Daniel sent us this one — and it's timely. He's asking about the US midterm elections coming up this November, specifically for listeners who aren't American. What's the actual significance of these elections? And there's a prediction floating around that the House will flip Democratic, creating a divided government. What would that mean in practice for the president's ability to lead, and how much historical precedent is there for this kind of arrangement? It's a good question, because midterms can feel like this opaque American ritual from the outside.
They absolutely do. I've had conversations with colleagues abroad where they sort of nod politely when midterms come up, but you can tell they're thinking — isn't the presidential election the one that matters? And the answer is, well, yes, but also no. Sometimes more no than yes, depending on what you care about.
The midterms are the constitutional plot twist nobody outside the country sees coming.
So let me lay out the basics, because the structure itself explains a lot. Midterms happen on the Tuesday after the first Monday in November, exactly two years into a four-year presidential term. This year that's November third. Every single one of the four hundred thirty-five House seats is up for election — all of them, every two years. And then thirty-four of the one hundred Senate seats are on the ballot this cycle. That's the class of senators elected back in twenty-twenty who are now up for reelection.
The entire lower chamber and a third of the upper chamber, all at once. That's not a small election by any standard.
It's not. And here's the thing that makes midterms structurally fascinating — they almost always punish the president's party. Since nineteen forty-six, the president's party has lost House seats in seventeen out of nineteen midterm elections. The only exceptions were nineteen ninety-eight, when the Clinton impeachment backlash actually helped Democrats gain seats, and two thousand two, when Bush had a post-nine-eleven rally effect.
The historical pattern is essentially a gravitational pull against whoever holds the White House.
The numbers are stark. Since World War Two, the president's party has lost an average of twenty-six House seats in midterms. That's the "midterm penalty." In this context, the current prediction that Democrats will flip the House — they need to flip just five seats — is actually modest by historical standards. The current House breakdown is two hundred eighteen Republicans, two hundred thirteen Democrats, and four vacancies. A net gain of five seats for Democrats would flip control.
That five-seat threshold is almost absurdly low. It's like the margin is a rounding error.
It really is. And the polling right now, according to FiveThirtyEight and RealClearPolitics averages, shows Democrats with about a six-to-eight-point advantage on the generic congressional ballot. That's the poll question that asks "would you rather vote for a Democratic or Republican House candidate in your district" without naming specific candidates. If that margin holds, history suggests a flip is likely.
That's the "what" and the "when." But I want to get at Daniel's core question — why should someone in Dublin or Delhi or Dakar care about this? What actually changes in their lives?
The short answer is: a lot more than most people think. US foreign policy, trade agreements, defense spending, immigration law — all of these shift dramatically depending on which party controls Congress, and none of them require a presidential election to change direction. A divided government can freeze or accelerate any of these.
The US president can't just...
This is the single biggest misconception people have about the American system. The president is not a prime minister. He doesn't command a legislative majority by definition. The Constitution separates powers deliberately. Article One, Section Nine gives Congress — specifically the House — the power of the purse. The president proposes a budget, but Congress appropriates the money. If the House doesn't want to fund something, it doesn't get funded.
Even if the president wants to, say, build a border wall or launch a new military initiative or fund a trade enforcement mechanism, a hostile House can simply refuse to write the check.
And that's just the spending power. The House also has subpoena power, investigatory power, and the sole power of impeachment. When the House and the presidency are held by different parties, the Speaker of the House becomes a de facto opposition leader with genuine constitutional leverage. It's not ceremonial. The Speaker can block the president's entire legislative agenda simply by refusing to bring bills to the floor.
That's the part I think gets lost in translation. People imagine the American president as this all-powerful executive, and in foreign policy and military command there is significant unilateral authority, but domestically the president is surprisingly constrained without Congress.
Let me give you a concrete example of what divided government looks like in practice. Nineteen ninety-five to nineteen ninety-six. President Clinton is a Democrat. The House flips Republican in the nineteen ninety-four midterms — Newt Gingrich becomes Speaker. What follows are two government shutdowns totaling twenty-seven days, because the Republican House refused to pass a budget that funded Clinton's priorities. And Clinton used the bully pulpit — that's the president's ability to dominate public attention — to blame the Republicans, which boosted his approval ratings and helped him win reelection in ninety-six.
The shutdown was a tactical failure for Gingrich but a demonstration of real power. The House actually shut the government down. That's not symbolic.
It's about as concrete as power gets. The federal government stops functioning. National parks close. Federal employees go home. It's a constitutional crisis as bargaining chip. And the mechanism is straightforward — the House simply refuses to pass appropriations bills. No money, no government.
Which brings us to the other lever: investigations and impeachment. A Democratic House in twenty twenty-seven would have subpoena power over the entire executive branch.
We've seen this play out twice in recent memory. Twenty nineteen to twenty twenty, a Democratic House impeached Trump while a Republican Senate acquitted him. The House hearings dominated the news cycle for months, they produced testimony from administration officials, they shaped the public narrative heading into the twenty twenty election. Divided government turns the House into what amounts to a permanent oversight committee with subpoena power.
It's a political theater mechanism, but that phrase undersells it. Political theater has real consequences for governance. When every cabinet secretary knows they'll be called to testify, when every policy decision might trigger document requests, it changes how the executive branch operates.
It slows everything down. The administration's political appointees spend enormous time preparing for hearings, responding to document requests, lawyering up. It's not just theater — it's friction. The founders designed the system this way.
Let me ask the question I think a lot of listeners are probably asking. If Congress blocks everything, can't the president just govern through executive orders?
This is the second big misconception. Executive orders are fragile. They can be overturned by the next president with a stroke of a pen — that's literally what happens on inauguration day, the new president signs an order rescinding the previous administration's orders. They can be blocked by federal courts, which has happened repeatedly. The Trump travel ban executive orders in twenty seventeen went through multiple court challenges and took months to partially implement. And critically, executive orders cannot appropriate money. The president cannot spend a single dollar Congress hasn't authorized. So you can't fund a border wall, or a new weapons system, or a climate initiative, or anything requiring actual dollars through executive order.
Executive orders are basically the president's ability to rearrange the furniture in a room where Congress controls the electricity, the plumbing, and the lease.
That's a very good way to put it. You can change how existing programs are administered, you can set enforcement priorities, you can direct agencies to emphasize certain things over others. But you can't create new programs, you can't spend new money, and anything you do can be undone by the next president or blocked by a judge.
Let's talk about the Senate side of this, because the current projections show a more complicated picture than just "Democrats take the House." What's the Senate outlook?
The Senate is currently fifty-three Republicans to forty-seven Democrats. Thirty-four seats are up this cycle — twenty held by Republicans, fourteen by Democrats. The current projections from Cook Political Report and Sabato's Crystal Ball suggest the Senate is likely to remain Republican, possibly at a fifty-fifty split with Vice President Vance as the tiebreaker. That's a crucial detail because it creates what I'd call a trifecta of tension.
The House passes bills — now controlled by Democrats in this scenario. The Senate — still Republican — kills them. Anything that somehow survives both chambers faces a presidential veto. And overriding a veto requires a two-thirds majority in both chambers, which almost never happens. So you have three separate choke points. Each chamber can block the other, and the president can block both.
The legislative pathway becomes essentially impassable. What actually gets done in that environment?
This is where the history gets interesting, because "nothing gets done" is itself a misconception. Major legislation has passed under divided government. The nineteen ninety-six welfare reform — the Personal Responsibility and Work Opportunity Act — was passed by a Republican Congress and signed by a Democratic president, Bill Clinton. No Child Left Behind in two thousand one passed with bipartisan support under a Republican president and a divided Congress. The FIRST STEP Act in twenty eighteen, which was criminal justice reform, passed under Trump with a Republican Congress but significant Democratic support.
Divided government doesn't mean gridlock by default. It means that the only things that pass are things that can build a genuine cross-partisan coalition.
And that's a feature, not a bug, in the constitutional design. The system is built to slow down change. James Madison wrote about this in Federalist Fifty-one — ambition must be made to counteract ambition. The idea is that dramatic policy shifts should require broad consensus, not just a narrow majority.
Which is philosophically coherent, but it also means that in practice, budgets become continuing resolutions instead of comprehensive bills. Judicial appointments slow down. The debt ceiling becomes a recurring hostage negotiation.
Let's talk about the debt ceiling, because that's one of the most important knock-on effect. The debt ceiling is a statutory limit on how much the federal government can borrow to pay obligations Congress has already authorized. It's not about future spending — it's about paying existing bills. But in divided government, it becomes a bargaining chip. The twenty eleven debt ceiling crisis is the textbook case. A Republican House, under Speaker Boehner, forced President Obama into the Budget Control Act, which created sequestration — automatic spending cuts that neither party actually wanted. It was a suboptimal policy outcome produced entirely by the leverage dynamics of divided government.
The phrase "hostage negotiation" is not an exaggeration. One side says "we will allow the United States to default on its debt — which would trigger a global financial crisis — unless you agree to our policy demands." That's the actual dynamic.
It's happened repeatedly. Twenty thirteen, twenty twenty-three. Every time we approach the debt ceiling under divided government, there's a standoff. The twenty twenty-three dynamic is instructive for what might happen in twenty twenty-seven. When Biden faced a Republican House under Speaker McCarthy, he used executive orders extensively — on student loan forgiveness, on climate policy — while Congress gridlocked on appropriations. The pattern is likely to repeat if the House flips.
The president shifts to what they can do unilaterally, the House uses its funding and investigatory powers to constrain and harass, and the Senate becomes either a firewall or a graveyard depending on which party controls it.
In this particular configuration — Democratic House, Republican Senate, Republican president — you'd have an unusual dynamic where the Republican Senate is protecting a Republican president from a Democratic House. That's different from the Trump impeachment scenario, where you had a Republican Senate protecting a Republican president from a Democratic House. The difference now is that the Senate would also be killing Democratic House bills that the president would veto anyway. It's almost redundant opposition.
But not quite. Because the House can still investigate. The House can still subpoena. The House can still hold hearings that dominate the news. The Senate can't stop any of that. The oversight power of the House is unilateral.
That brings us to foreign policy, which I think is the area where non-Americans feel the effects most directly. Divided government often produces more hawkish foreign policy. Both parties compete to appear tough on adversaries, and neither wants to be seen as weak on China or Russia or Iran. The nineteen ninety-eight Iraq Liberation Act is a perfect example. It passed under Clinton, a Democratic president, with a Republican Congress, and it made regime change in Iraq official US policy. That was bipartisan hawkishness produced by divided government.
The dynamic is: when nobody fully owns foreign policy, both parties try to out-tough each other.
It can accelerate things that might not happen under unified government, because there's no single party that can be blamed if it goes wrong. The responsibility is diffused. Trade policy is another area. A Democratic House could refuse to approve trade agreements negotiated by a Republican president. Defense spending becomes a bargaining chip — the House might demand domestic spending in exchange for military funding. Immigration policy, which affects visa programs and refugee admissions that have direct impacts on other countries, gets caught in the same gridlock.
Let me zoom out for a moment and ask about the historical pattern. You mentioned that the US has had divided government about sixty percent of the time since eighteen fifty-four. That's a lot.
It's the norm, not the exception. Since World War Two, we've had unified government for only about forty percent of the time. The most recent unified government was the first two years of the Biden administration — January twenty twenty-one to January twenty twenty-three — when Democrats held the White House, the House, and the Senate. Before that, the first two years of Trump, twenty seventeen to twenty nineteen. Before that, the first two years of Obama, two thousand nine to two thousand eleven. Notice the pattern?
Unified government is a brief window at the start of a presidency, and then midterms break it.
Almost every time. The president's party holds Congress for the first two years, passes what they can, and then loses the House in the midterms. The only recent exception was the twenty twenty-two midterms under Biden, where Democrats actually held the Senate and lost the House by a narrower margin than expected. But they still lost the House.
The prediction of a Democratic House takeover in twenty twenty-six is basically the system reverting to its historical mean.
And given the current polling and the historical pattern, it would be more surprising if the House didn't flip. The question isn't really whether divided government is coming — it's what kind of divided government, and what the specific configuration means for governance.
Let's talk about the specific configuration then. Trump in the White House, a Democratic House, a Republican Senate. What's the closest historical parallel?
The closest parallel is probably the second half of Trump's first term — twenty nineteen to twenty twenty-one. You had a Democratic House under Speaker Pelosi, a Republican Senate under McConnell, and Trump in the White House. That period saw impeachment, the longest government shutdown in US history at the time, and relatively little major legislation after the initial two years.
We've essentially run this experiment before, very recently.
And the results were: intense oversight and investigation from the House, legislative gridlock on most domestic priorities, and continued executive action on trade and foreign policy where the president has more unilateral authority. The wild card this time is that the Republican Party of twenty twenty-six is not identical to the Republican Party of twenty nineteen. The internal dynamics have shifted.
The GOP is more Trump-aligned now than it was in twenty nineteen, when there was still a meaningful faction of traditional conservatives who were skeptical of him. That faction has largely been marginalized or retired. So a Republican Senate in twenty twenty-seven would likely be more uniformly supportive of the president than the twenty nineteen Senate was. That changes the dynamic — there's less daylight between the White House and Senate Republicans, which means the House Democrats are the sole opposition, not part of a multi-sided negotiation.
Which could make things more confrontational, not less. If the Senate is just an extension of the White House, then the House becomes the only institutional check.
The Speaker of the House becomes, in effect, the leader of the opposition in a way that's almost parliamentary. The Speaker holds press conferences, the Speaker sets the agenda, the Speaker decides what to investigate and when. During the Pelosi speakership in Trump's first term, she was arguably as visible and influential as any figure in American politics other than the president himself.
For a non-American trying to follow this, the advice is: watch the Speaker race, not just the presidential race. The Speaker of the House is arguably the second most powerful person in American politics during divided government.
And that's a great practical takeaway. When the election results come in and we know who controls the House, the next question is who becomes Speaker. That person will shape the agenda, decide what bills get votes, and control the investigatory machinery. If it's Hakeem Jeffries, the current Democratic leader, he becomes the most important counterweight to the president.
Let me ask about a scenario that I think gets overlooked. What happens if the Senate ends up fifty-fifty? You mentioned the Vice President as tiebreaker, which would keep Republican control. But that's an incredibly fragile majority.
It's a majority of one person. And Senate rules require sixty votes to overcome a filibuster on most legislation. So even with fifty-one votes, you can't pass most bills without some Democratic support. The filibuster is the great complicating factor. It means that even "unified" government in the Senate isn't really unified unless you have sixty votes, which almost never happens.
The filibuster means the Senate is always, in practice, a divided government institution.
For legislation, yes. For judicial appointments and certain budget measures, there's a process called reconciliation that only requires a simple majority. That's how the Trump tax cuts passed in twenty seventeen and how the Biden climate and health care bill passed in twenty twenty-two. But reconciliation is limited — you can only use it for budget-related matters, and only a certain number of times per year.
The practical toolbox for a divided government is: reconciliation for budget stuff, executive orders for administrative stuff, and investigations and subpoenas for political combat. That's the menu.
Continuing resolutions to keep the government funded at previous years' levels, which is what happens when Congress can't agree on new appropriations bills. That's how the government increasingly operates — on autopilot, with previous years' priorities baked in, unable to adjust to new circumstances.
Which is a kind of slow-motion institutional decay that nobody campaigns on but everybody lives with.
It's the lossy compression of governance, to borrow a phrase. Every year of continuing resolutions and executive orders and reconciliation packages is a year where the government can't respond nimbly to anything. And that has real consequences. When a pandemic hits, or a financial crisis, or a major natural disaster, you want a government that can move quickly and appropriate money. Divided government makes that harder.
Let's talk about what to actually watch over the next five months. What are the indicators that tell you whether the Democratic flip is really happening or not?
The generic ballot is the broadest indicator, and right now it's favoring Democrats by six to eight points. But you also want to watch the number of competitive districts. Cook Political Report currently rates about thirty-five House seats as toss-ups. Most of those are held by Republicans in districts that Biden won in twenty twenty or that were very close. If the toss-up map expands, that's good for Democrats. If it shrinks, that's good for Republicans.
The Senate map?
The Senate map is actually tougher for Democrats this cycle. They're defending seats in states that Trump won, like Ohio and Montana. Republicans are defending seats in states like North Carolina and Maine that are more competitive. The math favors Republicans holding the Senate even if Democrats win the House.
The most likely outcome, based on current polling and the map, is a Democratic House and a Republican Senate. Divided government, but asymmetrical.
That asymmetry matters. A Democratic House can block the president's domestic agenda, launch investigations, and refuse to fund certain priorities. But a Republican Senate can confirm judicial appointments, approve executive branch nominees, and kill any Democratic House bills. The president can continue to shape foreign policy and trade through executive action. Each branch has its own lane, and they'll all be in conflict.
For someone outside the US trying to track this, what sources would you recommend?
For non-partisan race ratings, Cook Political Report, Sabato's Crystal Ball, and Inside Elections are the gold standard. They rate every competitive race on a scale from toss-up to safe for one party. For international perspective, the Economist's US politics coverage is excellent — they write for a global audience and don't assume you know the intricacies. FiveThirtyEight aggregates polling data and produces forecast models that are very accessible.
If you want to understand the implications for a specific issue area — trade, defense, immigration — follow the committee chairs. The House committees that will hold hearings and issue subpoenas under a Democratic majority will be chaired by Democrats. The House Foreign Affairs Committee, the Armed Services Committee, the Judiciary Committee — those chairmanships matter enormously.
That's an underappreciated point. The committee chairs set the agenda for oversight. They decide which administration officials get called to testify, which policies get investigated, which documents get subpoenaed. Those are powerful positions, and they flip when the House flips.
Let me bring this back to the core question. What does divided government mean for the president's ability to lead? The answer is: it depends on what kind of leadership we're talking about. In foreign policy and military command, the president retains enormous unilateral authority. In domestic legislation, the president becomes almost entirely dependent on what the opposition party is willing to tolerate. In administrative action, the president can do quite a bit through executive orders, but it's fragile and reversible.
In political terms, divided government transforms the presidency. The president shifts from being the head of a governing coalition to being the head of a opposition movement that happens to occupy the White House. The rhetorical posture changes. The legislative strategy changes. Everything becomes about positioning for the next election rather than governing for the present.
Which brings us to the open question I want to leave listeners with. The twenty twenty-eight presidential election will be heavily influenced by how the twenty twenty-six to twenty twenty-eight Congress performs. If divided government produces some bipartisan accomplishments — like the ninety ninety-six welfare reform — it could create a sense that the system is working. If it produces gridlock and shutdowns and debt ceiling crises — like twenty eleven — it could fuel anti-incumbent sentiment that shapes the presidential race.
The historical record suggests both outcomes are possible. Divided government has produced welfare reform, the FIRST STEP Act, and No Child Left Behind. It's also produced government shutdowns, sequestration, and impeachment. The variable isn't the constitutional structure — it's whether both parties see political advantage in cooperation.
That's the thing to watch. Not just who wins in November, but what they do with the win. Do they see it as a mandate for confrontation or a signal to negotiate? The answer to that question determines whether the next two years are productive or paralyzed.
Now: Hilbert's daily fun fact.
Hilbert: In the eighteen forties, a mathematician named Feodosy Osipovsky, working in the Caspian basin, proposed a theory of probability that treated every event as having only two possible outcomes — success or failure — and argued that all probability distributions could be reduced to this binary framework. The theory was widely taught in Russian academies for nearly two decades before being abandoned when it failed to account for dice rolls with more than two faces.
For twenty years, Russian mathematicians just... ignored dice with more than two sides?
The entire field of probability theory, temporarily reduced to coin flips.
This has been My Weird Prompts. Our producer is Hilbert Flumingtop. You can find every episode at myweirdprompts dot com. If you got something out of this, leave us a review wherever you listen — it genuinely helps people find the show. We'll be back next week.
