Regulatory Approaches — Cross-Cutting Comparison

This note compares the major paradigms for designing and implementing regulation — what economists, lawyers, and policy designers actually pick from when they decide how a given social problem should be regulated. Command-and-control vs market-based vs information-based; notice-and-comment rulemaking vs negotiated rulemaking vs case-by-case adjudication; principles-based vs rules-based supervision; federal preemption vs state experimentation; ex ante licensing vs ex post enforcement; voluntary standards vs mandatory regulation; administrative-law-judge adjudication vs Article III courts.

The frame: regulation is a design space, not a binary. Every domain (environment, securities, food, drugs, labor, consumer protection, financial services, AI) involves choosing instruments from this menu — and the choice has dramatically different costs, distributional consequences, and political durability.

See also

• administrative-law
• employment-and-environmental-law
• securities-regulation-and-ma-practice
• securities-regulation-deep
• eu-competition-and-regulation
• antitrust-and-competition-deep
• constitutional-law
• civil-procedure-and-evidence

1. The taxonomy

INSTRUMENT TYPE                        ENFORCEMENT VEHICLE
  Command-and-control                    Administrative agency (APA rulemaking)
  Market-based                            Article III court
  Information-based                       Agency adjudication / ALJ
  Liability-based (tort, private right)   Self-regulatory organization (FINRA, NAIC)
                                          Civil enforcement by AG / DOJ
                                          Criminal prosecution

PROCEDURAL CHOICE                      INSTITUTIONAL ALLOCATION
  Notice-and-comment (APA §553)          Federal preemption (uniform rules)
  Formal rulemaking (APA §556-557)        State experimentation (federalism dividend)
  Negotiated rulemaking (5 U.S.C. §561)   International harmonization (BIS, IOSCO, OECD)
  Adjudication-by-decision               
  Guidance / interpretive rules          REGULATORY POSTURE
  Direct final rules                      Ex ante licensing
                                          Ex post enforcement
                                          Mandatory standards
                                          Voluntary / safe-harbor standards
                                          Principles-based supervision
                                          Rules-based prescription

2. The six instrument families

Family	Mechanism	Canonical example	When it fits
Command-and-control	Direct rule: “Thou shalt not exceed X”	EPA NAAQS, OSHA PELs, FAA airworthiness	When externality is uniform; when measurement is straightforward; political legibility matters
Market-based	Price the externality	EU ETS, RGGI, CA cap-and-trade, Pigouvian carbon tax	When externality is heterogeneous; when efficiency gains from least-cost abatement are large
Information-based	Mandatory disclosure	Prop 65, SEC 10-K, FDA labels, Nutrition Facts	When information asymmetry is the core problem; when private market can act if informed
Liability	Tort, private right of action	CERCLA, products liability, securities Rule 10b-5	When ex ante regulation is infeasible; when private enforcement is efficient
Standard-setting	Coordinated technical standards	NIST CSF, IEEE, ISO/IEC, IFRS, GAAP	When network effects dominate; when industry expertise needed
Structural / institutional	Reorganize industry; mandate independence	Glass-Steagall, FERC unbundling, Sarbanes-Oxley	When conflicts of interest are systemic; when ex post fixes have failed

3. Command-and-control (CAC)

Mechanism

The regulator specifies what must (or must not) be done. Forms:

• Technology standards — “All facilities must install X technology” (BAT, BACT, MACT under Clean Air Act).
• Performance standards — “Emissions shall not exceed X kg/yr” (NAAQS, NSPS).
• Process standards — “You must follow this HACCP process” (FDA food safety).
• Outright prohibition — “Asbestos in new products is banned” (TSCA §6).

Origins

The dominant paradigm of 20th-century US regulation. Built around New Deal–Great Society agencies: ICC (1887, abolished 1995), FDA (1906/1938), SEC (1934), FCC (1934), NLRB (1935), CAA/FAA (1958), HUD (1965), EPA (1970), OSHA (1970), CPSC (1972), NHTSA (1970).

The Clean Air Act of 1970 is the modern archetype — Congress directs EPA to set National Ambient Air Quality Standards (42 U.S.C. §7409) for six “criteria pollutants” at levels “requisite to protect the public health” with “an adequate margin of safety.” Whitman v. American Trucking Ass’ns, 531 U.S. 457 (2001) — costs may not be considered in setting NAAQS.

Strengths

• Legible to courts, regulated parties, public.
• Predictable: compliance status is binary.
• Politically robust: bright lines are easy to defend.
• Works well when measurement is reliable and the problem is uniform across the regulated population.

Weaknesses

• Inefficient when abatement costs differ across firms — a uniform standard forces high-cost abaters to do as much as low-cost ones.
• Innovation-suppressing — once a technology is mandated, no incentive to do better.
• Captured — industry helps draft the standards (the “captured agency” problem; Stigler 1971).
• Brittle — fails when the world changes (e.g., emissions standards calibrated to existing tech can’t anticipate breakthrough alternatives).

Modern frame

CAC remains the workhorse of safety-critical regulation: aviation (FAA), pharmaceutical approval (FDA pre-market), nuclear (NRC), food (FSIS), workplace exposure limits (OSHA PELs). Areas with quantifiable, time-varying externalities increasingly shift to market-based supplements.

Cases

• Massachusetts v. EPA, 549 U.S. 497 (2007) — greenhouse gases are “air pollutants” under §202(a)(1) of the Clean Air Act; CAC framework extended to climate.
• Utility Air Regulatory Group v. EPA, 573 U.S. 302 (2014) — limited PSD/Title V to traditional pollutants; rejected EPA’s “Tailoring Rule.”
• West Virginia v. EPA, 597 U.S. 697 (2022) — “major questions doctrine”; agency cannot use minor statutory provisions for industry-restructuring rules (Clean Power Plan).

4. Market-based instruments

Mechanism

Price the externality so private decisions internalize it. Forms:

• Pigouvian tax — flat per-unit tax equal to marginal external cost (carbon tax, gasoline excise tax).
• Cap-and-trade — fixed quantity, tradable permits (EU ETS, RGGI, CA AB 32, the original SO₂ market under 1990 CAA amendments).
• Tradable performance standards — emission rate × output, with credit trading (CAFE, RPS).
• Tax credit / subsidy — negative externality of forgoing the good (45Q carbon capture credit; IRA §45Y/48E clean energy credits).
• Deposit-refund — bottle bills.

Origins

Theoretical roots: Pigou, The Economics of Welfare (1920) on taxes for externalities; Coase, “The Problem of Social Cost” (1960) on bargaining alternatives; Dales, Pollution, Property and Prices (1968) on tradable permits; Montgomery 1972 on equivalence of price and quantity instruments in static settings; Weitzman 1974 on price-vs-quantity under uncertainty.

US practice: Acid Rain Program (1990 CAA Title IV) is the canonical success — SO₂ emissions fell faster and at lower cost than projected. NOₓ Budget Trading Program, then CSAPR. RGGI (2009) — northeast cap-and-trade. California’s AB 32 / cap-and-trade (operational 2013). Federal carbon pricing has never passed Congress.

Strengths

• Least-cost abatement — firms with low marginal cost over-comply and sell credits to high-cost firms; total compliance cost minimized.
• Innovation incentive — every unit of emission costs money, so every reduction earns money.
• Revenue-raising — tax form is a tool of fiscal policy.

Weaknesses

• Distributional concerns — carbon taxes can be regressive without rebates (BC carbon tax dividend solves; federal “fee and dividend” proposals).
• Price volatility in cap-and-trade — see EU ETS in 2008 collapse.
• Quantity vs price tradeoff — Weitzman’s insight: cap-and-trade fixes quantity but lets price vary; tax fixes price but lets quantity vary. Under uncertainty about marginal costs, choose the instrument whose error is less costly.
• Hotspots — uniform pricing can permit local concentration; environmental justice critique.
• Banking/borrowing — temporal arbitrage requires explicit design.

Cases / programs

• EU ETS — Phase 1 (2005-07) collapsed when overallocation revealed; Phase 4 (2021-30) tightened; Market Stability Reserve.
• California AB 32 / Cap-and-Trade — auctioned allowances; revenues for high-speed rail, disadvantaged communities.
• RGGI — 11 northeast states; allowance proceeds fund energy efficiency.
• EU CBAM — Carbon Border Adjustment Mechanism, transitional 2023, full 2026. Effectively imposes EU’s carbon price on imports of steel, cement, aluminum, fertilizers, hydrogen, electricity.
• 45Q + IRA §45Y/48E — tax-credit-driven market for CCS and clean energy in the US since 2022.

Modern frame

Market-based has captured most of the post-1990 environmental policy frontier in mature democracies. Adoption blocked in the US at the federal level by political economy, but extensive subnational and sectoral. The 2024-2026 CBAM rollout is putting global pressure on carbon pricing through trade.

5. Information-based regulation

Mechanism

Mandate disclosure of information; allow markets (or regulators, or tort plaintiffs) to act on it.

Forms

• Securities disclosure — SEC 10-K, 10-Q, 8-K; prospectus requirements under §5 of the Securities Act of 1933; Basic Inc. v. Levinson, 485 U.S. 224 (1988) on materiality.
• Nutrition Facts — FDA 21 CFR 101; FDA 2016 revisions added added sugars.
• California Prop 65 (Safe Drinking Water and Toxic Enforcement Act of 1986) — businesses must warn of exposure to listed carcinogens or reproductive toxins; private enforcement creates a high-velocity disclosure regime.
• EPA Toxics Release Inventory (TRI) — EPCRA §313; facilities report releases; voluntary reductions followed disclosure.
• GHG Reporting Rule (40 CFR Part 98) — mandatory reporting >25,000 tCO₂e/yr.
• Carbon disclosure — SEC Climate Disclosure Rule (2024, stayed pending litigation); EU CSRD (Corporate Sustainability Reporting Directive, 2022); California SB 253 / SB 261 (2023, in force 2026).
• Right-to-know — OSHA Hazard Communication Standard (29 CFR 1910.1200); SDS sheets.
• Restaurant menu calorie labeling — ACA §4205 (FDA final 2014, in force 2018).

Theoretical basis

Disclosure addresses information asymmetry — the textbook market failure where buyers cannot observe quality. Originated in Akerlof’s “The Market for Lemons” (1970) on used cars; Spence’s signaling theory (1973); Stiglitz on screening.

Disclosure has appeal because it preserves private decisionmaking; doesn’t ban anything. The information-forcing function of disclosure also affects firm behavior: TRI led to voluntary reductions; Prop 65 led to reformulation rather than warning labels.

Strengths

• Preserves consumer autonomy.
• Lower regulatory burden — disclosure is cheaper than substantive mandates.
• Enables private enforcement (Prop 65 attorney general delegation, securities Rule 10b-5 class actions).
• Can address problems where ex ante banning is infeasible.

Weaknesses

• Information overload — endless mandatory disclosures cease to influence behavior (Ben-Shahar & Schneider, More Than You Wanted to Know, 2014).
• Warning fatigue — Prop 65 critique: every product has a warning, so nothing has a warning.
• Information without action — disclosing climate risk does not by itself reduce emissions.
• Strategic disclosure — issuers learn what triggers private litigation and avoid those topics.

Cases

• Basic Inc. v. Levinson, 485 U.S. 224 (1988) — fraud-on-the-market presumption.
• Halliburton Co. v. Erica P. John Fund, Inc. (Halliburton II), 573 U.S. 258 (2014) — defendants can rebut at class certification.
• Matrixx Initiatives, Inc. v. Siracusano, 563 U.S. 27 (2011) — adverse event reports can be material below statistical-significance threshold.
• NIFLA v. Becerra, 585 U.S. 755 (2018) — First Amendment limits on compelled disclosure (CA crisis pregnancy center act struck).
• Zauderer v. Office of Disciplinary Counsel, 471 U.S. 626 (1985) — purely factual / non-controversial commercial disclosure subject to rational-basis review.

6. Notice-and-comment vs negotiated rulemaking vs adjudication

Notice-and-comment rulemaking (APA §553)

The default.

• Notice of Proposed Rulemaking (NPRM) published in Federal Register.
• Comment period — typically 30-90 days; complex rules 120-180.
• Reasoned response to significant comments in preamble of final rule.
• Concise general statement of basis and purpose.
• Effective date — typically 30 days post-publication.
• Vermont Yankee Nuclear Power Corp. v. NRDC, 435 U.S. 519 (1978) — courts cannot impose procedures beyond what the APA requires.
• State Farm Mut. Auto. Ins. Co. v. Federal-Motor Vehicle Mfrs. Assn., 463 U.S. 29 (1983) — arbitrary-and-capricious review under §706(2)(A); the agency must consider relevant factors and explain its reasoning.

Formal rulemaking (APA §556-557)

Trial-type hearings with cross-examination of witnesses. Required only when statute uses “on the record after opportunity for an agency hearing.” Vanishingly rare since United States v. Florida East Coast Railway Co., 410 U.S. 224 (1973) restricted the trigger language. The 1969 peanut-butter rulemaking before the FDA, which took 9 years and produced a 7,736-page transcript, became the cautionary tale.

Negotiated rulemaking (Reg-Neg)

Established by the Negotiated Rulemaking Act of 1990 (5 U.S.C. §§561-570a). Convening:

• Convener identifies interests and constituencies.
• Federal Advisory Committee Act–compliant committee negotiates draft rule.
• Consensus draft submitted to agency, then through notice-and-comment.

When it works: small number of well-defined stakeholders; technical detail; rule will face complex implementation. When it fails: broad public interest; high-stakes distributional conflict; one party has overwhelming power. Examples of use: EPA reformulated gasoline rule (RFG, 1991); DOJ ADA accessibility regs; FAA noise rules. Coglianese 1997 study found that reg-neg did not significantly reduce post-promulgation litigation, a key claim of its proponents.

Adjudication by decision

Statutes that authorize the agency to develop policy through case-by-case decisions in individual proceedings.

• NLRB — historic exemplar; NLRB v. Bell Aerospace Co., 416 U.S. 267 (1974) confirmed broad discretion to develop policy through adjudication.
• SEC enforcement and administrative proceedings.
• FTC Section 5 unfair-method-of-competition adjudication.
• EEOC Title VII enforcement (though substantive doctrines mostly judicial).
• NLRB v. Bell Aerospace — agencies have broad choice of adjudication vs rulemaking.

Guidance / interpretive rules

Documents that “interpret” existing rules but do not have the force of law.

• Perez v. Mortgage Bankers Ass’n, 575 U.S. 92 (2015) — agencies need not use notice-and-comment to change interpretive rules; overruled the D.C. Circuit’s Paralyzed Veterans doctrine.
• Kisor v. Wilkie, 588 U.S. 558 (2019) — Auer / Seminole Rock deference to agency interpretation of own regulations survived but was constrained. Five conditions: regulation is “genuinely ambiguous,” interpretation is “reasonable,” it is the agency’s “authoritative” or “official” position, reflects “substantive expertise,” and is “fair and considered judgment.”

Strategic rule

Use notice-and-comment for substantive rules with general applicability; use adjudication for case-by-case policy development that benefits from factual concreteness; use negotiated rulemaking for technical rules with a small, well-defined stakeholder set; use guidance for clarifying interpretations (subject to Kisor limits on later judicial deference).

7. Principles-based vs rules-based

Principles-based supervision

• Set high-level standards; require firms to design their own compliance programs.
• Examples: UK FCA’s “Principles for Businesses” (PRIN 2.1.1) — 11 principles like “A firm must conduct its business with integrity”; Basel II/III on capital adequacy in qualitative terms.
• US examples: bank holding company “safety and soundness”; OCC heightened standards; SEC’s Reg BI requiring broker-dealers to act in client “best interest.”

Rules-based prescription

• Detailed rules covering specific transactions.
• US SEC historically rules-based.
• Volcker Rule (2014, simplified 2020); Dodd-Frank §619.
• Detailed CFTC rules for swaps, position limits.
• Solvency II — extensive technical rules vs UK FCA’s overlay.

Trade-offs

Dimension	Principles	Rules
Flexibility	high	low
Compliance cost	depends on firm	predictable
Loophole risk	low	high (compliance theater)
Predictability	low	high
Enforcement clarity	discretion-heavy	bright-line
Innovation friendly	yes	no
Suitable for:	sophisticated firms; supervised relationships	retail / mass-market; ex post enforcement

Modern frame

UK FCA’s principles-based regime was hailed pre-2007; the GFC tarnished it but did not displace it. US SEC is moving partially toward principles-based regulation in fiduciary contexts (Reg BI is principle-flavored but with detailed sub-rules). EU GDPR is principles-based in core architecture but with detailed annexes.

The genuine question is enforcement capacity: principles-based requires high-quality supervision and good-faith engagement with firms. Where regulatory capture is severe or staff turnover is high, rules win because they don’t require sophisticated supervisors.

8. Federal preemption vs state experimentation

Express preemption

Federal statute explicitly preempts state law. Examples:

• ERISA §514 — preempts state laws “relating to” employee benefit plans.
• FIFRA — pesticide labeling.
• Federal Cigarette Labeling and Advertising Act.
• Federal Aviation Act — economic regulation of airlines (preempted state regulation of routes, fares).

Implied preemption

• Field preemption — federal regulation so comprehensive that no room for state action (immigration; nuclear safety; maritime).
• Conflict preemption — impossible to comply with both, or state law obstructs federal objectives.
• Geier v. American Honda Motor Co., 529 U.S. 861 (2000) — federal vehicle safety standards preempt state common-law claims.
• Wyeth v. Levine, 555 U.S. 555 (2009) — FDA drug labeling does not preempt state failure-to-warn claims.
• PLIVA, Inc. v. Mensing, 564 U.S. 604 (2011) — generic drugs cannot be sued for failure-to-warn because they must use brand-name label.

State experimentation

• New State Ice Co. v. Liebmann, 285 U.S. 262 (1932) (Brandeis dissent) — “states as laboratories of democracy.”
• Examples: California environmental, energy, consumer-protection leadership (CARB on vehicle emissions has waiver under §209 of CAA); Massachusetts on health insurance (precursor to ACA); state cannabis legalization.

Cases

• American Trucking Ass’ns v. EPA, 533 U.S. 322 (2001) — federalism limits on NAAQS.
• Massachusetts v. EPA (2007) — states have standing to challenge under §307(b) of CAA.
• Murphy v. NCAA, 584 U.S. 453 (2018) — anti-commandeering; PASPA struck.

Modern frame

The 2020s have seen aggressive federal preemption (FAA dormancy on state drone rules; NLRA preemption of state labor laws) and aggressive state experimentation (California climate disclosure, AI regulation in NY and CA, abortion regimes post-Dobbs, gun regulation post-Bruen). The political-economy question is whether uniform national markets benefit consumers more than experimentation with feedback.

9. Ex ante licensing vs ex post enforcement

Ex ante licensing

• Pre-approval required before market entry.
• Examples: FDA drug approval (NDA / BLA); FCC spectrum licenses; nuclear plant operating licenses (NRC); state professional licensing (medicine, law, accounting, real estate, cosmetology, plumbing).
• Strengths: prevents bad outcomes; signals quality to consumers.
• Weaknesses: slow; expensive; captures rents; suppresses innovation.

Ex post enforcement

• Market entry free; bad actors pursued after harm.
• Examples: most consumer products under CPSC (recall and civil penalty, not pre-approval); securities under §5 of ‘33 Act + §10(b) of ‘34 Act; SEC enforcement; FTC unfair-or-deceptive-acts.
• Strengths: less rent-seeking; faster innovation; market discipline.
• Weaknesses: harm has occurred; enforcement may be uneven; small actors may exit before judgment.

Strategic rule

Use ex ante for catastrophic / irreversible harms (drugs, nuclear, aviation). Use ex post for reversible / repairable harms (consumer goods, advertising). Use ex ante for information-poor markets (drugs that consumers cannot evaluate). Use ex post for information-rich markets (most consumer goods).

The cost-benefit ratio is dominated by what fraction of harms are detectable ex post. If most damage is invisible (insurance contract terms, occupational safety patterns), ex post fails.

10. Voluntary standards vs mandatory regulation

Voluntary standards / safe harbors

• NIST Cybersecurity Framework — voluntary; widely adopted; cited in regulatory rulings as the benchmark.
• ISO 9001, ISO 27001, ISO 14001 — quality, security, environment management systems.
• GAAP / IFRS — accounting standards from private-sector standard-setters (FASB, IASB); the SEC has accepted both for public registrants.
• PCI DSS — payment card industry data security standard.
• OECD AI Principles (2019); Hiroshima AI Process (2023).

Mandatory regulation

• HIPAA — Privacy Rule and Security Rule are mandatory.
• GLBA — financial privacy.
• GDPR — EU; binding fines up to 4% global revenue.
• CCPA / CPRA — California; private right of action for data breaches.
• AI Act (EU) — phased in 2024-2027; risk-based categories with bans (social scoring) and obligations (high-risk systems).

The “private-sector-led standard becomes de facto mandatory” pattern

Many “voluntary” standards become de facto mandatory through:

• Insurance underwriting — cyber insurance requires NIST CSF.
• Procurement — federal vendors must follow FAR cybersecurity requirements; cascading down supply chains.
• Tort defense — meeting industry standards is a defense; not meeting them is negligence.
• Regulatory adoption — agencies cite voluntary standards as benchmarks (FTC §5 enforcement against firms failing NIST CSF).

Strategic rule

Voluntary standards work when industry expertise dominates regulator expertise. Mandatory regulation works when industry incentives diverge from public interest. Most modern frameworks combine: principle in statute, voluntary standard for implementation, agency oversight for enforcement.

11. Administrative-law judges vs Article III courts

ALJ adjudication

• Agency adjudicators with statutory protections (5 U.S.C. §7521 — removal “only for good cause”).
• Examples: SEC ALJs (enforcement); SSA ALJs (~1,300 nationally); FTC ALJs; NLRB ALJs.
• Lucia v. SEC, 585 U.S. 237 (2018) — SEC ALJs are “officers”; must be appointed under the Appointments Clause.
• Free Enterprise Fund v. PCAOB, 561 U.S. 477 (2010) — multiple layers of removal protection for inferior officers violate Article II.
• Jarkesy v. SEC, 603 U.S. 109 (2024) — SEC’s use of ALJs for civil penalty actions analogous to common-law fraud violates the Seventh Amendment; jury in Article III required.

Article III courts

• Life-tenured judges; juries available.
• Required for criminal penalties; required after Jarkesy for civil penalties of common-law-analogue claims.
• Slower; more procedural rigor; appeals to circuit and Supreme Court.

Modern frame

Post-Jarkesy, agencies must reassess every enforcement program that imposes civil penalties for fraud-like violations. The SEC has substantially shifted to filing in federal court rather than the agency adjudication track since the 2024 ruling; the FTC’s antitrust enforcement is under similar pressure; CFPB enforcement architecture under review.

This is a separation of powers shift, not merely procedural — the rebalancing toward Article III courts is reshaping the practical limits of administrative power.

12. The post-Chevron world (2024-2026)

Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024) overruled Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984). Courts now independently determine the best reading of statutes; agency interpretations get Skidmore respect (“persuasive based on the thoroughness, validity, and consistency”). Effect:

• Reduces agency authority to fill statutory gaps.
• Increases litigation — every ambiguous statute becomes a new fight.
• Empowers Congress — only Congress can rewrite the statute; agencies can only argue.
• Increases policy whiplash — without Chevron stability, each administration’s interpretation is challenged.

Combined with the major questions doctrine of West Virginia v. EPA (2022) — agencies cannot use minor provisions for major restructuring — and Loper Bright, the locus of regulatory authority is shifting from agencies back toward Congress and the courts. This affects every approach in this note: command-and-control rules now face more vigorous statutory challenge; cap-and-trade programs need explicit statutory authority; disclosure mandates must be tightly tied to statutory text.

13. How to choose — instrument by problem archetype

Problem	Best instrument	Why
Uniform environmental externality (criteria pollutant, CO₂)	Market-based (tax or cap-and-trade)	Heterogeneous abatement costs; least-cost solution
Catastrophic / irreversible harm (drugs, nuclear, aviation)	Ex ante licensing + CAC	Detection ex post too late
Information asymmetry (used cars, securities, food)	Disclosure (info-based)	Markets work if informed
Conflicts of interest (audit firms, broker fiduciary duty)	Structural reform + principles	Conflicts require institutional remedy
Network industry (telecom, payment systems, electric grid)	Standards + agency rulemaking	Coordination dominates
Fast-moving tech (AI, crypto, gene editing)	Principles-based + safe harbors + adaptive regulation	Rules-based outpaced by tech
Hyperlocal externalities (zoning, noise, smell)	State / local CAC; private nuisance	Federal-uniform inappropriate
Cross-border (CO₂, banking, securities)	International standards + national implementation	Bilateral / multilateral coordination
Workplace safety with known hazards	CAC (OSHA PEL) + record-keeping	Established hazards; uniform standards
Consumer fraud / deception	Ex post enforcement (FTC §5; state UDAP)	Cost of pre-approval would crush market
Health-care safety / quality	Mandatory + voluntary (CMS conditions of participation + Joint Commission)	Public payer leverage + industry expertise

14. Anti-patterns

1. Mandating a technology that becomes obsolete. EPA’s NSPS for power plants in the 1970s mandated scrubbers; locked in inefficient operations. Performance standards (output cap) would have allowed fuel-switching and innovation.

2. Disclosure without action. Climate disclosure without a tax or cap may move some capital but mostly produces glossy reports. Pair disclosure with substantive instrument when the underlying problem is too important to leave to markets alone.

3. Captured agency drafting rules with industry. Negotiated rulemaking risks this. Insist on broad public-interest representation on the negotiating committee.

4. Ex ante licensing where the harm is reversible. State licensing of “low-risk” professions (cosmetology, interior design) often serves rent-seeking incumbents rather than protecting consumers (Carpenter et al. 2012; Kleiner 2006).

5. Federal preemption that crowds out state innovation. ERISA §514 has been criticized for blocking state health-care reforms; FAA’s broad preemption of state drone rules has been argued to suppress local experimentation.

6. Rules-based with technology stuck in 1995. Volcker Rule’s original implementation was 71 pages of statute and ~70 pages of preamble; firms reported compliance costs >$1B before 2020 simplification.

7. Principles-based with no supervisory capacity. UK FCA’s principles only work because UK regulators have decades of staff continuity and qualitative expertise; importing the model where staff turnover is high produces under-enforcement.

8. Ignoring Loper Bright. Drafting a rule that relies on agency discretion to fill statutory gaps will face heightened review; tether the rule tightly to statutory text or expect litigation.

9. Using guidance to do what rulemaking should. Azar v. Allina Health Services, 587 U.S. 566 (2019) — substantive policy changes require notice-and-comment in Medicare context; Kisor’s constraints on deference.

10. Voluntary standards as cover. “Industry self-regulation” without enforcement leverage is hollow. Make adoption non-optional via insurance underwriting, procurement, or tort defense baseline.

15. Decision tree — pick the instrument

What's the problem?
├─ Externality (pollution, congestion, GHG)
│    ├─ Quantifiable, uniform across sources → Pigouvian tax or cap-and-trade
│    ├─ Uncertain marginal cost, fixed quantity goal → cap-and-trade
│    ├─ Uncertain marginal cost, fixed price ceiling acceptable → carbon tax
│    └─ Hyperlocal, hot-spot risk → CAC + monitoring
├─ Information asymmetry (food, drugs, securities, ESG)
│    ├─ Consumer can act if informed → mandatory disclosure
│    ├─ Consumer cannot evaluate (drug efficacy, professional credentials) → ex ante licensing + disclosure
│    └─ Sophisticated buyer (institutional securities) → disclosure + private enforcement
├─ Conflict of interest (audit, broker, rating agency, healthcare provider)
│    └─ Structural reform + heightened duties + supervision
├─ Catastrophic / irreversible risk (drugs, nuclear, aviation, BSL-4, AGI?)
│    └─ Ex ante licensing + CAC + ex post penalties
├─ Coordination problem (network, payment, grid, accounting)
│    └─ Industry-led standards + regulatory ratification
├─ Fast-moving technology (AI, crypto, gene editing, neurotech)
│    ├─ Risk-tiered, EU-style → AI Act framework
│    ├─ Innovation-first, sandbox → UK FCA sandbox; FDA breakthrough designation
│    └─ Adaptive, observe-and-learn → guidance + reporting
├─ Cross-border problem
│    ├─ Trade externality → CBAM; supply-chain diligence (EU CSDDD)
│    ├─ Banking, securities → BIS, IOSCO, FSB coordination
│    └─ Climate → Paris + national implementation
├─ Discrimination, civil rights
│    └─ Substantive prohibition + private enforcement + fee-shifting
└─ Consumer fraud / deception
     └─ Ex post enforcement (FTC §5; state UDAP) + private right of action

16. The 2024-2026 frontier

• AI regulation — EU AI Act (in force 2024, phased to 2027); US Executive Order 14110 (Biden 2023, partially rescinded 2025); state laws (Colorado AI Act 2024; California AI Transparency Act; NYC bias audits for hiring AI). The principles-based / rules-based debate is live: EU is risk-tiered prescriptive; US shifting to disclosure + sector-specific rules.
• Climate disclosure — SEC Climate Disclosure Rule (March 2024 finalized, stayed by Eighth Circuit); EU CSRD (operative 2024); California SB 253 / SB 261 (2026 effective). Securities + ESG + environmental + corporate-governance pressure converging.
• Cybersecurity — SEC Cybersecurity Disclosure Rule (2023); CIRCIA (Cyber Incident Reporting for Critical Infrastructure, 2022, CISA rules 2026); EU NIS2 (2024); EU CRA (Cyber Resilience Act, 2024).
• CBAM and supply-chain diligence — EU CBAM transitional 2023, full 2026; EU CSDDD (Corporate Sustainability Due Diligence Directive, 2024) requires supply-chain due diligence. Extraterritorial reach.
• Generative-AI copyright — voluntary opt-out registries; ETSI / W3C technical standards; courts and Congress still working through (Andersen, Authors Guild, NY Times v. OpenAI).
• Crypto / digital assets — SEC ALJ era constrained by Jarkesy; SEC litigation shifting to court; SEC v. Ripple Labs (D.N.J. 2023); CFTC dominance of derivatives; MiCA (EU Markets in Crypto-Assets Regulation, 2023-2025).
• Major questions doctrine continuing to constrain agency-led restructuring without explicit congressional authorization.
• State preemption battles — federal preemption of state AI laws debated; ERISA preemption of state PBM rules; HIPAA preemption of state health-privacy laws.

Adjacent

• Administrative law (T1) — administrative-law for APA, Chevron/Loper Bright, State Farm, agency procedures.
• Constitutional law — constitutional-law for non-delegation, major questions, separation of powers.
• Environmental — employment-and-environmental-law for CAA, CWA, RCRA, CERCLA.
• Securities — securities-regulation-and-ma-practice, securities-regulation-deep for SEC, disclosure, Jarkesy.
• Antitrust — antitrust-and-competition-deep for FTC §5, Sherman/Clayton, Chevron removal effects.
• EU comparison — eu-competition-and-regulation for DMA, DSA, GDPR, AI Act, CBAM.
• Civil procedure — civil-procedure-and-evidence for Jarkesy’s Seventh Amendment implications.
• Contracts & IP — contracts-and-ip for private enforcement of standards via contract.
• Litigation strategy — _compare_litigation-strategies for how enforcement choices interact with strategy.

When to pick what — one paragraph

Regulatory design is choosing instruments from a menu, not from a checklist. The default narrowing: if the externality is uniform and quantifiable, price it (cap-and-trade or Pigouvian tax); if the harm is catastrophic or irreversible, license ex ante (FDA, NRC, FAA model); if the problem is information asymmetry and consumers can act if informed, disclose (SEC, Nutrition Facts, Prop 65); if the problem is conflict of interest, restructure (Glass-Steagall, fiduciary rules); if the industry needs coordination, set standards (NIST CSF, GAAP/IFRS); if the technology is fast-moving, principles-based with safe harbors and sandboxes. The procedural fork — notice-and-comment vs adjudication vs guidance — turns on whether the policy is general (rulemaking) or fact-specific (adjudication) and how much Kisor and Loper Bright / major-questions exposure the agency wants. The 2024-2026 jurisprudence (Loper Bright, Jarkesy, West Virginia v. EPA) tightens the agency-discretion frontier and pushes substantive policymaking back toward Congress and Article III courts. The dominant cause of failed regulation is instrument-problem mismatch — mandating technology where pricing would work, banning where disclosure would suffice, leaving to ex post enforcement where ex ante licensing was warranted.

— end —