Democrats won control of both the House and the Senate on Tuesday, according to network projections, with a Republican in the White House. Both chambers of Congress changing hands at once is uncommon. Before this week it had happened at only four midterm elections since World War II: 1946, 1954, 1994 and 2006.
The natural question for an investor is what stocks did next. We pulled the daily closing prices and measured the S&P 500 three, six and twelve months after each of those four elections. We then compared those results with every other midterm since 1946, and looked at what the new Congress actually did and what conditions investors faced at the time.
The short answer is that the index was higher twelve months later every time, but only barely in one case, and the four paths have little in common. The composition of Congress explains very little of the difference between them.
S&P 500 price return after the last close before results
Four times both chambers flipped at a midterm. Four different outcomes.
1946
Truman (D). Republicans take both chambers.
1954
Eisenhower (R). Democrats take both chambers.
1994
Clinton (D). Republicans take both chambers.
2006
G.W. Bush (R). Democrats take both chambers.
Average
The four flips above
Others
The other 16 midterms, 1950 to 2022
Four observations cannot establish a pattern. The range tells the reader more than the average does.
Source: Commons Capital calculations from S&P 500 daily closes (Yahoo Finance and FRED). Starting point is the Election Day close, or the prior close in 1946 and 1954, when the exchange closed for the election. Price return only, excluding dividends and costs. The index is unmanaged and cannot be bought directly. The 1946 and 1954 figures use the S&P 90, the predecessor to the S&P 500. Educational only; not a recommendation to buy or sell any security. Past performance is not indicative of future results.
What the record shows
Across the four flips, the index gained an average of 6.8% after three months, 9.6% after six months and 16.8% after twelve months. Those averages hide most of the story. The twelve-month results run from +0.1% to +33.2%.
1946. Republicans took both chambers from a Democratic president. Stocks had already fallen about 20% between late May and the election. The index was down 3.1% six months later and finished the twelve months up 0.1%. Dividends mattered that year: the index yielded about 4.8% at the time, so a price-only figure understates what a shareholder earned.
1954. Democrats took both chambers from a Republican president, in the months after a recession ended in May 1954. The index never closed below its pre-election level in the following year and gained 33.2%. Even so, it fell about 11% from a high to a low along the way.
1994. Republicans took both chambers from a Democratic president, for the first time in four decades. The Federal Reserve was in the middle of a tightening cycle that ran into February 1995. The index gained just 3.3% in three months and 27.1% in twelve.
2006. Democrats took both chambers from a Republican president. The index gained 6.7% in twelve months, and that window hides what followed. It peaked in October 2007, about 13% above its election level. A recession began in December 2007, and by March 2009 the index stood about 51% below its election level.
Only 1954 and 2006 match this week's configuration of a Republican president and a newly Democratic Congress. Their twelve-month results were +33.2% and +6.7%. Two observations with outcomes that far apart do not describe a tendency.
Twelve months is a flattering window
A new Congress serves for two years, so the twelve-month mark covers only half of its term. At 24 months the four results spread much further apart: +0.7% after 1946, +47.8% after 1954, +56.9% after 1994 and -32.7% after 2006. The average of +18.2% describes none of them.
The 2006 case is the clearest warning about reading too much into a one-year number. An investor who looked only at the first year would have seen a modest gain. The financial crisis arrived inside the same Congress.
Compared with what?
The more useful test is whether a double flip changed anything relative to other midterms. We sorted all 20 midterms from 1946 through 2022 into three groups: both chambers flipped, one chamber flipped, and neither flipped.
Average S&P 500 price return, 1946 to 2022
More political change has not meant better or worse returns.
Both chambers flipped
4 midterms
One chamber flipped
6 midterms
Neither chamber flipped
10 midterms
Any starting day
All trading days since 1946
The one-chamber group includes 1986, whose following year contained the October 1987 crash. A single event can move an average this small.
Source: Commons Capital calculations from S&P 500 daily closes (Yahoo Finance and FRED). Same method as the chart above. Worst decline is the largest peak-to-trough fall in the 12 months after each election. The bottom row covers windows starting on every trading day from January 1946, using data through February 2026. Educational only; not a recommendation to buy or sell any security. Past performance is not indicative of future results.
The results do not line up in any sensible order. Midterms with no change in control averaged more than double flips at every horizon we measured, though only slightly at twelve months. Midterms where one chamber flipped did worst over twelve months, and that group's average is pulled down by a few weak years. If political change itself drove returns, we would expect a cleaner progression from none to one to both.
The bottom row of that table is the more durable finding. Any given twelve-month window since 1946 averaged about +9.0% and was positive about 74% of the time. The twelve months after a midterm averaged +14.4%, and the index was higher after all 20. That is a separate observation about the midterm calendar, and it holds regardless of who won. Twenty observations over 80 years are still too few to separate cause from coincidence, and all of them are drawn from a period when U.S. stocks rose substantially.
What each new Congress actually did
Divided government is often described as gridlock. The record is less tidy. Each of these Congresses produced something that mattered to investors, usually through conflict or compromise with the president.
After 1946. The new Republican majority passed the Revenue Act of 1948 over President Truman's veto, cutting individual income tax rates and allowing income splitting in community property states.
After 1954. A Democratic Congress and a Republican president produced the Federal-Aid Highway Act, signed by President Eisenhower in June 1956 and at the time the largest public works program in American history.
After 1994. Budget conflict with President Clinton led to two funding gaps in the winter of 1995 to 1996, one of them lasting 21 days. The same majority and the same president later agreed on the Taxpayer Relief Act of 1997, which lowered the top rate on long-term capital gains from 28% to 20%.
After 2006. In October 2008, a Democratic Congress and President Bush enacted the Emergency Economic Stabilization Act, authorizing $700 billion for the Troubled Asset Relief Program during the financial crisis.
None of these outcomes could have been read from the election result itself. A 21-day funding gap and a capital gains cut came from the same divided government within two years of each other.
What moved markets instead
If Congress did not set the course, what did? The conditions at each flip offer a better account than the election results do.
Conditions at each flip
The starting conditions differed far more than the politics did.
1946
Postwar inflation
1954
After a recession
1994
Late in a Fed tightening
2006
Late in an expansion
2026
Latest available
In 1946 earnings rose sharply and the index went nowhere. In 2006 earnings fell and the index still rose. Neither year followed from the election.
Source: Commons Capital calculations from Robert Shiller's dataset for November of each election year and the following November. Inflation is the 12-month change in CPI. Earnings are reported S&P composite earnings. 2026 figures are the October 2026 provisional CAPE, the November 2 10-year Treasury yield (Federal Reserve H.15) and September CPI. Educational only; not a recommendation to buy or sell any security. Past performance is not indicative of future results.
In 1946, reported earnings rose about 58% over the following year and the index went nowhere, with inflation running near 18% and investors unwilling to pay more for those earnings. In 1954, consumer prices were flat, yields were low and earnings rose about 31%. In 1994, the 10-year Treasury yield started near 8% and fell to about 6% over the next twelve months while earnings grew. In 2006, earnings fell about 13% and the index still rose, until the credit cycle caught up with it.
The common thread is that earnings, interest rates, inflation and the valuation paid at the start carried more weight than the party in control. Control of Congress can influence some of these at the margin, but a change in majorities does not set them.
The bottom row shows where 2026 sits. Valuation, measured by the cyclically adjusted price-to-earnings ratio, is higher than at any of the four earlier flips. The 10-year yield is above its 1946, 1954 and 2006 levels, and inflation is above its 1954, 1994 and 2006 levels. None of this is a forecast. CAPE has been a poor timing tool, and the index has stayed expensive by this measure for long stretches. It does mean that the comparison to earlier flips is imperfect, because no earlier flip started from these conditions.
What a divided government can and cannot do now
The president keeps the veto, and that frames most of the next two years. Allianz Global Investors notes that the new Senate majority falls well short of the three-fifths needed to break a filibuster and the two-thirds needed in both chambers to override a veto. The Tax Foundation writes that Democrats "will have limited maneuvering room in 2027 given Republican control of the White House," and that proposals such as a higher corporate rate or capital gains changes are limited by the veto.
That does not make the next two years quiet. Three items are worth watching. Allianz flags shutdown risk ahead of a December 11 funding deadline. The Tax Foundation says a debt limit debate in 2027 may give Democrats leverage. And several provisions of the 2025 tax law, including the tips, overtime and auto loan interest deductions, the enhanced senior deduction and immediate expensing of new factory buildings, expire on December 31, 2028. Those are the dates when divided government forces a negotiation.
Tariffs are a different case. Allianz does not expect the election to change their direction, since Congress has limited sway over trade policy in practice, and the Tax Foundation considers a veto-proof majority on the issue unlikely.
Where the sensitivity sits
A shift of this kind tends to affect particular industries more than the market as a whole. The first hours of trading illustrate the distinction. The S&P 500 was up about 1.1% Wednesday morning, a modest move at the index level, while individual groups moved in different directions: import-reliant retailers rose, some health insurers fell, and some AI-linked stocks opened lower. One morning proves nothing, but it shows where the exposure lies.
Questions worth asking now
For most households, none of this calls for a change in allocation. It may call for a few checks.
Does your plan assume any particular tax outcome? The prudent baseline is current law, with attention to what is scheduled to expire. If a tax decision only works under a specific legislative result, it is carrying more political risk than it appears to.
If you own a business, does any planned investment depend on provisions that expire at the end of 2028? Timing matters more when the extension of a provision depends on a negotiation.
What do you own through your index funds and employer stock? Exposure to health care, defense, AI infrastructure or trade-sensitive businesses is often larger than a statement suggests, and it can be concentrated in the same names across several accounts.
Could you ride out a 24-month result like 2006? Cash needs over the next two years should not depend on the direction of the market or the outcome of a budget fight.
How we approach a moment like this
We do not reposition portfolios around election results, and we do not predict how markets will respond to this one. History offers a narrow lesson here: a sound plan should not depend on which party controls Congress.
What a change in Washington can do is alter the assumptions behind specific holdings. Because Commons Capital builds portfolios from individual securities across asset classes, we can review those exposures one position at a time, against each client's tax picture and existing concentrations. That review is part of ordinary Investment Management work, and an election is only one of the reasons to do it. An index fund holder takes the exposure as the index delivers it, a point we covered in The Hidden Concentration in "Diversified" Portfolios.
Quick answers
How often have both chambers of Congress flipped at a midterm?
Four times from 1946 through 2022: 1946, 1954, 1994 and 2006. The 2026 election is the fifth.
How did stocks perform after those flips?
The S&P 500 averaged +6.8% after three months, +9.6% after six months and +16.8% after twelve months, on a price basis. Twelve-month results ranged from +0.1% to +33.2%.
What happened the last time a Republican president lost both chambers?
That was 2006. The S&P 500 rose 6.7% over the next twelve months and was down 32.7% after 24 months, during the financial crisis.
Do stocks do better after a midterm flip than after other midterms?
The record does not show a consistent difference. Double flips averaged +16.8% over twelve months, against +17.0% when neither chamber flipped and +8.4% when one did, on very small samples.
Can a Democratic Congress change tax law in 2027?
Only with the president's signature or a veto-proof majority. Analysts at the Tax Foundation see limited room for major change in 2027.
Does this history predict what stocks will do after 2026?
No. Four observations across eight decades cannot support a forecast, and current valuations are higher than at any earlier flip.
How we measured this
Returns are S&P 500 price changes, excluding dividends. Each starts from the Election Day close, or from the prior close in years when the exchange closed for the election, and ends on the same calendar date three, six, twelve and 24 months later, or the last trading day before it. Figures before March 1957 use the S&P 90, the predecessor to the S&P 500. Valuation, yield, inflation and earnings figures come from Robert Shiller's dataset, using November of each election year and the following November.
We counted control changes using the official House and Senate party division records. One-chamber flips are 1986, 2002, 2010, 2014, 2018 and 2022. We left out presidential election years, when the White House was also on the ballot, and midterms before 1946.
Sources
- CBS News, 2026 midterm election results, November 3 to 4, 2026
- National Constitution Center, "How midterm elections have changed Congress since 1946"
- U.S. House of Representatives, Party Divisions
- U.S. Senate, Party Division
- National Bureau of Economic Research, US Business Cycle Expansions and Contractions
- Robert Shiller, online stock market data
- Federal Reserve, H.15 Selected Interest Rates, November 3, 2026
- Tax Foundation, "What a Democratic Congress Means for Federal Tax Policy in 2027," November 4, 2026
- Allianz Global Investors, "Democrats take Congress, but sweeping change looks unlikely," November 4, 2026
- Congressional Research Service, Shutdown of the Federal Government: Causes, Processes, and Effects
- Congressional Research Service, The Taxpayer Relief Act of 1997: An Overview
- U.S. Senate, Federal Highway Act
- ABC News, "Dow surges 1,000 points as Democrats take control of Congress," November 4, 2026
- Al Jazeera, "US markets rise as Democrats sweep midterm elections," November 4, 2026
This material is educational and reflects our analytical framework. It is not a recommendation to buy or sell any security, and it expresses no view on any party or candidate. Past performance is not indicative of future results. Index returns are unmanaged, exclude dividends, fees and expenses, and cannot be invested in directly. The small number of historical observations limits any conclusion drawn from them. Descriptions of pending or proposed legislation reflect third-party analysis and may change. Holdings discussed may or may not be currently held in client portfolios. Consult your tax and legal advisors before acting.
Commons Capital is a fee-based fiduciary Registered Investment Adviser regulated under the Investment Advisers Act of 1940. Additional information, including our Form ADV Part 2, is available on the SEC's Investment Adviser Public Disclosure system.
