Money hoarded out of fear ends up buying at the top

The Liquidity Map · Macro report

Money hoarded out of fear ends up buying at the top

Americans have never been gloomier, corporate bankruptcies are at a 15-year high, and cash is piling up on the sidelines.

I've seen this kind of panic around every shmita year.

Each time, it was followed by two years in which money came back to every market.

November 1st, 2026 · 7 min read

Key takeaways

In the last three shmita windows, the S&P? lost between 15% and 58%, Bitcoin more than 75%, and oil between 46% and 77%.
In shmita years, the Fed?'s balance sheet stopped growing, and new money only came afterwards. Since December 2025, it's been growing again.
In May 2026, US consumer sentiment hit its record low, below the levels of 2008 and 2022.
Each time, within one or two years, the panic gave way to spending and to rising markets.
2027 and 2028 are years 5 and 6 of the cycle, historically the best. After them come roughly 18 hard months.

Everyone is tightening their belt and putting money aside. Pessimism is free. Sadly, it's the only thing that got cheaper.

I'm doing the opposite, and in this report I explain why.

I laid out the year-by-year numbers in the report on 80 years of shmita. Today I tie them to liquidity?: where money goes when people get scared, and when it comes back.

The markets

In every shmita window, money left the market

I took the last three shmita windows: 2007-08, 2014-15 and 2021-22.

For each one, I looked for the peak in the year before shmita or in the shmita year, then for the lowest point through the end of the year after.

No market escaped. When liquidity pulls back, it leaves everywhere.

Bitcoin fell the hardest, because it's the most sensitive to new money. The S&P had the smallest decline in 2015-16, but it went nowhere for almost two years.

Shmita window	S&P 500	Bitcoin	WTI crude
2007-08	-57.7%	didn't exist yet	-76.7%
2014-15	-15.2%	-86.9%	-76.3%
2021-22	-27.5%	-77.6%	-46.0%

The largest decline, from the peak in the year before shmita or in the shmita year to the lowest point reached by the end of the year after.

$152Bitcoin's low in January 2015, after a peak of $1,163 in November 2013.

The largest decline in the last three shmita windows, for the S&P 500, Bitcoin and oil.

Oil has a story of its own.

Oil

Oil fell hard in every shmita year

In the 2007-08 shmita year, oil fell 37% from its $145 peak in July 2008. By December it was at $34.

In 2014-15 it lost 60% within the year itself, and in 2021-22, 36% from its March 2022 peak.

The most violent drop came outside the cycle: on April 20th, 2020, US crude closed at minus $38.

The pandemic was a shock, not a phase of the cycle. But it brought the biggest injection of money in history, and within two years oil climbed to $124.

×2.2The pandemic money: the Fed's balance sheet from February 2020 to its April 2022 peak.

WTI crude since 2003, with the shmita years marked in red.

To see where money comes from, you look at the Fed.

The Fed

New money only came after the shmita year

The Fed's balance sheet shows how much money the central bank has put into the system.

In the 2007-08 shmita year, it stood still at about $900 billion. Only after the Lehman collapse, in September 2008, did it grow 2.4-fold in three months.

In 2014-15 it sat at $4.5 trillion and stayed there until 2017. In 2021-22 it grew only until March, and then the Fed started shrinking it. From April 2022 to December 2025 it fell 27%.

Since December 2025, the Fed's balance sheet has been growing again, for the first time in almost four years.

Interest rates aren't helping yet. On Wednesday, the Fed held them at 3.75-4.00%, and two members voted for a hike.

To me, the direction of the balance sheet matters more than any single meeting.

+$200BAdded to the Fed's balance sheet between December 2025 and early September 2026.

Total Fed assets since 2003, with the shmita years marked in red.

That's the market side. The other side is in our pockets.

Our pockets

Every crisis taught me a lesson

In 2008 I was finishing my economics degree in Constanța and had my first real job. Until 2011, almost nobody hired without two years of experience, and apartment prices where I lived fell by half.

In 2014 I graduated in petroleum engineering and made it to an offshore rig for a single rotation. Then oil lost 60%, and I lost my job.

In the first five months of 2015, US consumer prices were lower than a year earlier.

When something breaks in the economy, something else starts working.

In 2008-2009, my mother sold personal care products with home delivery, back when almost nobody delivered, and made an extra €1,000-2,000 a month.

In 2016, after losing my job, I opened a small coffee shop on the Black Sea coast with my best friend.

12The number of coffee shops I ended up with, starting in 2016, the year after a shmita year.

Today I see the same panic.

Now

The same panic, this time in 2026

Americans have never been gloomier. In May, consumer sentiment fell to the lowest level ever recorded. Five of the six weakest months in the survey's history came this year.

Companies are feeling it too: 785 large US companies went bankrupt in 2025, the most since 2010, and another 372 in the first half of 2026.

Meanwhile, money market funds, where cash waits on the sidelines, hit a record of nearly $8 trillion in September.

The money is there. It's just sitting idle, out of fear.

44.8US consumer sentiment in May 2026, the record low of the University of Michigan index.

US consumer sentiment since 1978, with the shmita years marked in red.

History tells us what comes next.

After the panic

One to two years after each crisis, people spend again

Panic doesn't last forever. US consumer sentiment climbed from 50 in June 2022 to 79.4 in March 2024.

Stocks got there first: within two years of each window's low, the S&P rose 98%, 45% and 67%.

Money doesn't disappear in a crisis. It hides, then comes back all at once.

Bitcoin did even more: 5.4-fold within two years of its 2015 low, and 6.4-fold from its November 2022 low.

$15,479Bitcoin's low in November 2022, eight weeks after the shmita year ended.

The S&P 500 in the two years after each shmita window's low, with the low = 100.

That's where my verdict for 2027 and 2028 comes from.

The verdict

2027 and 2028, the years money comes back

Year 5 of the shmita cycle began on September 12th, and year 6 runs from October 2027 to September 2028.

Across 12 cycles since 1938, they've been the best years for the S&P.

I think 2027 and 2028 will bring new money into every market, new businesses, and people spending what they're saving today.

A second clock says the same thing about 2027: it's year 3 of the US presidential term.

Since 1929, the US has had 15 recessions. Nine began in the first year of a term, five in the fourth and one in the second.

In year 3, none. The only borderline case is the 2008 crisis: the US economy peaked in December 2007, the last month of year 3, and the recession? began in January 2008, in year 4.

But the clock doesn't stop there. As I wrote on X on Friday, these two years are followed by roughly 18 months in which money leaves the markets again.

Cycle year	Dates	S&P history
Year 5	Sep 12, 2026 - Oct 1, 2027	+14.1% · 11 of 12
Year 6 (pre-shmita)	Oct 2, 2027 - Sep 20, 2028	+15.3% · 12 of 12
Year 7 (shmita)	Sep 21, 2028 - Sep 9, 2029	+1.3% · 6 of 12
Year 1 (year after)	Sep 10, 2029 - Sep 27, 2030	-1.9% · 4 of 12

The S&P's average return in that year of the cycle and how often it was positive, across 12 cycles (1938-2022).

0US recessions that started in year 3 of a presidential term, from 1929 to today. 2027 is a year 3.

US recessions by the year of the presidential term in which they started, since 1929.

This is where the paradox comes in.

The paradox

The money paradox

Most people do exactly the opposite. Right now, with plenty of things cheap, they put money aside out of fear.

In 2027-2028, when everything gets expensive, they take it out and spend it along with everyone else. And in 2029, when hard times come back, they have nothing left to invest.

Money should be put to work when everyone keeps it aside, and kept aside when everyone spends it.

I learned this the hard way. Between 2016 and 2019 I reinvested almost all my profit into new coffee shops, and in 2020 I had to close a few.

With money set aside, in 2020-2021 I could have picked up good locations cheaply from people who hadn't prepared. If you've read this far and your money is still under the mattress, the mattress thanks you. So does inflation.

2029The year when, if the pattern repeats, money set aside will matter again.

That's how I build my plan.

How we're positioned

Today's fear isn't pushing us out of the market: our money is at work, in stocks and in crypto. What we're adding today is a rule for profits.

As markets rise in 2027 and 2028, we take profits in stages and set them aside to ripen. Of the profits we take, we reallocate only 10-30%. The rest waits for the year after shmita, starting in fall 2029.

And we don't add leverage? toward the end of the cycle, exactly when it looks easiest. We'd change the plan if year 5 or year 6 closes lower.

Conclusions

Liquidity comes in waves, and people's fear has always been the sign that the next wave is getting close.

Money hoarded out of fear ends up buying at the top

Keep reading

Why November 3rd matters

JPMorgan moves onto Ethereum while the funds sell

21 out of 21. What comes after the US midterms