When the Watchdog Gets Defunded: How Aerospace Safety Science Could Protect Your Wallet
- the Institute
- Nov 12
- 15 min read
A community-driven accountability framework inspired by airplane safety principles might be the answer to the CFPB’s demise.
November 12, 2025
The federal agency that protects Americans from predatory lending, hidden fees, and financial fraud is running out of money. By early 2026, the Consumer Financial Protection Bureau — the watchdog created after the 2008 financial crisis — is expected to exhaust its funding after a court ruling declared its financing mechanism unconstitutional. Without congressional action, oversight of the $18 trillion consumer debt market could fade to near zero.
That crisis has left researchers, policymakers, and communities asking: What if the next era of consumer protection isn’t built in Washington at all? What if it’s built locally — through transparent systems, civic participation, and safety methodologies that have already proven themselves in one of the most unforgiving industries on earth: aerospace?
Here’s the puzzle that sparked NTARI’s latest experiment. Credit bureaus have spent decades tracking the behavior of individuals. Yet no system tracks how financial institutions treat people. Banks can study your credit history before offering a loan — but you can’t study their record before trusting them with your savings.
The Leveson-Based Trade Assessment Scale (LBTAS) proposes to change that imbalance. Developed by the Network Theory Applied Research Institute (NTARI), LBTAS is an open-source research tool — not a rating website or app — designed to test whether safety engineering principles can make financial systems more transparent and accountable.
Think of it as an experimental framework for public oversight: a set of algorithms, documentation, and data models that communities, credit unions, and regulators can use to measure and visualize trustworthiness in trade relationships. The code exists today on GitHub under an AGPL-3.0 license — but the system’s power depends not on software adoption alone, but on how communities choose to use it.
The MIT Professor Who Made Software Safe Enough to Fly
Nancy Leveson never set out to reform finance. She built the field of system safety engineering, creating methods that make airplane control software, nuclear plants, and spacecraft reliable enough for human lives to depend on.
Her insight was simple but revolutionary: most catastrophic failures aren’t caused by single mistakes — they arise from interactions within complex systems. Preventing them requires a transparent map of those interactions, continuous feedback, and explicit documentation of what each component is supposed to prevent.
In aerospace, this became the System-Theoretic Process Analysis (STPA) — a structured way to uncover hidden risks and design safeguards before tragedy strikes.
When NTARI researchers studied Leveson’s work, they saw a striking parallel between flight safety and financial safety. Both rely on trust in systems that are too complex for any individual to monitor. Both can collapse when feedback loops break. And both demand tools that reveal — rather than obscure — patterns of failure.
LBTAS began as an experiment:
Could the same rigor used to keep airplanes in the sky be used to keep communities financially safe on the ground?
Why 5-Star Ratings Don’t Work (And Never Did)
To understand why a system like LBTAS might be needed, it helps to see what’s wrong with the rating tools we already have.
Most people assume the 5-star review system—used on everything from restaurants to real estate apps—is a reliable way to gauge quality. In truth, it was never designed for anything serious.
The familiar “star” rating was invented in 1958 by Mobil Travel Guide to market hotels along new interstate highways. It was an advertising shorthand, not a scientific metric. A “five-star” hotel simply met the expectations of a traveling executive; a “three-star” did not. The model was one-directional: businesses were rated by consumers, not collaboratively assessed by any shared standard.
Decades later, that same convenience-based gimmick is still the default feedback mechanism for online life. We use it to judge doctors, financial advisors, and sometimes even public institutions. The simplicity hides the danger: stars compress complex experiences into meaningless averages, easily manipulated by marketing and mood.
When NTARI began exploring community-led accountability systems, the team found that nearly every modern rating platform suffers from four structural flaws:
Meaningless granularity. What separates three stars from four? No one can say. The difference is emotional, not technical.
One-way power. Customers can rate businesses, but businesses can’t safely or fairly rate customers. There’s no mechanism to establish mutual accountability.
Dependence on commentary. Star ratings alone convey almost nothing. To extract real insight, users must read subjective comments of uneven quality.
Susceptibility to manipulation. Public relations teams, bots, and legal pressure easily distort results. Negative reviews can be buried. Positive reviews can be bought.
The result: a noisy feedback economy that rewards popularity over performance.
Now imagine applying that model to financial services, where the stakes aren’t bad pizza or slow shipping, but people’s homes, retirements, and small-business capital. A 4.2-star lender could still be predatory. A low-rated borrower could simply be underbanked. The system tells us nothing useful about risk, fairness, or integrity.
This is the void LBTAS was built to study—not to replace consumer protection agencies, but to prototype a new layer of measurement that makes institutional behavior visible.
The LBTAS Difference: Measuring Safety, Not Sentiment
In the aviation world, every component—from wing rivets to control software—is rated on explicit, testable criteria. Engineers don’t “like” or “dislike” parts; they assess whether each one meets safety specifications.
LBTAS brings that discipline to trade relationships. Its 6-point Leveson Scale is not a popularity contest but a structured taxonomy of outcomes, ranging from “harmful” to “delightful.” Each level describes observable system behavior, not personal opinion:
+4 — DELIGHT: The system anticipates and prevents harm before it occurs.
+3 — NO NEGATIVE CONSEQUENCES: The interaction is designed to eliminate preventable loss or confusion.
+2 — BASIC SATISFACTION: It meets reasonable expectations and social standards.
+1 — BASIC PROMISE: It delivers exactly what was agreed—no better, no worse.
0 — CYNICAL SATISFACTION: The transaction barely clears the minimum bar; harm avoided, but trust eroded.
–1 — NO TRUST: Evidence of exploitation, malice, or systemic neglect.
Each category is derived from safety engineering logic: preventing failure before it cascades. The goal is not to punish but to map patterns of reliability and risk.
When NTARI researchers adapted this framework, they were careful not to assume that “users” already exist or that all evaluations would happen online. Instead, pilot programs focus on community-scale implementation—credit unions, housing cooperatives, and local watchdog groups testing whether structured feedback could strengthen financial transparency.
In practice, that might mean:
A regional cooperative using LBTAS forms to evaluate lenders serving their members.
A journalists’ consortium analyzing public records with the six-point model to classify corporate behavior.
A state regulator adapting the methodology for early-warning indicators of consumer harm.
In other words, the system is not a platform but a protocol—a set of shared definitions and open tools anyone can adapt.
Why Bidirectional Accountability Matters
Every contract, whether between a borrower and a bank or a buyer and a seller, is a relationship. Failures rarely happen in isolation; they happen when communication breaks down on both sides.
Traditional consumer rating systems are one-directional: the public judges companies, and companies react defensively. Credit bureaus invert the asymmetry but keep it one-sided: companies judge individuals, who have no equal access to data.
LBTAS experiments with a different premise: What if both sides evaluated each transaction under identical rules?
In the research model:
A lender rates the borrower’s reliability, clarity, and conduct.
The borrower rates the lender’s transparency, fairness, and responsiveness.
Both records are linked to the same verified transaction, stored as parallel data.
Neither side owns the record; both have visibility into their own behavior as part of a broader system.
This doesn’t require a single online “platform.” It can happen through:
Local credit unions collecting anonymized data.
Cooperative accounting software integrating a transparent ledger.
Paper-based community surveys digitized through NTARI’s open schema.
Bidirectional accountability is powerful because it prevents distortion from either direction. It disincentivizes revenge ratings and discourages institutional opacity. Over time, aggregate data could reveal recurring risk patterns—information that could inform regulators, journalists, or community decision-makers.
The principle is simple: shared data produces shared responsibility.
Symmetry and the End of the Information Black Box
Economists call it information asymmetry—the condition where one side of a transaction knows far more than the other. It’s what allows used-car dealers to hide defects and financial institutions to conceal risk. It’s a root cause of corruption, inefficiency, and distrust.
Credit bureaus were designed to reduce asymmetry for lenders: they can see your payment history and gauge your reliability. But the system stops there. Consumers see only marketing narratives, not institutional performance metrics.
LBTAS is the mirror image. It aims to document how financial entities behave, not just how individuals behave toward them.
In its mature form, a public LBTAS dataset might show:
Which lenders maintain high transparency scores.
Which institutions exhibit recurring patterns of “cynical satisfaction” or “no trust.”
How those patterns correlate with enforcement actions or consumer complaints.
It’s not about punishing companies with low scores—it’s about enabling informed decisions and encouraging systemic improvement.
To avoid digital centralization, NTARI’s research stresses federated data sharing: independent groups collect ratings and publish them to compatible ledgers rather than to a single centralized website. This allows local control and protects against political or corporate capture.
The Public Credit Bureau for Corporate Behavior
When the Consumer Financial Protection Bureau (CFPB) was created in 2011, it was meant to ensure that Americans could trust their banks, lenders, and credit card companies. For more than a decade it did — returning over $20 billion to consumers and enforcing standards of fairness that the private sector had ignored.
But as federal oversight weakens, the same structural problem that plagued finance before 2008 reappears: no feedback system for corporate behavior.
Credit bureaus track your reliability. Regulators audit the industry. Yet there is no public equivalent of a “credit score for companies.” When a financial institution mistreats its customers, patterns remain buried in lawsuits and complaint databases too fragmented for ordinary citizens to interpret.
LBTAS addresses that blind spot. It is not a regulator, and it is not a crowd-review platform. It is a framework for documenting institutional behavior — using the same kind of structured, safety-oriented evaluation that has guided aerospace for decades.
In this sense, it functions like a public credit bureau for trust.Instead of measuring how reliably people pay debts, it measures how reliably organizations honor obligations.
How Credit Bureaus and LBTAS Differ
Traditional Credit Bureaus | LBTAS (Prototype Framework) |
Track consumers for lenders | Track institutions for communities |
Operate as private corporations | Operate as public, open-source commons |
Proprietary algorithms | Transparent methodology |
Centralized data silos | Federated or locally governed data pools |
Designed to minimize lender risk | Designed to minimize community harm |
LBTAS doesn’t compete with credit bureaus — it complements them. The two form a conceptual pair: credit measures solvency; trust measures safety.
Just as credit data helps lenders avoid losses, LBTAS-style data could help communities avoid exploitation. Together, they create symmetry — both sides of the financial relationship illuminated by data, not advertising.
What Real Implementation Could Look Like
Because LBTAS is an open-source research framework, its application can vary by context. NTARI’s current focus is testing the methodology, not enforcing participation. The idea is to create a foundation that any community can build on.
Here are four pilot directions under study:
Credit Union Consortia – Local and regional credit unions could use the LBTAS schema to track their own service outcomes, publishing anonymized data for public accountability.
Consumer Advocacy Networks – State-level nonprofit watchdogs could use the framework to organize verified reports of customer experience into structured metrics, rather than freeform complaint narratives.
Academic & Civic Research – Universities and public-interest technologists could analyze LBTAS data to study correlations between transparency, customer satisfaction, and institutional integrity.
Regulatory Collaboration – State attorneys general or oversight boards could use LBTAS data as early-warning indicators of systemic risk or predatory patterns.
In all these cases, the data would remain decentralized—hosted by those who collect it, not pooled in a single corporate or government server.
This approach protects independence and ensures that no political decision can “defund” transparency itself. The software, code, and methodology are freely available. What matters is how people choose to use them.
Why a Safety-Based Model Works for Finance
In aerospace, safety systems operate on two assumptions:
Complex systems fail in predictable ways when feedback is weak.
Transparent feedback loops prevent small errors from becoming catastrophes.
Financial systems obey the same laws of complexity. Most scandals—subprime mortgages, overdraft fees, data breaches—begin as local failures of information sharing. Customers don’t see the full cost. Employees lack oversight. Regulators learn too late.
A Leveson-style approach treats those failures not as moral lapses but as systemic signal failures.Instead of waiting for collapse, it asks: Where did feedback stop flowing?
By measuring transactions and service outcomes through standardized categories, LBTAS gives communities an early-warning system. Patterns of “cynical satisfaction” or “no trust” can indicate weak safety controls before harm spreads.
This is the same reasoning that made aviation one of the safest industries in history: every incident becomes a data point for collective learning.
Why This Matters as CFPB Funding Collapses
The timing is no coincidence. The LBTAS research project accelerated in late 2025 after courts ruled that the CFPB’s funding structure violated constitutional separation of powers. Without a congressional fix, the agency could lose the ability to function by 2026.
That scenario would leave consumer protection largely to state agencies—many of which are underfunded—or to private litigation, which is slow and inaccessible to most people.
NTARI’s position is not that LBTAS can replace regulation. Rather, it offers an independent layer of transparency that can survive when regulatory funding fails.
Where the CFPB depended on centralized authority, LBTAS depends on distributed participation. Where the bureau enforces rules, the framework exposes behavior. Both can coexist; one should not be hostage to politics.
In that sense, LBTAS is regulatory infrastructure built from the bottom up. It equips states, cooperatives, and citizens with tools to monitor fairness and integrity—even if Washington can’t.
Limitations and Cautions
NTARI’s researchers stress that the framework’s success depends on context and collaboration, not on automation. There are important limits to acknowledge:
LBTAS cannot enforce laws. It can reveal problems but not compel restitution.
Participation must be verified. Data integrity relies on verified transactions and community governance, not anonymous submissions.
Local inequities persist. Offline communities and those without digital access must be included through cooperative or paper-based methods.
Human judgment remains essential. Quantitative scoring must support, not replace, professional investigation and public oversight.
In other words, LBTAS is not a silver bullet. It’s a toolkit for civic intelligence—a way to see systems more clearly, not to substitute for justice.
Hybrid Protection: From Federal Oversight to Community Transparency
If the CFPB disappears, consumer safety will depend on how well states, journalists, and citizens coordinate.
A pragmatic path forward combines three layers:
State Regulation: State attorneys general and financial regulators enforce laws and handle penalties.
Community Accountability: Local cooperatives, nonprofits, and unions use frameworks like LBTAS to collect transparent data.
Academic & Technological Support: Open research groups maintain shared infrastructure, validation tools, and data standards.
Together, these layers can replace centralized fragility with distributed resilience.
A defunded federal agency weakens oversight, but it also opens space for innovation. LBTAS is one possible blueprint: transparent, open, and resistant to capture.
Its goal isn’t to privatize regulation — it’s to democratize oversight.
Visit the LBTAS repository on Github
Support the Institute @NTARI.org
Research LBTAS WIth the Institute on Slack
The Vision: Credit Scores + Trust Scores
In a functioning economy, information should move in both directions.Right now, it doesn’t.
Every American adult is subject to the constant scrutiny of credit bureaus. Three private corporations—Equifax, Experian, and TransUnion—monitor virtually every loan, missed payment, and credit application. Their data determines whether you get a mortgage, rent an apartment, or sometimes even secure a job. But there’s no reciprocal mechanism. Consumers can’t easily access data about whether a lender behaves ethically, processes payments fairly, or resolves disputes in good faith.
That imbalance—information asymmetry—isn’t a flaw. It’s a design choice inherited from an era when banks were trusted by default. LBTAS suggests that it’s time to build the other half of the system: a public “trust score” that evaluates institutions, not individuals.
Unlike FICO or Equifax, however, this score wouldn’t be an opaque number computed behind closed doors. It would be derived from open, auditable criteria, the same way safety engineers document every control and hazard in a flight system.
The point isn’t to shame companies. It’s to provide symmetry—to allow both sides of the financial relationship to make informed decisions. Before you apply for a mortgage, you should be able to see a lender’s history of transparency and fairness. Before a lender approves you, they should see your record of reliability. That’s not a radical demand—it’s a completion of the feedback loop that already governs every other trust-based system we rely on.
From Abstract Code to Real-World Practice
While the LBTAS codebase exists today on GitHub under the AGPL-3.0 license, its purpose is educational and experimental. NTARI maintains it as a public commons, a reference implementation showing how the methodology could function in practice.
Communities interested in piloting it can already do so—on their own terms. That might look like:
A community development finance organization collecting structured feedback from borrowers after loan completion, using the six-point Leveson scale.
A local government testing LBTAS forms in procurement, rating contractor reliability and ethical conduct.
A university research lab using the open dataset schema to study correlations between transparency and financial resilience.
A mutual-aid cooperative maintaining its own small LBTAS ledger for internal accountability.
None of these require a single centralized “platform.” They rely on shared principles, documented methods, and open tools that others can adapt or improve.
This is what it means for governance to become infrastructural rather than institutional. The work doesn’t depend on one organization’s survival; it depends on distributed participation and shared literacy.
Democratizing Oversight Through Open Methodology
The deeper significance of LBTAS isn’t the software—it’s the method.
For most of modern history, transparency has been mediated by institutions. We depend on regulators to interpret data, journalists to investigate, and corporations to self-report. Those layers remain essential, but they are slow to evolve and easy to capture.
Open methodologies like LBTAS offer a different model: collective verification.If the underlying system of measurement is public, anyone—from state investigators to cooperative members—can apply it. The community itself becomes capable of producing structured accountability data.
This represents a shift from governance by permission to governance by protocol.
It doesn’t eliminate the need for regulation; it simply widens the circle of those who can practice it. In this vision, transparency isn’t a report you wait for—it’s an ongoing public process, accessible to anyone with the curiosity to participate.
Building Ethical Markets Through Visibility
Markets rely on trust, but trust without visibility decays into marketing.That’s why every stable market in history has evolved some form of standardized reporting: accounting for merchants, ratings for credit, flight logs for airlines.
What’s missing in finance is an equivalent ethical metric—a way to measure not just solvency, but integrity.
Imagine a future where transparency itself becomes a competitive advantage:
Local banks and credit unions publish their LBTAS-derived trust profiles as a sign of confidence.
Fintech startups integrate the methodology to demonstrate accountability to users and regulators alike.
Consumers compare institutions not just by rates and fees, but by fairness, responsiveness, and reliability.
State regulators use LBTAS data to prioritize enforcement, focusing on institutions showing consistent “no trust” ratings.
This is how markets evolve when visibility becomes systemic. Bad behavior doesn’t vanish overnight—but it becomes impossible to hide.
The result is a self-correcting ecosystem, where ethical conduct is rewarded and exploitation becomes economically unsustainable.
Beyond Platforms: A Network of Practice
A recurring misconception about LBTAS is that it’s a digital platform waiting for users. In truth, it’s more like a public laboratory—a set of methods and reference code designed to evolve through experimentation.
NTARI’s current focus is not scaling users but cultivating a network of practice:
Financial cooperatives testing governance models.
Software developers building verification tools.
Policy researchers refining scoring logic.
Educators translating the methodology for civic audiences.
Every implementation informs the next, creating a distributed learning process.
This approach mirrors how aerospace safety matured. The first flight data recorders were simple tape boxes, not global systems. It took decades of iteration and open investigation before “black boxes” became industry standard. LBTAS belongs to that same lineage: open, empirical, evolving. The code is the prototype; the community is the system.
From Bureaucracy to Systemic Literacy
The broader lesson of the CFPB’s collapse is not about one agency—it’s about dependence.When consumer protection is centralized in a single office, its survival depends on political goodwill. When transparency becomes a shared civic skill, it becomes permanent.
LBTAS points toward that future. It’s not “anti-government”; it’s post-fragile.
The goal is not to replace regulators, but to equip society with the literacy to understand its own systems.When people can see patterns of risk in real time, political action becomes informed rather than reactive.
In that sense, the real innovation isn’t technological—it’s cultural. It’s the reintroduction of engineering discipline into public life: observe, record, analyze, share, improve.
That’s the heart of safety science, whether you’re designing a flight-control system or a fair financial market.
A Note on Realism
For all its potential, LBTAS is not a cure-all. It will not stop financial misconduct by itself, nor guarantee honest participation.It can, however, lower the cost of truth.
By structuring how experiences are documented, it makes misbehavior legible sooner—before it becomes systemic. It turns anecdotes into data and data into prevention.
If aviation safety proved anything, it’s that transparency doesn’t eliminate error; it eliminates surprise.And in finance, surprise is often what makes harm catastrophic.
From Airplanes to Accountability
Nancy Leveson’s systems thinking changed how engineers build machines that must never fail. Her insight—that safety is not a product but a property of relationships—translates powerfully to the social systems that manage risk in daily life.
NTARI’s adaptation of that insight is a quiet but radical proposal: to treat economic fairness as a safety problem, not a moral one.Because when feedback disappears, harm is inevitable—whether in an aircraft or a credit market.
LBTAS is one framework among many that could help rebuild that feedback loop.Its value lies not in the code, but in the example it sets: that accountability can be open, decentralized, and unkillable. As public institutions face defunding and capture, such frameworks may become the last line of defense for democratic transparency.
The principle is timeless:
Make failures visible, and systems will improve.
The rest is simply engineering.
Learn More
LBTAS source code: github.com/NTARI-RAND/Leveson-Based-Trade-Assessment-Scale
NTARI: ntari.org
License: CC BY-SA 4.0 — share freely with attribution.
Disclosure: This article discusses open-source research software maintained by the Network Theory Applied Research Institute. The author has no financial interest in LBTAS adoption.
Comments