Public Trust

Term Limits for Congress — Lifetime

House Representatives are limited to 5 terms (10 years). Senators are limited to 3 terms (12 years). This is a lifetime ban — no cooling-off period and no return. A representative who enters office at 60, serves 10 years, and then waits out a 10-year cooling-off period would return to write policy in their 80s for a world they will not live in. Public service is a civic duty, not a dynasty.

This is enforced today through the party itself, not through law — Congress cannot currently impose term limits on itself by statute (the Supreme Court's 1995 ruling in U.S. Term Limits v. Thornton held that qualifications for federal office are fixed by the Constitution and cannot be added to by statute or state law). Until there is enough support in Congress to pass a constitutional amendment, term limits are a condition of the Sunset Party itself: a candidate who identifies as a Sunsetter commits to this standard as a condition of the party's support. A sitting member who abandons the commitment once in office answers to the same mechanism that got them there — a primary challenge. The perceived loss of legislative experience is outweighed by the current cost of entrenchment: the country already runs on unelected staff and lobbyists functioning as institutional memory for career legislators. That dependency does not go away by keeping legislators in office longer — it exists today, at current tenure lengths, and term limits do not make it materially worse.

Single-Use Bills Only

No hidden agenda. Every bill addresses one issue — no unrelated riders buried in omnibus legislation. Enforcement during a Sunset presidency is the veto. Beyond that, a censure mechanism holds representatives and senators accountable for violations. If a law is passed with hidden provisions, those responsible answer for it publicly.

Like term limits, single-use bills are enforced as a Sunset Party standard before they are federal law. A Sunsetter in Congress commits to introducing and supporting only single-issue legislation as a condition of the party's backing. Until there is enough support in Congress to formalize this into House and Senate rules — for example, a germaneness rule enforceable by a point of order, similar to existing mechanisms in both chambers — this is a Sunsetter-only commitment, not a claim that it currently binds anyone else.

Public Transparency Dashboard

Every dollar collected and spent by the federal government is accounted for in real time — publicly. Not for watchdogs and journalists. For every American. The dashboard lives under the Office of Budget and Accountability with proper independent funding. We live in an era where this is entirely possible. There is no excuse not to do it.

USAspending.gov already exists and demonstrates both the promise and the failure of this idea. Most federal spending is required to post within 30 days of an agency's certified submission. But Department of Defense procurement data — one of the largest categories of federal spending — carries a documented 90-day delay, acknowledged in multiple GAO reports dating back to 2017. GAO reports as recent as 2025 continue to find agencies inconsistent in addressing data completeness, timeliness, and accuracy. The goal is not to build something new from scratch — it is to close those specific gaps: push every agency, including Defense, toward real-time or near-real-time reporting, and hold agencies accountable for the completeness problems GAO has been documenting for over a decade.

The same principle applies to the judicial branch, where Washington already has a real, existing accountability mechanism that voters can't currently use: unlike federal judges, Washington's judges — all of them — are selected through nonpartisan elections, not lifetime appointment. The state's own Walsh Commission on judicial selection found the problem isn't an absence of accountability, it's that voters have almost no usable information when those elections happen. This platform's fix doesn't create a new disciplinary body or a new way to punish judges for individual outcomes — it makes the election that already exists actually informed.

Washington requires statewide use of a validated pretrial risk-assessment tool — the same kind of instrument (the Arnold Tool / Public Safety Assessment) Yakima County already uses, extended to every county rather than left as a patchwork where King County doesn't use one systematically. When a judge releases a defendant against a tool-flagged high-risk assessment, a brief written justification becomes part of the public record. All of this — release decisions, risk scores, deviations, justifications, and what happened afterward (reoffense during the pretrial period, failure to appear) — is published on the same public dashboard this platform already proposes for federal spending, organized by judge, searchable by any voter before they go into a judicial election.

One design detail that has to be built in correctly or the dashboard becomes misleading rather than clarifying: this can't be a simple raw count of 'how many people this judge released who reoffended,' because judges don't choose their own caseloads — a judge assigned mostly violent-felony calendars will show more high-risk releases than one hearing misdemeanors, through no difference in judgment at all. The dashboard has to show deviation rate relative to each judge's own caseload composition, not raw incident totals, or it unfairly punishes judges for the cases they're assigned rather than the decisions they make — and it would create a real incentive to lobby for easier calendars rather than judge well, which would be its own kind of corruption.

This is also, deliberately, a lighter legal lift than several other mechanisms in this platform: court proceedings and rulings are already public record. This isn't compelling new disclosure from a private party — it's the state organizing and publishing its own judicial branch's already-public records in a form voters can actually use, which sits on much more comfortable legal ground than, for example, the campaign finance disclosure mechanism elsewhere in this platform.

No Settlement Without a Public Admission of Fault

The Department of Justice does not accept a settlement — civil or criminal, from an individual or a corporation — without a public admission of fault. 'Neither admit nor deny' language, standard practice across federal settlements for decades, ends. If the government's case was strong enough to extract a payment, it was strong enough to extract the truth along with it.

This has real precedent, and the precedent tells you exactly how the fight goes. Judge Jed Rakoff rejected a proposed SEC-Citigroup settlement in 2011 specifically because its 'neither admit nor deny' language left him, in his words, no facts on which to judge whether the settlement was fair or in the public interest. SEC Chair Mary Jo White responded in 2013 with a policy requiring admissions in cases involving 'heightened accountability' — not a universal rule, a targeted one. In 2014, the Second Circuit reversed Rakoff's underlying reasoning on appeal, ruling that a judge can't withhold approval of a settlement merely because it lacks an admission — that's not the court's call to make. That reversal is the reason this has to be actual Department of Justice policy, not a hope that judges will force it — courts have already ruled they won't.

The reason companies fight admission requirements as hard as they do is the same reason this policy has teeth: an admission of fault in a government settlement becomes usable evidence in follow-on private litigation — shareholder suits, class actions, anyone else harmed by the same conduct. A company that admits wrongdoing to the DOJ has effectively admitted it to every other plaintiff waiting in line. That is precisely the accountability 'neither admit nor deny' was invented to avoid, and precisely why the Department of Justice does not offer companies that shield anymore.

Be direct about the tradeoff rather than pretend there isn't one: a universal rule, applied to every settlement rather than SEC's narrower 'certain cases' standard, means some defendants who would have quietly settled will choose to litigate instead rather than hand plaintiffs a ready-made admission. That means more trials, more DOJ litigation resources required, and slower resolution in some cases. The tradeoff is accepted deliberately: a slower process that produces the truth is worth more than a fast one that produces a check and a denial. This is already normal practice at the DOJ's own Antitrust Division for its most significant cases — and the Deere settlement this platform cites under Right to Repair (see Technology & the Future) is a live example of exactly the practice this policy ends: Deere paid, and Deere did not admit or deny wrongdoing. That doesn't happen again.

Show Your Work Before You Ask for More

Before a state can apply for additional federal discretionary or competitive grant funding in a given category, it submits a detailed accounting: how much federal funding it already received in that category, exactly how it was spent, and what outcomes resulted. That accounting becomes a major, heavily-weighted scoring criterion for the next funding cycle — not paperwork filed and forgotten, an actual factor in whether the state gets more.

Homelessness is the clearest illustration of why this matters. Federal homelessness assistance funding reached over $10 billion in the FY2025 budget, an 11% increase from FY2023 enacted levels. Over that same window, the number of people experiencing homelessness on a single night jumped 18% from 2023 to 2024 — the largest one-year increase in the history of the federal count, and a 27% increase since 2013. Spending is going up. The problem is going up faster. That gap is exactly what this policy is designed to surface before the next check gets written, not after.

This isn't a novel concern — GAO flagged data and oversight quality problems in federal homelessness spending back in 2020, in a report titled, plainly, 'Better HUD Oversight of Data Collection Could Improve Estimates of Homeless Population.' The problem has been known for years. This policy is what acting on it actually looks like.

One design detail that has to be built in correctly, or this accountability mechanism creates the same kind of unfair distortion the judicial accountability dashboard elsewhere in this platform was built to avoid: the metric cannot be a simple 'did the raw homelessness count go down,' because that count is affected by things no state government controls. GAO's own research found point-in-time counts likely undercount real homelessness in the first place. The 2024 national spike was partly driven by migration-related sheltering concentrated in a handful of cities under special funding allocations — a factor entirely disconnected from those cities' program quality. Hawaii saw an 87% single-year increase driven purely by the Maui wildfires. A metric that penalizes a state for a wildfire or an in-migration surge caused by another state's policy isn't measuring fiscal responsibility — it's measuring bad luck, and would create a real incentive to manipulate or discourage accurate counting rather than improve outcomes.

The fix is the same kind of program-level, already-collected data used elsewhere in this platform's accountability mechanisms: outcomes tracked through HUD's existing Homeless Management Information System — exits to permanent housing, returns to homelessness within 12 and 24 months of program exit, and cost per successful placement — rather than a jurisdiction's raw headline count. This data already exists; HUD already requires Continuums of Care to report it. The accountability failure isn't a lack of data. It's a lack of consequence for what the data shows.

The differentiation this policy would surface is already real and already measurable, even outside official channels: an independent tracking project (the National Homeless Information Project — a smaller outside effort, not a government metric, and cited with that caveat) ranks major city Continuums of Care by a 'housing efficiency' score using this same kind of program-level data. Boston and Baltimore rank near 99%. Several California CoCs — Los Angeles, Oakland, San Jose, Fresno — rank at the bottom, despite substantial federal and state funding. Whatever the exact right metric turns out to be, the underlying point stands: some jurisdictions are getting meaningfully more result per dollar than others, right now, and the current grant system doesn't weight that difference at all when handing out the next round.

This generalizes beyond homelessness to any category of federal discretionary grant funding to states — infrastructure, public health, education, disaster recovery. The mechanism is the same everywhere it applies: show the accounting, track outcomes through whatever program-level data infrastructure already exists for that category, and make that record count for something before the next application gets approved.

Panel Review Where Money Is on the Line

The FY2026 NDAA lets contracting officers withhold 5% of payments from an incumbent contractor whose GAO bid protest is dismissed for lacking 'any reasonable legal or factual basis.' Today, that determination is made by a single GAO attorney — one person, out of a pool of roughly 40 in GAO's Procurement Law Control Group, assigned to the case. That was a reasonable process when the only thing at stake was who wins the protest. It stops being reasonable once a specific dollar penalty is attached to the outcome.

Where the loser-pays forfeiture provision is actually in play, the decision moves from one attorney to a three-person panel — an odd number, so there's no tie. This mirrors how every other adjudicative body handling real financial or legal stakes is typically structured: three-judge appellate panels, three-arbitrator panels in significant commercial disputes, three-judge panels at the Patent Trial and Appeal Board. A single reviewer's judgment is good enough when nothing but the protest's outcome rides on it. It is not good enough when a specific financial penalty does.

This applies specifically to protests where forfeiture is a possible outcome — not GAO's full protest caseload. Requiring a panel for every one of GAO's protests, most of which carry no financial penalty at all, would add real overhead to a high-volume system and cut against the speed the loser-pays provision was meant to achieve in the first place. The extra rigor is scoped to match where the extra stakes actually are.

Each panelist discloses financial interests specific to that case before deliberation — stock holdings in either party, familial or professional connections to either party or their counsel, and any other relationship that would require recusal under the same standard federal judges already follow (28 U.S.C. § 455). A disclosed conflict results in recusal and replacement, the same way it would on a multi-judge court. This isn't a new category of federal ethics law — it's applying the standard federal judges are already held to, to a body that is now making decisions with comparable financial consequences.

No Sole-Source Contracts, Full Stop

Every federal contract, from every agency, is competitive and transparent. No exceptions carved out for convenience. The situations that are typically used to justify sole-source awards — a single manufacturer holding proprietary parts, a genuine emergency, classified work requiring cleared personnel, or a purchase too small to run a full competition — are all real constraints. None of them require abandoning competition and transparency. Each one gets solved by moving the competitive moment earlier, not by skipping it.

Existing single-source systems: lock in the markup before the government becomes a captive buyer. When a weapons system, software platform, or vehicle manufacturing system is purchased, the contract includes a binding memorandum of understanding specifying the maximum markup on every component of that system for its entire service life — negotiated while real competitive pressure still exists on the vendor, not after the government has no alternative supplier left.

This isn't a hypothetical problem. TransDigm, a major DoD spare parts supplier for aircraft including the Apache, C-17, F-16, and Chinook, built its business model around acquiring companies that already held sole-source government contracts, then raising prices with no competing bid to check them against. Two separate DoD Inspector General audits (2019 and 2021) found profit margins on individual parts ranging from 17% to as high as 3,850–4,400%, totaling roughly $37 million in documented excess profits from just two limited-sample audits. The Inspector General's own repeated finding, across audits dating back to 1998: contracting officers had no cost data to check the price against, because sole-source contractors below a certain dollar threshold aren't required to disclose it. Congress's response, twice, was to raise that disclosure threshold rather than close the gap — the FY2018 NDAA raised it to $2 million despite the Pentagon's own acquisition office finding that move 'unlikely to provide cost savings,' and the FY2026 NDAA raises it again, to $10 million. Locking the markup ceiling into the original contract — before the government is a captive buyer, not after — closes this regardless of who owns the sole-source position later or whether they ever disclose cost data going forward.

Disaster relief: close the urgency exception that still bypasses the pools FEMA already built. FEMA already runs Advance Contracts and Standby Contracts — competed and awarded before disasters occur, maintaining pools of pre-qualified contractors ready on short notice. That infrastructure exists and works largely as this framework would want it to. The gap is that it isn't the mandatory first resort: FAR 6.302-2's 'unusual and compelling urgency' authority lets agencies bypass those pools and award noncompetitively during active disasters, for up to 150 days at a time. In 2021, of FEMA's 3,011 disaster-related contracts, 526 — worth roughly $153 million, about 17.5% of both the count and the dollar value — were awarded this way instead of through the pre-positioned pools.

This isn't a hypothetical concern about an exception being used too often — the independent audit record shows it's already failing on its own terms. A 2015 GAO audit found more than half of the FEMA noncompetitive disaster contracts it reviewed exceeded the 150-day legal time limit without the required justification or approval. When a senator directly asked FEMA's own Administrator, during 2017 hurricane season testimony, which of FEMA's 355 noncompetitive contracts had exceeded that limit, FEMA could not answer — it didn't have that information about its own contracts. DHS OIG's review of FEMA's Katrina-era sole-source housing contracts found the required independent cost and price analysis was not consistently performed before award. And the Puerto Rico Whitefish/Cobra contract after Hurricane Maria was found procedurally compliant with noncompetitive procurement rules while DHS OIG separately found the actual costs may not have met the federal requirement that costs be reasonable — passing the paperwork test while failing the price test, the exact gap this framework is built to close.

This platform's position: the pre-positioned pool becomes the mandatory first resort, not an option running alongside continued urgency-based sole-source authority. Urgency justifies speed. It doesn't require abandoning a competitive pool that was built to provide exactly that speed in the first place. Performance against pre-negotiated terms is reviewed after every activation, with regulatory consequences — including removal from the pool — for contractors who don't hold to what they bid.

None of the evidence above should have taken independent audits and a congressional hearing to surface — that's the actual failure. A citizen shouldn't need GAO, a DHS Inspector General, and a Senate committee to piece together whether a noncompetitive contract stayed within its legal time limit or got a real price check. Every noncompetitive contract award — the FAR justification cited, the award date, the running 150-day clock, and whether the required independent cost/price analysis was actually completed — is published on the same real-time Public Transparency Dashboard this platform proposes elsewhere (see the Transparency Dashboard policy point above), the moment the contract is awarded, not discoverable years later in an audit report. If a contract blows through its time limit or skips the required price check, that shows up on the dashboard automatically — it doesn't wait for GAO to notice it in a sample review a decade on.

Classified and highly specialized work: build the competitive bench in advance, don't treat a shrinking pool as a permanent excuse. A standing procurement portal allows companies to apply for the security clearances and technical vetting needed for specialized national-security-relevant work on an ongoing, open basis — not only when an agency has an immediate need and one already-cleared incumbent happens to be available. This grows the competitive pool of cleared, qualified contractors continuously, rather than accepting an ever-narrower field of existing cleared vendors as a permanent justification for skipping competition.

Small purchases: standing vetted-supplier contracts with guaranteed markups, recompeted on a fixed cycle. Nobody runs a formal competition for $500 worth of paper — but that purchase happens under a standing supply contract with a vetted vendor at a pre-negotiated, capped markup, the same model GSA already uses for its Multiple Award Schedule contracts. The difference: every such standing supply contract — office supplies, electronics, sanitary supplies, and similar categories — goes back out for competitive bid on a fixed cycle, on the order of every five years, so a standing arrangement never quietly becomes a permanent, unreviewed sole-source relationship by default.

The common thread across all four: competition and transparency aren't abandoned in situations where a live, real-time bid is impossible. The competitive moment just moves to where it's actually achievable — at initial purchase, at the annual pool procurement, at the point of clearance vetting, at the recompete cycle — so real-time acquisition still happens under terms that were actually competed for, even when nothing is being competed for in that exact moment.

Dedicated Funds Stay Dedicated

When a fee is collected for a stated purpose, it funds that purpose — not whatever budget gap happens to need filling that year.

This platform doesn't need a new bureaucracy to fix this — it needs two mechanisms already in this platform actually applied to the problem. Single-use bills, already a Public Trust commitment, already prohibit exactly the kind of maneuver that makes fund diversions possible in practice: burying a dedicated-fund redirection inside a larger omnibus budget bill where it draws no independent scrutiny or vote. Any redirection of a dedicated fund has to stand as its own bill, voted on for what it actually is, not folded into must-pass budget language. And every dedicated fund's actual revenue and expenditure is tracked and published on the same real-time Public Transparency Dashboard already proposed elsewhere in this platform, so anyone who pays a dedicated fee can see exactly where that specific dollar went, not just trust an agency's word for it after the fact.

Data Center Tax Breaks Get the Same Accountability as Every Other Public Dollar

States and localities spent years competing to hand out data center tax breaks — bigger incentives, faster approvals, fewer questions. The bill is coming due, and it's large: Georgia estimates $2.5 billion in losses to data center tax breaks in a single year; Virginia $1.6 billion; Texas $1 billion. States are reversing course fast — more than a dozen have introduced moratorium bills in 2026 to pause new projects while they figure out what the actual return has been.

This is a local and state decision, not a federal one — tax incentive deals are negotiated and approved by state legislatures, governors, and local economic development boards. A Sunsetter serving in any of those roles brings the same standard this platform already applies to federal grant spending elsewhere ('Show Your Work Before You Ask for More'): before backing a new or renewed data center tax incentive, require a real accounting of what the previous incentive actually produced — jobs created, not promised; tax revenue generated net of the incentive's cost; and any community investment commitments actually delivered versus what was promised at signing.

There's a real, positive example worth citing rather than treating this as universally bad: Quincy, Washington has used data center tax revenue to fund a hospital and schools — a visible, spendable public benefit a resident can point to. That's the standard a Sunsetter holds every deal to, not the exception. A tax incentive that produces a hospital wing is defensible. One that produces an estimated multi-billion-dollar budget hole with nothing comparable to show for it is not — and the accounting requirement is what tells the difference apart before the next deal is signed, not after.

Benefit Corporations Get Real Preference in Federal Contracting

Most states recognize 'Benefit Corporation' as an actual legal corporate structure — not a marketing label, a real status filed with the Secretary of State that legally obligates a company to pursue a general public benefit purpose in every business decision, assessed against a third-party standard, with mandatory reporting and real shareholder enforcement. This is worth distinguishing clearly from 'B Corp' certification, a private designation from the nonprofit B Lab based on a proprietary scored assessment — voluntary, fee-based, and enforceable only through B Lab's own internal review, not a court. This platform's preference is built on the legal status, not the private certification, because a federal contracting advantage should rest on a real, government-recognized, legally accountable structure.

Federal procurement gives scored preference specifically to companies holding genuine Benefit Corporation status — not a lesser or differently-named substitute. Three states (Washington, California, and Florida) offer an alternative called a Social Purpose Corporation instead of or alongside Benefit Corporation status, and Washington currently has no Benefit Corporation statute at all — only the weaker Social Purpose Corporation option, where public benefit consideration is optional rather than required, no third-party assessment applies, and directors are shielded from monetary damages for failing to deliver the stated benefit. This platform does not lower the federal standard to accommodate that gap. A Washington company cannot qualify for this preference today, and that is the point: this is a deliberate incentive for state legislatures — including Washington's own — to adopt a real Benefit Corporation statute, the same way federal highway funding was historically used to push states toward adopting the 21 drinking age and seatbelt laws. The fix belongs at the state level, and a real federal incentive is what makes fixing it worth a state legislature's time.

This is the carrot half of a carrot-and-stick approach already used throughout this platform: rewarding a real, legally binding commitment to public purpose with real federal business, rather than treating corporate public-benefit commitments as symbolic marketing with no government-side consequence either way.

No Trading Stocks While Writing the Rules That Move Them — Instantly, By Name, In Public

A full ban on stock trading by elected officials needs something that isn't coming: 100% Congressional agreement to bind itself. This platform takes a different route to the same goal — instead of banning the trade, make every trade instantly and exactly visible to the public, by name, the moment it happens.

Here's how weak the current system actually is. Under the STOCK Act of 2012, members of Congress must file a Periodic Transaction Report within 45 days of a trade — not immediately, 45 days. The report doesn't even disclose the exact dollar amount, only a broad range ($1,001–$15,000, up to $1,000,001–$5,000,000+). It's an honor-system filing with no independent verification. And in 2013, one year after the STOCK Act passed, Congress scrapped its own plan to build a proper, searchable public database — every 'track a politician's stock trades' tool that exists today exists because private companies scrape and parse individual PDF filings the government never made usable itself.

This platform's fix: a mandatory federal reporting intermediary sits between a covered official's brokerage account and the public. When a trade executes, it's logged and published to a public dashboard immediately, under the official's real name — not 45 days later, not in a dollar range, not buried in a PDF nobody indexed. The exact security, the exact amount, the exact time, the person's actual name — the same names already public today, just instant and exact instead of delayed and vague.

This isn't unproven technology. FINRA's Trade Reporting and Compliance Engine (TRACE) already requires real-time public reporting of bond market transactions, and the SEC approved shrinking that reporting window from 15 minutes down to 1 minute in 2024. Stock trades are actually easier to capture this way than bonds were — equity exchanges already report trades to a consolidated public tape in near-real-time as a normal function of the market, unlike the bond market TRACE had to build transparency into from nothing. This platform doesn't invent new market infrastructure; it requires that trades belonging to covered officials get flagged and pushed to a dedicated public dashboard as they happen.

Coverage: elected members of Congress (House and Senate), Senators appointed by a governor to fill a vacancy (same legislative power and vote as an elected Senator, arguably an even stronger case for transparency since they haven't yet faced voters at all), Senate-confirmed executive branch appointees, and the spouses and dependent children of everyone covered — the same family-member scope the STOCK Act already uses today, just enforced instantly instead of on a 45-day honor system.

Prediction Markets: End the Tax Ambiguity, Keep Them Onshore, Require Real Oversight

Prediction markets — platforms where participants trade binary yes/no contracts on the outcomes of real-world events — are a genuinely useful information aggregation tool. The Federal Reserve published a study in February 2026 finding Kalshi 'a useful method for measuring macroeconomic expectations.' From 2021 to mid-2026, they've demonstrated documented precision on economic and financial forecasts. They work, when they're built right.

The problem is that right now they exist in a genuine regulatory and tax mess that benefits no one. On taxes: the IRS has issued zero guidance as of mid-2026 on how prediction market gains should be classified. Three treatments are all in play simultaneously — gambling income (ordinary rates up to 37%, losses largely non-deductible), capital gains (most common in practice), and Section 1256's 60/40 split (the most favorable, limited to CFTC-regulated exchanges and legally contested even there). Two traders with identical economic outcomes on the same platform can file completely differently and pay materially different tax bills, both with professional support for their position. That's not a tax system — it's a lottery based on which accountant you found.

This platform's position on taxes: prediction market gains are capital gains. The CLOB (central limit order book) mechanism that regulated prediction markets use makes this structurally coherent — every position is created when two real counterparties agree on a price through an exchange order book, the same mechanism used by stock and options markets. You're not wagering against the house; you're trading a financial instrument against another market participant, with a market-determined price and an objectively specified resolution condition approved in advance by a regulated exchange. That's a capital asset, not a bet, and the tax code should say so explicitly rather than leaving millions of taxpayers to guess.

On structure: only onshore business entities — US-incorporated, US-domiciled, CFTC-registered as Designated Contract Markets — may host prediction markets available to US participants. Polymarket's offshore blockchain-based model, where anyone can deploy a contract on a public decentralized network with no regulatory pre-approval of what gets listed, doesn't qualify. This isn't arbitrary: the capital gains classification is most defensible precisely when the instrument has objectively specified settlement conditions, approved by a regulated exchange before trading opens, resolved against a publicly verifiable data source — the Kalshi model. The Polymarket model, where a random user deploys a market at midnight on a blockchain, can't support the same capital asset characterization because the instrument's integrity depends on the platform's gatekeeping, and there is no platform doing the gatekeeping.

On market creation specifically: contracts on regulated prediction markets are created by the platform's own market team, not by users. Every proposed market goes through internal review and CFTC-standards compliance before going live — clear event definition, objective settlement data source, specified resolution date, permitted event category. Users can suggest ideas but cannot deploy contracts themselves. This gatekeeping is what separates a legitimate information market from an insider trading machine: if anyone could create a 'Will [specific thing only I know will happen] happen?' contract, the market isn't aggregating information, it's extracting it from people who don't have it yet.

One direct connection to the congressional trading transparency mechanism already in this platform: the same insider-information concern that motivated mandatory real-time disclosure of elected officials' stock trades applies to prediction market trades by those same officials. A Senator who knows a bill is about to pass before markets do shouldn't be able to trade that information on a prediction market any more than on a stock exchange. The real-time trading dashboard already proposed for elected officials and Senate-confirmed appointees covers prediction market positions on CFTC-registered platforms under the same real-name, immediate-disclosure requirement — not as a separate rule, as a natural extension of the same underlying principle.

Lobbying Reform: Mutual Disclosure With Real Consequences, and a Revolving Door That Actually Closes

Federal lobbying spending hit $5.3 billion in 2025 — a tenth consecutive annual record. The existing Lobbying Disclosure Act requires quarterly filings, a 20% time threshold before registration kicks in, and a one-to-two year cooling-off period before officials can lobby their former agencies. None of these are working. Quarterly filings mean a lobbying contact made in January isn't disclosed until April. The 20% threshold means a team of government affairs staff spending 19.9% of their time each on lobbying — collectively a full-time operation — reports nothing. And a two-year cooling-off period before someone can lobby their former colleagues isn't a door closing, it's a door propped open with a timer.

Real-time mutual disclosure. Every lobbying contact — any communication between a registered lobbying entity and a covered federal official or Congressional staff member — is disclosed by both parties within 24 hours of the contact occurring. Not quarterly. Within 24 hours. The disclosure is published immediately on the same real-time Public Transparency Dashboard already proposed elsewhere in this platform: who contacted whom, on behalf of which client, on which issue, at what time.

Both parties disclose. This is what gives the mechanism teeth that unilateral disclosure alone never has — a lobbying entity that doesn't disclose a contact the official already disclosed has nowhere to hide, and vice versa. Mismatches between the two disclosures are automatically flagged on the dashboard, visible to any journalist, researcher, or citizen who looks.

Consequences for non-disclosure are asymmetric by design. For a lobbying entity that fails to disclose a contact the official disclosed: immediate loss of federal lobbying registration for the firm, plus a fine. Loss of registration means the firm cannot legally conduct any further federal lobbying activity — that's the catastrophic financial consequence a fine alone doesn't produce for a well-resourced firm. The fine proceeds go to the general fund. For a government employee or official who fails to disclose a contact the lobbying entity disclosed: a mandatory ethics investigation that must begin within 30 days of the mismatch being identified and conclude within 90 days. Findings — including any finding of no wrongdoing — are publicly posted on the Transparency Dashboard. The investigation cannot be quietly shelved. The 30/90-day clock is statutory, not discretionary.

The revolving door: relationship-based, not time-based. The current revolving door rules — a one-to-two year cooling-off period before a former official can lobby their former agency — are time-based. Time passes, the door opens. This platform replaces the time-based rule with a relationship-based prohibition that is both more precise and genuinely more durable.

A former government employee is permanently prohibited from joining any lobbying firm that lobbies the specific agency they worked at, for as long as any person in their immediate organizational unit at that agency is still employed there. 'Immediate organizational unit' is defined as: anyone who reported directly to the former employee, anyone who was a peer in the same group or division, and anyone the former employee reported directly to. Not the entire agency — that would be both overinclusive and easy to game by structuring around — but the actual people whose relationships with the former employee create the specific conflict of interest.

When every person in that unit has left the agency, the prohibition lifts naturally. The clock isn't time — it's the persistence of the actual relationships that create the problem. A lobbyist who can no longer call a former colleague because that colleague has moved on has genuinely lost the access the rule is designed to prevent. A lobbyist who left two years ago but whose entire former team is still in place has not — and the current rule would let them walk back in anyway. This prohibition applies to joining any lobbying firm at all that lobbies the relevant agency — not to a specific issue or client — because the relationship is the problem, not the subject matter. A former EPA official whose entire former division is still at the EPA cannot join a firm that lobbies the EPA, regardless of which client or issue they'd be working on.

Non-Compete Agreements Are a Federal Contracting Disqualifier

The FTC tried to ban non-compete agreements nationwide in 2024. A federal court blocked it before it ever took effect, ruling the agency exceeded its statutory authority. The rule was formally struck from the Code of Federal Regulations in February 2026. The path to a federal ban runs through Congress passing actual legislation — and that legislation hasn't moved.

This platform's approach doesn't require a ban, a new enforcement agency, or a federal preemption of state contract law. It requires one thing: a company that imposes non-compete agreements on its employees cannot bid on federal contracts.

The federal government spends over $700 billion annually on contracts. That purchasing power is the largest single lever the executive branch controls over private business conduct — the same lever this platform uses elsewhere to prefer Benefit Corporations, remote-capable contractors, and companies with no sole-source pricing history. Here it's used to condition federal business on a company's treatment of its own workforce: if you want government contracts, you don't lock your workers out of the labor market when they leave.

Non-compete agreements affecting an estimated 30 million American workers — roughly one in five — restrict where those workers can go, what jobs they can take, and how much bargaining power they have at their current employer. They suppress wages, reduce labor mobility, and concentrate economic power with employers at the expense of workers. The FTC's own economic analysis, before the ban was struck, estimated the rule would have increased workers' earnings by $298 billion over 10 years and generated 8,500 new businesses annually from workers freed to start competing firms.

None of that requires the federal government to regulate private employment contracts directly. It requires the federal government to decide what kind of employer it does business with — which is a procurement standard, fully within existing executive authority, the same category of decision the federal government already makes about contractor safety records, labor law compliance, and environmental violations.

On October 8, 2026 — five days before this plank was written — Vice President JD Vance stood at a White House podium and announced Microsoft's suspension from the federal Permanent Labor Certification Program. His stated reason: companies that lay off American workers while simultaneously importing foreign workers on employer-dependent visas are replacing Americans with "foreign indentured servants." "You cannot lay off American workers and then replace them with foreign indentured servants," Vance said. This platform borrows that principle — not the immigration policy conclusion Vance drew from it — and applies it consistently: a worker on an H-1B visa who cannot change employers without losing their immigration status is economically captive to that employer; a worker under a non-compete who cannot take a competing job without being sued is also economically captive to that employer. The mechanism differs; the underlying condition is the same. If the current administration finds employer-imposed captivity objectionable enough to suspend Microsoft from federal programs over it, this platform finds it objectionable enough to disqualify companies that impose non-competes from federal contracting — applying the same principle without the immigration restriction.