 움직일 수밖에 없습니다. 항공안전·공역·도시계획은 정부 영역이지만, 항공기·운항·Software는 민간이 훨씬 잘합니다.

초기부터 UAM 승객수만으로 Vertiport의 Project Finance를 만들기는 쉽지 않을 겁니다.

Vertiport 하나만 떼기보다 철도역, 버스터미널, 주차, 상업시설, 전력·충전시설과 UAM을 합친 Mobility Hub(복합 모빌리티 허브)를 먼저 추진해보는게 어떨까 생각해봅니다. UAM 수요가 5년 늦어져도 주차·상업·환승수입이 있고, 시설을 다른 교통사업자가 쓸 수도 있습니다. 초기 UAM Demand Risk를 보완하기 위해 정부가 30년간 승객수 부족분을 보전하는 것보다 훨씬 안전한 구조입니다.

P3 금융에서도 같은 원칙입니다.

미래기술의 불확실성을 기존의 안정적인 Cash Flow와 섞어
Project Finance가 가능한 부분만 분리해내는 것입니다.

#66부터 여기까지 오고 나니 P3를 보는 기준도 조금 달라졌습니다

처음 미국 P3를 볼 때는 시설 종류를 먼저 봤습니다.

도로인가, 철도인가. 사용료를 받을 수 있는가. 정부지급금이 있는가.

그런데 분야를 계속 넓혀보니 P3에 적합한지를 결정하는 기준이 꼭 시설 이름에 있지는 않았습니다.

#66의 대학·담수화시설에서는 장기계약이 안정적인 Cash Flow를 만들 수 있는지가 중요했습니다.

#67의 데이터센터와 전력망에서는 민간자본이 있는데도 공공만 해결할 수 있는 병목이 있는지가 중요했습니다.

미래교통에서는 하나가 더 붙습니다.

기술이 바뀌거나 실패해도 공공이 투자한 인프라에 가치가 남는가.

앞으로 새로운 분야를 P3로 검토하면서 네 가지를 먼저 물을 것 같습니다.

민간이 혼자 해결하지 못하는 공공의 병목이 실제로 있는가.
기술은 장기계약을 맺을 만큼 성숙했는가.
계약이 미래의 더 좋은 기술과 경쟁을 막지 않는가.
그리고 실패해도 공공에 남는 자산이 있는가.

UAM은 이미 이 질문에 어느 정도 답할 수 있는 단계로 들어오고 있습니다.

자율주행·V2X도 공용 디지털 인프라부터 P3 적용을 생각해볼 수 있습니다.

Hyperloop는 아직 다릅니다. 2018년 P3로 타당성조사를 하는 것은 합리적이었지만, 현재 단계에서 바로 $25~30B(약 35조~42조원) 규모의 30년 DBFOM으로 넘어가는 것은 전혀 다른 이야기입니다.

여기서 이번 글의 답도 나옵니다.

미래교통에도 P3는 적용할 수 있습니다. 다만 미래기술 자체에 장기 베팅하는 P3보다,
여러 기술이 이용할 수 있고 기술이 바뀌어도 남는 인프라를 P3로 만드는 편이 훨씬 안전합니다.

미래교통 P3의 경쟁력은 민간자본을 얼마나 많이 끌어왔는지에서 나오지 않을 겁니다.

불확실성은 열어두면서도 필요한 인프라는 먼저 만들 수 있는 계약을 얼마나 잘 설계했느냐.

필자의 생각에는 그게 도로와 철도 다음 세대 P3가 풀어야 할 문제입니다.

핵심 참고자료

FAA(미국 연방항공청) – Advanced Air Mobility Implementation Plan FAA AAM Implementation Plan
FAA – Engineering Brief 105A / AAM Infrastructure FAA AAM Infrastructure
Florida Statutes §332.15 Florida Statutes §332.15
FDOT – SunTrax / Eve Air Mobility (SunTrax)
FAA – eVTOL Integration Pilot Program (연방항공청)
City of Orlando (Orlando)
WFTV·WESH (WFTV)
Orlando Economic Partnership – Lilium Lake Nona (News)
Reuters – eVTOL Financing (Reuters)
Archer Aviation·Reuters (Archer Aviation)
Joby Aviation (Joby Aero, Inc.)
FHWA – Utah Smart Roadways (Federal Highway Administration)
USDOT – V2X Deployment (교통부)
NOACA·HyperloopTT – Great Lakes Hyperloop (HyperloopTT)
Reuters (Reuters)

#67- 민간의 데이터센터·전력망에 왜 정부가 P3로 들어갈까?

foweuniverse — Sun, 30 Aug 2026 00:36:02 +0900

- 민간자본이 충분한 데이터센터·전력망에 왜 정부가 P3로 들어갈까요? Portsmouth AI 캠퍼스, DOE 송전촉진프로그램(TFP), Virginia 데이터센터 전력요금 사례를 통해 부지·인허가·초기계약·전력망 비용 같은 병목을 정부와 민간이 어떻게 나누는지 살펴봅니다. -

※ 본문의 원화는 사례 비교를 위해 $1=1,400원으로 단순 환산했습니다. 실제 사업 당시 환율과는 다릅니다.

#66에서는 조금 낯선 P3를 봤습니다.

시청에는 통행료가 없었지만 Long Beach시는 40년 동안 Service Fee(서비스지급금)를 지급했습니다. UC Merced에서는 Availability Payment(AP·가용성지급금)가 있었고, Carlsbad에서는 30년 Water Purchase Agreement(WPA·물 구매계약)가 금융을 가능하게 했습니다.

정부나 공공기관이 장기 Cash Flow(현금흐름)를 만들고 민간은 그 현금흐름을 담보로 Debt(부채)와 Equity(자기자본)를 조달하는 구조였습니다.

그런데 데이터센터로 넘어오면 상황이 달라집니다.

Microsoft, Google, Amazon, Meta 같은 기업은 정부가 AP를 지급하지 않아도 수십억 달러를 투자합니다. 발전회사와 송전사업자도 충분한 고객과 장기계약이 있다면 스스로 금융을 일으킬 수 있습니다.

그렇다면 정부가 왜 P3로 들어갈까요?

필자도 처음에는 이 질문이 가장 먼저 들었습니다.

민간이 돈을 벌 수 있는 사업이라면 굳이 정부가 끼어들 이유가 없는 것 아닌가?

자료를 따라가다 보니 지금 미국에서 나타나는 문제는 민간자본 자체의 부족과는 조금 달랐습니다.

돈을 투자할 기업은 있습니다. 수요도 있습니다.

그런데 5GW, 10GW짜리 데이터센터를 실제로 짓기 시작하려면 대규모 부지부터 전력, 발전소, 765kV 송전선, 가스관, 환경정화, 여러 기관의 인허가가 거의 동시에 준비돼야 합니다.

그중 하나라도 막히면 수십조원의 민간투자가 움직이지 못합니다.

이게 바로 최근 미국에서 P3가 새로운 분야로 확장되는 이유입니다.

정부가 민간의 수익을 만들어주는 것이 아니라,
민간만으로는 풀기 어려운 병목을 제거하는 쪽으로 역할이 바뀌고 있습니다.

AI가 데이터센터를 ‘건물’에서 ‘에너지 인프라’로 바꿨습니다

몇 년 전까지만 해도 데이터센터 정책을 이야기하면 부지와 세금이 먼저 나왔습니다.

지금은 전력이 먼저입니다.

Lawrence Berkeley National Laboratory, LBNL(로런스버클리국립연구소)의 최신 분석은 미국 데이터센터 전력소비가 2030년 649TWh까지 올라갈 수 있다고 봅니다. 미국 전체 전력소비의 "11.8%"입니다. 시나리오에 따라서는 "9.5~15.3%"까지 올라갑니다. (BIES)

전력소비의 10% 이상을 하나의 산업이 차지한다면 데이터센터 유치정책은 더 이상 건축허가나 산업단지 문제라기보다는 전력계통 의 정책문제가 됩니다.

실제로 2026년 8월 Reuters는 미국의 Grid Bottleneck(전력망 병목)이 심해지면서 기업들이 전기요금을 조금 아끼는 것보다 Energy Certainty, 즉 필요한 시점에 전력을 실제로 확보할 수 있다는 확실성을 더 중시하고 있다고 보도했습니다. 자체발전이나 Behind-the-Meter Power(계량기 뒤편의 현장발전)를 확대하는 것도 같은 이유입니다. (Reuters)

민간사업자의 입장에서 생각하면 이해가 쉽습니다.

“데이터센터에 $10B(약 14조원)를 투자할 수 있습니다. 그런데 2029년에 2GW를 실제로 공급받을 수 있습니까?”

발전사업자는 반대편에서 묻습니다.

“데이터센터가 정말 들어온다는 구속력 있는 계약이 있습니까? 그래야 발전소와 송전망에 먼저 투자할 수 있습니다.”

서로 사업성이 없다고 생각하는 것이 아닙니다.

상대방이 먼저 움직여야 자신도 움직일 수 있습니다.

이런 Chicken-and-Egg Problem(선후관계 교착)이 생기면 자본이 충분해도 사업이 진행되지 않습니다.

정부가 끼어들 공간은 여기서 생깁니다.

폐쇄된 핵시설이 세계 최대 AI 캠퍼스로 바뀌는 데 정부가 한 일은 무엇일까요?

2026년 3월 U.S. Department of Energy, DOE(미국 에너지부)와 Department of Commerce, DOC(미국 상무부)는 Ohio주 Pike County의 Portsmouth Gaseous Diffusion Plant(포츠머스 가스확산공장) 부지를 활용한 대규모 Public-Private Partnership(민관협력)을 발표했습니다.

여기는 과거 우라늄을 농축하던 연방 핵시설인데, 이 시설을 활용하기 위해 계획된 규모가 상당합니다.

SoftBank Group(소프트뱅크그룹)의 SB Energy는 최대 10GW AI 데이터센터와 이를 지원할 10GW 신규 발전설비를 계획했습니다. DOE 발표 기준으로 발전설비 가운데 최소 9.2GW는 천연가스 발전입니다. (The Department of Energy's Energy.gov)

대형 원전 한 기를 대략 1GW로 생각하면 발전설비만 원전 10기 수준입니다.

이렇게 큰 발전 설비면 전력을 옮길 설비도 따로 필요합니다.

SB Energy와 AEP Ohio는 $4.2B(약 5조8,800억원) 규모의 신규 송전망에 투자하기로 했습니다. 여기에는 765kV 송전선과 변전소 4곳이 포함됩니다. DOE 설명에서는 SB Energy가 이 비용을 부담해 일반 Ohio 전력소비자의 요금으로 전가하지 않는 구조를 강조하고 있습니다. (The Department of Energy's Energy.gov)

정부는 땅을 제공할뿐 10GW 데이터센터 건설비를 대신 내는 것은 아닙니다.

2026년 4월 DOE 자료를 보면 Pike County Portsmouth DC LLC가 Portsmouth 부지 가운데 약 189 acres, 약 76만㎡를 AI·Cloud Data Center 용도로 임차했습니다. (The Department of Energy's Energy.gov)

그런데 일반 산업단지를 빌려주는 것과는 달랐습니다.

과거 핵시설이기 때문에 환경정화와 연방정부 재산관리 문제가 따라붙었습니다.

DOE의 임대권한에는 42 U.S.C. §7256(c)~(f)가 사용됐고, 환경 관련해서는 CERCLA §120(h)와 10 CFR Part 770도 적용됐습니다. DOE의 2025년 부지임대 문서에 이 법적 근거가 명시돼 있습니다. (The Department of Energy's Energy.gov)

민간사업자 입장에서 이 사업의 매력은 “정부 땅이라 싸다” 정도로 보기 어렵습니다.

훨씬 중요한 것은 대규모 부지, 환경정화, 발전, 송전, 인허가를 하나의 프로젝트로 묶을 가능성이 높아졌다는 점입니다.

정부 담당자 입장에서도 계산이 있습니다.

폐쇄된 핵시설 부지를 계속 보유하면 관리·정화비용이 들어갑니다. 활용하면 민간의 발전·송전투자를 끌어오면서 지역 일자리와 경제활동도 만들 수 있습니다. DOE는 건설기간 약 10,000개, 운영단계 2,000개 이상의 일자리를 예상하고 있습니다. (The Department of Energy's Energy.gov)

필자는 이 사업에서 새로운 P3의 모습이 상당히 잘 보인다고 생각합니다.

정부는 민간사업자의 매출을 보장하지 않습니다. 민간이 혼자 확보하기 어려운 부지와 공공절차를 해결합니다.

인허가 기간도 이제는 금융조건입니다

Portsmouth 사업에는 2026년 6월 FAST-41이 적용됐습니다.

PORTS Technology Campus는 High-Performance Computing(고성능컴퓨팅) 분야 최초의 FAST-41 적용사업입니다. (Permitting Council)

FAST-41은 여러 연방기관이 따로 처리하던 환경검토와 인허가 일정을 하나의 체계에서 조정하고, Permitting Dashboard(인허가 공개관리체계)를 통해 일정과 책임기관을 관리하는 제도입니다.

법적 근거는 42 U.S.C. §4370m 이하입니다.

현행법은 적용분야에 Artificial Intelligence and Machine Learning(인공지능·머신러닝), High-Performance Computing, Advanced Computer Hardware and Software, Data Storage, Energy Storage까지 명시합니다.

일반적인 Covered Project(적용대상사업)는 NEPA(Environmental Policy Act·국가환경정책법) 검토 대상이고 총투자액이 $200M(약 2,800억원)을 넘으면서 간이절차 대상이 아닌 대형사업 등이 포함됩니다. (법률 정보 기관)

정부 담당자에게 인허가는 행정절차일 수 있지만, 민간사업자에게는 수십억, 수백억원이 움직이는 금융조건입니다.

Financial Close 이후 공사가 1년 늦어지면 IDC(Interest During Construction·건설기간이자)가 늘어납니다. Equity가 먼저 투입됐다면 IRR(내부수익률)도 떨어집니다.

AI 인프라는 기술세대가 빠르게 바뀌기 때문에 몇 년 늦게 시장에 들어가는 것 자체가 Technology Risk(기술위험)가 될 수 있습니다. 그러니 정부가 사업자에게 $500M을 직접 보조하는 것과 인허가 불확실성을 2년 줄이는 것 중 어느 쪽이 더 가치 있는 지원인지는 사업마다 달라질 수 있습니다.

이 부분이 앞으로 한국에서도 꽤 중요해질 것이라고 봅니다.

정부지원의 단위를 항상 “예산 얼마”로만 생각하면 실제 민간이 원하는 지원과 어긋날 수 있습니다.

송전망에서는 정부가 아예 ‘첫 번째 고객’이 됩니다

새로운 P3의 구조가 더 선명하게 보이는 제도는 Transmission Facilitation Program, TFP(송전촉진프로그램)입니다.

2021년 Infrastructure Investment and Jobs Act, IIJA(인프라투자 및 일자리법) §40106으로 도입됐고 현재 42 U.S.C. §18713에 규정돼 있습니다.

프로그램 규모는 $2.5B(약 3조5,000억원)입니다.

일반적인 보조금과는 구조가 다릅니다.

DOE는 적격 송전사업에 Capacity Contract(송전용량계약), Loan(융자), Public-Private Partnership 방식으로 참여할 수 있습니다. (The Department of Energy's Energy.gov)

그중 가장 흥미로운 것은 Capacity Contract입니다.

DOE가 민간 송전선의 최대 50% 용량을 최대 40년 동안 먼저 계약할 수 있습니다. (법률 정보 기관) 그렇다고 정부가 전기를 실제로 40년 동안 쓰겠다는 뜻은 아닙니다. 단지 정부가 Anchor Customer(앵커고객·최초 핵심고객) 역할을 하는 것입니다.

송전사업의 PF를 생각해보면 이유가 보입니다.

은행은 장기 이용계약을 원하는데, 전력회사나 발전사업자는 송전선이 실제로 완공될 확신이 생겨야 장기계약을 맺습니다.

송전사업자는 다시 고객계약이 있어야 금융을 받을 수 있습니다.

서로 기다립니다.

DOE가 먼저 일부 Capacity를 계약하면 이 연결고리가 끊어집니다.

법에도 목적이 상당히 분명하게 들어 있습니다. 정부 Capacity Contract는 추가 고객의 송전용량 계약을 유도하고 민간투자를 촉진하기 위한 장치입니다. DOE는 프로젝트가 충분한 민간계약을 확보하면 빠져나오는 것을 목표로 합니다. (The Department of Energy's Energy.gov)

법은 ‘정부가 없으면 안 되는 사업’에만 들어가라고 합니다

TFP에서 제가 가장 눈여겨본 조항은 42 U.S.C. §18713(i)입니다.

지원사업을 결정할 때 DOE는 단순히 “좋은 송전사업인가?” 보다 먼저 검토하는 요건이 있습니다.

공익성이 있어야 하고, 정부지원이 없으면 해당 사업이 적기에 건설되지 않거나 충분한 용량으로 건설되기 어려워야 하며, 장래 Revenue를 통해 정부 투입비용을 회수할 가능성도 있어야 합니다. (법률 정보 기관)

정부 담당자에겐 상당히 챌린지한 부분입니다.

왜냐하면 사업성이 아예 없는 사업에 정부가 들어가면 안 되지만, 반대로 사업성이 완벽하고 이미 모든 고객을 확보했다면 정부가 들어갈 이유가 없습니다. DOE도 TFP가 이미 Fully Subscribed(용량이 전부 계약된) 사업에는 적합하지 않다고 설명합니다. (The Department of Energy's Energy.gov)

지원의 대상은 그 사이입니다.

사업성은 있지만 시장이 아직 첫발을 떼지 못해 Financial Close가 막혀 있는 사업.

이 지점이 P3와 일반 보조금을 가르는 데 꽤 중요한 기준이라고 생각합니다. 민간사업자도 “우리 사업이 안 되니 정부가 돈을 달라”고 해서는 설득력이 약합니다.

오히려 이렇게 보여줘야 합니다.

“초기 Market Risk의 일부만 해결되면 이후에는 민간계약과 사업수익으로 스스로 굴러갑니다.”

정부와 민간이 바라보는 사업의 적정성 기준이 여기서 만납니다.

실제 TFP에서도 정부가 들어왔다고 모든 사업이 살아나지는 않았습니다

2024년 10월 DOE는 두 번째 TFP Capacity Contract 선정에서 네 개 송전사업에 약 $1.5B(약 2조1,000억원)을 조건부로 투입했습니다.

약 1,000 miles(약 1,609km)의 송전망과 7,100MW 신규용량입니다. (The Department of Energy's Energy.gov)

그런데 이후 일부 사업은 계약에서 빠졌습니다. Cross-Tie Transmission Line은 2025년 11월, SWIP-N은 2026년 6월 Facilitation Agreement 종료를 통보했습니다. (The Department of Energy's Energy.gov)

정부가 최초고객이 됐다고 사업성이 새로 생기는 것은 아닙니다. 사업비 상승, 토지확보, 인허가, 경쟁사업, 시장수요가 바뀌면 결국 사업이 멈출 수 있습니다.

P3는 Market Failure를 고칠 수는 있지만 Bad Project를 Good Project로 바꾸는 마법은 아닙니다.

이건 교통 P3에서도 계속 봤던 이야기입니다.

Virginia는 반대로 ‘너무 많이 유치해서’ 문제가 생겼습니다

데이터센터 정책에서 Virginia(버지니아)는 거의 반대편의 사례입니다.

오랫동안 세제혜택을 이용해 데이터센터를 적극적으로 유치했고 Northern Virginia(북버지니아)는 세계 최대 데이터센터 집적지 가운데 하나가 됐습니다.

법적 기반은 Virginia Code §58.1-609.3입니다. 일정 요건을 충족한 데이터센터의 Server(서버), Router(라우터), Cooling Equipment(냉각설비), Backup Generator(비상발전기) 등에 Sales and Use Tax(판매·사용세)를 면제해줍니다.

장기 초대형 투자에 대한 조건도 상당합니다. 현행 §58.1-609.3(19)는 $100B(약 140조원) 이상의 누적투자와 2,500개 이상 일자리 등의 요건을 충족하면 세제혜택을 2050년까지 연장할 수 있도록 합니다. (Virginia Law)

세제지원 효과도 작지 않습니다. Virginia Department of Taxation(버지니아주 세무국)의 2026년 보고서에 따르면 FY2024~2025 기간 세제면제와 관련된 State·Local Tax Revenue는 약 $5.346B(약 7조4,838억원)로 추정됐고, 세제혜택 약 $3.234B(약 4조5,276억원)을 빼면 순세수는 약 $2.112B(약 2조9,568억원)입니다. Revenue-to-Cost Ratio는 1.7이었습니다. (RGA Virginia)

과거 JLARC 분석에서는 세제혜택을 받은 데이터센터 투자 가운데 약 90%가 혜택이 없었다면 Virginia에서 이뤄지지 않았을 것이라고 추정했습니다. (RGA Virginia)

산업유치 정책으로만 보면 성공에 가깝습니다만, 문제는 너무 성공해서 전력인프라가 따라오지 못한다는 것이 었습니다.

JLARC(Joint Legislative Audit and Review Commission·버지니아 주의회 감사·정책평가기관)는 데이터센터 때문에 앞으로 대규모 신규 발전과 송전투자가 필요하고, Dominion Energy 일반가구의 발전·송전비용이 2040년 현재가격 기준 월 $14~37(약 1만9,600~5만1,800원) 늘어날 가능성이 있다고 분석했습니다. (Virginia's JLARC)

이렇게 새로운 상황이 발생하면 정부 담당자의 고민이 달라집니다.

처음에는

“어떻게 투자기업을 유치할까?”

였다면,

이제는

“그 기업 때문에 필요한 수십억 달러의 전력망 비용을 누가 부담해야 할까?”

가 됩니다.

처음부터 이러한 부작용을 고려해서 정책을 만들면 되지 않냐라고 생각할 수 있겠지만 담당자가 사람인 이상, 비용, 법령, 정책 우선순위라는 제약하에서 모든 것을 만족하면서 진행하기는 사실상 쉽지 않다는게 현실입니다.

Virginia가 만든 답은 ‘오래 쓸 것을 계약으로 증명하라’입니다

Virginia State Corporation Commission, SCC(버지니아주 공공서비스규제위원회)는 대규모 전력사용자를 위한 새로운 요금·계약체계를 만들었습니다. 2027년 이후 신규 대형 Load(전력수요)는 최소 14년 동안 계약을 유지해야 합니다.

실제 전력을 예상보다 적게 사용하더라도 Transmission·Distribution Cost(송전·배전비)의 최소 "85%"를 지급해야 합니다.

신용도가 부족한 고객에게는 계약기간 최소지급액의 최대 "60%"에 해당하는 Collateral(담보)을 요구할 수 있습니다. (Virginia SCC)

왜 이렇게까지 할까요?

전력회사가 특정 데이터센터를 위해 $2B짜리 송전망을 먼저 만들었는데 5년 뒤 데이터센터 사업자가 철수한다고 생각해보겠습니다. 남은 투자비가 일반 가정의 전기요금으로 넘어갈 수 있습니다.

그래서 사업자에게 말하는 것입니다.

“당신의 수요 때문에 장기 인프라를 건설하니, 당신도 그 수요가 실제라는 것을 장기계약과 담보로 증명하십시오.”

이건 전력요금제이지만 계약구조를 보면 P3에서 익숙한 Risk Allocation(위험배분)과 매우 비슷합니다.

Ohio에서도 같은 문제가 나타났습니다.

AEP Ohio의 Data Center Tariff(데이터센터 전용요금제) 시행 뒤 2026년 2월까지 사업자들이 Collateral을 제공하고 Binding Contract(구속력 있는 계약)를 체결한 수요가 5,642MW입니다. 기존 계약 12,219MW와 별도입니다. AEP는 후보수요 13,022.7MW 가운데 실제 금융·법적 약속을 할 사업자만 남기는 과정에서 최종 5,642MW가 계약됐다고 설명했습니다. (AEP Ohio)

이 숫자가 보여주는 것도 흥미롭습니다.

신청서에 적힌 수요와 실제 돈을 걸고 계약할 수요는 다릅니다.

정부와 Utility(공공전력회사)의 역할은 민간의 수요 전망을 그대로 믿는 것이 아니라
계약 가능한 수요로 걸러내는 것이기도 합니다.

시민들의 시선도 바뀌고 있습니다

산업정책에서 이 부분을 놓치면 안 됩니다.

Washington Post와 George Mason University Schar School(조지메이슨대 샤르스쿨)의 2026년 조사에서 자신의 지역에 새로운 데이터센터가 들어오는 데 편안하다고 답한 Virginia 유권자는 "35%"였습니다. 2023년에는 "69%"였습니다. (The Washington Post) 불과 3년 사이 분위기가 크게 달라졌습니다.

이유도 이해할 만합니다.

전기요금, 송전선, 물 사용, 대규모 토지개발 문제가 실제 생활권으로 들어오기 시작했기 때문입니다.

2026년 7월 Goochland County에서는 870 acres, 900MW 데이터센터 캠퍼스를 놓고 주민들의 공개적인 반대가 이어졌습니다. (Virginia Mercury) 같은 달 Prince William County는 최대 약 2,000 acres, 43 million square feet 규모 개발이 가능했던 Dulles Cloud South 데이터센터 계획을 부결했습니다. (Virginia Mercury)

민간사업자의 입장에서는 답답할 수 있습니다. 정부가 데이터센터를 유치한다고 해놓고 전력망과 지역 인허가를 제때 해결하지 못하면 수십억 달러 투자가 그대로 묶입니다.

지역주민 입장에서는 반대입니다. 기업의 수익을 위해 발전소·송전선과 환경부담을 지역이 떠안는데 실제 일자리나 생활편익은 기대보다 적다고 느낄 수 있습니다.

정부 담당자가 가장 어려워지는 지점입니다.

투자액이 크다고 좋은 사업이라고 할 수도 없고, 주민이 반대한다고 무조건 나쁜 사업이라고 볼 수도 없습니다.

P3가 장기간 유지되려면 결국 누가 편익을 얻고 누가 비용을 부담하는지가 납득 가능해야 합니다.

왜 하필 지금 이런 분야에서 P3가 필요해졌을까요?

저는 이 질문을 조금 길게 봐야 한다고 생각합니다.

과거 SOC에서는 공공수요가 먼저였습니다. 정부가 도로와 철도, 댐이 필요하다고 결정하고 인프라를 건설했습니다.

1990~2010년대 P3는 여기에 민간의 자본과 건설·운영능력을 결합했습니다. 최근 데이터센터와 전력인프라는 순서가 뒤집힙니다.

민간의 기술투자가 먼저 움직이고 공공 인프라가 뒤에서 따라갑니다.

AI기업은 몇 년 안에 데이터센터를 지으려 합니다. 하지만 발전소, 송전망, 변전소, 가스관, 환경평가, 토지보상은 그렇게 빠르게 움직이지 않습니다.

이렇게 되면 민간의 투자주기와 공공 인프라의 구축주기가 어긋납니다. 정부가 아무것도 하지 않으면 민간투자가 다른 지역으로 갑니다. 반대로 정부가 발전소·송전망·세금까지 모두 대신 부담하면 민간의 수익을 공공이 보조하는 구조가 됩니다.

결국 필요한 것은 그 중간입니다.

여기서 한 가지 선은 그을 필요가 있습니다. Virginia처럼 Tax Incentive(세제혜택)를 제공한다고 모두 P3라고 부를 수는 없습니다.

기업유치정책과 P3가 완전히 같은 것은 아닙니다.

공공과 민간 사이에 장기간의 역할·비용·위험 가운데 적어도 일부가 계약적으로 연결될 때
P3라는 표현이 더 적절하다고 봅니다.

그렇지 않으면 산업단지 조성부터 세액공제까지 전부 P3가 되어버립니다.

정부 담당자와 민간사업자는 같은 사업을 다르게 봅니다

이 글의 사례를 정부와 민간의 책상 위에 각각 올려놓으면 질문이 상당히 달라집니다.

정부 담당자가 묻는 것	민간사업자가 묻는 것
정부지원이 없으면 정말 사업이 안 되는가?	부지와 전력을 언제 확정할 수 있는가?
민간에게 넘길 수 있는 비용은 무엇인가?	정부가 책임질 인허가·정화·접속 위험은 무엇인가?
일반 국민에게 비용이 전가되지 않는가?	정책이 5~10년 뒤에도 유지되는가?
지원이 끝난 뒤 자립할 수 있는가?	Financial Close 가능한 계약이 있는가?
지역사회가 얻는 편익은 무엇인가?	주민·환경 리스크를 누가 조정하는가?

양쪽의 이해가 완전히 일치할 필요는 없습니다. 오히려 달라야 정상입니다.

문제는 각자가 통제할 수 없는 위험을 상대에게 넘기려고 할 때 생깁니다. 예를 들면, 정부가 기술수요를 보장해서는 안 되고, 민간이 정부만 해결할 수 있는 인허가와 공공부지 위험을 떠안는 것도 효율적이지 않습니다. 그래서 P3의 오래된 원칙 하나는 AI 시대에도 그대로 남습니다.

위험은 그 위험을 가장 잘 통제할 수 있는 주체에게 둬야 합니다.

제가 이 사례들을 보고 P3 대상시설을 보는 기준도 조금 바뀌었습니다

예전에는 어떤 시설이 P3에 적합한지를 볼 때 이런 질문부터 떠올렸습니다.

사용료를 받을 수 있는가.
정부지급금을 만들 수 있는가.
민간수익이 충분한가.

물론 여전히 중요한 질문입니다.

그런데 데이터센터와 전력망을 보고 나니 하나를 더 넣어야 할 것 같습니다.

“민간이 투자할 의사와 자본은 충분한데, 공공만이 풀 수 있는 병목 때문에 투자가 막혀 있는가?”

그렇다면 P3가 상당히 유용한 정책수단이 될 수 있습니다. 다만 정부가 민간 대신 사업을 해주는 방식이어서는 안 됩니다.

Portsmouth에서는 연방부지와 환경정화, 인허가가 공공의 역할이었습니다.

TFP에서는 초기 시장형성 위험 일부를 정부가 한시적으로 맡았습니다.

Virginia에서는 오히려 데이터센터 때문에 발생하는 장기 전력망 비용을 민간에게 다시 돌려주고 있습니다.

방향은 서로 다르지만 원리는 같습니다.

공공만 해결할 수 있는 문제는 공공이 해결하고,
민간사업에서 발생하는 기술·건설·운영·상업위험과 투자비는 가급적 민간에 남깁니다.

저는 앞으로의 P3에서는 “얼마의 민간자본을 유치했느냐”보다 이 질문이 더 중요해질 것이라고 봅니다.

정부가 개입해서 정확히 어떤 Market Failure(시장실패)를 없앴고, 그 대가로 국민은 얼마를 부담했는가?

이 두 숫자가 같이 나와야 정책의 성적표를 만들 수 있습니다.

그런데 미래교통으로 가면 이 질문이 훨씬 어려워집니다

데이터센터에는 이미 고객이 있습니다. 전력에는 이미 시장이 있습니다.

정부가 병목을 풀어주면 민간이 이후의 수익을 스스로 만들 가능성이 비교적 분명합니다.

그런데 UAM(Urban Air Mobility·도심항공교통), Autonomous Vehicle(자율주행차), Hyperloop(하이퍼루프) 같은 미래교통은 한 단계 더 불확실합니다. 기술 자체가 아직 변하고 있고 이용수요도 충분히 검증되지 않았습니다.

Vertiport(버티포트·수직이착륙장)는 먼저 지어야 할까요, UAM 이용자가 충분히 생긴 뒤 지어야 할까요?

자율주행차를 위해 도로와 통신 인프라를 정부가 먼저 깔아야 할까요, 자동차 제작사가 기술적으로 해결하도록 두는 것이 맞을까요?

Hyperloop처럼 아직 상업적 운영모델이 확립되지 않은 교통수단에 20~30년짜리 Project Finance를 붙이는 것이 가능할까요?

데이터센터에서는 민간이 투자하려는데 정부가 병목을 풀어주는 문제였다면 미래교통에서는 한 단계 더 나갑니다.

시장 자체가 아직 만들어지지 않은 상황에서 누가 먼저 인프라 위험을 부담해야 하는가라는 문제가 생깁니다.

그래서 #68에서는 이 질문을 따라가보려고 합니다.

핵심 참고자료

Lawrence Berkeley National Laboratory, LBNL(로런스버클리국립연구소) LBNL 2025 Data Center Energy Usage Update
U.S. Department of Energy, DOE(미국 에너지부) DOE Portsmouth AI Partnership
DOE Portsmouth Fact Sheet DOE Portsmouth Fact Sheet
DOE NEPA 자료 DOE Portsmouth 189-acre Lease 자료
42 U.S.C. §7256 / DOE Portsmouth Lease DOE Portsmouth Lease Legal Basis
Permitting Council(미 연방 인허가개선위원회) PORTS Technology Campus FAST-41
42 U.S.C. §4370m 42 U.S.C. §4370m
42 U.S.C. §18713 – Transmission Facilitation Program 42 U.S.C. §18713
U.S. Department of Energy – TFP DOE Transmission Facilitation Program
DOE TFP Selections DOE TFP Selections
Virginia JLARC(버지니아 주의회 감사·정책평가기관) JLARC Data Centers in Virginia
Virginia State Corporation Commission, SCC(버지니아주 공공서비스규제위원회) Virginia SCC Data Center Rules
Virginia Department of Taxation(버지니아주 세무국) Virginia Data Center Tax Exemption Report 2026
AEP Ohio(AEP 오하이오) AEP Ohio Data Center Tariff Update
Washington Post-Schar School Poll (The Washington Post)
Virginia Mercury(버지니아 공공정책 전문언론) (Virginia Mercury)
Reuters (Reuters)