Costco: Buying This Dip


The Price Club merged with Costco in 1993 to form the company now known simply as Costco Wholesale Corporation (NASDAQ:COST). The full name is important and is discussed briefly later. Since formation, this is the comparison of COST versus the SP 500’s leading ETF (NYSEARCA:SPY):

SPY data by YCharts

SPY has done well, up 4.8X since 1993, plus dividends.

COST is up 17.2X, plus dividends.

Since 1993 was long ago, and because I have certain relative valuation data to present that does not go back that far, the 10-year comparison of COST and SPY is shown next:

ChartSPY data by YCharts

Over the past 10 years, meaning from just around the market peak pre-Great Recession, COST is up 148%, whereas SPY is up just 63%. That’s before dividends. While COST’s regular payout is lower than the dividend yield from SPY, $7 special dividends in 2012 and 2017 and a $5 special dividend (per share) lead me to say that, via rough estimates, COST may have also out-yielded SPY by about 1% per year over the past decade.

So: as has happened more often than not, Warren Buffett’s faith in COST has been rewarded. Now the question is whether COST is too mature to provide alpha versus SPY, as has happened to some of Mr. Buffett’s other old faves.

Introduction – COST as a relative value play on defensive-oriented growth (the best kind)

My views are similar to those laid out in my other COST article, published on Seeking Alpha this past June 22 in Costco: Store Of Value, where the bullet points asserted:

  • COST, one of the great stocks of our time, faces P/E challenges from Amazon’s (NASDAQ:AMZN) expansion.
  • Yet COST has many strengths that may insulate it from its rapidly-expanding neighbor.
  • This article reviews COST as an ecosystem company with a high degree of customer loyalty and provider of substantial value.
  • My conclusion is that COST’s high P/E limits its attractiveness, but given paltry alternatives and its many strengths, it may still provide alpha.

When that article was submitted the morning of June 22, COST had closed June 21 at $163.15, down from an all-time of $182.72 in early June. Even though COST had entered a correction, down a little more than 10% from the high, I thought it looked good long term. I had bought a small starter position and bought more as the stock dropped to $150 at the July bottom, and have traded around the core position ever since.

This article continues the same point of view expressed in the June article, with new information generally favorable in my view.

The following sections summarize first the positive reasons that lead me to again buy the dip in COST, and buy more if it drops some more; then a discussion of weaknesses follow.

First, valuation comments.

History of COST’s relative value versus SPY

Since the end of the 2001-2 bear market period, COST has outperformed notably when its relative P/E has been in the range of 10% or so above that of the stock market.

There are just three periods to discuss. These are annual averages using Value Line (VALU) data and represent the average P/E of COST during the year versus that of the market. Whether Value Line used one of its indices, that of SPY or some other index is not known to me. In any case, these are its data for relative P/E of COST over some broad stock market index:

  • 2004-7: 1.06, 1.19, 1.19, 1.11
  • 2008-16: 1.39, 1.30, 1.27, 1.39, 1.39, 1.31, 1.32, 1.35, 1.54
  • 2017: About 1.1 (see below).

It turns out that COST’s outperformance came before the Great Recession and during it.

Even the relatively low average P/E in 2010 for that long run of very high premia to the market was associated with depressed COST prices throughout almost all of that year, then significant alpha in 2011 and 2012.

What’s the relative P/E now?

Whether one uses the $6.08 TTM GAAP EPS or adjusts down for one-time gains last fiscal year, looks forward to FY 2018, or uses GAAP EPS of the SP 500 as of Q2 or uses estimates for the 12 months ending in Q3 2017 of $107, I get roughly COST at a 10% premium to SPY.

The exact ratio does not matter. COST is well within the range where it has begun from in generating its alpha versus SPY since 2004. So I look at it as cheap to the market, though not cheap on absolute terms.

Who’s afraid of AMZN? Evidence COST is healthy

Costco resolutely sticks to its operating plan that just keeps on providing goods and services to its members near COST’s cost (excepting the membership fee), plus growth initiatives. The best evidence that AMZN, in full swing now in e-commerce in COST’s territories, has not harmed it noticeably comes from the Q4 details. This period is not truly a “quarter,” as COST divides its year into three 12-week periods and a fourth, 16-week period, which this year was a 17-week period.

Comparable sales for the 17-week fourth quarter, the 53-week fiscal year, and the five-week September retail sales month were as follows:

Comps were also strong when excluding gasoline sales and forex, with the same trends.

COST’s business strengthened as the year wore on, and strengthened further in the September five-week retail period.

EPS for the full year was $6.08. Many adjust that downward to $5.82 due to $0.26 per share of two discrete one-time gains. It is off the $5.82 number that this fiscal year’s consensus EPS of $6.44 is made. Per ETrade (ETFC), the Street is thus projecting the following EPS growth path, around 9-10% per year for the following fiscal years:

  • 2017: $5.82
  • 2018: $6.44
  • 2019: 7.04
  • 2020: $7.62
  • 2021: 8.30.

Because COST reports GAAP, I expect that except for FY 2017, the numbers used above are GAAP projected EPS.

Because forward projections for earnings of SPY are always too high

Why COST can meet growth expectations

The company is slowing the pace of store openings this year, and increasingly it is opening stores where it already has a presence, such as near its headquarters outside of Seattle. This sort of cannibalization is profitable with low risk, but it’s not especially growth-oriented. However, it does thin the crowds out at existing stores, making them more attractive for locals to shop in while preparing for population growth. So I’m going to assume a slowdown in retail square footage to 2.5% per year.

I get to about 10% EPS growth without financial engineering (of which COST is not especially fond, preferring dividend payouts) this way.

Base 7% growth comprised of these two assumptions (with stable margins):

  • 2.5% square foot growth in selling space
  • 4.5% annual sales gains per SF of selling space (= GDP growth).

The other 3% would come from the following:

  • E-commerce, including sales to customers living far from a COST store
  • Expansion of the COST ecosystem of services
  • Vertical integration, such as increased food production by COST for COST
  • A move into China soon

Most businesses fail to meet growth expectations, but I think that COST could well be an exception.

As of Q3, COST had increased its fulfillment center base for e-commerce from 7 to 19 yoy, mostly by expanding adjacent to existing facilities.

Based on the company’s comment in the Q3 conference call that e-commerce comprised a little less than 3.5% of COST’s sales and was annualizing above $4 B, the latest ramp to 30% yoy growth in the latest five-week period should take e-commerce to 4% of sales and above $5 B in annual sales.

Basically I expect that the strongest retail operators will continue to take share from the weak chains and mom-and-pop operators.

The three major worries about COST don’t bother me

I think the following are priced in, and possibly overly priced in:

  • Whole Foods belonging to
  • Margin decline at COST
  • Renewal rates down at COST

Taking these in order…

Whole Foods

In the Q4 conference call, COST notes that it has been competing with, and tracking, Whole Foods for many years. In response to a question from Karen Short, this was the relevant part of the interchange:

…But do you have any color just specifically on the performance of your stores that are in close proximity to Whole Foods since the price reduction were — took place at Whole Foods.

Richard Galanti

Yeah. We essentially overlap everywhere. And I’m not trying to be cute, but other than reading about it on the news and the paper on Wall Street, we have not seen — we recognized – I read yesterday that there are some specialties brick and mortar retail stores that are impacted more than others. We don’t believe we’ve seen an impact from it.

Earlier, in response to a question from Michael Lasser, COST replied in part:

As it relates to the publicity and the news and the noise around Amazon Whole Foods, all we can do is perform. When we look at the value proposition, our view is our value proposition got better.

There are always worries, but I suspect that AMZN will use the Whole Foods stores to sell more electronics gear, to keep Amazon Prime sign-ups growing,and to do a lot of learning about running a large but not giant brick-and-mortar operation.

Margins and renewals

COST sends its CFO as the only participant from COST to the QA, and he was asked about each issue. For each response, he had cogent answers and expressed confidence that there was no fundamental change in either parameter. For example, COST sold more gasoline by volume, and at a higher price, in Q4; gasoline has lower margins than COST’s other products. Other factors were named.

As far as renewals go, the company referenced the credit card changeover from AmEx (AXP) to Visa (V). Because Mr. Galanti of COST provided so much info, I’ll take the risk of telling you what you know and pick out one paragraph from his prepared remarks about renewals:

At year end, paid executive memberships totaled 18.5 million, an increase of 274,000 since third quarter end, which is about 16,000 per week increase in the quarter. Executive members are about 38% of our member base and about two-thirds of our sales. In terms of renewal rates, at year end, business members renewed at 94%, Gold Star members at 89.3%, these are numbers for the US and Canada combined, which is over 80% of our company and total US and Canada 90.0%, and worldwide, 87.2%, a slight tick down of a tenth or two from the last quarter.

Executive memberships are growing at a strong pace, driven by an extraordinary renewal rate, and are the most profitable for COST.

Given how forgiving traders have been over huge misses at AMZN, I find the idea of taking COST down hard over almost random changes in renewal rates for part of its business to likely be a contrarian buy signal.

Overall, as said, there are always risks in business, but I see no actual evidence that COST is in trouble.


The macro stock market picture is short term strong, intermediate term (a few years) iffy just as it was in similar strong periods such as 1964 and 1997. Valuations are very high, and the support for valuations from falling interest rates may be reversing.

Within COST’s business, everything from competition from countless sources, changes in buying habits from its warehouse format to more upscale venues, inability to source enough products, and other factors present just some of the things that could go wrong fundamentally. Please see COST’s regulatory filings for a more complete list of risks.

Concluding comments – COST as an enterprise value and ecosystem play

COST is the US, North American and possibly ex-North American leader in warehouse shopping, with accelerating e-commerce sales. Its strengths include immense loyalty from consumers and employees, giving it in my view large enterprise value. In addition, and tying in with its very high customer satisfaction rate, it has a horizontal ecosystem wherein it provides some higher-margined, steady-growth businesses within its warehouse. These include hearing and vision-related services, and pharmacy. The warehouse concept allows it to vary what goods and services it provides, as the tides of competition and consumer preference dictate. As gasoline use continues strong, it drives sales by selling gas. If gas sales fade away and electric vehicles take over, perhaps COST will provide electric charges for shoppers while they shop. Or, perhaps the gas pumps will be ripped out and instead the space will be used for customers to pick up product, inventory storage, or even additional selling space.

Overall, I see COST as having expanding further within North America, substantial global expansion, the ability to provide or direct Costco customers to more ancillary services, vertical expansion into food and perhaps production of other products, and e-commerce expansion.

Thus I am thinking of 10% annual sales and EPS gains for years to come, which can occur with payment of the regular dividend and an occasional special dividend. Assuming that this continues to average about a 3% annual dividend yield, that would suggest 13% annual total returns if the P/E remained stable. Since I expect that if the Fed continues to reverse its QE program, the trend in P/Es is more likely lower than stable or higher, my expectation for COST is also tempered. Thus I hold COST as a permanent asset so long as it performs at a high level, and buy dips such as occurred after the earnings report. In a time of high anxiety about all things AMZN, I am reminded of the brief period in July when it was reported that the Sears (SHLD) appliance brand Kenmore was going to be listed on AMZN. Just from that, mighty Home Depot (HD) dropped as much as 10 points, or about $12 B in market cap. From Kenmore!

As it stand now, that was a great buy point for HD (I bought that dip, adding to my permanent HD holdings) and its smaller competitor Lowe’s (LOW). HD is up over 15% from its panic Kenmore low.

Something similar may lie in store for COST going forward after Friday’s smash in the stock.

In summary, I’m playing for COST to regain its P/E premium over the market while continuing its relatively steady, rapid growth which has attractive, defensive characteristics. Thus I think the odds favor COST as a source of alpha versus other large-cap names, with a time frame of months for traders to many years for long-term investors.

Thanks for reading and sharing any comments you wish to contribute.

Disclosure: I am/we are long COST, HD.

I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Additional disclosure: Not investment advice. I am not an investment adviser.

Do you have an unusual story to tell? E-mail