A Record Year Built on Refunds: AutoZone's Core Earnings Went Nowhere

Results vs. Consensus

AutoZone's fourth quarter spans 16 weeks and ended August 29, 2026. It closed a fiscal year in which the company passed $20 billion of sales for the first time. The print itself repeated the pattern of the preceding quarter: revenue short of the Street, earnings ahead of it.

Published consensus sat in a narrow band: EPS estimates ran $54.08 to $54.30 and revenue estimates $6.70 billion to $6.71 billion. The scorecard uses the higher pair, which is the one cited in post-print coverage. The direction of the beat and the miss is the same under every estimate we collected. No published consensus exists for gross margin, operating profit, same-store sales or free cash flow, so those lines are shown as not available rather than estimated against a number we would have had to invent.

Year-over-year comparison

Fiscal year comparison

The full-year figures tie to the four reported quarters. Nine-month revenue of $13,743.7 million plus the fourth quarter's $6,594.9 million equals the filed $20,338.6 million, and nine-month net income of $1,641.2 million plus $931.6 million equals the filed $2,572.8 million exactly. The fourth quarter carried an outsized share of the year: 32.4% of revenue in 30.8% of the weeks, and 36.2% of net income.

Sequential comparison, on a weekly rate

The fourth quarter contains 16 weeks and the third contains 12, so a dollar sequential comparison is meaningless for this company. The weekly sales rate is the comparable measure, and margin ratios compare directly regardless of quarter length.

The weekly sales rate is our calculation from the filed revenue and the disclosed quarter lengths, not a company-reported figure. The sequential margin improvement is almost entirely the tariff refund and the LIFO comparison landing in the fourth quarter; the comp lines moving the other way are the more informative part of this table.

Quality of the beat

Two of those rows need their arithmetic shown. Management stated that excluding this year's $15 million LIFO charge and last year's $80 million charge, EPS grew 8.5%. Working back from that figure, this year's ex-LIFO EPS is $56.05 plus the disclosed $0.69, or $56.74, so the prior-year ex-LIFO base is $56.74 divided by 1.085, or $52.29, implying $3.58 per share of prior-year LIFO drag. The share-count step is this quarter's net income divided by each year's diluted share count, a difference of $1.83. The residual of $2.68 is what is left over, and it is a plug rather than a company figure. Order matters in a bridge like this, and a different sequencing would move a few cents between rows. The conclusion does not depend on the sequencing.

Revenue. The 5.6% increase is better than it looks on comps and worse than it looks on quality. Total company same-store sales rose 2.7% as reported but only 1.5% in constant currency, because exchange rates added more than 900 basis points to the international comp. Store growth did the rest: 374 openings during the year took the base to 8,031 from 7,657. Within the comp, roughly 5% ticket growth was offset by falling transactions in DIY. Management described mid-single-digit like-for-like same-SKU inflation. That means the reported top line is substantially a price outcome at a time when unit demand is contracting, and it is the third consecutive quarter in which that has been true.

Margins. The 182 basis points of gross-margin expansion decomposes cleanly and unflatteringly. Each year's LIFO charge as a share of that year's sales gives 128 basis points in the prior fourth quarter and 23 basis points in this one, a 105 basis point swing that reproduces the filing's stated figure exactly. The tariff refund added a further 145 basis points. Those two items sum to 250 basis points against 182 reported, which leaves roughly 68 basis points of underlying pressure, attributed in the filing to higher commercial mix. Below the gross line, SG&A grew 8.9% against sales growth of 5.6% and deleveraged 101 basis points. The prior quarter had produced 25 basis points of leverage on 7.6% expense growth. On an underlying basis, then, this quarter's margin structure got worse in both directions at once.

EPS. The reported 15.1% increase is accurate and uninformative. Management's own ex-LIFO measure is 8.5%. Removing the tariff refund as well takes the growth rate to roughly zero, and removing the currency benefit takes it slightly negative. None of this is a criticism of the reporting, which was unusually clear and gave the per-share impacts without being asked. It is a statement about what the quarter proves. On the evidence of this print, AutoZone's operating earnings power was flat at best year over year, and the headline grew because of a tariff policy reversal, an inventory accounting comparison, the peso and the buyback.

Channel Performance

AutoZone does not publish a profit statement by channel. Commercial sales are reported inside domestic sales, and the only channel-level profitability disclosure is qualitative. The table below therefore organises the quarter by operating channel rather than by reportable segment.

Domestic commercial: the engine is intact, the slope is lower

Commercial generated $1,913.0 million, 34% of domestic auto parts sales and 29% of consolidated revenue. Growth of 8.6% missed the 10% to 12% range we carried into the quarter after last quarter's 10.4%. The cadence matters more than the average: the first three months of the quarter ran just over 8% and the final month ran well over 9%. For the full year commercial grew 10.6% to $5,762.4 million.

Network expansion again supplied part of the growth. The program count rose to 6,443 from 6,098, with 87 net additions in the quarter and 345 for the year, and commercial programs now sit in 94% of domestic stores. Average weekly sales per program reached $18,700, up 2.7%. That per-program growth rate is the number to watch: it decelerated from 4.5% last quarter and 9.0% in the year-ago quarter. Growth is increasingly coming from adding programs rather than from selling more through each one.

"There are approximately 2,000 commercial programs that are linked to a Mega Hub network. These programs sell 16% more annually than the balance of the commercial programs in the chain."
— Jamere Jackson, Chief Financial Officer

That is a new disclosure and a useful one. With 172 Mega Hubs at year end against a target of roughly 300 within three years, and roughly 2,000 of 6,443 programs currently inside the network, the lift is quantified and the runway is countable for the first time. Sixteen percent applied to the un-linked majority is a real arithmetic opportunity rather than a slogan.

Assessment: Commercial remains the growth engine and the Mega Hub disclosure strengthens the mechanism behind it. We lower the near-term growth assumption to the high single digits for the first half of fiscal 2027, in line with the guide, and note that per-program productivity rather than program count is now the swing factor.

Domestic DIY: the pillar broke on its own terms

The DIY comp was negative 0.6%. It ran negative for the first three months of the quarter and was essentially flat in August. Average ticket grew roughly 5% against same-SKU inflation that management characterised as mid-single-digit, which implies transactions fell by more than 5%. Management confirmed the magnitude in question and answer, and that figure is worse than the 3.6% traffic decline we recorded in each of the prior two quarters.

"We see evidence of deferrals and trade down in DIY, particularly with the most financially challenged DIY customers. This is not a new phenomenon for our industry, and we expect it to recover over time."
— Jamere Jackson, Chief Financial Officer

We have carried DIY resilience as a thesis pillar on the explicit basis that sales remained positive even while traffic fell. That is no longer the case. Management's framing, that a 5% transaction decline is not typical and should normalise toward the historical 1% to 3% range, is plausible on long history and consistent with an aging vehicle fleet. It is still a forecast. The quarter itself showed the ticket tailwind failing to cover the traffic gap for the first time.

Assessment: This pillar moves from at risk to broken. It can be repaired quickly, because the arithmetic only needs transactions to improve by a few points against a 4% to 5% ticket, and the August exit rate points that way. Until it is repaired, the domestic business is not contributing comp growth.

International: a genuine inflection, reported through a currency lens

International constant-currency comps grew 1.3%, slightly below the roughly 1.6% management indicated last quarter. The reported comp of 10.7% is a translation artifact and should not be read as demand. Mexico remains the constraint, and management was direct that the macro environment in the country is the cause rather than execution.

"Importantly, we saw an uptick in sales in the last four weeks of the quarter, which gives us confidence that sales may inflect upward in Q1."
— Phil Daniele, President and Chief Executive Officer

Store growth continued at pace: 68 openings in Mexico took the country to 1,001 locations, a milestone, and 10 in Brazil took that country to 167. International is now roughly 15% of the store base. The infrastructure to support it advanced materially during the year, with the relocated Monterrey distribution centre now more than twice its previous size and fully operational, a completed Brazil distribution centre already reducing in-country supply chain cost, and ground broken in August on a third Mexican facility due online late in fiscal 2028.

Against that, the fiscal 2028 store target was cut from roughly 500 openings to about 430, and the reason given was international.

"Specifically, we will slow the pace of our Brazil expansion to concentrate on the U.S. and Mexico in the near term."
— Phil Daniele, President and Chief Executive Officer

Assessment: The pillar stays at risk, with the balance of evidence improving on Mexico and deteriorating on Brazil. The Q1 guide of low double digit reported growth and slightly better constant-currency growth is the first forward indication of recovery we have had in three quarters, and the distribution infrastructure is now in place ahead of the demand. Brazil becomes a smaller part of the story, which reduces both the optionality and the risk.

Key KPIs

The Mega Hub count for the year-ago quarter is our derivation: 172 at year end less the 39 opened during fiscal 2026 gives 133 at the start of the year. Inventory grew 10.1% against quarterly sales growth of 5.6%, which is the intended consequence of the growth initiatives rather than a demand misread, but it shows up in turns falling to 1.3x from 1.4x and in payables funding a smaller share of the stock.

Key Topics & Management Commentary

Overall Management Tone: Management was more forthcoming than in any call of the past year, volunteering the per-share impact of every unusual item and publishing the new-store return curve unprompted. The posture on the domestic consumer was candid rather than defensive, conceding that the quarter fell below plan before describing the August recovery. Where the call was least convincing was on returns: the response to direct pressure on what the investment programme has earned stayed at the level of store-level averages and did not engage with the 550 basis point decline in group ROIC.

1. The tariff refund is a real $96 million and a real one-off

The single largest item in the quarter was the refund of tariffs collected under the International Emergency Economic Powers Act. Management sized it precisely and told investors what it was worth per share, which removes any ambiguity about the quarter's underlying earnings.

"Our results included a $96 million tariff refund and a $15 million LIFO charge this quarter. Excluding the $15 million LIFO charge in this year's quarter and the $80 million LIFO charge in last year's quarter, EBIT grew 4.4% and our EPS grew 8.5%."
— Jamere Jackson, Chief Financial Officer

Note what that formulation does and does not exclude. The ex-LIFO growth rates of 4.4% at EBIT and 8.5% at EPS still contain the full $96 million refund. Removing it as well takes EBIT growth below zero and EPS growth to roughly flat. Total IEEPA tariffs were described as a little over $100 million, most of it recovered in the fourth quarter on inventory already sold, with the remainder expected to arrive across fiscal 2027 in an unquantified amount.

Assessment: Full marks for the disclosure, which is more than many companies would have given. The analytical consequence is that fiscal 2026's record EPS carries roughly $4.43 that does not repeat, and the fiscal 2027 gross-margin guide has to absorb that lap.

2. The LIFO comparison turns from headwind to tailwind, and is already spent

LIFO has distorted this P&L for two years. The charge was $80 million in the year-ago fourth quarter and $15 million in this one, and $192 million for fiscal 2026 as a whole. Management gave the forward path with unusual specificity: a $40 million charge in the first quarter of fiscal 2027, roughly $15 million in each of the following three, and $85 million to $90 million for the year.

"In total, we are modeling $85 million to $90 million in LIFO charges for FY 2027. As a reminder, we recognize $192 million in LIFO charges for all of fiscal 2026."
— Jamere Jackson, Chief Financial Officer

That is roughly $104 million of pre-tax relief, worth about $5.06 per share at our fiscal 2027 tax and share assumptions. It is also, on the company's own gross-margin guide, already committed. Our FQ3 recap assumed a $75 million fiscal 2027 LIFO base, so the $85 million to $90 million range is marginally worse than we carried, while fiscal 2026 at $192 million came in better than the roughly $207 million outlook we were working from.

Assessment: The LIFO cadence pillar holds. The distortion is shrinking on a known schedule, which is progress. Investors should not count the relief twice, because the gross-margin guide spends it.

3. DIY transactions are the whole fiscal 2027 question

The domestic retail comp was negative for the quarter and transactions fell more than 5%. Management's case is that this is a cyclical overshoot driven by inflation, particularly fuel, and that history says it reverts.

"What we believe is, the 5%+ decline in transactions, particularly on DIY, will kind of start to moderate back to a normal trend of, historically, over long periods of time, it's been down 1% or 1%-3% in transactions, and you've gotten average ticket growth that's been higher than that."
— Phil Daniele, President and Chief Executive Officer

The supporting argument is structurally sound. Vehicle age keeps rising as new and used car prices and financing costs stay elevated, miles driven have not fallen despite higher fuel prices, and failure-related categories are close to inelastic. Management also noted that the two-year stacked ticket is up about 9%, which is the measure of how much price the DIY customer has absorbed.

Assessment: We accept the direction and discount the timing. Every element of the domestic outlook, the comp guide, the SG&A leverage point and by extension the earnings trajectory, rests on transactions improving from here. The quarter provided an exit rate rather than a trend.

4. The new-store return curve, published for the first time

Last quarter we flagged the absence of cohort-level return disclosure as one of the four things management was not saying. This quarter they said it, in detail and without being asked.

"With an average investment of around $2.9 million per store, we assume that return on invested capital is roughly zero in the first year. By the end of the fourth year, we expect to achieve approximately 15% ROIC. By the end of the sixth year, we expect ROIC of over 20%, which should continue to improve well after that."
— Phil Daniele, President and Chief Executive Officer

The sales curve was given alongside it: roughly $1.7 million in year one ramping to roughly $2.7 million by year six, with double-digit growth in year two and roughly 5% in each of the next four, and commercial sales more than doubling across the six years. Management described the current cohorts as slightly ahead of that plan on both sales and EBIT.

One implication deserves stating, because it is not obvious. A greater than 20% after-tax return on a $2.9 million investment against $2.7 million of year-six sales would require a store-level operating margin far above the group's 18.3%. The reconciliation is that invested capital per mature store falls well below the initial outlay, because payables fund the inventory: net inventory per store is negative $107 thousand, and accounts payable stand at 111.1% of inventory. The new-store return curve is therefore a function of supplier financing as much as of store productivity, which ties it directly to a ratio that moved the wrong way this year.

Assessment: This is the most valuable thing on the call and it upgrades the new-store pillar from a qualitative claim to an auditable one. It also hands investors a specific thing to monitor, because the payables ratio fell from 114.2% to 111.1% over the year.

5. SG&A discipline stopped working, and the guide says it stays that way

For three quarters the expense line was the reliable part of this story. Operating expenses grew 7.6% last quarter against 8.4% sales growth and produced 25 basis points of leverage. This quarter they grew 8.9% against 5.6% sales growth and cost 101 basis points. Per store, expenses grew 4% against 3% last quarter.

"We expect SG&A per store and total growth to be in a similar range in the first quarter of FY 2027, which is up roughly 4% on a per store basis or 8.5%-9% on a total growth basis."
— Jamere Jackson, Chief Financial Officer

For the full year the guide is roughly 3% per store and about 8% in total. Set that against the sales growth the comp guide implies, which management agreed is somewhere in the 2% to 5% comp range plus roughly 3.5 points of unit growth, and the expense line is guided to grow faster than revenue for a second consecutive quarter and probably for the year.

Assessment: This pillar moves from on track to at risk. Management's position, that SG&A can be managed down if sales disappoint, is credible from this team's record. The guide as given does not assume they will need to.

6. The fiscal 2027 gross-margin guide is tighter than it reads

The headline guide sounds benign and is not.

"Given the lumpiness of gross margins this year, excluding LIFO, you should expect gross margins to be flat to up 25 basis points on a GAAP basis in FY 2027."
— Jamere Jackson, Chief Financial Officer

Question and answer clarified that the range is the GAAP figure, inclusive of both the LIFO improvement and the tariff-refund lap. Work the two known pieces. Fiscal 2026's $192 million LIFO charge was 94 basis points of sales; $87.5 million on our fiscal 2027 revenue estimate is about 41 basis points, a tailwind of roughly 54 basis points. Fiscal 2026's tariff refunds were worth 48 basis points and do not repeat. Netting those two leaves about 6 basis points of improvement before any operating change, inside a guided range of zero to 25 basis points. On that arithmetic the underlying merchandise margin is guided roughly flat, with commercial mix pressure continuing.

Assessment: The guide is honest and the lap is real. It also means the one mechanical tailwind available to fiscal 2027 gross margin is already allocated, and nothing in the range implies underlying margin expansion.

7. ROIC fell 550 basis points, and the denominator explains most of it

Adjusted after-tax ROIC declined to 35.8% from 41.3%. Management addressed it by committing to a floor rather than a recovery.

"We will hold our ROIC in the mid-30s area and remain committed to driving higher ROIC over time as new stores continue to mature."
— Phil Daniele, President and Chief Executive Officer

The filing's own reconciliation table is more illuminating than the commentary. The adjusted after-tax return rose 2.9%, from $3,246.5 million to $3,342.1 million. Invested capital rose 18.6%, from $7,869.1 million to $9,333.0 million. Decomposing that $1,464.0 million increase: average debt fell $63.4 million, rent capitalised at six times rose $221.9 million, average financing leases rose $19.4 million, and the average stockholders' deficit narrowed by $1,286.1 million. That last item is 88% of the increase.

Why the deficit narrowed matters. AutoZone has spent decades repurchasing more stock than it earned, which drives a deepening negative equity balance and mechanically shrinks the invested-capital denominator. Fiscal 2026 repurchases of $2.0 billion ran below net income of $2.6 billion, so retained earnings grew faster than treasury stock and the deficit shrank. The reported ROIC decline is therefore mostly the unwinding of a balance-sheet flattery, not a collapse in asset productivity.

Assessment: The supply-chain and store investment cycle is genuinely pressuring returns, but far less than the headline implies. Investors marking AutoZone down for a 550 basis point ROIC decline are reacting substantially to a denominator effect. The honest reading is that returns on the operating assets are broadly holding while the capital base normalises, which is a better outcome than the number suggests and a worse one than 41.3% ever represented.

8. The investment cycle is ending, but the cash relief is not

Management walked through the state of each long-running programme, and the message was consistent: nearly done.

"Supply chain investments that started in earnest in 2020 are now in the final stages of completion."
— Phil Daniele, President and Chief Executive Officer

Two United States distribution centres and two direct import facilities are complete, the Brazil facility is operational and cutting in-country cost, Monterrey is rebuilt at more than twice the size, and the technology platform work is described as predominantly cloud-based and largely finished. Yet capital spending is guided to roughly $1.65 billion in fiscal 2027 against $1.496 billion in fiscal 2026, because the mix shifts from infrastructure to stores, and a third Mexican distribution centre arrives in late fiscal 2028.

Assessment: The pillar stays at risk. Free cash flow was $1,806.6 million for fiscal 2026, up 0.9% while capital spending rose $131 million, and management expects it higher in fiscal 2027 without quantifying by how much. The composition of spending improves this year; the quantum does not.

9. A record store year, and a lower long-run ceiling

AutoZone opened 175 stores in the fourth quarter and 374 for the year, the most in its history and above both the 355 to 365 range in the prior release and the roughly 365 indicated on the last call. First-quarter openings are guided to roughly 65 against 53, and the full year to roughly 400.

The counterpoint arrived in the same breath. The fiscal 2028 target came down from roughly 500 openings to about 430, with the composition given as roughly 300 in the United States, about 120 in Mexico and about 20 in Brazil. Management framed this as nearing a long-term run rate rather than as a retreat, and the domestic and Mexican components are unchanged.

Assessment: The store opening pace pillar moves from at risk to on track on delivery. The ceiling reduction is a 14% cut to the fiscal 2028 unit plan and is worth a lower terminal growth assumption, though it is concentrated in the market we infer contributes least to profit today.

10. Capital returns continue at a price that now looks high

The company repurchased $697.5 million of stock in the quarter and $2.0 billion for the year, retiring 579 thousand shares, and ended with $1.6 billion of remaining authorisation against $632.3 million a year earlier. Leverage finished at 2.5 times EBITDAR, unchanged.

The execution price deserves a comment that management did not make. Fiscal 2026 repurchases averaged $3,496 per share and fourth-quarter repurchases averaged $3,125, against a post-print close of $2,894.73. The year's buyback was therefore executed roughly 21% above where the stock now trades. That is not a criticism of the policy, which has compounded value for a quarter of a century, but it is the cost of a programme that buys steadily rather than opportunistically.

Assessment: The capital return pillar holds. The replenished authorisation and the unchanged leverage target mean the share count keeps shrinking roughly 3% to 4% a year, which is worth about $6.18 of the fiscal 2027 EPS increase in our model. It is also the single largest contributor to that increase, which is the problem with the earnings story rather than with the buyback.

Guidance & Outlook

AutoZone does not guide earnings. It does give an unusually complete set of line-item assumptions, and this call gave more than most.

Implied company comp: management accepted an analyst's sum-of-the-parts calculation putting the full-year comp somewhere between 2% at the low end and 4% to 5% at the high end. Adding roughly 3.5 points of unit contribution from the fiscal 2026 and fiscal 2027 opening programmes gives total sales growth of roughly 5.5% to 8.5%, before currency. We underwrite 6.0%.

Street at: the published consensus price target cluster sat around $3,656 to $3,887 before the print, which is 26% to 34% above the reaction close. The direction of revisions into the print was downward on comparable sales and on the multiple, and at least one framework was cut from the low twenties to the mid-to-high teens.

Guidance style: AutoZone guides inputs conservatively and has a long record of delivering the store programme it commits to. This year it exceeded the opening range and came in under the capital spending indication. The weak spot is the comp, where the Q4 commercial outcome of 8.6% landed below the 10% to 12% the prior call pointed us toward, and where the international constant-currency figure of 1.3% came in under the roughly 1.6% indicated.

Implied margin trajectory: this is the part of the guide that drives our downgrade. Gross margin flat to up 25 basis points, against SG&A growing about 8%, produces operating margin contraction of roughly 20 basis points at the 7% midpoint of implied sales growth, and roughly 50 basis points at our 6% underwrite. Fiscal 2026 already contracted 75 basis points. Fiscal 2026's realised 75 basis point contraction followed by a further implied decline in fiscal 2027 is not consistent with the premium multiple we were carrying.

Analyst Q&A Highlights

Whether the tariff refund repeats in fiscal 2027

The first question of the call went straight to the quarter's largest unusual item and to whether a second tranche arrives through the supplier channel rather than directly. Management sized the total programme and was clear that the bulk is already recognised, while declining to quantify the remainder.

Q: "On the IEEPA contribution, is most of the refund in the fourth quarter? Obviously, your direct import refunds come to you, but do you have suppliers who are going to be trickling out rebates to you as we get into 2027?"
— Brett Jordan, Jefferies

A: "Yeah. Most of that refund was in the fourth quarter. As we talked about, we had a little over $100 million that were in IEEPA tariffs. We got most of that in the fourth quarter, and that is basically on the inventory that we have sold. We expect the remainder of that to sort of trickle out over FY 2027."
— Jamere Jackson, Chief Financial Officer

Assessment: a straight answer that narrows the uncertainty without removing it. A little over $100 million in total with $96 million recognised leaves a residual in the single-digit to low-double-digit millions, immaterial to fiscal 2027. The exchange effectively confirms the $4.43 per share does not repeat.

The cadence of same-SKU inflation as the tariff comparison laps

Because ticket is carrying the entire comp, the inflation path is the comp path. The question probed the gap between the 5% ticket indicated for the first quarter and the 4% indicated for the year.

Q: "Then question on same-SKU inflation. I think you called out 5% in Q1, but you called out a 4% ticket contribution for fiscal 2027. Could you sort of talk about how you see the cadence of same-SKU inflation as we sort of roll over that peak tariff impact in late calendar 2025?"
— Brett Jordan, Jefferies

A: "Yeah. We think the first quarter will be in the similar zip code as the fourth quarter of this year, probably closer to 5%, and then we expect it to moderate over time, and we expect to average around 4%. What I will say is that it is a pretty volatile environment. In the near term, there is certainly pressure on categories like oil that are driving higher inflation."
— Jamere Jackson, Chief Financial Officer

Assessment: the honest reading of this answer is that ticket growth decelerates through the year while the comp guide improves, which requires transactions to supply the difference. That is an internally consistent forecast and a demanding one.

What the sum of the parts implies for the full-year comp

A recurring line of questioning tried to convert the channel-by-channel framework into a single company comp number, since no company comp was guided. Management confirmed the arithmetic without endorsing a point estimate.

Q: "I guess if you just do the weighted sum products, you'd get somewhere between a 2% at the low end of comp guide to something like a 4% or a 5%, and I know you're not guiding, but you're giving us a framework. Is that fair?"
— Simeon Gutman, Morgan Stanley

A: "Yeah. So, you're in the right zip code on the comps when you sort of do the math on the sum of the parts, if you will. I think the important thing for us is we're calling what we see today based on that ticket inflation being elevated and some of the trends that we're seeing in our business."
— Jamere Jackson, Chief Financial Officer

Assessment: useful, because it converts a vague qualitative guide into a 2% to 5% range the model can use. The low end of that range would not cover the guided expense growth, which is the crux of the fiscal 2027 earnings risk.

Where the SG&A leverage point sits

The most important exchange on the cost structure asked directly what comp rate the expense plan requires in order to leverage. Management declined to name one and reframed the question as a matter of flexibility.

Q: "As you think about the SG&A leverage point, is the model that you're building predicated on 4% comps to drive SG&A leverage, understanding that you have these upfront investments, supply chain and technology and people and process, but at the same time, you're going to have sort of these more immature stores that are burdened by labor and rent and so forth."
— Christopher Horvers, JPMorgan

A: "Yeah, I wouldn't say that it's built on a 4% leverage point. Clearly, the way that we think about it is that we have the ability to grow SG&A in line with sales. In environments where those sales are a little bit lower, then we have the ability to go into the P&L and manage the SG&A downward."
— Jamere Jackson, Chief Financial Officer

Assessment: a capability claim offered in place of a number. The claim is supported by history, including the expense leverage delivered as recently as last quarter. It sits awkwardly against a guide that explicitly plans expense growth above the implied sales growth, and no leverage threshold was disclosed.

Whether DIY trade-down and deferral are still deteriorating

With the retail comp negative, the question was whether the consumer behaviour behind it had worsened within the quarter. The answer supplied the transaction magnitude that the release had omitted.

Q: "I wanted to follow up on the DIY side. I was curious if you could just give a little bit more of an assessment of what you are seeing. You talked about softer transactions. I think there are so many moving pieces with the consumer at the moment. Have you seen some of that activity, whether it is trade down or deferral, get worse? How should we think about DIY from a quarterly cadence over the course of FY 2027?"
— Steven Zaccone, Citi

A: "The 5% decline or north of that in transactions is not typical in this industry, and we would expect those to moderate back into a more normal traffic rate. So we saw that improving coming out of the quarter. We like the exit rate on DIY, and we think that momentum will kind of continue."
— Phil Daniele, President and Chief Executive Officer

Assessment: the most consequential number of the call came out in question and answer rather than in the release, which did not mention traffic. A transaction decline above 5% against roughly 5% ticket is the arithmetic of a negative comp, and it is two points worse than the trend of the prior two quarters.

Whether the fiscal 2027 outlook depends on a transaction recovery, and what happens if it does not arrive

This was the sharpest exchange on the call. It put the central assumption to management explicitly and asked for the downside protection if the assumption fails. Management confirmed the assumption and answered the second half only by reasserting the first.

Q: "The heart of the assumption is that transactions improve from here. So A, is that right? And B, if your comps do not fall into, let us call it a 3%-4% range, is there enough flexibility for you to manage the P&L to still arrive at the earnings number that you are implicitly guiding for FY 2027?"
— Michael Lasser, UBS

A: "What we believe is, the 5%+ decline in transactions, particularly on DIY, will kind of start to moderate back to a normal trend of, historically, over long periods of time, it's been down 1% or 1%-3% in transactions, and you've gotten average ticket growth that's been higher than that. We expect that to continue, and it's similar to what we saw in the latter half of Q4."
— Phil Daniele, President and Chief Executive Officer

Assessment: the first half of the question was answered and confirmed. The second half, on P&L flexibility if the comp lands below the range, was not engaged. That is the open question coming out of this call, and it is the reason we treat the fiscal 2027 earnings path as a forecast with a soft floor rather than a guided outcome.

Quantifying what the investment programme has returned

A pointed challenge asked management to dimension the return on two years of elevated capital and expense, and framed it against a competitor comping well ahead. Management conceded the comp point and redirected to store-level performance.

Q: "How would you quantify the return on the investments that you have seen over the last couple of years? Those have come in the form of accelerated CapEx, accelerated SG&A. As outsiders, what we see is the comps are moderating. You have a player in the industry that is driving same-store sales growth well ahead of AutoZone, so it is hard for us to dimension the return that you are generating on these investments that you have been making."
— Michael Lasser, UBS

A: "But I would agree with you that our comps have maybe not been as high as we would have liked. We do feel that we are operating as well as we have. We like the performance of our new stores. We ultimately had to do some work in our supply chain to make sure we were able to service our base stores that we had and do that well and efficiently."
— Phil Daniele, President and Chief Executive Officer

Assessment: a candid concession on the comp and no answer on the return. The new-store curve given in the prepared remarks is a partial response, but it describes the economics of an average new store rather than the incremental return on the aggregate programme. The competitive share question went unaddressed entirely.

Whether the fiscal 2027 gross-margin guide absorbs the tariff-refund lap

A follow-up sought confirmation that the margin range is struck against a base that includes this year's refunds, with no separate drag called out for lapping them. Management confirmed it and offered a reason why the lap should be smaller than it looks.

Q: "Just a quick follow-up, just a confirmation here for the gross margin outlook for FY 2027. The 0 basis points-25 basis points of FIFO gross margin expansion, I guess that's inclusive of the baseline benefit that you've captured with tariff refunds, right? There's no expectation for a cycling to drag on the FIFO gross margin profile."
— Steven Forbes, Guggenheim

A: "That's right. So what we gave you when we talked about gross margins is we gave you the LIFO assumptions for this year. We gave you the LIFO assumptions for last year. Obviously, we had $190+ million last year. We've got $85 million-$90 million this year. So if you just look at that on a GAAP basis, we should be flat to up 20. As it relates to tariffs specifically, while the tariff refunds may appear to be a windfall, the reality is that the refund environment was deflationary, and the industry likely would have raised prices faster if both retailers and suppliers hadn't received cost relief from those refunds. So, we're expecting to lap those tariff benefits and not have that be a call-out."
— Phil Daniele, President and Chief Executive Officer

Assessment: the single most clarifying answer of the call, and the one that most constrains the fiscal 2027 model. The argument that the refunds were economically offset by foregone price increases is reasonable. It does not change the reported arithmetic, which is that 48 basis points of fiscal 2026 margin must be replaced and the available LIFO tailwind is the thing replacing it.

How much comp the new-store cohort contributes as it enters the base

The final question tested a common bull argument, that the accelerated opening programme produces a mechanical comp tailwind as the cohort matures into the comparable base. Management damped the expectation with arithmetic.

Q: "How are you thinking about or modeling the higher comp contribution that you should get from the new store acceleration, which I guess started about a year ago? Can we assume you are building a growing lift as the year progresses, as more stores roll into the comp base?"
— Scot Ciccarelli, Truist Securities

A: "Yeah. As you might expect, the newer stores have higher comps in the near terms. It does have some impact on our comps going forward. Remember that if you take domestically, for example, we added 236 stores on a base of 6,500 or 6,600 stores. While it does have some impact, it does not move the needle as much as you might think."
— Jamere Jackson, Chief Financial Officer

Assessment: management argued against its own bull case here, which is worth noting. Last quarter we listed the unquantified new-store comp contribution as something they were not saying; the honest answer turns out to be that it is small. That removes a convenient offset to fading inflation and puts the weight back on mature-store transactions.

What They're NOT Saying

1. No P&L flexibility answer if the comp lands below the range. The question was asked directly and only the first half was answered. Management has said it can manage expenses down, but has given neither a leverage threshold nor a contingency, while guiding expense growth above implied sales growth.
2. No quantification of the residual tariff refund. A little over $100 million in total with $96 million recognised, and the remainder described only as trickling out across fiscal 2027. The amount is probably small, but the fiscal 2027 margin guide treats tariffs as a non-item and that treatment is unverifiable.
3. DIY transactions were not quantified in the release. The press release did not mention traffic. The figure above 5%, the worst in our coverage, surfaced in question and answer. The same pattern held last quarter, when 3.6% came from the call rather than the release.
4. No ROIC recovery path or date. The commitment is to hold the mid-30s, which is a floor rather than a trajectory, after a 550 basis point decline. The filing's own table shows the narrowing equity deficit drove most of it, and management did not make that point, which would have helped their case.
5. No response on the competitive share gap. Asked about a peer comping well ahead, management conceded its own comps were lower than hoped and moved to new-store performance. The industry share position was reasserted at 5% to 6% of an approximately $100 billion market without addressing why a competitor is taking more of it.
6. No explanation of what Brazil needs. The fiscal 2028 opening target fell about 14% because of a slower Brazil build, described as "things that we need to go do down there." Neither the problem nor the remedy was specified, and no Brazil comp or profitability figure has ever been disclosed.
7. No quantified fiscal 2027 free cash flow target. Cash flow is guided only as higher, while capital spending rises to roughly $1.65 billion. Fiscal 2026 free cash flow grew 0.9%. The combination of a rising capital programme and an unquantified cash commitment is how free cash flow came to sit near $1.8 billion for two consecutive years.
8. No international profitability disclosure. Returns in Mexico and Brazil were characterised as strong, with roughly 15% of the store base now outside the United States. No international operating margin, EBIT contribution or country-level return has been published.
9. No cohort-level actuals against the new return curve. The curve itself is a genuine advance. What accompanies it is the phrase "slightly ahead," without the fiscal 2025 or fiscal 2026 cohort's actual first-year sales against the $1.7 million model.

Market Reaction

• Pre-print setup: the stock closed at $2,803.25 on September 21, down 17.3% year to date, down 5.2% over the trailing 30 days and down 32.0% over twelve months. That close was also the low end of its pre-print 52-week closing range of $2,803.25 to $4,290.24. The S&P 500 was up 13.4% year to date. The stock had declined after seven of its eight previous earnings reports, having missed revenue consensus in seven of the prior eight quarters.
• Reaction session, September 22: AutoZone reports before the open, so the print day is the reaction session. The stock opened at $2,894.25, a 3.2% gap, traded between $2,844.80 and $2,999.30, and closed at $2,894.73, up $91.48 or 3.3%.
• Volume: roughly 0.8 million shares against a trailing 30-day average near 0.3 million, about 2.8 times normal.
• Market comparison: the S&P 500 was unchanged in the same session, so the move was company-specific.

An eighth revenue miss in nine quarters produced a 3.3% gain. That is a change in how this stock is trading, and the most defensible explanation is positioning rather than content. AutoZone entered the print at a 52-week closing low after a 32% twelve-month decline, with the sell-side cutting price targets and at least one house compressing its multiple framework from the low twenties into the mid-to-high teens. Into that setup, a 182 basis point gross-margin headline and 15.1% EPS growth were enough.

We would caution against reading the reaction as validation of the quarter. The stock rose because expectations had been reset far enough that a print carrying lower core earnings cleared them. That is a useful signal about the risk of being short, and a weak signal about the risk of being long. Our interpretation of what the market responded to is an inference from the setup and the headline figures, not a measured decomposition of the move.

The more durable observation is the valuation that the reaction left behind. At $2,894.73 the stock trades at 19.0 times the fiscal year just reported and 17.9 times our fiscal 2027 estimate, against the 22x to 23x framework we were carrying into this print and the low-twenties multiples the sell-side used before this year's revisions. That de-rating is the reason this remains a Hold rather than an Underperform.

Street Perspective

Debate: is the earnings quality problem cyclical or structural?

Bull view: the items flattering this quarter are the mirror image of the items that depressed the prior two years. LIFO charges of $192 million and tariff costs were a self-inflicted drag on reported earnings through fiscal 2025 and fiscal 2026; their reversal is a normalisation, not a windfall. Underlying earnings power was understated then and is being revealed now.

Bear view: whichever direction the accounting runs, the operating facts are a negative domestic retail comp, commercial per-program productivity decelerating for a second quarter, 101 basis points of expense deleverage and two consecutive years of operating margin contraction. Core earnings fell about 5% in a quarter that reported 15.1% growth.

Our take: the bull argument is correct about the symmetry and wrong about the conclusion. Normalising both directions still leaves fiscal 2026 operating margin down about 63 basis points and a fiscal 2027 guide that implies another 20 to 50, depending on where sales land in the implied range. The quality problem is cyclical in origin and is being compounded by an investment programme whose expense is landing ahead of its revenue.

Debate: does the accelerated store programme create value or consume it?

Bull view: the newly published return curve settles this. A $2.9 million investment reaching 15% returns by year four and above 20% by year six, with current cohorts running ahead of plan, is a high-return use of capital, and 400 openings a year compounds it. Group ROIC is depressed by immature stores, which is the expected cost of a growth programme.

Bear view: the programme has coincided with group ROIC falling from 41.3% to 35.8%, free cash flow stuck near $1.8 billion for two years, capital spending rising to $1.65 billion and the fiscal 2028 target being cut 14%. Management's own answer is that the cohort contributes little to comp. The curve is a model, and the only reported outcome so far is lower returns.

Our take: both cases overstate. Our decomposition of the filing's ROIC table shows 88% of the invested-capital increase came from the narrowing stockholders' deficit, so the headline decline substantially measures a balance-sheet effect rather than store productivity. The bulls are right that the returns scare is overdone. The bears are right that two years of free cash flow near $1.8 billion with capital spending still rising is the relevant investor outcome, whatever the return curve says.

Debate: is the multiple cheap or correctly re-rated?

Bull view: 17.9 times forward earnings for a business with 35.8% returns, a 25-year buyback record shrinking the count 3% to 4% a year, and a nearly completed investment cycle is the cheapest this stock has been in years. The price-target cluster 26% to 34% above the current price reflects that.

Bear view: the multiple compressed because the fundamentals deserved it. Earnings growth is now mostly buyback and accounting, the domestic comp is negative, and the long-run unit target has come down. A high-teens multiple is the right price for mid-single-digit earnings growth.

Our take: we were on the wrong side of this debate last quarter, carrying a 22x to 23x framework while the market de-rated the stock by a further 7%. We now use 19.0x to 20.5x. That is not a capitulation to price: it reflects fiscal 2026's realised margin contraction, a further decline implied by the fiscal 2027 guide, and a lower terminal unit plan, all of which are new information since our last framework was set. At the revised multiple the stock is close to fair rather than cheap.

Our Estimates & Valuation

The earnings bridge to fiscal 2027

Rather than present a fresh set of assumptions, it is more useful to show what moves our fiscal 2026 actual to our fiscal 2027 base, because the composition is the argument for the downgrade.

Of the $9.45 of fiscal 2027 EPS growth in our base case, $6.18 is the buyback and $0.33 is the net of two accounting and policy items. Core operating growth supplies $2.94, or 1.9% of the fiscal 2026 base. That is the whole case in one line: on management's own guided inputs, AutoZone's operating earnings grow at about 2% next year and the share count does the rest.

Fiscal 2027 model detail

The bear case assumes the domestic comp lands at the low end of the guided range, gross margin holds flat rather than expanding, expenses grow as guided, and the buyback slows to protect the leverage target. The bull case assumes the transaction recovery arrives, commercial holds a low-double-digit rate, and management finds the expense restraint it says it can. All three assume no dividend, which AutoZone does not pay.

12-month return cases

Weighting the base case at 50% and the tails at 25% each gives an expected 12-month return of roughly 11%, against our illustrative 8% return hurdle for the S&P 500 over the same horizon. An excess return near three points is not enough to carry an Outperform through a year in which core operating earnings are guided to grow about 2% and the downside case is a 13% loss. That is the arithmetic behind the downgrade.

What would change the view back. Two things, and they are both measurable within two quarters. First, domestic transactions moderating toward the historical 1% to 3% decline while ticket holds near 4%, which would put the domestic comp in the 2% to 3% range and restore the expense leverage point. Second, evidence that free cash flow is genuinely inflecting rather than holding near $1.8 billion for a third year, since the capital programme is guided higher again. Either one on its own would lift the base case; both together would restore Outperform.

What would make it worse. A first quarter in which the domestic comp fails to reach flat despite the guided 5% ticket, implying transactions have not improved off the fourth quarter's rate; adjusted ROIC falling below the mid-30s floor management committed to; or a slowdown in repurchases, which would remove the largest single contributor to our fiscal 2027 EPS growth.

Thesis Scorecard Post-Earnings

The ten pillars below are the standing thesis carried since the second quarter, graded against what this print and call revealed. Three status tags moved.

Overall: the thesis is weaker on earnings and stronger on disclosure. Two pillars that carried the Outperform have given way: DIY resilience broke on its own definition, and expense discipline reversed into deleverage that the fiscal 2027 guide extends. Against that, the store programme delivered a record year, the Mega Hub mechanism and the new-store return curve are now quantified rather than asserted, and our own decomposition of the ROIC decline shows most of it is a balance-sheet effect rather than a productivity failure. The business is in better shape than its reported earnings quality suggests and in worse shape than a 22x multiple required.

Action: Downgrade to Hold at $2,894.73, with a $3,080 to $3,320 12-month fair-value range and conviction of 6 out of 10. We were too slow to cut the multiple last quarter and are cutting it now for reasons that are new rather than price-driven: a second consecutive year of operating margin contraction and a 14% reduction in the fiscal 2028 unit plan. The next two quarters settle it. If domestic transactions moderate toward the historical range while ticket holds near 4%, the expense leverage point returns and so does the rating.