Stanley Druckenmiller predicts: (1) A major U.S. financial crisis will occur sometime in the 2025–2035 period, driven by entitlement costs as baby boomers age; (2) By roughly 25 years from 2023 (i.e., by around 2048), spending on seniors will reach about 60% of all U.S. tax revenues; (3) Combined U.S. federal spending on Social Security plus Medicare and Medicaid will rise from about 12% of GDP in 2023 to about 24% of GDP by around 2048.
“Stanley Druckenmiller gave a speech at USC... and he expressed concern about the financial crisis that occur could occur in the 2025 to 2035 period due to the baby boomers turning 65 and the impact on entitlements. He predicted that in 25 years, spending on seniors will grow to 60% of all taxes... And it's predicting here that those combined will go from what looks like 12% today, up to 24% of GDP.”
View on YouTube
Explanation
These are long-horizon (through roughly 2048) demographic and fiscal projections that cannot yet be evaluated this early, though early trend data (rising entitlement spending share) tracks in the predicted direction.
U.S. federal debt-to-GDP will continue to rise from its 2023 level over the coming years rather than being reduced, and this increase will not cause a systemic break or cessation of the functioning of the U.S. economy.
“I think the reality is that debt to GDP will continue to increase... So as a practical matter, this thing will go up and I don't think the economy will stop.”
View on YouTube
Explanation
US federal debt-to-GDP has continued rising since 2023 without causing a systemic break in the economy's functioning.
Starting from the 2023 level, U.S. federal debt-to-GDP will reach 200% before it ever falls to 50%.
“Okay, I will bet you that debt to GDP gets to 200 before it gets to 50.”
View on YouTube
Explanation
US debt-to-GDP has continued rising (well above 100% by public-debt measures) but has not yet reached either the 200% or 50% threshold as of mid-2026, so this wager remains unresolved.
If U.S. federal debt rises to around 200% of GDP with interest rates around 4% (implying roughly $2 trillion in annual interest expense at current GDP levels), then to sustain that debt service the U.S. will eventually raise overall tax burdens so that effective tax rates exceed 70% of government income (i.e., taxes will need to be raised to historically high, roughly 70%-plus levels).
“You're using more than half of the government's income based on the current tax rates to fund the interest payments on your debt. That's not even to pay for social services... That's just more than half of the income... That's why you have to see taxes go up to over 70% because it's the only way you can. You got to tax everything in order to fund that.”
View on YouTube
Explanation
US debt-to-GDP had not reached 200% as of mid-2026, and effective federal tax rates have not risen anywhere near 70%; interest expense, while growing substantially, remains well short of the described scenario.
The U.S. Treasury will in the future issue sovereign bonds with maturities longer than 30 years, specifically 50-year and 100-year U.S. government bonds.
“I think the thing you guys have to be open to is the fact that we've never really tested the ability for the US to borrow durations beyond 30 years... I do think that they'll be able to get durational assets that are that far out on the yield curve... So we'll have 50 year US bonds. We'll have 100 year US bonds.”
View on YouTube
Explanation
The US Treasury has not issued 50-year or 100-year bonds as of mid-2026; the longest standard maturity remains 30 years, despite periodic Treasury study of ultra-long bonds.
If and when the U.S. issues 100-year Treasury bonds, the yield on those 100-year bonds will be below 1%, making the long-term borrowing cost for that debt effectively near-free for the U.S. government.
“So I, again, am less concerned about the debt wall here, because I think you'll be able to push maturities out. You'll be able to refi a bunch of short term obligations into the future... I would be very surprised if 100 year rates, if they priced a bond weren't somewhere sub 1%. So I do think it becomes effectively free money for the United States.”
View on YouTube
Explanation
Since no 100-year US Treasury bond has been issued as of mid-2026, this specific yield prediction has not been tested, and current interest-rate levels (well above 1% even for 30-year bonds) make a sub-1% outcome look unlikely if issued today.
Donald Trump will win the next U.S. presidential election following this May 2023 town hall, i.e., he will be elected president in the 2024 U.S. election.
“And then I got the sense that CNN's management wants this. This is like a ratings bonanza for them, and I think they secretly want him... And then I thought, oh, he's going to get elected.”
View on YouTube
Explanation
Donald Trump did win the November 2024 US presidential election.
As of May 2023, Friedberg predicts that Donald Trump has a real chance of being reelected president in the 2024 U.S. election.
“So I think he's got I think, I think the reality is he's got a real shot at getting reelected here.”
View on YouTube
Explanation
Trump did have and ultimately capitalize on a real chance at reelection, winning the 2024 election.
As of May 2023, Sacks predicts that Trump’s CNN town hall performance increases the likelihood that Trump will win the 2024 Republican presidential nomination.
“I do think it makes him more likely to be the nominee.”
View on YouTube
Explanation
Trump went on to win the 2024 Republican presidential nomination decisively.
Sacks predicts that during the 2024 campaign, if Trump is the Republican nominee, President Biden will largely run a low-key ‘Rose Garden’ campaign, appearing roughly once a week to respond to Trump, and will not actively campaign at a high-energy pace due to lack of vigor.
“He'll do a Rose garden campaign where once a week, he goes in front of the microphones and responds to whatever Trump's latest outrage is. He doesn't have the vigor to campaign, and he won't.”
View on YouTube
Explanation
Biden did run a relatively low-profile campaign through mid-2024, broadly consistent with a 'Rose Garden' strategy, though this became moot when he withdrew from the race in July 2024 before facing Trump directly in the general election.
Sacks predicts that by approximately late 2024, after about 18 months of Trump dominating media coverage, voter fatigue with Trump will lead enough Americans to choose Biden despite concerns about his cognition, resulting in Biden’s reelection if the matchup is Biden vs. Trump.
“I think that it's it's quite possible here that after 18 months of Trump and the media beating each other up, the American people just say, you know what? This Biden guy is totally senile. But I'm like, so tired of the the Trump show. I've got Trump fatigue again. I'm just going to have to go with Biden. And I think I think this is how Biden gets reelected.”
View on YouTube
Explanation
Biden was not reelected; he withdrew from the race in July 2024 amid concerns about his age and cognition following a poor debate performance, and Trump went on to defeat Kamala Harris in November.
As of May 2023, Friedberg predicts that there is a meaningful (non-zero) probability that President Biden will decide not to run for reelection in 2024 and will exit the race before the election.
“I think there's a non-zero chance Biden actually doesn't run for reelection at this point.”
View on YouTube
Explanation
Biden did ultimately not run for reelection, withdrawing from the race in July 2024, confirming the predicted non-zero probability.
Sacks predicts that by the end of 2023, the Russia–Ukraine war will be an even worse debacle for Ukraine than it appeared in May 2023, implying significantly deteriorated conditions or outcomes for Ukraine.
“So this war is turning into a debacle. I think it could be an even worse debacle by the end of the year.”
View on YouTube
Explanation
Ukraine's military situation remained difficult through the rest of 2023 with a disappointing counteroffensive, though 'even worse debacle' is a subjective escalation that is only partially confirmed by the actual, more gradual deterioration.
Sacks predicts that the U.S. banking turmoil evident in May 2023 will continue and develop into a major fiasco (a serious, extended banking crisis) over the ensuing period (through at least the next several months).
“The economy has a banking crisis going on. It's turning into a big fiasco.”
View on YouTube
Explanation
The March 2023 US regional banking turmoil (SVB, Signature, First Republic) was largely contained by mid-2023 through federal intervention and did not develop into a broader systemic fiasco.
Sacks predicts that if the 2024 U.S. presidential election is a Biden vs. Trump matchup, the race will be extremely close, effectively a toss-up with no clear favorite heading into the election.
“I think this thing's going to be a nail biter. I think it's going to be a toss up if it's Biden versus Trump.”
View on YouTube
Explanation
A direct Biden-vs-Trump general election never occurred since Biden withdrew in July 2024, so this specific matchup prediction was never tested; the eventual Trump-vs-Harris race was competitive but Trump won by a clearer-than-toss-up margin.