Nvidia Just Wrote the Biggest Buyback Check in History. Here’s What History Says It Means for Owners

By Dominick Paoloni, CIMA® | Founder & Chief Investment Officer, IPS Strategic Capital | September 29,
2026

Every time a company announces a big buyback, somebody on television calls it financial
engineering: a sugar high for the stock price that does nothing for the business. Sometimes that
critic is right. Most of the time, over long periods, the data says otherwise, and the difference
between the two comes down to one variable almost nobody talks about.

On Monday, Nvidia’s board authorized another $150 billion, the largest buyback increase any
company has ever announced. It tops the $110 billion Apple approved in 2024 and lifts Nvidia’s
remaining authorization to $235 billion, which the company expects to use through fiscal 2028.
The stock rose about 2% on the news.

My view is that this is good news for long-term shareholders. But not for the reason the
headlines give. The buyback itself creates nothing. The price paid for the shares is where the
value is made or destroyed, and on that score Nvidia is doing something unusual.

It’s not magic, it’s math: the pizza

Picture a pizza cut into 100 slices, owned by 100 people. The company takes cash it has already
earned, buys back 4 slices, and throws them away. The pizza is exactly the same size. But each of
the remaining 96 slices is now about 4% bigger. That is all a buyback is. No new pizza gets
baked.

Now the real numbers. Nvidia’s $235 billion of remaining authorization is roughly 4% of a
company valued at about $5.4 trillion. If Nvidia spends all of it at today’s prices and earns not
one dollar more, each remaining share’s claim on those earnings rises by about 4.5%. CNBC
estimates the share count could fall by about 4% if the full amount is used. Two honest caveats:
an authorization is permission, not an obligation, and new shares issued to employees will offset
part of the reduction.

So the question is never “Is a buyback good?” The question is “Did they buy the slices for less
than they’re worth?”

What 40 years of research says

The landmark study is Ikenberry, Lakonishok and Vermaelen, published in the Journal of
Financial Economics in 1995. They tracked every open-market buyback announced from 1980
to 1990 and followed the stocks for four years. On average, the announcing companies beat
comparable stocks by 12.1% over those four years. The market, in other words, underreacted: it
treated the announcement as a one-day story when it was really a four-year story.

The detail that matters most is inside the average. Companies that looked cheap, the ones most
likely buying back stock because it was undervalued, beat their benchmarks by 45.3%. Richly
valued “glamour” companies showed no lasting edge at all. The buyback wasn’t the signal. The
price was.

Three case studies: the good, the patient, and the ugly

Apple, the poster child. Apple started buying back stock in 2012 and, by the middle of 2026,
had repurchased more than $880 billion of it. The share count has fallen by about 44%, which
means a share bought before the program now owns roughly 80% more of Apple’s business than
it did then. The stock is up nearly 12-fold since the start of fiscal 2013, and over the ten years
ending in early September it compounded at about 29% a year against roughly 15% for the S&P
500. Crucially, Apple paid for it out of free cash flow and never starved research, capital
spending or acquisitions to do it.

But here’s the part nobody puts in the highlight reel: in 2013, the first full year of the program,
Apple gained about 8% while the S&P 500 gained about 32%. Buybacks are a compounding
engine, not a catalyst. They reward owners measured in years, not in quarters.

Meta, buying when everyone else was selling. In February 2023, after a brutal 2022,
Meta added $40 billion to its buyback and paired it with deep cost cuts in what Mark
Zuckerberg called the “Year of Efficiency.” The company was buying its own stock at a moment
when few others wanted it. The shares nearly tripled over 2023, and a year later Meta
authorized another $50 billion and paid its first dividend.

Bed Bath & Beyond, the warning. From 2004 on, Bed Bath & Beyond spent about $11.8
billion buying back its own shares, at an average cost of more than $44 each, often spending
more than the business generated in free cash flow. It retired roughly three-quarters of its stock.
In 2023 it filed for bankruptcy and the remaining shares were worth nothing. A buyback cannot
fix a business that is losing its customers. It just speeds up the arithmetic.