Home Depot Q2 (18 August 2026): Comps Beat at 1.7%, but Transactions Fell 1.0% and Ticket Did All the Work

Home Depot beat on every headline line and told you almost nothing new about the consumer. Second-quarter sales came in at $47.9 billion, up 5.7%, against consensus of roughly $47.2–47.5 billion. Comparable sales rose 1.7% against expectations near 1%, US comparable sales rose 1.3%, and diluted earnings per share of $4.79 cleared the roughly $4.71 expected. Guidance was reaffirmed unchanged. But the composition is the same machine that has been running all year, only more so: comparable average ticket rose 2.8% while comparable customer transactions fell 1.0%. Fewer visits, each one worth more — and with management having described roughly 3% of pricing as already in the market, close to the entire ticket gain is explained by price rather than by bigger baskets. The demand line did not turn. It got repriced.

What actually happened

The figures below are from the company's second-quarter results release filed with the SEC on 18 August 2026, covering the 13 weeks to 2 August.

Against the consensus set out in the table further down — roughly $4.71 of earnings per share on $47.2–47.5 billion of revenue — every headline line cleared. McPhail's comment in the release was that "our second quarter results exceeded our expectations. We saw broad based demand across the business as customers continued to engage in smaller projects."

That last clause is the one to hold onto. It is the same description management gave in May, and it is the opposite of the recovery a comp beat might be taken to imply.

Ticket did the work, and ticket is mostly price

A comparable sales figure is an identity, not an observation. It decomposes into how many transactions happened and how much each was worth, and those two components moved in opposite directions again.

Comparable customer transactions fell 1.0%. Comparable average ticket rose 2.8%. In absolute terms, 443.2 million transactions against 446.8 million a year ago, at an average of $92.50 against $90.01. Traffic did not return in the biggest selling weeks of the year; it declined, and by more than the 0.4% it declined in the same quarter last year.

The one thing that did improve is the rate of decline. Transactions fell 1.0% against 1.3% in the first quarter, and the comp itself more than doubled from 0.6% to 1.7%. A category that grew 6.0% nominally across May to July, per the retail sales data discussed below, was always going to lift the comp somewhat. What it did not do is bring people back through the door.

Where the margin actually moved

The margin story runs the opposite way to the one usually assumed when a retailer discusses tariffs.

Gross profit rose 6.5% on sales growth of 5.7%, which lifted gross margin to 33.7% from 33.4% — an expansion of roughly 27 basis points, and comfortably above the approximately 33.1% embedded in full-year guidance, though the second quarter is seasonally the strongest mix of the year. So the cost of the goods themselves was not the pressure point in this quarter.

Operating margin nevertheless fell, to 14.3% from 14.5%. The reason sits one line lower: selling, general and administrative expense grew 8.5%, comfortably faster than the 5.7% sales line, and depreciation and amortisation grew 5.7%. Total operating expenses grew 8.2% against gross profit growth of 6.5%. That is operating deleverage, and a business absorbing the running costs of acquired distribution branches alongside the fuel and energy costs management has flagged.

The earnings-per-share growth, for once, is not financial engineering. Diluted share count was 996 million against 994 million — slightly higher, so buybacks did not flatter the per-share line this quarter. The effective tax rate rose to 24.5% from 24.2%. Net interest was a modest tailwind, falling 4.7%. Net earnings grew 4.7% and diluted earnings per share grew 4.6%, which is close to the 4.3% growth in operating income. What produced the earnings growth was the operating business.

That matters as a contrast with the first quarter, when adjusted earnings per share actually fell year on year, $3.43 against $3.56. Across the first half, adjusted earnings per share now sit at $8.35 against $8.24, up 1.3% — inside the flat-to-4% full-year guide, but only just, and with half the year gone.

What the 14 August retail sales report had already told you

Four days before Home Depot reported, the Census Bureau published the closest thing to a public preview of the quarter — and the way it was headlined was almost the opposite of what it contained for this company. With the result now in hand, it holds up: the category grew, and so did Home Depot, just by considerably less.

Advance retail and food services sales for July 2026 came in at $763.6 billion, down 0.6% on the month against expectations of a small gain, and up 5.0% on the year. That monthly fall is the largest since May 2025 and it is statistically real: the 90% confidence interval on the change is ±0.4 percentage points, so zero is excluded. June was left unrevised at +0.2%.

But a total is a weighted average of very different things, and the composition is where a single-category retailer lives.

The 0.6% decline was overwhelmingly autos and online. Motor vehicle and parts dealers fell 1.8%, and within that, auto and other motor vehicle dealers fell 2.0% after a 2.4% June gain — the shape of a pull-forward unwinding rather than a consumer stopping. Nonstore retailers fell 2.2% while still running 10.4% above the year-ago quarter. Strip out autos and the total fell 0.3%; strip out autos and gasoline and it fell 0.2%, a change the Census Bureau itself flags as not statistically distinguishable from zero.

Two caveats keep this from being a forecast, and they matter. First, these figures are nominal — the Census Bureau adjusts for seasonality and trading days but explicitly not for price changes. With merchandising head Billy Bastek describing roughly 3% of pricing already in the market and July CPI at 3.4% headline, deflating a 6.7% nominal category gain leaves real volume growth somewhere around 3%: real, but roughly half the headline. Second, NAICS 444 is a category, not a company. It contains Home Depot, Lowe's, and every independent lumberyard and garden centre in the country, and Home Depot's reported total also carries SRS and GMS, which sit in wholesale distribution rather than retail. A category growing 6% is consistent with Home Depot comping anywhere from negative to mid-single-digit depending on share.

What the data does establish is a negative: whatever weakened the July consumer, it was not home improvement demand. The category that is shrinking on a year-on-year basis is furniture and home furnishings, down 1.2% — the discretionary, move-triggered purchase that sits right beside a remodel in a household's decision order. That divergence is the deferral thesis showing up in third-party data: maintenance and smaller projects continue; the things people buy when they move do not.

Who signed the results

On 12 August, Home Depot announced interim management plans while its chief executive takes a temporary medical leave. The company said it expects Ted Decker to return within the next few months. In the interim, and in line with Decker's own recommendation, senior executive vice president Ann-Marie Campbell provides oversight of day-to-day operations while executive vice president and chief financial officer Richard McPhail provides oversight of financial management and the Pro subsidiaries. Independent lead director Greg Brenneman chairs the board during the leave, and said of the arrangement that both executives "are strong, seasoned executives who have worked together for more than 20 years."

The arrangement was visible in the results release itself. Where a quarterly release would ordinarily carry a quote from the chief executive, this one carried two: McPhail on the numbers, and Campbell thanking the associates. Decker's name does not appear in the results commentary.

The practical consequences are narrow and worth stating without embroidery. The guidance framework and the storm-activity assumption discussed below were set by this management team and remain the company's stated plan; nothing in the 12 August release revised them, and the 18 August release reaffirmed guidance unchanged. McPhail is the executive who supplied most of the mechanical detail on the first-quarter call — the SRS comp drag, the tariff-refund assumption, the fuel headwind — and he now also carries oversight of the Pro subsidiaries where SRS sits. That an interim arrangement produced a beat and an unchanged outlook is the least eventful outcome available, and on the evidence of the release, it is what happened.

What the quarter was expected to deliver

Home Depot reported before the market opened on Tuesday 18 August and held its second-quarter earnings conference call at 9:00 a.m. Eastern. The quarter ended on 2 August and covers May, June and July. The bar it had to clear is worth keeping visible, because it frames how much of a beat this was.

That period matters more than any other in the fiscal year. On the May call, management repeatedly noted that the largest selling weeks were still ahead — spring and early summer are when outdoor projects, garden, patio and seasonal categories concentrate. A first quarter landing in line with plan is a weak signal, because most of the year's demand had not yet been tested. A second quarter landing in line is a much stronger one.

Here is where expectations sit against what the company has already delivered.

Two details framed the bar. Adjusted earnings per share fell year on year in the first quarter — $3.43 against $3.56 — while full-year guidance calls for adjusted earnings growth of flat to 4%. And the comparable sales figure had been running below the midpoint of the flat-to-2.0% full-year range. Both implied the back half of the year was carrying an above-average share of the plan.

The second quarter relieved some of that. Adjusted earnings per share swung from a 3.7% decline to 5.1% growth, and the comp at 1.7% landed at the top of the full-year range rather than below its midpoint. But guidance was reaffirmed rather than raised, which is management declining to bank the beat — and the first-half comp of 1.2% is still only mid-range, so the back half continues to carry the plan.

The sales line and the demand line have separated

Home Depot's reported revenue growth and its underlying demand growth are now measuring genuinely different things, and conflating them is the most common error in reading this company.

In the first quarter, total sales rose 4.8% to $41.8 billion while comparable sales rose 0.6%. Comparable sales measure stores and branches that have been in the base for more than a year. The roughly four-point gap is acquisitions and new stores: SRS Distribution, bought in 2024, which delivered $4 billion of sales in the quarter on its own; GMS, which SRS acquired on 4 September 2025 for an enterprise value of about $5.5 billion; and the HVAC distributor Mingledorff's, added since. Ted Decker described the resulting network as more than 2,360 stores, 325 customer-facing warehouses and over 1,300 SRS branches.

The second quarter did not close that gap. Total sales grew 5.7% against a 1.7% comp — four points of difference, essentially unchanged from the first quarter's 4.2 points. The balance sheet shows why the gap persists: goodwill stood at $22.9bn against $19.6bn a year earlier and acquired intangibles at $10.5bn against $8.8bn, with $1.3bn of cash paid for businesses in the first half against $233m in the same period last year. At quarter end the company operated 2,364 retail stores and over 1,340 SRS locations.

The composition of the comp is more revealing than its level. In the first quarter, comp average ticket rose 2.2% while comp transactions fell 1.3%; in the second, ticket rose 2.8% while transactions fell 1.0%. Fewer visits, more spent per visit, in both. Big-ticket comparable transactions — purchases over $1,000 — were positive 0.8%, and nine of sixteen merchandising departments posted positive comps. Management's own summary of the shortfall was specific: "larger discretionary projects remain under pressure."

Why housing turnover, not confidence, sets the ceiling

The channel from interest rates to Home Depot's income statement does not run mainly through the cost of borrowing. It runs through how often houses change hands.

Large renovation projects cluster around moves. People redo a kitchen when they buy, prepare a bathroom when they sell, or draw on equity after a refinancing. When turnover is depressed, the pipeline of project-triggering events shrinks regardless of how healthy household balance sheets are.

Freddie Mac's weekly survey put the 30-year fixed-rate mortgage average at 6.67% in the week of 13 August 2026, down from 6.69% the previous week and against 6.58% a year earlier, with the 15-year average at 5.96% from 6.01%. That was the first decline in six weeks, and it is the right size to notice and the wrong size to matter: two basis points lower than a week ago and nine higher than a year ago is a rate that has not unlocked the market in either direction. The National Association of Realtors reported June existing-home sales at a seasonally adjusted annual rate of 4.09 million, down 2.4% on the month and up 2.8% on the year, with a median price of $440,600 and 4.6 months of supply.

The industry forecast says the same thing from a different direction. The Leading Indicator of Remodeling Activity from Harvard's Joint Center for Housing Studies projects homeowner improvement and repair spending edging from about $517 billion in the second quarter of 2026 to roughly $519 billion by mid-2027 — growth decelerating to about 0.5% year on year, below inflation. The Joint Center's Rachel Bogardus Drew noted that "growth in remodeling permitting and retail spending on building products have flattened recently", and managing director Chris Herbert put the dependency plainly: "Until home sales rebound from current low levels, remodeling expenditures are likely to stay at this pace."

Deferral or deterioration — the distinction that decides everything

Asked in May whether persistent high mortgage rates were destroying demand or merely postponing it, Decker's answer drew the distinction that matters. He pointed to a consumer he described as "remarkably resilient" — core customers who own their homes, saw substantial gains in home values over recent years, and are supported by employment and wage growth. The constraint, in his framing, was not capacity to spend but willingness to commit: "the main thing is just this uncertainty that's holding them back from taking on large projects." He then added the mechanical constraint on top: "with the higher rates, housing turnovers remain low", with new construction starts and sales also trending down.

That is the difference between a category that is coiled and one that is impaired. Deferred demand accumulates; deteriorated demand does not. The first-quarter data leaned toward deferral — big-ticket transactions were modestly positive and nine of sixteen departments comped positively, while the specific weakness sat in large discretionary projects.

The 14 August sentiment data sharpened that distinction rather than settling it. The University of Michigan's preliminary August reading put the Index of Consumer Sentiment at 51.0, down 7.6% on the month from 55.2 and down 12.4% on the year, with current conditions at 51.8 and expectations at 50.6. Survey director Joanne Hsu noted that sentiment "fell about 8% this August, ending two consecutive months of improvement", and that "only 8% expect their income growth to exceed inflation in the year ahead, down from 18% in December 2024." Year-ahead inflation expectations sat at 4.3% against 4.2% previously, with the five-year measure unchanged at 3.3%.

Set that against a building-materials category growing 6.7% year on year in the same month and the picture is specific: households reporting a bleak outlook are still buying materials. That is the signature of deferral rather than deterioration — sentiment surveys capture willingness to commit to the large, optional, financed purchase, which is precisely the line management described as under pressure, while the smaller repair and maintenance spending that makes up the bulk of the category carries on. It also carries a warning. An expectations index at 50.6, with only 8% of households expecting to outpace inflation, is not the backdrop against which a deferred kitchen becomes a booked kitchen. Deferral can persist for a very long time without ever converting.

The second quarter is a harder test, because the seasonal excuse is gone. This is the quarter when the projects actually get done.

The cost side: fuel up, tariff refunds pending

Two cost stories run underneath the margin line, pulling in opposite directions.

Fuel is the headwind. McPhail noted that Home Depot carries considerable transportation expense in its profit and loss account and that higher fuel prices hit both directly and through input costs, while cautioning that fuel's effect on consumer demand is hard to separate from the broader interest-rate environment.

Tariff refunds are the assumed offset, and the second-quarter release promoted that assumption from a call remark to a stated component of guidance. The release says in terms that fiscal 2026 guidance "includes IEEPA tariff refunds, which are expected to partially offset unplanned fuel, energy, and other product input costs throughout the fiscal year" — language that also confirms the fuel and energy costs are running above plan. In May, McPhail had said the company filed for the refunds, had received an immaterial amount to date, and had "assumed that that could provide a significant offset to those costs."

The quarter's margin split is consistent with that: gross margin expanded 27 basis points to 33.7% while operating margin contracted 20 basis points to 14.3%. Whatever is squeezing this business in the current quarter is sitting in operating expense, not in the cost of goods. That places Home Depot inside a much larger corporate cash-flow story — one this site covered in why Q2 was the peak tariff-refund quarter for US corporate margins. Whether that assumption has begun converting into cash is a legitimate question for Tuesday's report.

On pricing, Bastek said in May that the price increases already put into the market had "settled in", describing the level as around 3%, and that the company was lapsing the earliest tariff-driven cost pieces. The sourcing position behind that: in May 2025, McPhail told CNBC the company intended to "generally maintain our current pricing levels across our portfolio", and said that within a year no single country outside the United States would represent more than 10% of purchases.

That 3% pricing figure connects directly to the inflation data. July CPI came in at 3.4% headline and 2.5% core, with core goods turning positive for the first time in three months — the detail examined in the July CPI breakdown. Home improvement is an unusually goods-heavy, import-exposed basket, which makes a retailer of this scale one of the more direct places to observe whether tariff costs are reaching shelf prices or being absorbed in margin. That question feeds back into the rate expectations scoring the US dollar on the meter.

The scenario map, resolved

The point of a scenario map is not to pick one. It is to know in advance which observation distinguishes them, so the report can be read in the first ninety seconds rather than the first ninety minutes. Here is the map this piece carried into the print, marked against what the release showed.

The last row is the one that survives the beat. A category growing 6.0% nominally across May to July and a company comping 1.3% in the US are not the same story, and the distance between them is not explained by demand — the demand was there in the category data. It is explained by share, mix, and the fact that a nominal category figure is not deflated while a company comp is competing against its own prior-year base.

And the single most important framing carried in still stands, because a guidance reaffirmation did nothing to change it. Decker stated that the higher second-half comparable sales embedded in guidance is "solely driven by a return to normal storm activity", and that the company is "not looking at a marked improvement in underlying demand." Guidance is therefore not a forecast of consumer recovery. It is a forecast of weather reverting to normal after a quiet storm season, plus market-share gains. Anyone reading Tuesday's reaffirmation as a statement of confidence in the consumer is reading something management did not say — and the transaction count is the line that says so most plainly.

What it reads across to — and what it does not

Home Depot is a component of the Dow Jones Industrial Average and the S&P 500, but it is not among the mega-caps that dominate a daily index move, so the read-across is informational rather than arithmetic.

What it genuinely informs is narrow and useful: discretionary big-ticket spending by homeowners, a cohort with above-average balance-sheet strength. What it does not inform is the consumer in aggregate. Walmart reports on 20 August 2026, and with Home Depot's figures now on the table, the comparison between a staples retailer and a discretionary one is where the actual signal lives. Steady staples alongside soft projects describes postponement. Both soft together describes something broader — and would sit alongside the July control-group data as a genuine demand question rather than a housing-turnover one. The same distinction ran through June's retail sales report, where a control-group beat kept the dollar firm despite a soft headline: the composition, not the total, carried the information both months.

Home Depot's own answer to that comparison is a specific one to carry into Thursday: spend per visit up, visits down, management describing "smaller projects". If Walmart shows the same shape — ticket carrying the comp while traffic softens — the read is economy-wide pricing rather than a home-improvement quirk. If Walmart's traffic holds while Home Depot's falls, the read is that the deferral is specific to the financed, move-triggered purchase, which is exactly where housing turnover binds.

For a currency reader, the transmission is indirect and worth stating honestly rather than inflating: a single retailer's comparable sales do not move the dollar. What moves it is the rate path, and this report is one small input into how the market reads consumer resilience at current rates. That is the whole claim. The mechanism runs rates → turnover → projects → this income statement, and it runs far more strongly in that direction than in reverse. More on how the site frames these channels is on the about page.

Educational macro context only — not investment advice.