Midterms: A Test of Checks and Balances
The first 19 months of Donald Trump’s second term have been explosive, both metaphorically, with a head-spinning series of radical decisions (tariffs, ICE, slashing of government agencies, universities and research) and literally (endangering international alliances, unconditional support for Israel’s campaigns in Gaza and Lebanon, the Iran war). Although the committed MAGA community for now maintains its support, results have been largely unpopular not only abroad but also domestically. According to The Economist, by the beginning of June, Donald Trump’s net approval rating was the lowest since the magazine started the survey in 2009. With midterm elections in November, it is natural to ask what will change.
Midterm Elections, a Difficult Test
Midterm elections — elections that arrive at the midpoint of the president’s term and renew the entire House of Representatives and about one third of the Senate — are always a difficult test of a presidential mandate. With the exception of George W. Bush in 2002 (only one year after the September 11 attack on New York), every recent president has seen his party lose seats in the midterm and lose control of one of the houses of Congress. Clinton lost both the House and the Senate in 1994, and so did George W. Bush in 2006; Obama lost the House in 2010 and the Senate in 2014; Trump lost the House in 2018 and so did Biden in 2022.
As of this writing (June 2026), the quasi-totality of polls predict the Democrats will win the popular vote. As of 14 June, out of 495 polls tracked by The New York Times, 412 favor Democrats. Although in most polls the advantage is in the single digits, and Republicans have scored recent victories in the redistricting battle, on betting markets, the margin in favor of a Democratic House is high (83% on Polymarket as of 14 June).
If the Dems Take the House
The question then becomes: supposing Democrats win the House, what difference will that make to US policy decisions? Long experience with midterm political reversals has resulted in a few lessons.
Predictably, divided government leads to fewer major legislative decisions than unified government. While the impact on total legislative activity is ambiguous, the literature agrees in identifying a decline of about 30–40% in the number of non-trivial, significant laws.
The Lesson of History
The reduction in legislative activity is typically decried as an example of the immobilism created by parliamentary divisions. If there are fewer pieces of legislation, however, they may be of superior quality — and thus more durable — because they are less subject to future amendments or repeals. Recent research has studied this hypothesis with contradictory results, but measuring the quality of laws is difficult. The hypothesis aligns with intuition and experience: an ongoing study finds that checks and balances reduce the negative effects of polarization and dampen the impact of populism. Divided government is a plausible form of checks and balances.
Thus, history teaches us that a Democratic House is likely to issue fewer landmark laws, although possibly better and more durable ones. But how relevant is history for this most unusual of administrations, Donald Trump’s second term? Can we be sure that the impact would be more than minimal?
The Use of Executive Orders
The reason to wonder is the unprecedented style of the current government, and in particular, Trump’s exceptional reliance on tools of unilateral presidential power, executive orders most visibly. In the US, contrary to Italy, executive orders do not need to be approved by Congress within any specified delay and remain in force until either superseded by laws, invalidated by the courts or canceled by the next president. In the most recent 16–17 months since coming to power, Donald Trump has issued 265 executive orders, or an average of about one every two days. This is more than three times the number issued by any of the preceding five administrations over the same length of time, and much higher than the number issued over the entire length of these administrations. In addition to their number, Trump’s executive orders stand out for the variety and importance of the matters they address: immigration and border enforcement; federal agencies; trade and tariffs; energy and environmental policy; foreign affairs and national security; cultural and education issues; technology and artificial intelligence.
How much are midterms likely to matter then, even if the Democratic Party does indeed win a majority in the House? The answer cannot abstract from the use (and abuse) of unilateral presidential actions. A divided government typically increases the relative frequency of presidential measures, but even if the pace were to remain constant or even slightly decrease, the president’s recourse to unilateral action would leave much of his power undiminished.
Two Possible Effects
In our opinion, there would be two main effects. The first is institutional, but would express itself only in the longer term. Under divided government, converting executive actions into laws would be harder. This would deprive the president of some room for constraining future governments: a future president can easily cancel Donald Trump’s executive actions, but is much less likely to be able to reverse measures that have been codified as laws. The second effect, more likely to be felt in the short term, is political rather than institutional. Translating the decline in popularity captured by the polls into electoral losses would make vivid the risks to the Republican Party of its current supine position for the next presidential election. In the wake of the Iran war, some of this reckoning appears to have started. The midterms could solidify it.
