The headline reads like a verdict on Washington. The full data set supports a narrower one. The drop is not uniform, it runs almost entirely through one party, and it sits beside a figure that barely moved: 92% of Americans still say the federal government has the power to make a positive difference in their lives. The public has not concluded that government is powerless. It has concluded, in this survey at least, that government is underperforming. This connects to our earlier piece, 100 Percent Drug Tariffs Set: What the Pharma Proclamation Does. This connects to our earlier piece, 100 Percent Drug Tariffs Set: What the Pharma Proclamation Does.

Government News coverage often treats single-year polling moves as trend lines. This one deserves more caution.

How big is the drop, really?

Eight points is a real move in a probability-based survey. But the survey measured four institution types, and the federal government was the only one to fall sharply. Perceived effectiveness of businesses fell four points, to 56%. State and local government lost three points on trust. Charitable and advocacy organizations were unchanged. A decline that hits one institution while three others hold roughly steady says something about that institution's year, not about a general collapse in institutional faith.

The survey fielded May 4-11, 2026, with 3,270 U.S. adults drawn from the probability-based Gallup Panel. That is a solid methodological footing for the snapshot. What it cannot settle is why the number moved, and Gallup's write-up does not claim to know.

Who moved the number?

Here is the part that should temper any sweeping interpretation. Among Democrats, trust in the federal government fell from 33% to 21%, and effectiveness ratings dropped from 54% to 41%. Among Republicans, trust rose from 36% to 41%, and their effectiveness ratings held steady. The eight-point national decline is, in effect, a Democratic decline offset partly by a Republican gain.

According to Gallup's 2026 findings, Democrats' trust fell while Republicans' rose over the same year, so the national decline reflects a partisan split in this survey rather than a uniform drop. A second opinion on this data should say so plainly. If the drop mostly reflects the party out of power reassessing, the finding is less a referendum on government's actual performance than on who is running it. That does not make the number meaningless. It makes it a different story than "Americans have lost faith." For related coverage, see FTC Puts Numbers on Who Controls US Ethanol Production.

The counterargument is that the decline is real regardless of its source, because the number itself is what matters for governing. A 43% effectiveness rating constrains any administration, whichever party benefits from the next shift. That reading has force. But it argues for watching the partisan composition over time, not for treating one year's swing as durable.

The 49-point gap is the more interesting finding

The widest gap Gallup measured is not between parties. It is within each respondent. Ninety-two percent say the federal government has some or a great deal of power to improve lives. Only 43% say it is effective at doing so. That 49-point spread is the largest of any institution measured; state and local government shows a 35-point gap, businesses 32 points.

Charitable and advocacy organizations are the only group whose effectiveness rating, 77%, matches or exceeds its perceived power, 74%. Everywhere else, Americans believe the capacity exists and the delivery falls short.

That gap cuts both ways for anyone interpreting the data. It supports the pessimistic read: the public expects much and rates delivery poorly. It also supports the optimistic one Gallup itself offers: Americans believe better outcomes are possible, which is a very different posture from resignation. A public that thought government was powerless would not produce a 49-point gap. It would produce agreement between the two numbers.

What does this change for Washington?

For the machinery of government, the practical implication is modest but real. The numbers point to an expectation problem rather than a legitimacy problem. Congress and federal agencies still operate with public permission to act; what has eroded is confidence that action produces results. That distinction matters for how any administration sells its work, and it is a harder sell than simply being trusted.

It also matters for readers trying to place these numbers in context. In this survey, trust in the federal government, at 26%, sits far below trust in state and local government, at 47%, and far below charitable organizations, at 80%, according to Gallup.

The honest takeaway is this: Gallup has documented a sharp, partisan-tilted one-year decline in perceived federal effectiveness, set against unchanged beliefs about federal power. Whether that becomes a durable shift depends on what happens after May 2026, and no one, including this publication, can know that yet. What the evidence supports saying is that Americans still think their government could do the job. They just do not think it is doing it.

So how does early voting work in practice? A voter shows up at a designated location during a set window before Election Day, checks in, receives a ballot, marks it, and feeds it into a tabulator or a secure ballot box. A mail voter requests a ballot, receives it at home, marks it, signs an envelope, and returns it by mail or at a drop-off point. From that point forward, the two ballots travel remarkably similar roads.

The word itself is simple. Something that is early happens at or near the beginning of a period of time, or before the usual or expected time, as Vocabulary.com defines it — and that is exactly what early voting does for Election Day: it moves the act of voting to the front of the calendar. The mechanics behind it, though, are anything but casual. They are built around verification, storage, and counting rules that apply to every ballot regardless of when it was cast.

What does early voting actually involve?

Early voting is in-person voting that takes place before Election Day at locations a state or county designates. The voter goes through the same check-in as on Election Day: identity is confirmed against the voter roll, a ballot is issued, and the ballot is cast on the spot. Many jurisdictions use the same machines for early voting and Election Day voting.

The details vary by state. Some states offer a long early voting window; others offer a short one or none at all, directing voters to absentee or mail options instead. Hours, locations, and dates are set by state law and administered by local election offices. What does not vary is the principle: an early ballot is a normal ballot, cast early and counted with all the others.

For readers who want the broader machinery of how election rules are set and enforced, the site's elections coverage tracks the full calendar, from primaries to certification.

How does a mail ballot differ from an early in-person ballot?

A mail ballot — often called an absentee ballot — is issued to a voter who requests it, and in some states automatically to all registered voters. The voter marks it at home, places it in a return envelope, and signs that envelope. The signature is the key step: it is the identity check that replaces the in-person check-in at a polling place.

Once the ballot arrives back at the election office, officials verify the signature against the voter's record on file. If the signature is missing or does not appear to match, many states run a cure process that lets the voter confirm or fix the problem. The site has a separate explainer on how states fix a defective mail ballot that walks through that process. This connects to our earlier piece, Absentee Ballot Cure: How States Fix a Defective Mail Ballot.

After verification, the mail ballot is stored securely until it is counted — often by a bipartisan team, and often starting before Election Day where state law allows. Early in-person ballots, by contrast, are usually tabulated on the spot. That is the main mechanical difference between the two.

What security steps apply to both?

Both systems rely on the same foundation: the voter roll and the chain of custody. In every case, officials must confirm that the person voting is the person on the roll, that each voter casts one ballot, and that the ballot is stored and counted without alteration. Early voting does this at check-in. Mail voting does it through the signed envelope and the verification review.

Other shared safeguards include:

• Bipartisan handling — teams that include members of different parties commonly process and count ballots together.
• Secure storage — cast ballots are kept in locked, tracked containers with access logs.
• Ballot tracking — many states let mail voters follow their ballot online, from mailed out to accepted to counted.
• Post-election audits — many states check a sample or all ballots against the machine tallies before certifying results.

The audit step deserves special attention. Risk-limiting audits compare hand-counted samples of paper ballots to the machine results and can trigger a larger recount if the numbers do not line up. The site's explainer on risk-limiting audits explains the statistics behind that check. For related coverage, see Risk-Limiting Audits: The Statistical Check on Election Results.

Why do states run these systems at all?

The practical reasons are straightforward. Early voting spreads demand across multiple days, which shortens lines on Election Day and gives voters who cannot take time off on a Tuesday a real alternative. Mail voting serves voters who are away from home, ill, disabled, or serving in the military, and in some states it serves everyone by default.

There is also an administrative logic. A longer voting period gives election offices a steadier workload instead of a single-day surge, and it gives voters more time to resolve registration or identification problems before the deadline. The same logic shapes other parts of the election calendar, including how jurisdictions staff the polls for a multi-week voting period rather than a single day.

What does this mean for voters?

Our analysis, based on how the two systems are structured: the choice between early in-person voting and a mail ballot is mostly about convenience, not about which ballot counts differently. Both are verified, both are stored under custody rules, and both are counted in the same tabulation as Election Day ballots. The security debate around mail ballots centers on the signature check and the return process; the debate around early voting centers on staffing and machine capacity. Neither system skips the verification steps that apply to any ballot.

The strongest counterargument — that extending the voting period increases the surface area for error or dispute — has a fair point in one respect: a longer window means more ballots in transit and more envelopes to verify. The answer election administrators give is that the verification and audit steps exist precisely to catch those errors, and that the cure process gives voters a chance to fix mistakes rather than lose their vote silently.

For voters, the practical takeaway is to check the local rules early. Deadlines for requesting a mail ballot, early voting windows, and ID requirements are set state by state, and the site's state-by-state guide to what identification counts where covers one of the most common stumbling points. If a mail ballot arrives damaged or a signature is questioned, acting quickly matters — cure windows are short.

What the structure of both systems points to is this: the reliability of early and mail voting rests not on any single safeguard but on the stack of them — roll checks, signature verification, custody logs, tracking, and audits. Remove any one layer and the argument over the system sharpens; keep the stack and the ballots, early or late, count the same way.

The scale gap is the first thing the data shows. BrightSpring generated about US$14.4 billion in revenue, split between roughly US$12.7 billion from Pharmacy Solutions and US$1.7 billion from Provider Services, with a market value near US$13 billion. Clover earns about US$2.5 billion from its insurance segment and carries a market value near US$2.5 billion. agilon manages about US$5.9 billion in Medicare-centric arrangements against a market value near US$1.5 billion. Same policy exposure, three very different revenue-to-value profiles.

This is analysis of what the screen reveals about where policy risk concentrates. It is not investment advice, and the figures below are the company metrics as reported in the Simply Wall Street article.

What does BrightSpring actually sell to the government?

BrightSpring delivers home and community-based care and pharmacy services to government-backed and insured patients, rather than running a consumer-facing retail model. Per the screen, approximately 80% of its revenue comes from specialty and institutional pharmacy services. That includes infusion therapies delivered at home, long-term care pharmacy services, and specialty drug distribution for chronic and rare disease patients.

The policy lever here is reimbursement. Pharmacy reimbursement rates are set by a mix of federal program rules and payer contracts, and a single policy-driven reimbursement lever moving faster than the market expects could reshape the company's margins. The screen's framing is worth noting: it does not claim the lever will move, only that the company's risk-reward profile depends on where it lands.

Why does Medicare Advantage matter so much to Clover?

Clover runs Medicare Advantage plans, the privately administered alternative to traditional Medicare, plus a physician software platform. Both tie directly into US Medicare funding and benefit design. Its model is built around policy-linked enrollment and benefits rather than discretionary consumer spending, which means the federal government effectively sets the size of its addressable market year after year.

The company's stated case, as quoted in the screen, is that its Clover Assistant platform positions it to benefit from the industry's shift toward technology adoption and data-driven, value-based care, supporting lower medical costs and higher medical margins. The open question the screen raises is the split: if a small policy adjustment changes how technology-driven care savings are shared between Clover, providers, and Medicare, the margin profile changes with it. The savings exist; who keeps them is a policy decision.

How does agilon's model depend on federal reimbursement rules?

agilon health runs a senior-focused primary care platform built around value-based contracts, which pay providers for outcomes rather than the traditional fee-for-service volume of visits. It manages senior care for primary care physicians through a per-member-per-month platform, earning about US$5.9 billion from Medicare-centric arrangements in the United States.

The screen quotes the company's growth argument directly: the accelerating growth of the senior population, especially among the 80+ age group, fuels multi-year demand, and rapid adoption of Medicare Advantage plans continues to favor the business. But the same mechanism that drives growth transmits risk. Regulatory tweaks to Medicare reimbursement flow straight through the per-member-per-month economics, changing how much the company keeps on each enrolled senior.

What does the comparison show about policy exposure?

Set side by side, the three companies illustrate three distinct transmission channels for federal healthcare policy:

• Reimbursement rates — BrightSpring's pharmacy economics move with what payers reimburse for specialty and institutional drugs.
• Benefit design and enrollment — Clover's market size is set by Medicare Advantage funding and benefit rules.
• Value-based contract mechanics — agilon's per-member economics depend on how federal rules score risk and outcomes.

The pattern generalizes. A business earning US$14.4 billion almost entirely from US government-backed and insured patients is not exposed to healthcare policy in the abstract. It is exposed to specific levers, on specific schedules, set by specific agencies and committees.

Where does the uncertainty actually sit?

The screen's own context is the 2026 election cycle, with heavy political spending, polling that barely moves, and policy risk that keeps grinding higher. That framing matters because healthcare policy rarely turns on campaign drama alone. It turns on rulemaking, scoring and benefit design — the machinery readers can follow in How a Federal Agency Writes a Rule and in What a CBO Score Decides for a Bill. A reimbursement change that looks sudden in a stock chart usually took months of proposed rules and cost estimates to arrive.

The strongest counterargument to the screen's thesis is that policy exposure cuts both ways and is hard to time. The article itself concedes the point, noting that its three highlighted stocks are a sample set from a larger screen of 15 companies spanning renewables, healthcare, and regulated utilities, and that fresh policy stories often move first and fast, then slow once the crowd notices. That is a fair description of the risk: by the time a policy link is obvious enough to publish, part of the repricing may already have happened.

The answer is that the exposure is real but specific. Investors and analysts tracking these firms do not need to predict elections. They need to read the reimbursement rule, the benefit design notice, and the contract mechanics — the same documents that determine what a Medicare change means for how Medicare drug negotiation works more broadly.

What does this change?

The evidence here establishes one thing with unusual clarity: for these three companies, the distance between a Washington decision and a cash flow line is short. BrightSpring's US$14.4 billion, Clover's US$2.5 billion and agilon's US$5.9 billion all sit downstream of federal payment rules. What remains unknown is which lever moves first and how fast. The screen frames that as opportunity; the honest reading is that it is concentrated risk, measurable from public documents, that rewards readers who follow the machinery rather than the headlines.

The menu is shorter than the headlines suggest. Courts can order structural relief, meaning a breakup or divestiture. They can order conduct relief, meaning rules on how the company behaves. Or the parties can settle on a consent decree that mixes the two. Each tool works, and each fails, in predictable ways. Consider the long arc of Europe's Android case: Ground News reported that the European Court of Justice dismissed Google's appeal in July 2026, leaving in place a €4.1 billion fine over Android restrictions that required phone makers to pre-install Google apps — on a system running about 70% of mobile devices. A fine is the simplest remedy of all, and even it took years to survive appeal.

What can a court actually order after a monopoly finding?

A court can order three broad things. Structural relief changes what the company owns: divestitures, spinoffs, or a full breakup into separate firms. Conduct relief changes what the company does: bans on exclusive contracts, requirements to share access, or rules against tying one product to another. Monetary relief — fines in Europe, damages in private US suits — transfers money rather than changing structure.

Judges generally prefer the narrowest order that fixes the problem. That instinct protects defendants from overreach, but it also means the boldest remedies tend to arrive only after the record shows the conduct rules cannot work. Remedies are calibrated to the violation, and appeals can trim them further.

How does a structural remedy — a breakup — actually work?

A breakup order tells the defendant to divest assets: sell a division, spin off a unit to shareholders, or split into independent companies. The court names what must go and sets a timetable. If the company cannot find buyers on acceptable terms, the judge can appoint a trustee to sell the assets anyway.

Structural relief has one great virtue: it is self-enforcing. Once two firms exist, no monitor is needed to keep them apart. It also has one great cost: courts must draw the lines, and markets do not always match the org chart. A divested unit can emerge weaker than the parent, or the pieces can keep cooperating in ways the order never anticipated. That is why breakups are rare and usually reserved for cases where the monopoly itself, not just one practice, is the problem.

What is a conduct remedy, and why do regulators reach for it first?

A conduct remedy regulates behavior. The order might prohibit exclusive dealing, force the company to deal with rivals on equal terms, or bar it from punishing partners who choose competitors. The Android case shows the shape: the European Commission objected to Google tying its apps to Android and restricting rival Android versions, and the remedy aimed at those specific practices rather than at Google's existence. This connects to our earlier piece, Metals Tariff Rejig Takes Effect August 6: Who Wins, Who Pays.

Conduct remedies are faster and easier to tailor. But they need policing. Someone — a monitor, a regulator, a court — must read the compliance reports, hear complaints from rivals, and decide what the order means when a company innovates around the edges. Enforcement can stretch for a decade, and a well-resourced defendant can litigate every ambiguity. The fee, in a sense, sits with the enforcer, and the enforcer eats it.

What is a consent decree, and who approves it?

A consent decree is a settlement filed with a court. The defendant agrees to specific obligations without admitting liability, and the decree carries the force of a court order: violating it can bring contempt sanctions. In federal antitrust cases, proposed decrees go through a public comment period before a judge approves them, and the final document typically includes a compliance monitor and a term of years.

Decrees trade breadth for speed. The agency gets enforceable commitments now instead of a judgment after years of appeals. The company avoids a liability finding that plaintiffs could piggyback on. Critics call the trade too generous; defenders call it the only realistic way to close a case. Both are right, depending on the terms — which is why the text of the decree, not the press release announcing it, is the document that matters.

What this means: how remedies play out over years

Every remedy runs on a clock measured in years. Liability takes years to establish. Appeals add more. Conduct remedies then live through their supervision period, with compliance disputes standing in for the litigation the settlement avoided. Structural remedies take years to execute even after the order, because separations of that size are corporate surgery.

Our analysis of the pattern: the remedy is where an antitrust case is really won or lost, and public attention almost never follows it there. Two practical checks help a reader evaluate any headline about a remedy. First, ask what the order actually requires — a promise to "share access" means little until the terms, pricing, and dispute process are written down. Second, ask who enforces it and for how long; an order without a monitor and an end date tends to fade. Readers who want the machinery behind the cases can follow how agencies set their competition rules in the first place, including the current review of rules for competitor collaborations, or how the government screens deals before they close through the merger review process. Readers following this should also see DOJ and FTC Open Review of Rules for Competitor Collaborations.

Money remedies look decisive and often are not. The Android fine is Europe's largest, and Ground News reports Google has accumulated close to €11 billion in EU antitrust fines over the past decades — a sum that, whatever its deterrent effect, left the company's position in mobile search and apps intact. Fines punish; they do not restructure. That is the core reason regulators on both sides of the Atlantic keep reaching for conduct and structural orders in the biggest platform cases, and why the fights over those orders will outlast the verdicts that produced them.

Where the remedy debate goes from here

The live question in competition policy is no longer whether large platforms can be challenged — courts have answered that — but what courts are willing to order against them. Structural proposals face the line-drawing problem described above. Conduct proposals face the enforcement-burden problem. The likely equilibrium, judging by what regulators have actually sought, is long, detailed conduct orders with independent monitors, backed by the threat of structural relief if compliance fails.

For businesses, the practical takeaway is procedural: remedy design is now a battleground in its own right, from the remedies phase of litigated cases to the commitments negotiated in settlements. For consumers, the effect arrives indirectly — through prices, defaults, and which apps come pre-installed on the next phone. The order that changes those defaults may be a paragraph buried on page 40 of a decree nobody reads. That paragraph is the real story.

Here is who sends those alerts, and how the system works. The design has not changed much over the years. It is built on zones, criteria, and shared language. The alerts are plain by design. That is a feature. This connects to our earlier piece, What Is the National Debt and Who Owns It?.

Who Issues the Alerts

The National Weather Service is a government agency within the Department of Commerce. It operates under the National Oceanic and Atmospheric Administration. According to Wikipedia's guide to NWS severe weather terminology, the agency runs national centers plus 122 local Weather Forecast Offices. Those centers include the Storm Prediction Center and the National Hurricane Center. Some products come from the national centers. Most local alerts come from the forecast offices. The layers work together on the same storms.

Each forecast office covers a set area, known as a county warning area. Those areas are split into smaller forecast zones. That lets alerts target the neighborhoods actually at risk, rather than a whole region. The work runs around the clock, every day of the year.

Watch Versus Warning

The system sorts alerts by timing and threat level. A watch means conditions are favorable for dangerous weather. It is a heads-up, issued hours ahead. A warning means the hazard is happening or imminent. It rests on radar or trained spotters. Advisories sit in between, for events that are serious but less severe. The difference comes down to timing.

The color codes and radio codes may look like trivia. They are not. Phones and radios use those codes to sort alerts. The same codes let one alert reach many devices at once.

• Watch: favorable conditions; usually covers a large area for hours.
• Warning: a hazard detected or imminent; act immediately.
• Advisory: notable but lower-threat conditions; stay aware.

The Case of Tornado Alerts

Tornado watches cover large areas. The Storm Prediction Center in Norman, Oklahoma issues them, and they are usually valid for five to eight hours. A tornado warning works differently. It is triggered when Doppler radar shows strong rotation in a storm. A trained spotter or a police officer who sights a tornado can trigger one too. Spotters are volunteers trained through a program called SKYWARN. The radar sees motion. People see the funnel. Both paths can set off the same alert.

Warnings are drawn as polygons around the threat, not blanket county areas. They usually last about 30 minutes, though they can run as long as one hour. For the most dangerous setups, forecasters can issue a "particularly dangerous situation" tornado watch. That signals that a major tornado outbreak may be possible.

More Than Storms

The alert system covers far more than tornadoes. The NWS defines products for winter storms, fire weather, flooding, and coastal hazards. It also covers marine conditions and tropical cyclone threats. Coastal areas get their own set of products. And it issues alerts for hazards without rain, such as extreme heat, extreme cold, dense fog, and high winds. Each product has set criteria, standard wording, and radio codes. The standard wording is the point. A flood warning means the same thing in every state. That shared language saves time when minutes matter. For related coverage, see Why Gas Prices Vary So Much by State.

Conclusion

The system works because it is standardized. The same words carry the same meanings, in every office, every time. Learn the watch-versus-warning line for your region. Then the next alert will arrive with a plan already built in. One more step helps. Check which alerts your phone can send. Turn on the ones for your area before storm season starts.

Each layer of the price has its own story. Here is how they fit together.

The Federal Floor

Every gallon of gasoline sold in the United States carries a federal excise tax. Wikipedia's entry on U.S. fuel taxes puts it at 18.4 cents per gallon for gasoline, and 24.4 cents for diesel. That federal rate was last raised on October 1, 1993. It is not indexed to inflation, so its real value has worn away over the decades. That aging floor matters. Congress has debated raising it many times, without passing a change.

Federal fuel tax money partly supports the Highway Trust Fund. A transportation secretary stated in 2007 that about 60 percent of federal gas taxes go to highway and bridge construction. Much of the rest goes to earmarked programs, including mass transit. Readers following this should also see How Congress Kills a Federal Rule Without Writing New Law.

Where States Diverge

On top of the federal tax, each state adds its own excise tax and fees. According to Wikipedia, state and local taxes and fees added an average of 34.24 cents per gallon of gasoline as of April 2019. That brought the combined national average to 52.64 cents per gallon. The spread between states is where the price gaps begin. Weights matter in that average. Many of the states with high taxes also have large populations. That pulls the national average up and leaves most states below it. This connects to our earlier piece, Trump's AI Order Puts State Laws in the Crosshairs.

The history is long. Oregon was the first state to tax fuel. It introduced a one-cent-per-gallon tax on February 25, 1919. Within a decade, every state and the District of Columbia had followed. Many states later redesigned their systems. As of 2016, nineteen states and the District of Columbia had fuel taxes that moved with prices, inflation, or fuel economy. Some states lean on fuel taxes more than others.

Taxes Are Not the Whole Story

Several states collect full or partial sales tax on gasoline, in addition to the excise tax. Sales tax moves with the price itself. Environmental rules can require special cleaner-burning blends in some regions, and those cost more to make. Distance from refineries and pipelines adds transport cost. Local competition among stations shapes the final sticker too. Crude oil costs move everywhere at once. But the local layers on top of that cost differ from place to place. That is why the base price can be stable while pump prices are not.

Why the Pump Price Looks the Way It Does

Unlike most goods in the United States, the advertised pump price already includes all taxes. There is no extra tax added later at checkout. That is why one sign can show one complete price. It is also why comparing signs across a border compares whole tax systems at once. Signs hide the mix too. Two stations a mile apart can sit in different zones for fuel rules and delivery. Their prices can differ with no tax change at all.

• Federal excise tax: the same everywhere, unchanged since 1993.
• State and local taxes: the largest source of variation between states.
• Fuel rules and shipping distance: quieter but real cost drivers.

Conclusion

State-by-state price gaps mostly come down to tax policy, fuel rules, and distance from supply. The next time the price jumps at a border, you will know where to look. It is rarely about the gasoline itself. It is about the system selling it. The pump price is a stack of decisions, not one number. Prices will still move. Now you know why.

Here is what the national debt is, how it grows, and who actually owns it.

What the National Debt Is

The national debt is the total amount owed by the federal government to the holders of its Treasury securities, according to Wikipedia's article on the national debt of the United States. At any moment, it equals the face value of all outstanding Treasury bills, notes, and bonds. These are issued by the Treasury Department and federal agencies. The debt does not generally include future promises, such as benefits that have not come due yet.

The debt grows through deficits. When the government spends more than it takes in through taxes, the gap is called a deficit, and borrowing fills it. The debt is the running total of those borrowed amounts. Congress caps the total through the debt limit, the same article notes.

The Two Big Buckets

The debt splits into two main parts. The first is debt held by the public. That is money owed to investors outside the federal government, including individuals, companies, the Federal Reserve, and foreign, state, and local governments, according to the article. The second part is intragovernmental debt. That is debt the government owes to its own programs, such as the Social Security Trust Fund, which holds special Treasury securities in its accounts. We covered a connected angle in New Order Tightens Pentagon Rules on Foreign Critical Materials.

The split matters for the answer to "who owns it." A little less than half of the total national debt is owed to the Federal Reserve and intragovernmental accounts, according to the same article. In plain terms, a huge share of the debt is money one part of the government owes to another part.

Foreign Holders of the Debt

Foreign investors and governments own a large slice too. As of December 2020, foreigners held 33 percent of publicly held US debt, about $7 trillion out of $21.6 trillion, according to the article. Of that amount, $4.1 trillion belonged to foreign governments and $2.8 trillion to foreign investors.

The largest foreign holders are well known names. As of December 2020, the top three holders of American public debt were Japan, China, and the United Kingdom, according to the same source. Japan held $1.2 trillion, China held $1.1 trillion, and the United Kingdom held $0.4 trillion. Foreign ownership has moved over time. It grew from 13 percent of the public debt in 1988 to 34 percent in 2015, the article states. In recent years it has pulled back, in share and in dollars.

Why the Debt Keeps Climbing

The debt rises and falls with the budget. It typically grows when spending goes up and shrinks when tax receipts run strong, according to the article. History shows a pattern too. The debt-to-GDP ratio has tended to jump during wars, pandemics, and recessions, and then ease afterward. Because deficits keep being financed with new borrowing, the total has climbed over the long run.

For now, the scale is simple to state. The national debt exceeds the total annual output of the US economy, according to the same article. That is why the number gets so much attention from economists and lawmakers.

Conclusion

The national debt is not owed to a faceless void. It is held by two big groups. One is the government itself, including the Federal Reserve and program trust funds. The other is the public, which includes American investors and foreign lenders such as Japan and China. The total grows when deficits add new borrowing, and its size shifts with wars, recessions, and recoveries. Knowing the owners changes the debate. Paying interest on the debt sends money to those holders, and policy choices decide how much new debt gets issued. The debt is complex, but its owners are not a secret, and the government publishes the total every day. For related coverage, see How Congress Kills a Federal Rule Without Writing New Law.

Most people can name a president. Fewer can name the person who runs the department that inspects that president's food supply. Yet the secretary's decisions often matter more to a household budget than anything said at a press conference. This piece explains where a secretary's power comes from, what its limits are, and how it reaches from a desk in Washington to a school lunch tray.

What does a cabinet secretary actually run?

A cabinet secretary is the head of an executive department. There are fifteen of them, from State and Treasury to Education and Veterans Affairs. Each one is a permanent bureaucracy created by law, unlike the White House staff, which serves the president directly.

The secretary sits at the top of that bureaucracy. Below the secretary sit deputy and undersecretaries, then assistant secretaries, then the career civil servants who do most of the day-to-day work. The secretary does not personally write every rule. The job is to decide what the department works on, how it interprets its legal authority, and who answers for failures.

Think of it like a very large company. The chief executive does not design every product. The chief executive chooses the strategy, hires the leadership team, and decides which fights are worth having. A secretary does the same, except the 'customers' are the entire public and the 'products' are rules, benefits, and enforcement.

Where does the power come from?

Two sources. The first is Congress. Statutes are often broad. A law might say the department must keep food safe or air travel safe without spelling out every detail. The department fills in those details through regulations, which carry the force of law. The secretary steers how that filling-in happens.

The second source is the president. The secretary is a presidential appointee, confirmed by the Senate, and serves at the president's pleasure. The president sets broad direction; the secretary translates it into department priorities, budgets, and personnel choices.

There is a third lever that is easier to miss: discretion over enforcement. A department cannot prosecute or inspect everything it is legally allowed to. Choosing where to point inspectors and investigators is itself a policy decision, and it belongs to the secretary.

How does that reach your kitchen table?

Follow one example. A food-safety rule written by the agriculture department determines how often meat processing plants are inspected. Inspection frequency affects recall risk. Recall risk affects what lands in your grocery cart and what it costs. No single rule moves prices much. The accumulated effect of hundreds of them does.

The same chain runs everywhere. Transportation rules shape highway design and vehicle standards. Education rules shape how federal school funding flows and what conditions attach to it. Financial rules, written with other regulators, shape what a bank may charge and what it must disclose. Our coverage of Banks Face July 21 Deadline as CFPB's Reg B Rewrite Is Challenged shows one of these rulebooks in motion, and the fight over it.

Even tariffs, which sound like a president's tool, run through department machinery. Trade actions under Section 232, which we covered in Metals Tariff Rejig Takes Effect August 6: Who Wins, Who Pays, depend on department investigations and recommendations before any proclamation is signed.

What are the limits on a secretary?

Three big ones. First, statute. A secretary cannot rewrite a law by disagreeing with it. If Congress says benefits must be paid on a certain schedule, the department pays on that schedule. Courts enforce this line, and they do so regularly.

Second, the rulemaking process itself. Major regulations generally require public notice, a comment period, and a reasoned response to comments. That process takes months or years. It also creates a record that courts can review, which makes hasty or poorly explained rules fragile.

Third, other institutions. Inspectors general audit the departments. Congress holds oversight hearings and controls the budget. Courts review agency action. None of these bodies runs the department, but each can stop it. Our explainer on What the President Can and Can't Do by Executive Order covers the parallel limits on presidential action, and most of them apply to secretaries too.

The strongest counterargument to secretary-centered analysis is that career staff matter more than the political appointee, because appointees turn over quickly while the bureaucracy persists. There is real force in that. A secretary with an eighteen-month tenure may leave before a major rule is finalized. But the counterargument has its own limit: the secretary picks which proposals rise, which get slowed, and which die quietly in a drawer. Slow-walking is a decision, and it is one only the top of the department can make.

Why confirmation and turnover matter

Because the job is presidential, transitions change everything. A new secretary arrives with new priorities, and the department's attention shifts. Rules in progress can be pulled back, finalized early, or left to expire on the calendar. Delays we have reported on, such as the one in EPA Delays Key TCE Ban Provisions Pending Court Review, show how timing decisions at the department level change outcomes for businesses waiting to comply.

Confirmation matters for a second reason: the Senate hearing is often the only moment the public gets a detailed look at how a nominee reads the department's legal authority. The answers given there, and the commitments made there, become the yardstick for judging the secretary later.

Practical steps: how to track a secretary's decisions

You do not need a Washington contact list. You need three habits.

1. Read the Federal Register entries for the departments you care about. Proposed rules, final rules, and comment deadlines are all published there.
2. Watch the department's budget request. The budget is the priority list in numeric form. What goes up, what goes flat, and what disappears tell you what the secretary values.
3. Follow the oversight record. Inspectors general publish findings, and congressional committees publish hearing testimony. Both are free and searchable.

For readers who want the wider machinery, our Government News section tracks rulemakings and enforcement actions as they move, and our broader policy coverage follows the arguments behind them.

What this means

The evidence points to a simple conclusion: a cabinet secretary is not a figurehead and not a monarch. The job is to steer a machine that Congress built, within limits that courts police, under direction from the president. The steering is real. A change in inspection priorities, enforcement focus, or rule timing can alter costs and protections for millions of people without a single new law passing.

What remains unknown, in any given case, is which of the secretary's choices will survive contact with the courts and the comment process. That is by design. The system trades speed for reviewability. Readers who understand where the levers sit can see, months before a headline, which everyday rules are about to move.

That limit is the part most people miss. An order can tell federal agencies how to interpret a rule, reorganize how they work, or set priorities for enforcement. It cannot raise a tax, fund a program Congress never funded, or command a private citizen to do anything. When a president tries, the checks kick in: courts can strike the order down, and Congress can cut off the money or pass a law that supersedes it.

The office itself is bounded in ways that are easy to forget. As the Wikipedia list of presidents of the United States notes, the president is head of state and head of government, leads the executive branch under the Constitution, and serves as commander-in-chief of the armed forces — and since the Twenty-second Amendment was ratified in 1951, no person may be elected president more than twice. Those boundaries apply to every order a president signs.

What is an executive order, exactly?

It is an instruction from the president to the people who work for the president. The Constitution never names the instrument. Presidents derive the authority from Article II, which vests executive power in the office, and from statutes Congress has passed that delegate specific duties. George Washington set the practice in motion, and every administration since has used orders to direct the machinery of government.

The practical test is simple: an executive order tells federal employees what to do with the authority they already have. It does not manufacture new authority. Think of it as a management memo with the force of law behind it — binding on the executive branch, and largely inert everywhere else.

Orders are numbered, published in the Federal Register, and they stay on the books until a later president revokes them, a court invalidates them, or Congress legislates past them. A new administration routinely rescinds its predecessor's orders, which is why policy made this way can flip with each election.

What can a president actually do by executive order?

Within the executive branch, quite a lot. A president can direct how agencies enforce a law, within the bounds the law sets. A president can reorganize offices, create advisory bodies, set hiring priorities, impose sanctions on foreign actors under existing statutory authority, and declare emergencies that unlock powers Congress has pre-approved in statute.

Trade is a live example. Tariff actions in recent years have run through proclamations and orders that invoke delegation statutes — Congress long ago wrote laws letting the president adjust imports under specified conditions. Our coverage of what the pharmaceutical tariff proclamation does and the metals tariff changes that took effect August 6 shows the pattern: the president pulls a statutory lever, and the order's reach is whatever that lever allows, no more. For related coverage, see Metals Tariff Rejig Takes Effect August 6: Who Wins, Who Pays.

Foreign affairs is the widest lane. The president commands the military, conducts diplomacy, and sets enforcement priorities for sanctions and export controls. Courts tend to give the presidency more room here, though not unlimited room — emergency powers still trace back to statutes, and a court can still find the emergency thin.

What can't a president do by executive order?

Three big things, and they account for most of the public confusion.

• Make law. Article I gives Congress the legislative power. An order can implement a statute; it cannot fill the space where no statute exists with rules binding on the public.
• Spend money. Congress holds the purse. An order cannot fund a program that appropriations law does not fund, and it cannot redirect money Congress appropriated for one purpose to another.
• Override the Constitution. No order can suspend free speech rights, due process, or any other constitutional limit. The president's oath is to the Constitution, not to his own directives.

There is a fourth limit that surprises people: an order generally binds only the executive branch. It cannot compel state governments to act, dictate how private companies run their businesses, or legislate for citizens directly. When an order appears to do those things, it is usually riding on an underlying statute — and the statute, not the order, is doing the work.

How do the courts check an executive order?

Anyone sufficiently affected by an order can sue, and federal courts decide whether the order stays within the president's legal authority. The tools are familiar. A judge can issue an injunction blocking the order from taking effect while the case proceeds. A court can vacate the order entirely if it exceeds statutory or constitutional limits. Appeals run upward, and the Supreme Court has the final word.

The pattern repeats across policy areas. A Texas court struck down an expanded merger filing form that a commission had adopted, our report on the court striking the expanded merger form explains, because the procedural vehicle did not hold up — a reminder that process failures sink ambitious government action regardless of which branch issues it. Regulators have also been barred from using reputation risk against banks by court action. Executive-branch directives get the same scrutiny. This connects to our earlier piece, Regulators Barred From Using Reputation Risk Against Banks.

Timing matters too. An order can take effect and shape agency behavior for months before a court rules. By the time litigation concludes, an administration may have partially achieved its goal — or a successor may have revoked the order, mooting the case.

How does Congress check an executive order?

Money is the sharpest tool. Congress appropriates funds, and an order without appropriations behind it is a promise, not a program. Congress can also pass a law that supersedes an order outright, since a statute outranks a presidential directive. And it can hold oversight hearings, subpoena officials, and legislate to narrow the delegations a president is invoking.

There is even a formal mechanism aimed at this. The Congressional Review Act lets Congress disapprove certain federal rules by joint resolution. It applies to agency rules rather than orders themselves, which tells you something: the framers of that statute knew the real regulatory weight usually sits in the rulemaking process, and orders often work by steering that process.

The political check is blunter. Orders that stretch authority invite legislation, litigation, and — when a president defies a court order — the constitutional remedy of impeachment. As the US Presidents List at PresidentsUSA.net records, three presidents have been impeached by the House — Andrew Johnson, Bill Clinton, and Donald Trump twice — and none were convicted by the Senate. The remedy exists; it has never removed a president.

What this means for reading the news

When a headline says a president "signed an executive order" on some policy, the right questions are: which statute is this riding on, does money already exist to do it, and who is likely to sue. An order that directs agencies inside clear statutory lines tends to stick. An order that tries to spend unappropriated money, rewrite a statute by interpretation, or reach into areas Congress reserved for itself is an opening bid in a court fight, not a finished policy.

Our analysis: the order is best understood as a steering wheel, not an engine. The engine is the authority Congress and the Constitution have already granted; the order decides where that power points. That is why the same instrument produces both durable changes — reorganizations, sanctions programs, enforcement priorities — and high-profile orders that evaporate in litigation. The instrument never changed. The legal footing under each one did.

For readers tracking a specific order, the durable facts live in the underlying documents: the order text in the Federal Register, the statute it cites, and the docket of any challenge. Everything else is commentary, including this article. More coverage of how federal decisions reach the economy sits in our Government News section, and broader policy work in our policy coverage.

The argument, in short: analysis is not a machine that produces the right answer. It is a discipline for asking what stands in the way of policies that are more effective, more efficient, and more just, and then asking whether better design can beat those obstacles or whether political mobilization has to do the work. That framing sounds obvious now. It was not obvious in the early 1970s, and Bardach is a large part of the reason it became so.

Policy analysis as a formal field was young when Bardach arrived at the Goldman School of Public Policy at Berkeley. According to Mendeloff, who studied there starting in 1972, the introductory policy analysis course was co-taught by Bardach, a political scientist, alongside Bill Niskanen, a Chicago-trained economist. The question on the table was blunt: what role did a political scientist play in a room built around economic reasoning?

Why did a political scientist belong in an economics-led field?

Because the economics of a proposal is only half the question. Mendeloff's account is clear on the point: policy decision-making requires a normative stance, and the hard questions are about obstacles to better outcomes, whether design can overcome them, and when a breakthrough in political mobilization is needed instead. An economist can price a policy's effects. A political scientist asks whether the machinery to carry it out actually exists and who has reason to block it.

That is the same question anyone who has watched a bill become a rule asks. The gap between a statute's intent and an agency's enforcement is where most policy quietly fails, which is why the mechanics of how a federal agency writes a rule matter as much as the text Congress passes. Bardach spent his career mapping that gap.

What did his regulatory work actually argue?

His 1982 book with Robert A. Kagan, Going by the Book: The Problem of Regulatory Unreasonableness, took on a problem that still looks current: how do you get standard-setters and inspectors to enforce rules in a way that pushes firms to take safety more seriously, while still leaving room for flexible and efficient responses? Rigid enforcement can be counterproductive. Lax enforcement forfeits the point of the rule. Bardach and Kagan treated that tension as the central design problem, not as an excuse to pick a side.

Anyone who has followed the long argument over regulatory enforcement style will recognize the stakes. The same tension shows up in debates over how long a federal permit takes, where speed and thoroughness pull against each other and neither side accepts the trade-off. Bardach's contribution was to treat the trade-off itself as the object of analysis.

Why did a small textbook matter so much?

A Practical Guide for Policy Analysis began as a little book and grew through more than seven editions, with later editions co-authored with Eric Patashnik. Mendeloff compares it to William Strunk's The Elements of Style: short, opinionated, and full of do's and don'ts. That format is the point. Policy analysis is mostly learned by doing, badly, and a book of prescriptions gives a new analyst something to fail against.

Yet Bardach was not doctrinaire about the conclusions. Mendeloff attributes this to his awareness of the practical difficulties involved, and quotes the governing idea: "the best of the feasible was the best." In a field that often grades proposals against an ideal, that is a discipline. It forces the question of what is actually achievable, the same discipline that drives the scoring of legislation, where a CBO score prices what is real rather than what is hoped.

What kind of thinker was he?

Mendeloff's portrait is unusually specific. Bardach was never afraid to present himself as puzzled about his own views, and when he explained why he was unsure, the issue turned out to be more complicated than the listener had believed. He enjoyed debate but engaged in a way that treated others as fellow-seekers of the truth. And his reasoning, Mendeloff writes, sometimes seemed to start in left field and often ended up scoring with a close play at the plate.

That trait is rarer than it sounds. A field that quantifies everything rewards confidence. Bardach's method rewarded stated uncertainty, on the theory that a puzzled analyst finds the complication a confident one papers over. The strongest counterargument to his style is that doubt can become an excuse for paralysis, that someone must eventually recommend. His answer, embedded in the feasible-standard, was that a recommendation is exactly what analysis owes, so long as it is honest about what it cannot know.

What does his passing change?

Mendeloff's conclusion is that the field has lost one of its giants but that, through his work, Bardach's intellectual and policy impacts will remain long-lived. The remembrance supports that in a specific way: according to Mendeloff, the textbook grew through more than seven editions and the regulatory scholarship took on a problem that still looks current, work that continues to shape how analysts frame the gap between intention and implementation. What the evidence does not establish is who carries the method forward with the same authority, and that is the open question his death leaves.

The practical implication, for anyone who works in or around government, is that the habits Bardach codified are the part of the field most at risk of being taken for granted. Ask what the obstacle is. Ask whether design can beat it. Recommend the best of the feasible, and say plainly what you are unsure of. That is a small set of instructions. Most of the field runs on them.

So how does the electoral college work in practice? Each state gets electors equal to its full congressional delegation: its House seats plus its two senators. Washington, D.C. adds three electors under the Twenty-third Amendment, bringing the total to 538. In 48 states, the winner of the statewide popular vote takes all of that state's electoral votes, which is why a candidate can win the most votes nationwide and still lose the election. USAGov states plainly that the Electoral College is not a place but a process: selection of electors, their meeting, and the counting of their votes by Congress.

The system comes straight from Article Two of the Constitution, and each state legislature decides how its electors are chosen. That detail matters more than almost anything else in the process, because it is the reason winner-take-all rules exist at all. This explainer walks through the timeline, the allocation math, the two states that split their votes, and what happens when nobody reaches 270.

Where does the number 538 come from?

The total is simple arithmetic. There are 435 voting members of the House of Representatives, 100 senators, and 3 electors for the District of Columbia. That sums to 538. Every state is entitled to at least three electors no matter how small its population, which gives voters in the least populous states more weight per person than voters in large states.

Because House seats are reapportioned after each census, a state's electoral vote count can change every decade. The current allocation, based on the 2020 census, applies to the presidential elections held in 2024 and 2028. As summarized on Wikipedia's article on the United States Electoral College, that allocation gave Kamala Harris 226 electoral votes and Donald Trump 312 in the 2024 presidential election.

One more rule from the founding era still applies: federal office holders, including senators and representatives, cannot serve as electors. Each state's political parties nominate their own slates of potential electors, and who ends up on those slates varies by state.

What happens between Election Day and the January count?

The process runs on a fixed calendar with three main stops. We covered a connected angle in The Certification Calendar: When Election Results Become Official.

1. November: On the first Tuesday after the first Monday in November, voters cast ballots. State officials count the popular vote and determine which candidate's slate of electors won. The governor then prepares a Certificate of Ascertainment, the official document listing the winning electors' names.
2. December: On the first Tuesday after the second Wednesday in December, the winning electors meet in their state capitals. They cast separate ballots for president and vice president, a requirement added by the Twelfth Amendment. The votes are recorded on a Certificate of Vote and sent to the President of the Senate.
3. January: In a joint session of Congress, the certificates are opened and counted in alphabetical order by state. The candidate with at least 270 electoral votes becomes president-elect.

The January count is where the vice president, presiding as president of the Senate, formally reads the results. The Electoral Count Reform Act of 2022 clarified that this role is purely ceremonial: the vice president cannot reject electoral votes. The broader certification machinery, from county canvasses to the Electoral College itself, is covered in our piece on when election results become official. For related coverage, see The Electoral Count Reform Act: What Congress Changed in 2022.

Why can the popular vote winner lose?

The divergence comes from winner-take-all allocation. In 48 of the 50 states, state law gives every electoral vote to whoever wins the statewide popular vote, even by a single ballot. Only Maine and Nebraska differ. They award two electoral votes to the statewide winner and one to the winner of each congressional district.

Winner-take-all is not in the Constitution. Article Two leaves the method to state legislatures, and states adopted the general ticket system during the 19th century because it maximized a state's influence over the outcome. The consequence is that votes are effectively weighed state by state rather than nationally. A candidate who racks up huge margins in some states while losing narrowly elsewhere can lose the national popular vote and still win the Electoral College.

It has happened. According to USAGov, a candidate has won the Electoral College while losing the popular vote in 2016, in 2000, and three times in the 1800s, five instances in total. Benjamin Harrison in 1888 is one of those five. The split votes in Maine and Nebraska can matter too: Nebraska divided its electoral votes in 2008 and 2020, and Maine did so in 2016 and 2020.

What is a faithless elector?

The Constitution does not require electors to vote for the candidate who won their state's popular vote, though most do. An elector who breaks that pledge is called a faithless elector, and such a vote has never changed the outcome of a presidential election.

States have not waited for Congress to act. Most states have adopted laws enforcing the electors' pledge, and the penalties vary. According to USAGov, a faithless elector may be fined, disqualified and replaced by a substitute, or prosecuted by the state. Some state laws flatly prohibit faithless electors. The practical effect is that the December meeting, though constitutionally the moment the votes are cast, almost always simply ratifies the November result.

What happens if no candidate reaches 270?

If no candidate wins a majority of electoral votes, the election goes to the House of Representatives, which chooses the president. The Senate chooses the vice president. This is called a contingent election, and it has happened twice: after the 1800 election, when the House chose Thomas Jefferson, and after the 1824 election, when it selected John Quincy Adams.

The scenario is remote but not hypothetical in design. With three or more credible candidacies, or with enough disputed electors, a majority can fail. That is the structural reason the 270 threshold dominates every campaign's strategy.

What does this mean for how campaigns run?

Winner-take-all rules concentrate the contest. Political science research finds that winner-take-all allocation tends to concentrate presidential campaign visits and advertising in a small number of competitive swing states, while noncompetitive states receive comparatively little attention. If you live in a state that reliably votes one way, the campaign largely passes you by, whatever your own preference.

Our analysis: the Electoral College is best understood as two systems layered together. The constitutional layer, electors, certificates, the January count, is stable and mechanical. The political layer, winner-take-all rules adopted by state legislatures, is what produces the swing-state map and the popular-vote divergences. Changing the first layer would take a constitutional amendment; changing the second requires only state legislatures acting one at a time.

The debate is old. More resolutions to amend the Electoral College mechanism have been submitted than for any other part of the Constitution. An amendment to abolish the system passed the House in 1969 but failed in the Senate. Supporters argue the system requires candidates to build broad appeal across the country; critics argue it does not reflect the popular will of the nation. Public opinion polling has found that about two-thirds of Americans support replacing the Electoral College with a national popular vote. Nothing in the current rules is likely to settle that argument, and the next census will redraw the map underneath it.

That gap is the door. It does not mean the field is easy to enter, and it does not mean every employer skips degree requirements. It means a candidate who can prove competence has a genuine path. This piece lays out that path: what the work actually involves, which credentials carry weight, how to build proof at home, and which first jobs get you in.

What does the work actually involve?

Cybersecurity is the practice of protecting systems, networks and data from digital attacks. According to Cisco, those attacks usually aim to steal or destroy sensitive information, extort money through ransomware, or disrupt normal business operations. The day-to-day jobs behind that definition vary more than most newcomers expect.

Some roles are analytical. A security operations center analyst watches alerts, sorts real threats from noise, and escalates what matters. Some roles are hands-on technical: firewall rules, endpoint protection, patching. Microsoft describes a typical defense stack that includes firewalls, identity and access management, and SIEM systems, which collect and analyze security data across an organization's infrastructure. Each tool in that stack is a job family.

And some roles are not technical at all. Policy, awareness training, compliance documentation and vendor management all sit under the security umbrella. A person who can write clearly and read a regulation has skills the field needs. Knowing which layer fits you is the first decision, and it costs nothing to make.

Which certifications actually open doors?

Certifications are the closest thing the field has to a trade license. They are portable, verifiable and cheaper than a degree. Hiring managers treat them as proof you studied the material and passed an independent exam.

The widely recognized entry sequence starts with fundamentals: a general cert covering security concepts, then a networking or systems credential, then a hands-on security certification. After that, specialization. Cloud security, incident response and governance each have their own advanced credentials. The right order depends on the role you picked in the last section. An aspiring analyst studies detection and monitoring. A future compliance specialist studies frameworks and auditing.

Two cautions. First, a certification alone rarely lands a job; it qualifies you to compete. Second, exam fees and study materials add up, so plan the sequence before buying anything. One credential earned and understood beats three collected and forgotten.

How do you build proof without a job?

This is where self-taught candidates win or lose. Employers want evidence you can do the work, and a home lab provides it. A home lab is simply a small practice environment you run yourself — old hardware, free virtualization software, or low-cost cloud instances — where you install security tools, break things safely, and fix them.

Practical steps that carry weight in an interview:

• Set up a small network with a firewall, a few virtual machines and logging. Learn what normal traffic looks like on it.
• Practice on legal training platforms that host deliberately vulnerable systems. Never probe systems you do not own or have written permission to test — that line is the difference between a portfolio and a felony.
• Document everything. A short write-up of a problem, your steps and the fix is a portfolio piece. Three of these beat a resume full of adjectives.
• Learn some scripting. Automating a repetitive task, even a simple one, demonstrates the mindset hiring managers look for.

Free and low-cost learning options exist across the industry. Vendors publish fundamentals material at no charge, and community colleges run certificate programs that cost far less than a four-year degree. Treat the budget like a farm input decision: spend where the return is measured, not where the brochure is loudest.

Which entry-level roles get you in the door?

Few people start as a security specialist. Most enter through adjacent IT work and move over. The common on-ramps:

1. IT help desk or desktop support. You learn how systems break, how users behave and how tickets work. Security teams notice support staff who handle incidents calmly.
2. Network or systems administrator. You manage the infrastructure security teams defend. Understanding the terrain is half the job.
3. SOC analyst, tier one. The classic first security title. You triage alerts under supervision. It is shift work, it is repetitive at times, and it teaches detection faster than any course.
4. GRC roles — governance, risk and compliance. A less technical entry for people with writing, auditing or policy backgrounds. Documentation work here builds toward security program management.

The pattern is simple: get any legitimate IT job, do it well, and angle toward security from inside. Internal moves are far easier than cold applications. Public-sector and critical-infrastructure employers are worth watching too — as we covered in Why US Water Systems Face Cyber Risk, utilities and local systems face persistent threats and steady hiring needs, often with less competition than big tech firms. Federal workforce trends also shape the market, as seen in How the Military Fixed Recruiting, where service branches rebuilt their pipelines with concrete incentives rather than slogans.

What does this mean for the hiring market?

Our analysis: the degree requirement is loosening from the employer side, not just the candidate side. The spending numbers explain why. IBM reports that IDC projects global security spending will reach $377 billion by 2028, and that the average cost of a data breach rose to $4.99 million in 2025, a 12% spike. When breach costs climb that fast, organizations cannot afford to filter out capable people over parchment. The Fortinet glossary makes the structural point plainly: cybersecurity is not one discipline but a convergence of network, information, cloud, endpoint and application security. Convergent fields hire from many backgrounds.

The counterargument deserves its say. Degree-holders argue that a diploma signals four years of sustained, verified effort, and that some employers — large financial firms, federal agencies requiring specific qualifications — still screen for it. That is true, and candidates should read posted requirements honestly rather than assume every door is open. But the same postings increasingly list 'degree or equivalent experience.' The equivalence is what a certification, a lab and two years of help desk work construct together.

What the evidence points to: the field needs people faster than universities can graduate them, the entry routes are documented and inexpensive relative to a degree, and the first job matters more than the first credential. Start where the work is, prove what you can do, and let the portfolio argue for you.

This distinction matters because the two lists are not symmetrical. A responsibility you skip has a concrete consequence — a fine, a missed summons, a penalty. A right you never use simply sits there, intact. And the strongest rights in the American system are not rewards for good behavior. They attach to persons, not to merit.

What follows is a plain-language map of the two categories, where each one comes from, and where the line between them gets blurry.

What is a right, exactly?

A right is a legal, social, or ethical principle of freedom or entitlement — a rule about what is owed to people under some legal system or ethical theory. That is how the Wikipedia entry on rights frames it, drawing on the Stanford Encyclopedia of Philosophy's observation that rights "structure the form of governments, the content of laws, and the shape of morality."

Philosophers split rights into useful categories. Natural rights are said to come from human nature itself and to apply to everyone, everywhere. Legal rights come from a society's statutes and customs — the right to vote is the standard example, because it exists only where a legal system grants it. The same source also distinguishes claim rights, which impose a duty on someone else, from liberty rights, which simply mark what you are free to do without anyone owing you assistance.

The claim-versus-liberty line explains a lot of everyday confusion. Your right to speak freely is, at bottom, a liberty: the law does not forbid you to speak. It does not mean anyone must hand you a microphone or listen. But your right to a fair trial is a claim: it obliges the government to actually provide a tribunal. Knowing which kind of right you are holding tells you what to expect when you invoke it.

Where do American rights come from?

The Constitution and its first ten amendments — the Bill of Rights — are the core source. They guarantee freedoms of speech, religion, and assembly; protections against unreasonable searches; due process of law; and more. Later amendments added the equal-protection guarantee of the Fourteenth Amendment and voting protections in the Fifteenth, Nineteenth, and Twenty-Sixth. We covered a connected angle in Due Process in Immigration Detention: The Time Limits and Safeguards That Apply.

One point worth stating plainly: most constitutional rights protect everyone on US soil, not only citizens. Non-citizens get due process. Citizens get a few extras — the vote, federal office-holding, jury service eligibility. Coverage of this distinction runs throughout our rights section, including how due process works in immigration detention and how birthright citizenship has been litigated under the Fourteenth Amendment. For related coverage, see Birthright Citizenship: The Litigation Line Running Through the 14th Amendment.

There is also an international layer. The Universal Declaration of Human Rights, proclaimed by the UN General Assembly in Paris on 10 December 1948, set out fundamental rights "to be universally protected" and has been translated into over 500 languages, according to the United Nations. It is not itself a binding US statute, but it shaped the human rights treaties that followed and remains the reference point for what rights are understood to be worldwide.

What are the actual responsibilities?

Some responsibilities are legal obligations. Others are expectations of civic life that carry no penalty when ignored. The two often get lumped together in civics materials, so it helps to separate them.

• Obey the law. The baseline. Breaking it triggers the enforcement machinery — courts, fines, incarceration.
• Pay taxes. Federal, state, and local governments fund themselves through legally required assessments. Nonpayment is enforceable.
• Serve on a jury if summoned. Jury service is a legal duty tied to the Sixth Amendment's promise of trial by jury. Courts can penalize ignored summonses, though excusals exist for hardship and other reasons.
• Register for Selective Service. Most male US citizens and immigrants must register within 30 days of their 18th birthday, per the Selective Service System's published requirement.
• Vote. Here is the wrinkle: voting is usually described as a responsibility, but in the United States it is a right, not a duty. No law compels you to show up. Turnout is a choice.
• Stay informed and participate. This one has no statute behind it. It is the soft part of the list — the part civics teachers mean when they talk about citizenship as a practice.

Notice the pattern in the hard list: each item is a claim the government holds against you. The pattern in the soft list is the reverse — those are things the system works better when people do, but nothing breaks if an individual declines.

Where the line gets blurry

Some items resist sorting. Jury service is the clearest example: it is a legal duty, but it is also described as a civic safeguard, because the jury is the citizen's point of contact with the justice system. Serving is both a burden and a protection — your duty today protects your claim tomorrow.

Voting is the mirror image. It is a right, and rights are usually things you hold against the government. But voting is also the mechanism by which citizens hold the government accountable, which is why civics materials list it under responsibilities too. It sits on both sides of the ledger, and that is not sloppy thinking; it is accurate.

Free speech has its own blur. The right protects you from government punishment for your views. It does not protect you from a private employer's response, from a platform's moderation rules, or from social consequences. A right defines the floor the law sets. It does not define the whole room.

What this means for how you use your rights

Our analysis: the practical value of this map is knowing what to demand and what to expect. If you are detained, questioned, or facing a government decision, you are in claim-right territory — the government owes you process, and there are doctrines that decide when you can sue if it fails. If you are summoned, billed, or asked to register, you are on the responsibility side, and the question is what the consequence of noncompliance is and whether an exemption applies.

The second takeaway is that rights do not maintain themselves. Every guarantee in the Bill of Rights has been contested, narrowed, or expanded by litigation — the shape of the Fourth Amendment today looks very different after the courts addressed location data, and the scope of who can sue police turns on qualified immunity. Rights are settled only until the next case.

The third is about the soft responsibilities. Nothing enforces staying informed, showing up to town meetings, or reading a bill text before forming an opinion. But the evidence of history, as Britannica's account of human rights notes, is that rights language grew out of "resistance to religious intolerance and political and economic bondage" — rights were won by people who showed up. The responsibilities that carry no penalty are the ones that keep the penalized ones rare.

The bottom line for citizens

Rights are what the law owes you. Responsibilities are what the law or civic life asks of you. The hard responsibilities — taxes, jury duty, legal obedience — are enforceable claims the government holds. The soft ones — participation, attention — are the maintenance work a self-governing system needs. And the rights themselves, whether grounded in the Constitution or in the broader human rights tradition the Universal Declaration codified, are guarantees that attach to persons, not prizes for good citizenship. Knowing which side of the line you are standing on at any given moment is most of what a civics education actually needs to deliver.

USA Daily News 24 is an independent online publication covering policy — not legal advice, and not a broadcaster.

What does Section 702 actually permit?

Section 702 is a surveillance authority that targets non-Americans outside the United States, using compelled assistance from American telecom and internet companies, without individual warrants. The catch has always been collateral: Americans texting, emailing, or calling a foreign target get swept in, and their communications sit in databases that agencies can later query. The Director of National Intelligence's 2023 annual statistical transparency report counted roughly 250,000 "new targets" of 702 collection in 2022 — a number that measures foreign targets, not Americans incidentally collected, whose totals the government says it cannot count.

What did the 2024 renewal change?

The RISAA, signed April 20, 2024 as Public Law 118-63, made three structural moves. First, the sunset: two years, shorter than prior reauthorizations, forcing the next Congress to revisit it. Second, new FBI discipline: a criminal penalty of up to eight years for knowingly misusing 702 data, tightened internal approval chains for US-person queries, and a court-ordered escalation path when the FBI seeks Americans' communications in certain national-security investigations. Third, expansion: the number of entities beyond the FBI permitted to receive 702 information grew, including, per the statute's text, immigration and border agencies — a point civil-liberties groups flagged immediately and sponsors did not dispute.

Did it require warrants for searching Americans' data?

No — and that was the fight's centerpiece. A bipartisan amendment to require a warrant before querying the US-person database failed on the House floor in April 2024 by a 212-212 tie vote, one of the stranger procedural outcomes of that Congress. The RISAA instead relies on internal approvals and audits. The FBI's compliance record is the evidence both sides cite: the 2019 FISA court opinion documented misuse in campaign and media-related searches, while the FBI's 2023-2024 compliance reports document corrective measures and falling error rates. Both records are public; both sides quote the half that helps them.

What is the strongest argument for the renewal as written?

The intelligence community's case, made on the record by the FBI director and the DNI in 2023-2024 testimony, is that 702 produces intelligence no other authority can — the FBI attributed specific counterterrorism and counterintelligence cases to it, and US allies' traffic routed through American infrastructure falls under it rather than bilateral agreements. The shorter sunset, sponsors argued, is a real accountability mechanism: the authority now expires mid-term, in April 2026, guaranteeing another public fight rather than an eight-year quiet slide.

What is the strongest argument against?

The opposition, led in the Senate by Ron Wyden and in the House by a left-right coalition, holds that internal approvals are not judicial warrants, that the expansion to immigration enforcement widens the pipeline for a discipline problem the FBI has already had, and that the two-year sunset is a concession, not a safeguard. The failed warrant amendment had a whip count that came within a single tied vote — evidence, opponents argue, that the coalition will be larger in 2026.

What does this change going forward?

The documents point to three implications. The April 2026 expiry lands in a presidential midterm period, meaning the next reauthorization fight will happen with the failures and expansions both on the record. The statutory role for immigration agencies in 702-derived information will get its first real test cases, likely in immigration courts where the provenance of evidence must be litigated. And the FISA court's oversight role is now written into the escalation path itself, a structural change from the purely internal approvals that governed before. Those are the directions the statute points; what Congress does with them is not something anyone can know in advance.

USA Daily News 24 is an independent online publication covering policy — not a broadcaster.

What did the communique actually commit allies to?

Per the Hague summit declaration, allies pledged to move to 5 percent of GDP annually by 2035, split into 3.5 percent core defense requirements and 1.5 percent for defense- and security-related investment, with a review checkpoint in 2029. The trajectory is not flat: national plans due under the pledge are expected to front-load the core 3.5 percent by 2032. Spain publicly sought flexibility on the target, and the summit's final language on national circumstances — noted in reporting by Reuters and AP from the summit — is the trace of that negotiation.

How does this change the 2 percent era?

The Wales 2014 benchmark of 2 percent was met by only a handful of allies for most of a decade; NATO's own pre-summit 2025 estimates put around two-thirds of members at or above 2 percent that year, largely driven by Russia's 2022 invasion of Ukraine. The 5 percent pledge formalizes a different problem: not whether allies will spend more, but what counts. The 1.5-point infrastructure category has no precedent in NATO's defense-planning rubric, and its definitions — set out in the communique's implementing language — will be written by national governments first and audited later. The comparison:

What is the US position on record?

The US pressed publicly for the 5 percent figure through spring 2025, with the president and the defense secretary both tying US force posture in Europe to allied spending levels in on-record statements. The communique's language meets that demand on paper. What the document does not establish is any binding consequence for missing it — the pledge, like Wales before it, rests on peer pressure, national plans, and NATO's annual review of them. The strongest counterargument to the skeptics' "numbers on paper" reading is the Wales precedent: the 2 percent target was dismissed the same way in 2014, and European defense spending roughly doubled in the decade after, per NATO published figures. The answer to that counterargument is that the doubling followed an invasion, not a communique.

What does this change?

The documents point to three consequences. National budgets through 2032 now carry a visible NATO benchmark that opposition parties can cost and attack — Spain's budget arithmetic, reported by Reuters in June 2025, is the first test case. The infrastructure category will pull transport and energy ministries into defense planning for the first time, since their budgets can now earn pledge credit. And the 2029 review, written into the communique, is the date the split's definitions get their first audit — the moment this pledge either acquires teeth or becomes the Wales target with a bigger denominator.

USA Daily News 24 is an independent online publication, not a broadcaster, and this is an explainer of public law — not a prediction about any pending vote.

Why does the October 1 deadline exist at all?

Federal fiscal years run October 1 through September 30, set by the Budget and Accounting Act of 1921 and later law. Congress is supposed to pass 12 appropriations bills before the year turns. It rarely does on time — the last time all 12 cleared before October 1 was fiscal 1997, per the CRS record on appropriations action. When the deadline passes without either full bills or a CR, authority to spend lapses: that is a shutdown, and it is a different legal animal from a CR.

What does a CR actually do?

Three things, per the standard CR template the CRS documents in its continuing resolutions reports. It extends funding for a set period. It sets the rate at mostly prior-year levels — the familiar freeze. And it carries an "anomaly" list: specific accounts Congress exempts from the freeze because flat funding would break something, from census operations in CRs of 2020 to contract deadlines in defense CRs. The anomalies are where the real decisions live; they are negotiated quietly, line by line, and they routinely favor whichever committee wrote them.

What can't agencies do under a CR?

The freeze blocks new starts. An agency generally cannot begin a new program, award a new multi-year contract, or ramp an account up or down — the Anti-Deficiency Act and CR language together make prior-year rates the ceiling and the floor. The Pentagon feels this hardest: service secretaries have testified since 2017 that CRs delay ship maintenance blocks, aircraft procurement, and military construction. Procurement lead times don't pause because the budget did, so money unspent in a frozen quarter is often money re-scoped later.

How long can CRs run?

Any length Congress writes. Recent history runs from days to, in fiscal 2017, a full-year CR covering the whole year — the worst case for planners, because it froze an entire budget year at fiscal 2016 levels, per the CRS. The pattern since fiscal 2010, per the CRS's count, is at least one CR every single year. The stopgap has quietly become the normal way the federal budget works, with full-year bills on time as the exception.

What is the difference between a CR and a shutdown?

Everything. Under a CR, government runs normally at frozen rates. Under a shutdown — a lapse, in budget language — agencies must halt non-excepted work, furlough staff, and open only functions that protect life, property, or lawful execution (the categories come from DOJ shutdown guidance developed since 1980 and updated since). The 35-day lapse of December 2018 to January 2019, the longest on record, was a lapse after a CR expired; the distinction is the single most common point of confusion in coverage.

What does the machinery point to?

Two implications the record supports. First, because anomalies have become the working budget inside every CR, the committees that draft them hold more effective spending power than the headline process suggests — a structural fact, not a scandal. Second, defense and non-defense planners now budget for the CR as the base case, holding back first-quarter obligations. The CRS notes this shifts work toward year-end contract surges, which the Government Accountability Office has separately flagged as an acquisition-risk pattern. The stopgap was built to buy time; used every year, it quietly sets policy.

USA Daily News 24 is an online publication, not a law firm, and it publishes information, not legal advice. Anyone navigating a resentencing or a juvenile transfer should rely on counsel in the sentencing jurisdiction.

What did Miller actually decide?

Miller consolidated the cases of two 14-year-olds sentenced to mandatory life without parole under state statutes that left judges no choice after conviction. A 5-4 Court held that children are constitutionally different — less mature, more vulnerable to peer pressure, more capable of change — and that a mandatory life without parole scheme deprives the sentencing court of the chance to weigh those differences and the child's circumstances. The Court stopped short of banning the sentence outright: life without parole remains available for juveniles whose crimes reflect irreparable corruption, but only through individualized sentencing that considers youth and its hallmarks. The decision sat in a line with Roper v. Simmons in 2005, which barred capital punishment for minors, and Graham v. Florida in 2010, which barred life without parole for nonhomicide offenses.

What did Montgomery and Jones add?

Montgomery resolved the retroactivity question in 2016: the Court held 6-3 that Miller announced a substantive rule, so the roughly 2,000 prisoners serving mandatory juvenile life without parole sentences could seek relief in state courts, whatever their states' usual limits on reopening final convictions. Justice Anthony Kennedy's opinion framed the rule as one that categorically prevents a certain class from receiving a punishment no one may constitutionally impose. Jones then answered what happened next: in 2021, a 5-4 Court held that a sentencing judge may impose discretionary life without parole on a juvenile without first making a separate factual finding of permanent incorrigibility — the judge's denial of a defendant's request to find incorrigibility was enough to show the mandatory life sentence rested on individualized consideration. Justice Brett Kavanaugh wrote for the majority; Justice Sotomayor's dissent argued the ruling hollowed out Miller's promise.

Why does youth matter constitutionally?

The underlying science is cited by the Court itself in Roper: adolescents' brains, particularly the prefrontal regions governing impulse control and long-term judgment, continue developing into the early twenties. The decisions translate that into legal doctrine — diminished culpability for children, heightened prospects for rehabilitation, and therefore an Eighth Amendment principle that punishment must account for the difference. Whatever one thinks of the neuroscience, its legal effect is fixed: the cases rest on the categorical distinction between minors and adults, and Jones reaffirmed rather than weakened it. That foundation also explains why the doctrine has proven durable across changes in the Court's composition.

How have states responded?

State legislatures have done most of the recent work. Pennsylvania, which once held more juvenile lifers than any state, enacted resentencing legislation in 2019 after Montgomery; dozens of formerly mandatory sentences there have been converted to parole-eligible terms. States including California, Iowa, West Virginia, and Massachusetts have barred or effectively eliminated juvenile life without parole by statute or court decision. By the mid-2020s, per Sentencing Project analyses, well under half the states still permitted the sentence in any form, and imposition nationwide had fallen to a handful of new sentences a year — the practical measure of the doctrine's effect.

What happens in a resentencing hearing?

The Montgomery cohort enters court years after conviction, and the hearing centers on capacity for change: testimony about childhood trauma, brain development, conduct in prison, education, and family ties; expert evidence on adolescent development; and the sentencing judge's discretion to impose parole eligibility, a term of years, or in rare cases life without parole again. The defendant's conduct over decades in custody carries more weight than the original offense's circumstances, which is why these hearings function less like appeals and more like new trials on the question Montgomery itself posed — who, exactly, is irreparably corrupt? Outcomes vary widely by state and judge, and post-Jones, prosecutors in some states again argue for life sentences at these hearings with the incorrigibility finding folded into the argument rather than the verdict.

What happens at the parole stage after resentencing?

Resentencing to a parole-eligible term does not mean release; it means a hearing before a parole board that many of these defendants first face decades into custody. States differ on whether Montgomery resentences make parole presumptively available at a set fraction of the term or leave eligibility to the original sentence structure. Boards weigh institutional conduct, programming, and release plans, and research on the resentenced cohort — Pennsylvania's is the largest studied — has shown low reconviction rates among those released, evidence reform advocates cite and opponents rarely rebut. The parole stage, not the courtroom, is where the Montgomery promise is actually cashed out.

What does this change going forward?

The doctrinal floor as of mid-2026 is stable: no mandatory juvenile life without parole, retroactive relief for the original cohort, and a discretionary sentence that survives Jones but is used sparingly. The remaining fights are practical — how many of the Montgomery cohort have actually been resentenced, with pending litigation in a few slow-moving states; whether extreme term-of-years sentences, like 60- and 80-year terms, evade Graham's logic through arithmetic; and whether the trend toward abolition at the state level continues into the remaining permissive states. The three cases changed the question from whether children can receive a state's harshest sentence to whether any specific child deserves it — a shift in burden that has, in practice, been the whole story. A sentence that was once the statutory default for a 14-year-old murderer is now a rare, argued, individually justified exception, and every actor in the system — prosecutor, defender, judge, parole board — now plans around that default.

That prevalence matters because a CR is not a smaller budget — it is the same dollar figure moving through a narrower set of rules. Programs cannot grow, start, or restructure. Contracting officers cannot sign the multiyear deals that lower unit costs. The result, GAO found after surveying 74 acquisition programs, is not abstract: about 49 percent reported schedule delays, and the costs of working around the freeze land on facilities contracts, vehicle programs, and training exercises months after the shutdown threat that triggered the CR has faded from the news.

What Does a Continuing Resolution Actually Prohibit?

A CR prohibits new starts and accelerated activity, full stop — that is its defining legal mechanism, not a side effect. Section 102 of the FY2026 act bars the Defense Department from obligating stopgap funds for "new or accelerated production" of projects, for "other new starts and activities," and for initiating new multiyear procurements, per the Senate Appropriations Committee's section-by-section summary of the act.

Congress can still carve out exceptions inside that freeze, and the FY2026 text does: it authorizes up to $154 million to keep the Virginia-class submarine program moving and up to $200 million for the E-7 Wedgetail airborne early-warning aircraft, both name-checked in the same section-by-section document. Those carve-outs are the exception that proves the rule — every other program not explicitly listed stays locked at last year's funding profile, with last year's scope, until a full appropriations bill replaces the CR.

How Much Do the Delays Actually Cost?

The costs are line-item and traceable, not speculative. GAO's review found a facilities-sustainment contract at Joint Base San Antonio rose from an initial estimate of $579,000 to $1,445,000 after CR-related delays pushed the award into a period of currency and market shifts. The Marine Corps' Amphibious Combat Vehicle program absorbed an additional $17.7 million across fiscal years 2022 through 2024 tied to the same kind of delay-driven timing exposure.

The pattern extends to how money gets spent once it is finally available. In fiscal years with short CRs, research and development accounts had obligated roughly 41 percent of their total funding by the end of the second quarter; in years with long CRs, that figure fell to about 31 percent, per the GAO report. Spending does not disappear — it compresses into the back half of the fiscal year, which is itself a worse way to buy weapons systems and services than a steady pace would be.

Administrative overhead is the least visible cost and, by GAO's account, one of the largest in staff time. F-35 program officials told auditors that roughly 20 percent of their financial-management staff's time goes toward replanning budgets to fit CR constraints — work that produces no additional capability, only compliance with the freeze.

Is the Damage Overstated? The Counterargument, Answered

The strongest case for continuing resolutions is that they are the alternative to something worse. A CR keeps the government funded, avoids a shutdown, and — as the FY2026 act's own drafters note — is written to "protect employee compensation" so uniformed and civilian personnel keep being paid while Congress finishes its work. Lawmakers can and do carve out named exceptions, as the submarine and Wedgetail funding shows, meaning the mechanism has some flexibility built in rather than being a blunt, universal freeze.

That argument holds for payroll and shutdown avoidance, which is exactly what CRs are designed to protect. It does not hold for acquisition and training, where GAO's own data shows the freeze has a cost that shutdown-avoidance framing does not capture: canceled training events tied to funding gaps, delayed contract awards on nearly half of surveyed programs, and a spending curve pushed toward the end of the fiscal year regardless of how carefully the CR is written. Protecting paychecks and protecting program timelines are separate problems, and the FY2026 act solves only the first one by default.

What Does This Change?

The recurrence is the story more than any single CR is. With the Pentagon under stopgap funding in 37 of the last 49 fiscal years — and long CRs, those exceeding three months, in eight of the last 15 — the "temporary" measure functions as a standing operating condition for defense budgeting rather than an occasional interruption. A 2021 Breaking Defense analysis of Army modernization programs under a prior CR illustrates why that recurrence compounds: new-start prohibitions delayed funding for programs including a $287 million Mobile Protected Firepower effort and $108 million in B-21 Raider advance procurement, pushing schedules that, once slipped, are difficult to recover.

What the evidence points to is a defense budget process where the announced top-line number and the money actually available to move programs forward are two different things for months at a stretch. Congress's own appropriators can narrow that gap with named carve-outs, as they did for the submarine and Wedgetail lines in FY2026. Absent those carve-outs, GAO's data indicates the compliance and delay costs of a CR are not shutdown-adjacent — they show up in program schedules, industrial-base planning, and financial-management staff time long after the immediate funding lapse has been avoided.

FAQ

How long can a continuing resolution last?

There is no fixed limit; duration is set by whatever Congress passes. GAO found that in eight of the last 15 fiscal years, the Pentagon operated under CRs exceeding three months, and the FY2026 stopgap itself was written to run until a full-year defense appropriations act is enacted.

What exactly counts as a "new start" that a CR blocks?

Under Section 102 of the FY2026 act, it covers new or accelerated production, other new starts and activities, and the initiation of new multiyear procurement contracts — programs that were not already funded and underway in the prior fiscal year.

Can specific programs be exempted from a CR's restrictions?

Yes, when Congress names them. The FY2026 continuing resolution explicitly authorizes up to $154 million to continue Virginia-class submarine work and up to $200 million for the E-7 Wedgetail program, exceptions carved directly into the bill text rather than left to agency discretion.

The rule is the first time Title II of the Americans with Disabilities Act has been paired with a named, enumerated technical standard rather than a general nondiscrimination mandate. Before 2024, DOJ treated web inaccessibility as a Title II violation case by case, without specifying which guidelines satisfied the law. The final rule, published in the Federal Register on April 24, 2024, closed that gap: it named WCAG 2.1 AA — the Web Content Accessibility Guidelines maintained by the World Wide Web Consortium — as the measurable bar for compliance, in a DOJ fact sheet on the rule.

The rule sorts covered entities by population. State and local governments serving 50,000 or more residents originally had until April 24, 2026, to comply; smaller jurisdictions and special districts had until April 26, 2027. On April 20, 2026, DOJ published a notice in the Federal Register extending both deadlines by one year: large entities now have until April 26, 2027, and small entities and special districts until April 26, 2028.

What does WCAG 2.1 AA actually require?

The standard covers how web content and mobile apps present information to users with disabilities. DOJ's fact sheet describes the requirement as covering captions for video, transcripts or alternative descriptions for audio, text that works with screen readers, and interfaces that function without a mouse. WCAG itself, maintained by the World Wide Web Consortium's Web Accessibility Initiative, organizes those requirements under four principles: content must be perceivable, operable, understandable, and robust across assistive technology. Level AA sits in the middle of WCAG's three conformance tiers, above the minimum Level A and below the more demanding Level AAA, which DOJ did not adopt for this rule.

The rule applies whether a government entity builds its own digital services in-house or contracts them out to a vendor — outsourcing does not exempt an agency from the requirement, and the underlying obligation runs to the government entity even when a third party wrote the code. "Web content" under the rule covers text, images, sound, video, and documents; "mobile apps" covers downloadable software built to run on phones and tablets. Both categories are covered whether the government entity hosts them directly or arranges for someone else to provide them on its behalf.

The rule does carve out specific exceptions. Archived web content kept solely for reference or recordkeeping, unchanged since it was archived and clearly labeled as such, is exempt. So are preexisting conventional electronic documents — word-processing files, PDFs, spreadsheets, and presentations — posted before an entity's compliance deadline. Content that members of the public post independently, such as comments on a government social media page, is not itself covered, though the platform hosting it must be. Individualized, password-protected documents tied to a specific person's account, property, or benefits are also excluded, as are government social media posts made before the compliance date.

Why did DOJ push the deadline back?

DOJ's stated rationale, laid out in the April 2026 Federal Register notice, is that the department “overestimated the advancement and availability of technology” available to covered entities. The notice cites resource and staffing constraints among state and local governments and states that generative artificial intelligence tools cannot yet reliably automate accessibility remediation at the scale the rule demands. The extension applies only to the compliance timeline — it does not change the WCAG 2.1 AA standard itself, the population thresholds, or the underlying legal obligation under Title II. DOJ also left a public comment period open on the extension, running through June 22, 2026, meaning the new dates are not necessarily final.

School and local-government administrators had pressed for the delay. Sasha Pudelski of AASA, the School Superintendents Association, told NPR that districts already operating under financial strain were struggling to afford the remediation costs of retrofitting existing digital systems — older course-management platforms, archived instructional video, and legacy student portals — on the original timeline.

What does the pushback from disability advocates argue?

Disability rights organizations opposed the extension on the record, and their objections form the strongest counterargument to DOJ's timeline. Corbb O'Connor of the National Federation of the Blind of Minnesota said the delay meant “the blind have been told to wait to live on terms of equality,” noting that web accessibility guidelines have existed in some form since 1999 — longer, he argued, than the technology excuse can account for. Katy Washington, president of AHEAD, an association of higher-education disability service providers, said the delay “slows critical momentum and leaves institutions without the clarity needed to fully realize equitable access.” Jennifer Mathis, who worked on the original rulemaking, called the postponement “mindless and cruel” given what she described as sixteen years of process behind the rule.

Both sides are arguing about timing, not substance: no party in the public record disputes that WCAG 2.1 AA is the correct standard, or that state and local governments are ultimately bound by it. The dispute is entirely over how much runway compliance requires, and DOJ's own notice acknowledges the criticism is about pace rather than principle by keeping the standard and thresholds intact while moving only the calendar.

What does this change?

For the roughly three years between now and the new deadlines, the legal obligation exists but the specific technical deadline has not yet arrived, so enforcement leverage rests on whatever a complainant can establish under Title II's general nondiscrimination standard rather than a missed WCAG deadline. DOJ directs residents who encounter an inaccessible government website or app toward its ADA Information Line or its online complaint process — that channel does not change with the extension, and it predates the 2024 rule.

The bigger signal is precedential. This is DOJ's second look at implementation costs for a rule it had already finalized after extensive rulemaking, and the department chose to revisit a deadline rather than the underlying standard. Whether the June 2026 comment period produces further schedule changes, or whether April 2027 and April 2028 hold, is what the record supplied so far does not resolve. What the record does establish is that the technical bar those dates lead to — WCAG 2.1 Level AA, on every covered government website and app — has not moved since 2024.

This explainer covers the federal rule governing government websites specifically; it does not address the separate body of Title III litigation over private-sector websites, which proceeds under different, less specific case law, or answer whether a particular website currently complies.

The mechanism starts with Congress's basic constitutional job. Federal agencies cannot spend a dollar that lawmakers have not appropriated. When Congress and the president fail to enact the twelve annual appropriations bills, or a stopgap measure standing in for them, before the fiscal year's funding runs out, a gap opens. The Government Accountability Office defines that gap precisely: a shutdown is "a period of time between the expiration of an appropriation and the enactment of a new one," and GAO's own accounting shows it has happened twenty times between 1977 and 2019, lasting anywhere from one day to 35.

What law actually forces agencies to stop working?

The trigger is not a shutdown order from the White House. It is the Antideficiency Act, which GAO describes as requiring that when a funding lapse occurs, "agencies generally stop their operations." The statute exists to protect Congress's constitutional power of the purse by making it illegal for the executive branch to spend or commit money it has not been given. In GAO's language, the law "prohibits agencies from incurring obligations or making payments in advance or in excess of an appropriation."

That prohibition reaches further than payroll. Agencies "generally may not spend money during a shutdown, including for employee salaries," and federal staff cannot even volunteer their labor for free, since the law bars the government from accepting "services without payment—except in very limited circumstances." In practice, this is why furloughed employees are legally barred from checking work email or logging into government systems during a lapse, not just discouraged from it.

Why do some offices stay open anyway?

If shutdowns froze every federal function, air traffic control towers and Social Security checks would stop along with passport offices and national parks. They do not, because GAO lays out what amounts to a two-step legal test agencies apply once funding lapses. First, any activity with its own funding source outside annual appropriations can continue — GAO's example is Social Security, which "is funded through a permanent appropriation, and they may continue" paying benefits regardless of a lapse elsewhere in government. Second, for everything else, only work that fits narrow emergency exceptions survives, limited to functions "necessary to protect human life and government property."

That second category is why air traffic controllers, active-duty troops, and federal law enforcement keep reporting for duty during a shutdown while regulatory reviews, grant processing, and routine inspections stop. GAO is explicit that the exceptions are not a loophole for keeping government running as usual: "the ongoing, regular functions of government may not continue during a shutdown," even where a case could be made that the work matters.

Do federal employees get paid during a shutdown?

No — and this is the detail most frequently misunderstood. Whether an employee is furloughed or classified as "excepted" and ordered to keep working, both groups go unpaid while the lapse continues. As Government Executive's guide to shutdown pay and benefits puts it, "neither furloughed federal workers nor those deemed essential and forced to work during a shutdown will be paid during a shutdown." Excepted employees do eventually receive premium pay for the hours they worked without a paycheck, but only "after the government reopens," not while the shutdown is underway.

What changed after the 35-day shutdown that ended in early 2019 is that back pay is no longer optional. Government Executive notes that in earlier lapses, "Congress needed to approve back pay for furloughed workers following each shutdown" — a separate political fight layered on top of reopening the government. Since the law enacted at the close of that 2018-2019 shutdown, "all federal workers will be automatically granted back pay once funding has been restored," removing that extra round of negotiation.

That guarantee is also the strongest counterargument to treating shutdowns as pure fiscal discipline: if furloughed employees are made whole automatically, the leverage a shutdown creates falls almost entirely on agency operations and the public waiting on services, not on the workforce's paychecks in the long run — even though the missed paychecks during the lapse itself are real and immediate for households budgeting week to week.

How do stopgap bills prevent, or just delay, a shutdown?

Most years, Congress avoids finishing all twelve appropriations bills on time and instead passes a continuing resolution, or CR — what GAO calls "temporary spending bills that allow federal government operations to continue when final appropriations have not been approved by Congress and the President." A CR generally freezes funding at the prior year's levels rather than setting new amounts, buying time without resolving the underlying budget disputes. GAO notes that Congress has completed the full regular appropriations process before the new fiscal year began only three times in the last 47 years, which is why CRs have become the default tool for keeping agencies open.

A CR is not a permanent fix, and GAO warns it carries its own costs: stopgap funding produces "administrative inefficiencies and limited management options in areas such as hiring and travel" even when it succeeds at averting a lapse. The real risk moment is what happens when a CR itself expires without a follow-on bill. GAO's review of the 2010-2022 period found 47 CRs enacted, but on three occasions — fiscal years 2014, 2018, and 2019 — no CR was in place when funding ran out, and each produced an actual shutdown.

What this means the next time a deadline approaches

A shutdown fight in Washington is often covered as political theater, but the underlying process is mechanical and legally constrained. Once a funding gap opens, the Antideficiency Act does not leave agencies discretion — it dictates what must stop, what narrowly qualifies to continue, and that employees on both sides of that line go unpaid until Congress acts. Understanding that structure is what separates a genuine countdown to consequences from routine last-minute deal-making that resolves before any lapse occurs. The clearest signal to watch for is not rhetoric but the calendar: whether a CR or final appropriations bill is signed before the prior one's funding runs out.