This is investment research, not personal financial advice.
Walmart (NASDAQ:WMT) fell 9.15% on 20 August, from $114.30 to $103.84, after reporting second-quarter FY2027 results and lifting its full-year outlook. The decline erased about $83.2 billion of equity value even though adjusted earnings per share rose 19.1% and the new FY2027 range moved up to $2.80-$2.87 (Walmart 2026b; Yahoo Finance 2026).
The contradiction disappears once the tariff refund is separated from the recurring business. A $2.9 billion refund raised second-quarter gross margin by part of a 96-basis-point gain, but management has already committed most of that cash to lower prices. Its Q3 outlook calls for just 2%-4% adjusted operating-income growth, against 3%-3.75% sales growth (Walmart 2026c; CNBC 2026). The final 9.15% reaction looks broadly proportionate rather than excessive. At the close, Walmart still carried a 36.6-times multiple on the midpoint of its raised FY2027 earnings range, while the next quarter removes the result's largest non-recurring aid.
The raised guide was not the surprise
The headline numbers were strong. Revenue rose 5.9% to $187.9 billion, or 5.1% in constant currency. Reported operating income increased 28.8% to $9.38 billion, and adjusted operating income in constant currency increased 17.4% to $9.2 billion. Adjusted EPS reached $0.81, up from $0.68. Global eCommerce revenue grew 23% and represented 24% of total sales (Walmart 2026b; Walmart 2026c).
Management also raised three full-year ranges. Constant-currency sales growth moved from 3.5%-4.5% at the start of the year to 4%-5%. Adjusted operating-income growth moved from 6%-8% to 7%-8.5%. Adjusted EPS moved from $2.75-$2.85 to $2.80-$2.87. None of that resembles a conventional earnings disappointment.
The market reacted to the bridge between Q2 and Q3. The Q3 base period contained $177.8 billion of sales, $7.3 billion of adjusted operating income and $0.62 of adjusted EPS. Against it, Walmart expects $0.62-$0.64 of adjusted EPS and operating-income growth of only 2%-4%. The low end implies no EPS growth. Capital expenditure for the full year was also revised from about 3.5% of sales to about 4%, which adds roughly $3.5 billion on a revenue base around $710 billion before allowing for growth (Walmart 2026c).
That combination changes the reading of the quarter. Q2 proved that Walmart can turn a discrete cost recovery into customer-price investment and still raise annual expectations. It did not prove that 17.4% adjusted operating-income growth is the new run rate.
The $2.9 billion refund belongs to customers now
Walmart's first-quarter 10-Q said it had not recognised any tariff-refund claim at 30 April because timing and recovery remained uncertain. Less than three months later, the company had received about $2.9 billion after the United States invalidated tariffs imposed under emergency economic powers. The Q2 filing shows the refund passing through cost of sales, with price investment and higher distribution fuel costs taking back part of the benefit (Walmart 2026d; Walmart 2026b).
The arithmetic helps frame its scale. Dividing $2.9 billion by about 7.96 billion shares gives $0.36 per share before tax, but that is not an earnings contribution estimate. Walmart did not disclose how much was recognised in gross profit after lower prices, fuel costs and timing effects. The company said only about $100 million remained unallocated, according to CNBC's account of management's briefing. Almost all of the recovery has therefore been assigned to merchandise prices or other business uses rather than retained as a permanent margin step (CNBC 2026).
Gross margin rose from 24.5% to 25.4%. The 96-basis-point expansion was the largest of the past five quarters. Yet adjusted operating expenses rose 40 basis points to 21.4% of sales as depreciation, employee health costs and self-insured claims increased. The clean recurring gain was smaller than the gross-margin chart suggests. Advertising and membership mix helped, but they shared the quarter with a refund that cannot recur at the same size.
There was also a $2 billion-plus fuel cost headwind in distribution and fulfilment, which Walmart did not fully pass through to customers. That choice is consistent with its everyday-low-price model. It is also why the refund belongs in an event bridge, not in a forward margin base. The company used a cost windfall to protect price gaps while absorbing another cost shock. Economically, that may strengthen customer trust. Financially, the resulting Q2 margin cannot travel intact into Q3.
A grocery network with three extra tollbooths
Walmart is no longer just a low-margin merchandise chain. The physical network remains the foundation: thousands of stores, 192 U.S. distribution facilities and a private truck fleet put inventory within same-day reach of a large share of American households. Less than one-third of what the company carries in the United States is imported, which moderates direct tariff exposure relative to a general-merchandise specialist (Walmart 2026e; Walmart 2026d).
Grocery creates visit frequency. General merchandise raises basket profitability when demand is healthy. Pharmacy, optical and wellness services add another reason to return, although federal maximum-fair-price regulation and generic substitution reduced Walmart U.S. Q2 comparable sales by about 125 basis points. The total comparable-sales increase was 2.6%; excluding health and wellness it was 3.4%. Transactions excluding fuel rose 1.5%, while average ticket rose 1.2%. Those figures show demand was not manufactured solely by higher prices (Walmart 2026c).
Three businesses now collect higher-margin revenue from that traffic. Walmart Connect collects advertising spend from brands. The marketplace collects fees from third-party merchants. Walmart+ and Sam's Club collect membership income. Global advertising grew nearly 50% in Q2, U.S. marketplace sales rose 20%, and global membership-fee revenue increased 17%. These revenue streams do not require Walmart to own every item of inventory, and their economics are better than the core grocery basket.
The reinforcement loop is visible. Stores make rapid delivery economical. Delivery adds digital traffic. Traffic attracts marketplace inventory and advertisers. More selection and lower fulfilment cost can improve the customer offer, bringing traffic back to stores and apps. Global eCommerce growth of 23% at 24% of company sales is large enough to affect consolidated mix rather than sit in an experimental division.
There is a cost attached. Depreciation is rising because Walmart is automating distribution, remodelling stores and adding technology. Capital spending has climbed much faster than revenue since FY2023. The loop widens only if digital contribution profit and labour productivity outrun depreciation and fulfilment expense.
Five years show the margin repair and the capital bill
The table uses filed figures in US$ millions. Free cash flow is Walmart's reported non-GAAP measure, operating cash flow less capital expenditure. The ROI column is Walmart's reported return on investment, mapped to the schema's roic_pct field; it is not a standard NOPAT-based ROIC. Net debt to EBITDA is author-computed from cash, short-term borrowings, long-term debt and operating income plus depreciation and amortisation. H1 FY2027 is a six-month flow row, while ROI and leverage use trailing-twelve-month inputs.
| Period | Revenue | Net income | OCF | Capex | Reported FCF | Walmart U.S. comp | Reported ROI | Net debt / EBITDA* |
|---|---|---|---|---|---|---|---|---|
| FY2023 | $611,289m | $11,292m | $28,841m | $16,857m | $11,984m | 7.0% | 12.7% | 0.97x |
| FY2024 | $648,125m | $16,270m | $35,726m | $20,606m | $15,120m | 5.5% | 15.0% | 0.79x |
| FY2025 | $680,985m | $20,157m | $36,443m | $23,783m | $12,660m | 4.8% | 15.5% | 0.71x |
| FY2026 | $713,163m | $22,270m | $41,565m | $26,642m | $14,923m | 3.1% | 15.1% | 0.77x |
| H1 FY2027 | $365,688m | $12,019m | $19,710m | $14,181m | $5,529m | 2.6% | 15.4% | 0.82x |
Sources: Walmart FY2024-FY2026 Forms 10-K, Q1 FY2027 Form 10-Q and Q2 FY2027 release and presentation (Walmart 2024; Walmart 2025; Walmart 2026b; Walmart 2026c; Walmart 2026d; Walmart 2026e). * Author calculation.
Revenue compounded at 5.3% a year from FY2023 to FY2026. Net income nearly doubled, partly because FY2023 carried opioid and restructuring charges and later years included investment-value movements. Walmart U.S. comparable-sales growth slowed from 7.0% to 3.1% as inflation and pandemic-era comparisons faded. The cleaner achievement is the ROI recovery from 12.7% to roughly 15%, even as the capital base expanded.
Capital expenditure tells the other half. It rose from $16.9 billion in FY2023 to $26.6 billion in FY2026, a 16.5% annual rate. Revenue grew about one-third as fast. In H1 FY2027, another $14.2 billion of capital spending consumed 72% of operating cash flow. Reported free cash flow fell to $5.5 billion even though operating cash flow increased by $1.4 billion (Walmart 2026b).
The 15.4% reported ROI therefore matters more than the Q2 refund. Its numerator adds back depreciation and rent to operating income, while its denominator restores accumulated depreciation and removes non-interest-bearing operating liabilities. It is generous compared with conventional after-tax ROIC, but it is internally consistent through time. The year-on-year increase says operating improvement is still keeping pace with a rapidly expanding asset programme. A fall below 14.5% would say the opposite.
The cash bridge gets harder after the income statement
FY2026 operating cash flow was $41.6 billion. Subtracting $26.6 billion of capital spending produced $14.9 billion of reported free cash flow. At the post-result market value, that is a 1.8% historical free-cash-flow yield. H1 FY2027 free cash flow annualised mechanically would produce only a 1.3% yield, though Walmart's working capital and fourth-quarter seasonality make that extrapolation crude.
A second bridge estimates owner earnings by separating maintenance from growth capital. Depreciation and amortisation was $14.2 billion in FY2026. If that amount approximates maintenance spending, an upper-bound owner-earnings proxy is:
| FY2026 owner-cash bridge | US$m |
|---|---|
| Operating cash flow | 41,565 |
| Less: total capital expenditure | (26,642) |
| Reported free cash flow | 14,923 |
| Add: capex above depreciation and amortisation | 12,439 |
| Owner-earnings proxy | 27,362 |
This is an author calculation, not a company measure. Depreciation can understate maintenance needs, especially when store automation changes the asset mix, so $27.4 billion is an upper bound rather than a clean distributable number. It produces a 3.3% yield on the market value. The gap between 1.8% reported free-cash-flow yield and 3.3% owner-earnings yield is the value presently trapped in the investment programme.
The balance sheet can carry it. Author-computed net debt was about $34.0 billion at January 2026, or 0.77 times operating income plus depreciation and amortisation. By July, comparable net debt had risen to about $38.9 billion and the ratio to 0.82 times. Walmart also reported $11.5 billion of cash and $57.2 billion of total debt on its broader definition (Walmart 2026b). There is no near-term solvency problem. The analytical issue is whether debt and retained cash are financing assets that earn more than their cost.
Working capital needs context. Walmart had a $22.6 billion working-capital deficit at January 2026 because customers pay immediately while suppliers are paid later. That negative working capital is a funding advantage when inventory turns quickly. Q2 inventory rose 6.7%, faster than 5.9% revenue growth. One quarter does not establish deterioration, but inventory growth above sales growth deserves attention if it repeats.
Scale is widening online, but Costco keeps the price discipline honest
The scale moat looks wider in digital distribution. Walmart U.S. gained share across categories and income groups in Q2, with upper-income households leading the gains. Twenty-three percent global eCommerce growth on a vast base is stronger evidence than app downloads or management claims. Store density lowers last-mile distance, and the marketplace adds selection without matching inventory investment item for item.
The counter-evidence comes from Costco. In the 12 weeks to 10 May 2026, Costco's U.S. comparable sales excluding gasoline and currency grew 7%, against Walmart U.S.'s 3.4% excluding health and wellness in the July quarter. Costco's digitally enabled comparable sales grew 21%, close to Walmart's pace, while fee revenue rose 11% and U.S./Canada renewal held at 92.2% (Costco 2026). The periods and definitions differ, so this is directional rather than a like-for-like contest. It still shows that Walmart does not own value retail or digital grocery momentum.
Costco also states openly that it protects pricing authority rather than maximising short-term merchandise margin. Walmart's decision to return the tariff refund through price follows the same logic. The moat is customer trust backed by cost scale, not the temporary retention of a supplier or government windfall.
Macro demand was adequate but not effortless. U.S. retail and food-service sales in July fell 0.6% from June and rose 5.0% from a year earlier. The Census Bureau cautioned that its monthly estimate is not adjusted for price changes and carries sampling uncertainty (Census 2026). Walmart's 2.6% U.S. comp and 1.5% transaction growth therefore came against a mixed monthly tape, with pharmacy deflation reducing its own reported comparison.
A wider moat does not automatically justify any valuation. Advertising, membership and marketplace revenue must become visible in cash return faster than capex and depreciation rise. Costco's performance keeps that test demanding.
The new chief spent $3 billion at $117.61
John Furner became Walmart's chief executive in February 2026 after running Walmart U.S. The transition matters because the current programme combines larger capital commitments, AI and automation spending, aggressive price investment and shareholder distributions. These are linked choices, not separate initiatives (Walmart 2026e).
Walmart repurchased 25.7 million shares for $3.0 billion in Q2 at an average $117.61. The 20 August close was 13.3% lower. Marking only those shares to the close produces a roughly $354 million difference. That does not settle whether the repurchase was sensible over a multi-year period, but it illustrates the cost of retiring shares while the equity carries a premium multiple (Walmart 2026c).
Year-to-date repurchases were $5.1 billion. Dividends and Q2 repurchases together returned about $5.0 billion during the quarter, while H1 free cash flow was $5.5 billion. Capital distributions have nearly absorbed the cash left after investment. Net debt has also risen. With leverage below one turn of EBITDA, the company has room, but the hierarchy deserves scrutiny: stores and automation first, price trust second, distributions third. Repurchases at more than 40 times trailing owner cash can dilute the economics of excellent operating execution.
Management's strongest evidence is the reported ROI. It increased to 15.4% despite a larger asset base, and the full-year operating-income range moved up. The anti-thesis to the capital-allocation concern is straightforward: if automation lowers fulfilment labour, while ads and membership raise mix, today's capex is building a higher-return retailer. FY2027 and FY2028 cash statements will show whether that claim survives the spending phase.
What $103.84 requires from the next four years
The post-result price is 36.6 times the $2.835 midpoint of FY2027 adjusted EPS. That multiple is the starting fact. A mature grocery and general-merchandise base cannot support it through sales growth alone. Walmart needs a larger share of earnings from advertising, marketplace, membership and automated fulfilment, plus continued share gains in the core business.
The scenario model capitalises FY2031 adjusted EPS and discounts the terminal equity value to August 2026. It is a multiple-based present-value model because Walmart's current free cash flow is heavily affected by an investment programme whose maintenance/growth split is not disclosed. The owner-cash bridge above acts as a separate cross-check.
| Case | FY2027 EPS base | EPS growth to FY2031 | FY2031 multiple | Discount rate | Present value per share |
|---|---|---|---|---|---|
| Severe downside | $2.75 | 3% | 22x-25x | 9.5% | $47-$54 |
| Bear | $2.80 | 6% | 27x-31x | 8.5% | $69-$79 |
| Base | $2.835 | 10% | 32x-36x | 8.0% | $98-$110 |
| Bull | $2.87 | 13% | 36x-40x | 7.5% | $126-$140 |
These ranges are author estimates, not management forecasts. The base produces FY2031 EPS of about $4.15 and places the $103.84 close near the middle of $98-$110. The severe case assumes the mix shift stalls and the multiple moves toward a conventional large-retailer level. The bull case needs four years of 13% EPS growth, a 36x-40x terminal multiple and a lower discount rate. It represents sustained digital profit and stronger capital productivity, not a repeat tariff refund.
A two-variable sensitivity holds the FY2027 base at $2.835 and the discount rate at 8%:
| Four-year EPS growth | 30x FY2031 | 34x FY2031 | 38x FY2031 |
|---|---|---|---|
| 8% | $85 | $96 | $108 |
| 10% | $92 | $104 | $116 |
| 12% | $98 | $112 | $125 |
The reverse valuation is plain. At a 32x FY2031 multiple and an 8% discount rate, the current close requires about 11.7% annual EPS growth for four years. A 35x endpoint lowers that requirement to about 9.2%. Either path is faster than the core revenue rate. The margin mix must do real work.
The FY2026 owner-cash cross-check is stricter. Even the upper-bound $27.4 billion owner-earnings proxy yields only 3.3%. Reported free cash flow yields 1.8%. Those yields can coexist with the market price only if current capital spending creates future cash growth or if the required return remains unusually low.
The anti-thesis is a better mix, not another refund
The bearish reading is easy to state: Q2 margin was flattered by a discrete $2.9 billion recovery, Q3 profit growth slows sharply, comparable sales are normalising, capital intensity is increasing, and the equity still carries a mid-30s earnings multiple after a large decline. If advertising growth cools or automation savings arrive late, the multiple has little protection from the cash yield.
The strongest opposing evidence is not management optimism. It is the operating data. Global eCommerce grew 23%, advertising nearly 50%, membership fees 17% and U.S. marketplace volume 20%. Walmart U.S. gained share across income groups. Reported ROI increased to 15.4%, and the company raised its full-year operating-income range even after setting aside almost all of the refund for customers (Walmart 2026c).
Pharmacy also makes the 2.6% U.S. comp look weaker than the merchandise engine. Excluding health and wellness, the comp was 3.4%. The roughly 125-basis-point pharmacy drag came from regulated price reductions and brand-to-generic transfers. Those changes reduce revenue per prescription without necessarily indicating lost customer demand. As the comparison resets, reported comp growth can recover even if transactions merely remain steady.
This is the reaction asymmetry. The market is right that the refund does not recur. It may be too harsh if it treats all 96 basis points of gross-margin expansion as temporary. Advertising, membership and business mix contributed too, and those lines are growing far faster than merchandise. Q3 will separate them.
Q3 is the first clean read
The next report has three jobs. First, it must show gross margin without a large new refund. A level below 24.5%, absent another discrete item, would suggest Q2's recurring margin improvement was thin. Second, adjusted operating income needs to reach the stated 2%-4% growth range while adjusted EPS stays within $0.62-$0.64. Third, Walmart U.S. comparable sales excluding health and wellness need to remain above 3% so the pharmacy reset does not obscure a broader slowdown.
The longer test is capital return. Reported ROI below 14.5% while capex stays near 4% of sales would indicate that assets are entering the denominator faster than profit is entering the numerator. Net debt to EBITDA above 1.0 times, or full-year free cash flow below $10 billion, would show distributions and investment pressing harder on the balance sheet. Neither condition is present now.
Source notes and confidence
Verification is full for the filed history, trigger, market close and event context. The Finance API sidecar resolved the identity and filings but had not ingested the 20 August close when checked; it stopped at $114.30 on 19 August. The final $103.84 close, $114.30 previous close and 9.15% decline were therefore reconciled to the fetched Yahoo daily bars. Shares are the market-cap-implied count from the same snapshot. The amount of tariff refund retained in Q2 earnings remains undisclosed, so this article does not estimate it.
Walmart's $83.2 billion results-day loss corrected an earnings multiple that had left little room for a mixed quarter. Yet the post-fall price still assumes roughly 9%-12% annual EPS growth, depending on the terminal multiple. The November report will show whether advertising, membership and automation can supply that growth after the refund has passed through the checkout.
References
- CNBC 2026, "Walmart raises full-year outlook after a tariff refund boosts profits," 20 August 2026.
- Census 2026, Advance Monthly Sales for Retail and Food Services, July 2026, 14 August 2026.
- Costco 2026, Costco Wholesale Corporation Q3 FY2026 Form 10-Q, 3 June 2026.
- SEC 2026a, SEC company ticker registry for Walmart Inc. (WMT), accessed 20 August 2026.
- Walmart 2024, Walmart Inc. FY2024 Form 10-K, 15 March 2024.
- Walmart 2025, Walmart Inc. FY2025 Form 10-K, 14 March 2025.
- Walmart 2026a, Walmart Inc. Form 8-K, 20 August 2026.
- Walmart 2026b, Walmart Inc. Q2 FY2027 earnings release, 20 August 2026.
- Walmart 2026c, Walmart Inc. Q2 FY2027 financial presentation, 20 August 2026.
- Walmart 2026d, Walmart Inc. Q1 FY2027 Form 10-Q, 5 June 2026.
- Walmart 2026e, Walmart Inc. FY2026 Form 10-K, 13 March 2026.
- Yahoo Finance 2026, Walmart Inc. daily market data, 20 August 2026.