Public Storage Q3 2026: EPS fell 35% to $1.69 on NSA deal costs; core FFO per share rose to $4.34

Public Storage's third quarter was the first with National Storage Affiliates (NSA, closed July 22) and Public Storage Canada (closed September 1) inside the numbers, and the two deals pulled the headline figures in opposite directions. Total revenue rose 11.6% to $1.37 billion, almost all of it from newly acquired stores. Net income per diluted share fell 35.5% to $1.69 because the company booked $114.6 million of deal and integration costs and $122.3 million more depreciation on the bigger property base. Core FFO per share, the cash-earnings measure REIT investors use (explained below), edged up 0.7% to $4.34. Underneath, the stores the company has owned since before 2024 grew rent per square foot again for the first time this year, and management raised its full-year guidance.

Key figures

Two definitions carry most of this report:

• FFO (funds from operations) is net income with real-estate depreciation and gains or losses on property sales added back. Buildings are depreciated on paper every year even though storage properties usually hold or gain value, so REITs (real estate investment trusts) use FFO as their main earnings measure. Core FFO is Public Storage's own version that also strips out items it treats as non-recurring, such as deal costs and currency gains.
• Same-store results cover only the 2,755 facilities (192.1 million square feet, about 71% of the US portfolio) the company has run at a stable level since January 1, 2024. They show how the existing business is doing without the effect of buying new stores. NOI (net operating income) is property revenue minus the cost of running the properties. Occupancy is the share of rentable space that customers are actually renting.

The existing stores: occupancy and rent both edged up

Same-store revenue rose 0.5% ($4.8 million). The 10-Q gives the reasons: realized rent per occupied square foot rose 0.2% and average occupancy rose 0.3 points to 92.2%, partly offset by a 0.5% drop in late charges and administrative fees. The rent gain came from new customers: average contract rent for people moving in was $14.66 per square foot a year, up 9.3%. The 10-Q puts the occupancy gain down to "lower move-out activity and reduced customer churn"; churn (units vacated as a share of units occupied at the start of the period) fell to 20.8% from 21.9%.

That improvement had a cost. Move-in volume was down 7.1% by square footage, and promotional discounts given to new customers rose 5.6% to $16.5 million. Marketing spending rose 12.5% to $25.8 million, which the 10-Q attributes to "a higher volume of online paid search programs to attract new tenants". Repairs and maintenance rose 17.4%. Direct property costs rose 2.0% overall, yet same-store NOI still grew 0.5% because indirect costs (supervisory payroll and central management) fell 11.2% to $27.0 million.

Rent for the new customers is still well below what departing customers paid: $14.66 per square foot for move-ins against $19.85 for move-outs. Public Storage's model relies on raising rents on existing customers after they move in, so that gap closes over time, but it is why same-store revenue growth is measured in tenths of a percent rather than whole points.

The acquisitions

NSA brought 1,057 properties in 37 states and Puerto Rico. The total price was $6.44 billion: 11.2 million new Public Storage common shares ($3.49 billion), 4.1 million operating-partnership units ($1.22 billion), preferred shares and units ($385 million) and $1.35 billion of cash. Of the NSA stores, 313 went into a new joint venture in which Public Storage owns 20% and lent the venture $237 million at 10.22%. Public Storage Canada (68 properties, 5.3 million square feet) cost US$1.2 billion including assumed debt, paid partly with 2.76 million operating-partnership units.

Self-storage NOI from acquired facilities was $113.7 million in the quarter, up from $45.8 million, and the non-same-store pool's revenue rose 90.8%. That is where nearly all of the 11.6% revenue growth came from. Management expects $110 million to $130 million of annual run-rate savings and revenue gains from NSA within three to four years; the quarter shows the costs of the deal, not those gains yet.

Funding the deals raised debt. Total debt was $12.2 billion at September 30, net debt to EBITDA rose to 3.4x from 3.0x, and the weighted average interest rate rose to 3.6% from 3.0%. Interest expense rose $22.8 million (28.7%) to $102.5 million.

What the headline numbers hide

• The EPS drop is mostly deal accounting. The 10-Q lists the drivers of the $0.93 fall in EPS: $122.3 million more depreciation, $114.6 million of transaction and integration costs, $22.8 million more interest and $9.5 million more G&A, partly offset by $76.5 million more self-storage NOI and $19.6 million more ancillary NOI (mostly tenant insurance). The deal costs are one-time. The extra depreciation will recur, but it is a non-cash charge based on the price paid for the buildings.
• One-off items also helped. GAAP net income included a $27.0 million gain on property taken through eminent domain proceedings and a $16.7 million deferred tax benefit from releasing a valuation allowance as a result of the NSA deal. Both are excluded from core FFO. Netting them against the deal costs still leaves about $71 million of one-off drag on GAAP profit.
• Core FFO's other adjustments cut it rather than pad it. Besides the $114.6 million of deal costs, core FFO adds back $7.5 million of executive severance and CEO transition costs and $8.7 million of amortization of acquired non-real-estate intangibles (new this quarter, from NSA), but removes a $22.8 million tax benefit and a $9.9 million currency gain. Net of the deal costs, the remaining adjustments lower core FFO by about $14 million, so the adjusted figure is not inflated by its exclusions.
• The same-store cost line was helped by property tax refunds. Same-store property tax fell 0.8%, which the 10-Q says was "a result of refunds received during the quarter"; for the nine months it rose 1.4% on higher assessed values. The refund amount is not disclosed. Had the line risen at the nine-month pace of 1.4%, it would have been about $2 million higher and same-store NOI growth about 0.2% rather than 0.5%.
• Cash flow tracks core FFO once deal costs are counted. Operating cash flow for the quarter (nine-month $2,272.1 million minus the first half's $1,563.2 million) was $708.9 million, down 19.0% from $875.1 million. Adding back the $114.6 million of deal costs brings it to about $823 million, close to core FFO of $822.8 million, so the core figure is backed by cash.
• Share issuance, not buybacks. Diluted weighted-average common shares rose 5.2% to 185.1 million (7.6% to 189.6 million including operating-partnership units). The company bought back no shares this quarter and has signed forward sale agreements for another 1.23 million shares (about $399 million). Dilution is why an 8.2% rise in core FFO dollars became a 0.7% rise per share.
• Dividend timing. The $3.00 per share common dividend declared August 4 (about $561.0 million) was paid on October 6, after the quarter ended, so the quarter's cash distributions to shareholders were only $34.0 million against $575.9 million a year earlier. This is a payment-date shift, not a dividend cut: $583.3 million of dividends payable sits on the September 30 balance sheet.

On balance the quarter is understated: GAAP EPS fell 35.5% mainly because of $114.6 million of one-time deal costs (about $0.60 per share) and acquisition-driven depreciation, while core FFO per share, backed by operating cash flow, rose 0.7%.

Takeaway: The 35% fall in EPS is the cost of buying NSA, not a sign the business weakened. The more important change is in the existing stores: rent per occupied square foot rose for the first time this year, on 9.3% higher move-in rents and lower churn, though it took more marketing spending and discounts to get there. Per-share growth will stay slow while the new shares issued for NSA are absorbed.

Did last time's read hold up?

Our Q2 report said to watch three things in Q3: whether realized rent per square foot would stop falling, whether property tax and marketing inflation would ease, and whether any of the NSA synergy target would start to show. Rent did stop falling (+0.2% after -0.8%). Costs were mixed: property tax fell 0.8%, but only because of refunds, and marketing rose 12.5%, faster than in Q2. The NSA synergies have not shown up yet; the quarter carried $114.6 million of deal costs instead. Our point that heavy 2026 borrowing would weigh on per-share results held: interest expense rose 28.7%.

Outlook

Management raised its 2026 guidance for the second quarter in a row:

Core FFO for the first nine months was $12.74 per share, so the new range implies $4.16 to $4.36 for the fourth quarter, a midpoint of $4.26, below Q3's $4.34. That fits a full quarter of extra shares and interest from the NSA and Canada deals before cost savings arrive. CEO Tom Boyle said the company continues "to post encouraging trends on churn, occupancy, and move-in rates."

Our read: same-store results are now growing again, but slowly, and the 9.3% rise in move-in rents is the figure to follow, because it determines how quickly new customers' rents catch up with those of the customers leaving. In Q4 and 2027, the things to watch are whether same-store revenue growth moves above 0.5% without more promotional discounts, whether marketing costs keep rising faster than revenue, and whether NOI from the NSA stores starts to rise as Public Storage applies its pricing and operating systems to them. Until then, per-share growth depends on the acquired stores earning more than the cost of the shares and debt used to buy them.