Both Sides Keep Losing on Wall Street
One reform bill. Too prescriptive AND not far enough. That’s not a contradiction — it’s the pattern.
When a conservative economist writes that Elizabeth Warren’s Wall Street reform bill is “chock full of excessive interventions and yet, ultimately, does not go nearly far enough,” something interesting is happening.
That’s Oren Cass — former Mitt Romney advisor, director of American Compass, free-market conservative — analyzing the Stop Wall Street Looting Act in his report Confronting Coin-Flip Capitalism. And he’s right. Not partially right. Structurally right.
Progressives see the problem clearly: private equity loads companies with debt, extracts value through fees and dividends, then walks away when the leveraged buyout collapses — leaving workers unemployed and communities hollowed out. The Stop Wall Street Looting Act (SWSLA) tries to fix this. It targets PE firms specifically, restricts dividends for two years after acquisition, limits interest deductions when debt-to-equity ratios exceed certain thresholds, and creates new liability rules.
And a conservative economist says it’s both too much and not enough.
Before we get to that question, let’s be clear about what’s actually at stake. When firms are heavily leveraged, downturns don’t just hurt investors. Leveraged companies cut jobs faster, freeze investment harder, and amplify recessions across entire communities. This isn’t incidental damage — it’s structurally incentivized. The system is designed to produce it.
How does that happen? How does a reform effort end up simultaneously over-engineered and insufficient?
It happens every time. Healthcare. Campaign finance. Banking regulation after 2008. The same pattern, different domain.
This isn’t a coincidence. It’s a structural problem — and it has a structural solution.
The Two Ways Reform Always Fails
The SWSLA suffers from two failure modes that appear contradictory but actually cause each other.
Failure Mode One: Too Prescriptive
The bill doesn’t say “protect workers from predatory financial practices.” It locks specific engineering mechanisms into law. Debt-to-equity ratios. Two-year waiting periods for dividends. Rules that apply only to PE firms, specifically defined.
The problem with this isn’t the intent. The problem is that financial firms don’t sit still. They restructure to avoid the label. As Cass observes, the main effect of narrow PE-specific rules would likely be to encourage firms to recast their activities in other forms — same economic substance, different legal wrapper. The rules become a maze to navigate, not a barrier to predation.
Arbitrary thresholds make it worse. Section 204 of SWSLA limits interest deductibility only when debt-to-equity ratios exceed 1. That specific number creates a bright line to game. Structure your deal to land at 0.99 and you’re fine. The rule adds complexity, distorts incentives around that threshold, and doesn’t address the root cause: the tax code subsidizes all debt, not just PE debt.
The two-year dividend restriction has the same problem. Cass is blunt: the restriction is hard to see as accomplishing anything beyond prompting PE firms to build up cash reserves inside their acquired companies and release them the moment the restriction expires. Amateur engineering creates workarounds. Every time.
Failure Mode Two: Didn’t Go Far Enough
While SWSLA over-engineers solutions for a narrow slice of the problem, the actual structural problems are much broader and go completely untouched.
The tax code still subsidizes debt across the entire economy. Interest payments are tax-deductible. Dividends are not. That single asymmetry tilts the system. CBO modeling of effective marginal tax rates on capital income found that debt-financed investment can carry a negative effective rate (around –6%) while equity-financed investment faces a strongly positive rate (around 38%). These aren’t statutory rates — they reflect how the deductibility rules change the actual return on investment. That incentive structure doesn’t just affect PE — it shapes how every major corporation in America is financed. SWSLA focuses narrowly on PE and leaves this foundational distortion in place.
Bankruptcy law still treats workers as unsecured creditors in any bankruptcy, not just PE-driven ones. Workers lose jobs without priority claims on assets in any industry, any context, any type of ownership structure. SWSLA tries to create PE-specific worker protections while the broader bankruptcy framework that creates worker vulnerability goes unreformed.
Regulatory capture — the process by which financial sector interests come to dominate the regulators meant to oversee them — is untouched entirely. Pay-to-play campaign finance, the revolving door between regulators and industry, the structural incentives that make capture inevitable: SWSLA doesn’t address any of it.
Both failure modes — over-prescriptive in the narrow lanes it addresses, and fundamentally insufficient at the structural level — exist in the same bill. That’s not a drafting error. That’s what happens when policy tries to do architecture’s job.
The Missing Distinction
We have confused two different kinds of work, and the confusion is destroying our ability to solve hard problems.
Policy is democratic decision-making about goals and values.
Should we protect workers from financial strategies that privatize gains and socialize losses? Should we prioritize long-term capital investment over short-term extraction? How do we balance market freedom with economic stability? These are normative questions. They have no objectively correct answer. They require representation, debate, and democratic legitimacy. Different people have legitimate disagreements about them, and the answers should evolve as society changes.
Democracy is the right tool for this. Compromise and representation are features here, not bugs. A Congress that debates and votes on whether to protect workers from predatory finance is doing exactly what it should do.
Architecture is professional design of institutions and mechanisms.
How do you actually protect workers from predatory finance in a way that can’t be gamed? What bankruptcy structure achieves that goal? What tax treatment creates the right incentives across the entire economy, not just for one class of actors? What accountability mechanisms prevent regulatory capture? These are technical questions. They have better and worse answers. They require systems thinking, understanding of incentives, knowledge of how complex institutions fail.
You don’t democratically compromise on whether the bridge supports the load. You engineer it to spec, then democratically decide which bridges to build.
When Congress tries to answer both questions simultaneously — setting the goal AND designing the mechanism — you get what you always get: compromises that water down the mechanism while political pressure narrows the scope. The architecture gets corrupted in the design process. What emerges is half-measures that don’t solve the structural problem and rigid rules that create workarounds.
This isn’t a critique of Congress. It’s a recognition that these are two different kinds of problems requiring two different kinds of work.
What It Should Look Like
Here’s what proper separation of concerns produces — and it turns out a conservative and a progressive can agree on most of it once you strip away the amateur engineering.
What Congress should provide: A clear policy mandate. “We want workers and communities protected from financial strategies that privatize gains and socialize losses. We want capital allocated to productive investment, not extraction.” That’s it. Democratic legitimacy for the goal. No debt-to-equity thresholds. No two-year timelines. No narrow definitions that financial engineers will immediately route around.
What professional governance architecture should design:
Bankruptcy reform with real teeth. Workers should receive priority claims — something equivalent to six months of wages — ahead of financial creditors in any bankruptcy. Not just PE-driven bankruptcies. Any bankruptcy. This applies broadly, makes creditors wary of excessive leverage across all ownership structures, and actually protects the workers it’s meant to protect. This isn’t ideological — it’s how you make the stated goal achievable.
Tax treatment that doesn’t subsidize leverage. Eliminate the interest deduction advantage that makes debt financing artificially attractive compared to equity financing across the entire economy. This addresses the root cause rather than policing symptoms. If you remove the subsidy driving overleveraging, you change incentives for every firm, not just the ones clever enough to call themselves PE.
Transparency requirements with real accountability. Private funds should register benchmarks before they’re deployed, not after. Fees, performance, and returns should be publicly reported. This doesn’t prescribe outcomes — it creates information infrastructure so markets and regulators can actually function.
Reduce financial engineering’s artificial advantages. Return stock buyback rules to something closer to what existed before 1982. Apply financial transaction taxes that make high-frequency trading pay the same costs that other economic activity pays. These aren’t punitive — they’re adjustments that remove artificial advantages currently baked into the system.
These proposals address root causes rather than symptoms. They apply broadly rather than narrowly, making them harder to route around. They can be adjusted based on outcomes rather than locked into political fights about whether to revisit the original legislation.
Notice something: these proposals come largely from Cass — the conservative. He’s not defending Wall Street. He’s not arguing for less regulation. He’s proposing structural architecture that actually works. A progressive and a conservative, applying structural thinking to the same problem, arriving at largely compatible proposals.
That’s what happens when you separate “what goals to pursue” from “how to achieve them reliably.”
Why This Keeps Happening
This isn’t a financial regulation problem. It’s a governance architecture problem that shows up everywhere financial regulation is just the current example.
The pattern is identical in healthcare. The Affordable Care Act had a clear policy goal — expand coverage — and tried to achieve it by writing specific mechanisms into legislation: mandates with specific penalty structures, exchanges with specific designs, Medicaid expansion with specific thresholds. Every one of those mechanisms became a political target. Every stakeholder got carve-outs. Amateur engineering left gaps that were immediately exploited. The architecture was negotiated by people whose job is building coalitions, not designing robust systems.
The pattern appears in banking regulation after 2008. Dodd-Frank’s policy intent was clear: prevent another financial crisis. But implementation was delegated to regulatory agencies whose capture by the financial sector it was designed to prevent. The architecture didn’t address the capture problem, so the architecture got captured.
It appears in infrastructure, in education, in every domain where we ask democratic policy-making to simultaneously design the implementation. The failure modes are always the same: too prescriptive where political fights force specificity, not far enough where structural root causes go unaddressed.
This is a predictable outcome of a governance architecture that doesn’t separate the two kinds of work.
The Governance Design Agency
What I would build is an institution that makes this separation explicit and durable. I’ve laid out the full design in The Governance Design Agency and its institutional and legal framework — the short version is this:
The Governance Design Agency — modeled on how the Federal Reserve handles monetary policy — would provide the professional architecture layer that democratic policy-making currently has to do badly or not at all. Congress sets the mandate: protect workers from predatory finance, ensure capital allocates to productive investment. The GDA designs the institutional mechanisms that actually achieve it: the bankruptcy structure, the tax treatment, the transparency requirements, the accountability systems.
The GDA isn’t making policy. It’s doing engineering. It’s designing the bridge to the spec Congress provides.
Democratic accountability runs in both directions. Congress confirms GDA leadership. Congress can reject proposed architecture and send it back. Congress retains the power to change the mandate when democratic preferences evolve. What Congress doesn’t do is also design the engineering — because that’s not what democratic deliberation is designed to do well.
The obvious objection: won’t the GDA just get captured by the financial sector faster than Congress does? That’s a real risk, and it’s why the design matters. Unlike regulatory agencies that depend on industry expertise and rotate staff with the firms they oversee, the GDA’s mandate is architectural — it designs systems for all sectors, not just finance, which breaks the single-industry capture dynamic. Its output is publicly reviewable before Congress approves it. That’s not a perfect firewall, but it’s structurally better than what we have now, where capture is baked into the current architecture with no accountability mechanism at all.
This is why we have a Federal Reserve Chair for monetary policy rather than asking Congress to set interest rates through legislation. Not because monetary policy is unimportant — because professional design of monetary mechanisms produces better outcomes than political compromise on the mechanisms themselves.
Governance design is no different. We just haven’t built the institution yet.
The Broader Point
When Oren Cass — a free-market conservative — and Elizabeth Warren — a progressive champion of worker protection — both look at Wall Street and see a structural problem, and then produce reform efforts that fail in opposite directions, the failure isn’t partisan. It’s architectural.
Both sides are trying to solve a real problem through a process that systematically produces the wrong kind of solution. Too prescriptive in the mechanisms it does address. Not far enough in the structural causes it leaves untouched. Amateur engineering that financial sophistication routes around within years.
The solution isn’t better Democratic or Republican proposals. It’s a governance system that does the two kinds of work separately: democratic legitimacy for policy goals, professional design for institutional mechanisms.
Financial reform keeps failing because we keep asking policy-making to do architecture’s work.
We don’t lack intelligence or capital or political will. We lack correct incentive architecture. That’s a design problem. Design problems have design solutions.
That’s not a people problem. That’s a design flaw — one we know how to fix.
Share the loading symbol. Demand professional governance architecture. Let’s build this.