Description
AutoZone – no stranger to VIC -- certainly looks cheap. Industry leader + very good long term business + strong owner-orientation + Over 25% after-tax returns on capital + 6% square footage growth for years to come + no serious SEC/Legal/Liquidity issues does not seem to = less than 11 times this year’s earnings. Of course, if looking cheap were all that mattered, I wouldn’t need to invest in the stock market. In fact, there are some potential reasons to think AutoZone isn’t anywhere near as cheap as it looks at 11X earnings. I’m here to tell you why those reasons are wrong.
Brief History
AutoZone was founded in 1979 as a division of grocery wholesaler Malone and Hyde, which I believe was the subject of the first-ever cash LBO. KKR spun AZO out of M&H in 1987 (FYI) and brought it public in 1991. Central to the thesis, AutoZone was originally named Autoshack, but caved to the wrath of Radioshack’s IP lawyers and changed its name. Recently, AutoZone turned around and sued Tandy for implementing “Powerzones” within Radioshacks. AutoZone lost again. Okay, not central to thesis, but that really does seem like slap in the face.
AutoZone has always been a terrific business with store economics at the very top of the automotive after-market industry.
In 1991, AutoZone had 598 stores, did $818m in sales, $79m in EBIT, $44m in net income and $0.33 in split adjusted EPS. In FY04 ended 8/04, AutoZone had 3420 domestic stores, did $5.64b in sales, $999m in EBIT, $566m in net income and $6.56 in EPS on diluted shares of roughly 81 million. Sales, EBIT, net income and EPS have grown at respective 13-year compounded rates of 16.0%, 21.6%, 21.7% and 25.6%. The stock, despite currently trading at 11X earnings, has compounded at just under 20% since its IPO.
Even Briefer Industry Overview
With its 3500 stores, AutoZone is the largest retailer in the Do-it-Yourself (“DIY” – selling parts to consumers) aftermarket auto parts industry. Its largest competitors in the DIY business are Advanced Auto (AAP), CSK Auto (CAO), O’Reilly’s (ORLY) and Pep Boys (PBY), though ORLY is 50% Do-It-For-Me (“DIFM” – selling parts to service shops), while PBY has a significant service bay business. Of the three big DIY shops, the mix is roughly 85% DIY and 15% DIFM, with all three having pushed into the DIFM market in recent years. Most estimates have the large DIY chains accounting for 30-40% of the market, with the remainder consisting of a declining base of smaller chains and mom and pops. Wal-Mart has been in the category for many years, but has not been anywhere near as significant a force as it is in many other categories.
Promotion or markdown driven competition is very unusual in the industry. Most participants believe that parts consumers are relative inelastic, at least insofar as responding to big sales promotions, and thus rarely compete on that basis. This is an extremely slow turn business, with intensive inventory breadth (but little depth) required in the hard parts area. While industry growth has picked up slightly, as the fastest grower (ORLY) has gotten larger and AAP has shifted from acquisitions or organic growth (5-6%) net industry growth is still probably only 1-3% when mom and pop attrition is included.
Current Situation
Soon after Eddie Lampert took a big stake in AutoZone in the late 1990s, new management came on – led by CEO Steve Odland and CFO Mike Archibold -- and the company immediately began to improve operating margins from good to great. In the last five years, operating margins have jumped from 11-13% to around 17%, despite little if any benefit from comp-driven SG&A leverage. In fact, absolute comps have worsened, and in recent quarters have materially lagged AutoZone’s auto aftermarket competitors.
In March, AutoZone announced that CEO Steve Odland was jumping ship for Office Depot, and that comps had declined 7% for the first four weeks of its third quarter (mid Feb to mid March). The stock has since dropped from roughly $98 to $83. Subsequent to the announcement, there has been some speculation (confusion) over how strong the 2004 comparison period was, and thus how exactly how bad the 7% decline really was in context. Consensus estimates of $7.44 for the year-ended August have not come down following the announcement.
Eddie Lampert is a Cow Milking, LBO Aping, REIT lover
Given its capital returns, operating history and reasonable growth prospects, AutoZone’s current market valuation is really only justifiable if (1) earnings are unsustainable or (2) leverage is distorting the multiple.
Margins
Has AutoZone’s concurrent margin expansion and same store sales compression over the last few years left it with bloated and unsustainable operating margins? Is ESL helping to drive a cow milking strategy here, or is something else at play that will naturally erode AZO’s peak margins as comps continue to fall? Will AutoZone sell Martha Stewart hood ornaments?
At roughly 17%, AutoZone’s operating margins are far superior to competitors, its own historical record, and in fact are maybe the 5th or 6th best in all of retail. And margins continue to improve despite negative comps, which is very unusual in a non-turnaround retailer. For example, excluding one-time items, EBIT margins actually increased 120 basis points (15.1% to 16.3%) in the first half of FY05, despite comps being down 1-2% for the period. AutoZone has credited its 5-year margin improvement to category management (including significantly increasing private label penetration) and some low hanging fruit in expense reduction. Notably, competitors have also improved their margins since 2001, which they, too partially credit to category management.
After a decade of consistently posting positive mid-single digit comps, AutoZone began to see SSS ease in 2003, and since then they have continued to worsen. 2003 SSS were +3% (after +9% in 02), dropping to flat in FY 2004, including a –3% Q4 summer. Comps were again –3% in Q105 (last fall), and flat in Q2 ended February. The only other data available is the –7% four week period after Q2, which is worse than any quarterly comp in AZO’s public history. While competitor comps did worsen somewhat in recent quarters, they remain ahead of AZO’s numbers. Most directly, AAP’s comps in the DIY business were +2.8% in its Calendar 04 (compares to roughly -1% for AZO over the same period) and +5.6% in the December quarter (again compared to –1% or so for AZO). CSK, which had a strong 2003 comp-wise, ran more closely to AZO for 2004, at around –1%.
One reasonable possibility and occasional sell-side accusation is that AutoZone’s expense scrimping is simply coming home to roost, and that under-investing in things like labor relative to competition is starting to show up in customer traffic. This seems validated by AutoZone’s substantial operating margin advantage (17%) over AAP (9%) and CSK (7%). But those numbers are misleading.
For one, AutoZone owns 58% of its locations, whereas both AAP and CSK lease the vast majority of theirs. This obviously distorts comparisons at the EBIT margin level. If AutoZone leased all of its stores, rent expense would likely have been higher by $140-$150mm in FY04, lowering margins from 17% to 14.4%, already making the gap much smaller than it appears. AZO also has a small amount (my guess is $75-$100m) of very high margin software sales form their ALLDATA diagnostic product, which probably brings comparable retail operating margins close to the 13-14% range. Of course, that still leaves a hefty margin advantage. But you would expect such an edge given that AZO’s DIY business is still vastly more productive than any of its competitors. DIY sales per square foot at AZO still run in the $250 range, versus $190-$210 psf for AAP and CSK. In addition to productivity AZO has a natural gross margin advantage in two huge private label brands (Valuecraft and Duralast) which account for more than 50% of sales, a figure which no competitor even approaches. This issue becomes even more clear when you look at cash SG&A, ex rent, spent per square foot of retail space. AZO at $77 per square foot actually spends more than AAP ($73) and roughly the same as CSK – and this includes corporate overhead, where AZO has the most scale.
In sum, I don’t believe AZO is under-spending. While margins are higher than competitors, they ought to be, and the gap is not as wide as it appears. I (and the company) do not see any reason why operating margins should decrease in the future, particularly given the general lack of promotion and markdown pricing as a competitive strategy in the DIY business. In fact, AZO still believes it has some opportunity to improve its cost efficiency, albeit at a slower pace than recent years.
So Why Are Comps So Poor
AutoZone blames its absolute comp performance on gas prices, and its relative comp performance on store maturity. Historically, auto parts stores have opened at 60%-65% of full volume and ramped to maturity over 4-5 years. At the end 2004, only 22.8% of AZO’s stores were less than five years old, compared to 50% in 1999. In addition, any improvements AZO made in its Chief Auto and Auto Palace acquisitions in 1998 were long baked in. So some of AZO’s comp compression over time is simply a function of store maturity. When I model typical a store ramp and the changes in AZO’s store maturity profile, I estimate that aging has taken roughly 3% off the natural comp rate in recent years. AutoZone strongly believes the rest of its comp deterioration has been caused by the increase in gas prices, which both directly impacts its customers’ wallets and tends to restrain miles driven – a demand driver. While gas prices have obviously increased markedly and intuitively should hurt their business somewhat, it’s tough to nail down the impact with any precision from the outside.
As to peer comparisons, it’s true that AZO’s average store is older than competitors, including AAP. But excluding ORLY, AZO’s percentage of “immature” stores is actually similar to its competitors, making the comp disparity less simple to explain. While average store age is likely playing a role, I believe the bigger difference actually comes from competitors (AAP and, to a lesser extent, CSK) improving their operations. Both CSK and AAP are in significant part roll ups of other chains. AAP had acquired 600 Western Auto and Parts America store from Sears in ’98 and 671 Discount Auto Parts store in 2001 – this is roughly half of their store base. Both chains have historically had a much larger percentage of under-performing stores than AutoZone, and AAP has done a nice job bringing the laggards up to par in the Advance format.
Of course, competitor improvement is not exactly the hallmark of a great investment, but I think there are two important points to made here.
One, while competitors are getting the benefit of some natural mean reversion, they are still very far from AutoZone’s industry leading DIY psf productivity. I believe that while AAP and to some degree CSK have done well to improve their businesses, the ultimate level of reversion is likely limited as AZO has some permanent advantages. In addition to having the best known brand in the industry, AZO also has two of the leading parts brands (Duralast and Valucraft) as private label exclusives. They have the most clout with suppliers, as shown by their industry leading vendor financing (payables at 92% of inventories, far ahead of competitors), a unique database of store-specific warranty information allowing particularly customized inventory planning, and, maybe most importantly, a base of 3500 locations that largely represent intelligent site selection by a very successful, single organization as opposed to a conglomeration of various defunct chains.
Two, AutoZone’s comp numbers should be put in some perspective. While they have worsened recently, the overall 1-2% decline in the last twelve months has in fact come during the greatest YoY gas price increase since it’s been a public company. This after a dozen years of consistently positive comps. And AutoZone’s two years of depressed store openings – 2000 and 2001 – meant a maturity mix tilted away from ramping, easy comp stores in 2004 and 2005. There is no evidence that long term demand in the industry, driven by miles traveled and vehicles 7+ years old, is departing from its 4% annual growth. The SUV craze is just beginning to enter its prime aftermarket years, and SUVs are particularly strong drivers of this business. Overall industry square footage is not growing rapidly, and there has been no palpable increase in supply pressure from non-auto chains (i.e. discounters).
Admittedly, -7% is an ugly number. But this came during a four week period that is (1) the lightest volume period in the third quarter; (2) a kind of inflection period where winter transitions to spring and is perhaps unusually sensitive to weather; and (3) apparently was comp’ing against difficult good-weather 2004 four week comparisons. The entire third quarter year-ago was a +2%, versus flat in the second quarter. Thus, some decline from Q205 flat comp wasn’t surprising. If the rest of the quarter picked up from the –7% as I expect, and comps comp in at –3% to –4% for the quarter, I think a scary announcement turns into a distant hiccup. Beginning the fourth quarter, AZO’s biggest as it is the summer and 16 weeks long, comparisons get markedly easier as last summer marked the begging of gasoline price pressure and a –3% comp.
Leverage
Apart from peak margins and lagging comps, there has been some criticism that AZO has juiced EPS by buying back shares financed in part with increased leverage. AutoZone carries $1.9b of (almost entirely fixed) debt, which gives it a debt/equity ratio of your standard GSE. Over last seven years, it has spent $3.7b repurchasing more than half its stock (82.5m shares), leaving it with 79.8m shares outstanding, and so has begun to look like a slow going LBO. But this company is nowhere near as levered as it looks. By owning more than half its stores, meaning including the land underneath, AutoZone is simply exchanging financial leverage for the lease leverage that most supposedly debt free retailers (and aftermarket competitors) carry. I believe that AutoZone’s owned properties alone, including land undervalued on the balance sheet, could generate sale-leaseback proceeds that could pay down as much as 2/3 of its debt. Suddenly you’d see debt decrease to a very small number relative to EBITDA, margins recede toward “normal” as rent replaced interest expense, and a lot of cheering from the bleachers… but no economic benefit. As it is, AutoZone manages its debt to maintain an investment grade credit rating, which means a 2.1X Debt/EBITDAR (including capitalized rent in debt). I don’t believe AutoZone is anywhere near dangerously levered given its consistent earnings history and store ownership profile, and if fact I don’t believe it is more levered than most putatively debt free retailers who lease all their stores. As a result, I think the bottom line earnings power to equity is a perfectly appropriate measure of value to capitalize, and I think financial risk is low.
Odland’s Departure
Steve Odland brought a very quantitative, return on capital based approach to the business, and helped generate significant shareholder value. He came from the grocery industry, and helped really drive working capital reduction (AZO basically runs with negative working capital now) which has helped turn the economics of an AutoZone store from very good to terrific. His replacement, Bill Rhodes, is a long time AutoZone operations guy, though he does have an accounting background. Happily, Bill also drinks from the return on capital chalice, and plans to run the business for profits (quaint) and returns on capital. AutoZone’s managers at multiple levels have been and will continue to be compensated based on various EVA and capital returns measures. They will continue to use free cash flow to open stores at 6% year, and then repurchase stock if it’s priced reasonably. Bill has worked in just about every area of the company in his tenure, and does not believe there is any need to increase expenses or investments, such as labor costs.
Obviously, owner-orientation is buoyed by ESL’s significant stake in the company (roughly 27% -- originally bought in 98-01, then sold some in Dec 03 at $100, then bought more last fall in the mid $70s) and Eddie Lampert’s board presence. While AutoZone may not be the first ticker on Eddie’s My Yahoo page these days, it is unlikely that this company would do anything obviously contrary to shareholder interests. Finally, I would point out that I have met several members of AutoZone’s management team, and the talent ran much deeper than Steve Odland. The CFO Mike Archibold, in particular, is a very helpful, data driven manager focused on capital returns. Finally, it's reasonable to wonder why Odland would leave a pretty good thing for a seemingly sideways move to another retailer that isn't exactly akin to taking over for Nardelli or Sinegal. But ODP does offer a chance to repeat his success with Autozone with the Eddie Lampert disciple asterisk in the records, and of course a relocation from Memphis to Delray Beach.
What It’s Worth
AutoZone earned $540 million last year after subtracting one-time warranty benefits. In the first half of this year, it was roughly $23mm ahead of last year’s pace. Given the rough start to Q3, but the easier comparisons in the large Q4, I think it will probably earn somewhere in the neighborhood of $580mm this year. While this equates to $7.30 to $7.50 in EPS depending on the amount and timing of repurchases, it might be easier to take the Treasury Method out of the equation. At $580mm in net income plus $110mm in depreciation and $40mm from working capital reversion, AZO would have OCF of $650mm for the year, less $200mm for Capex. (Maintenance cap ex that isn’t expensed, incidentally, only runs $30mm-$40mm a year versus $110mm in depreciation.) That leaves $450m to repurchase stock, which will essentially all come in Q3 and Q4. If we assume that AZO repurchases 5 million shares during the fiscal year, they would end August with roughly 75 million basic shares outstanding. Thus, sitting in August trailing earning per existing share would be roughly $7.75. In FY06, AutoZone again expects to again grow square footage by 6%, and of course will have easier sales comparisons. If they can grow net income by 8%, to $630mm, forward earnings ending Aug06 would be $8.40 (using Aug 05 share-count). The current price is less than 10X that number. Even if AutoZone did suffer a 300 bp permanent hit to operating margins in a fairly bad case scenaior, AZO’s current price would only be a bit over 12X FY06 adjusted earnings (again, using August 05 sharecount).
AutoZone is earning greater than 25% after-tax (un-levered) returns on its new stores. They believe, based on their site selection analysis, they can open more than two thousand new stores in the United States alone, which would account for ten more years of the current 6% square footage growth rate. You can run your own numbers on what a business with a reliable earnings history, 6% top line growth and 25-30% un-levered incremental returns on capital is worth, but it’s not 10 or 11 times earnings. I think 15-16X earnings is much more the ballpark. Moderate growth at excellent returns on capital is worth a lot. And this excludes any value-based benefits from opportunistic share purchases, excess depreciation, continued margin improvements that management expects or a significant pickup in the commercial business.
Speaking of the Commercial business, I have omitted a discussion of AZO’s DIFM business (around 13% of sales), which has been struggling both absolutely and relative to competitors in recent quarters. This isn’t for lack of caring, but is a rare whim of mercy and brevity…and because I think the company is cheap without respect to whether they ultimately right their commercial woes. I would be happy to discuss the DIFM business in the follow up thread.
Catalyst
AutoZone uses its free cash to repurchase shares, so if this company is in fact cheap, the value will accrue to shareholders soon enough. Beginning the fourth quarter, AutoZone laps the biggest gas price increases and the steepest slope thus far in its comp declines. Of course, gas prices aren’t exactly showing signs of easing, and you could shy away from this business for that reason alone. While constantly increasing gas price will obviously be bad for all businesses targeting lower income consumers, it will be worse for the aftermarket shops. What happens if gas prices simply plateau at a high level is an open question, though European and Canadian consumer behavior indicates that related consumption likewise tends to stabilize.