Big Tech has quietly spent a sum that rivals the GDP of a mid-sized country building AI infrastructure, but the real financial reckoning has not yet arrived on any income statement. One accounting schedule will determine whether these bets pay…
This post may contain links from our sponsors and affiliates, and Flywheel Publishing may receive compensation for actions taken through them.
The sticker shock in Big Tech earnings this year came from the capital expenditure line. Microsoft (NASDAQ:MSFT | MSFT Price Prediction) spent $115.95 billion on property and equipment in the fiscal year ended June 2026. Alphabet (NASDAQ:GOOGL) laid out $44.92 billion in a single quarter, watched free cash flow flip to negative $5.86 billion, and suspended its buyback. Amazon (NASDAQ:AMZN) spent $54.21 billion in the June quarter alone. Meta Platforms (NASDAQ:META) guided full-year capital spending to $130 to $145 billion. Yahoo Finance pegged the four-company AI data center bill at roughly $1 trillion in an August 4, 2026 analysis.
A $100 billion data center is a capital asset, expensed in slices over the years the equipment is in service. The cash leaves the company now, but the accounting expense arrives in slices over the years the equipment is expected to be useful. That yearly slice is called depreciation for physical assets and amortization for intangibles. If a server has a useful life of six years, roughly one sixth of its cost lands on the income statement each year, reducing reported earnings per share.
Why the Timing Gap Is the Real Story
Depreciation is where the trillion dollar bet actually shows up in profit margins. And it shows up on a schedule, whether the customers arrive or not. Meta is the early warning. Operating margin fell to 31% from 43% a year earlier. CFO Susan Li said the year-over-year expense growth came from “higher depreciation, data center operating costs, and third-party cloud spend.” Free cash flow collapsed to $784 million, from $8.55 billion a year earlier. EPS missed at $6.18 against a $7.2173 estimate, breaking a six-quarter streak. The stock is down 22.27% over the past year.
Companies do have a lever here, and it matters. The longer the assumed useful life, the thinner each year’s depreciation slice. On its July 29, 2026 call, Microsoft CFO Amy Hood told investors that “effective at the start of FY27, we are extending the estimated useful life of our data centers and office buildings from 15 to 25 years.” She described the impact as a “minimal benefit to FY27 operating income.” The change flatters near-term earnings. It does not change how much cash left the building.
The offset the bulls point to is contracted backlog. Microsoft’s commercial remaining performance obligations reached $678 billion, up 84%. Amazon’s AWS backlog sits at $496 billion. Andy Jassy framed the mechanics plainly on the July 30, 2026 call: “Data center capital is spent starting two years before we can put servers into them to start monetizing.” That gap is the risk.
Line Item Retirees Should Actually Watch Next Quarter
On the next set of quarterly reports, ignore the headline capex figure. Look for depreciation and amortization inside the cash flow statement or the cost of revenue footnote. Then compare its growth rate to the growth rate of the cloud or AI segment it is meant to support. If Azure grows 45% and depreciation attributable to AI infrastructure grows faster, operating margins compress further. If AWS grows 37% and depreciation grows slower, the buildout is paying. That single comparison, quarter by quarter, will tell you whether the $1 trillion bet is working long before any analyst does. The other side of the trade sits with the power, cooling, and networking suppliers cashing those checks, which we profiled in a free report on seven of them here: 7 Stocks Powering the AI Boom (That Aren’t Chipmakers).
Contact [email protected] for any questions or corrections.
