Description

I’ve gotten a few QVC related questions on the baileyb’s prior LINTA writeup and figured there have been enough changes since December 2013 that it was worth re-posting the idea.  

QVC is a truly great business, run by great people, growing predictably, and trading for 10x free cash flow.  Historically that combination of factors has worked very well for my investments.

If you already know the story and just need a refresher, skip down to the bottom where I update the financials and give estimates / thoughts on the Zulily transaction.  If the only thing you know about QVC is they sell stuff on TV, please start at the top and work your way down.  If after reading this you aren’t completely convinced on QVC, hit me up on the message board and I will spend as many hours as it takes to get you over the hump.

An Introduction to QVC

Before jumping in, pull up the most recent QVC investor presentation from April 2015, which you can access here:

http://files.shareholder.com/downloads/AMDA-GY7JI/5583556x0x820968/DE3A5BC3-0DAD-4360-A8FF-A3212E3DA948/QVC_Group_Euro_Roadshow_2015.pdf

QVC was founded in 1986 and has become the largest television centric retailer in the world, with a presence in the US, Germany, Japan, the UK, Italy, France and China (through a minority owned JV).

While QVC is technically in the retail business, the sticky nature of their customer relationships and consistent customer behavior actually point more towards a subscription model than a transactional retail business – some facts:

• 89% customer retention rate, very consistent by year and geography

• Average customer buys 2+ items per month (in the US)

• Average customer spends $117 per month (in the US)

• Customer acquisition is extremely efficient, driven primarily by TV viewership, which is increasing over time (believe it or not)

QVC has several structural advantages versus traditional retail.  First, they have EXCELLENT customer data going back decades.  One of the eye-openers for me with QVC was the first time I saw this slide (#12 if you’re following along):

What this shows is the purchase behavior in 2014 of every vintage / cohort since inception in 1986.  How many retailers could tell you how much money customers who began shopping with them in 1990 spent last year?  The answer is not that many.  Maybe none.  

Not surprisingly, the ’07 / ’08 / ’09 years were a little lean – since then the business has been acquiring new customers at a pretty good (and accelerating) clip.

One question that this slide might bring up is the average age of the shopping population.  It stands to reason that if we have a significant percentage of customers who’ve been with us for more than 20 years, our average customer age must be high and going higher.  Some facts:

• Average customer is in her 50’s, and getting younger every year

• She has a household net worth of $280k

• She is married and owns her home

• She has a high propensity to shop at Saks and Neiman’s

QVC knows a lot about its customer.  The company has enough data to segment customers by platform (TV vs Mobile vs PC) and by age.  Slide 10 shows the percentage of existing customers by age and by platform compared against the percentages for new customers.  New customers are coming to QVC either through their phone or online and are ~10 years younger on average that the current customer:

To see how the Existing Customer has changed over time, QVC put up this slide in its 2013 analyst day (slide 49 of the presentation found here):

The flatter curve means more equally distributed customers across age cohorts (and by definition,  younger customers).

Another structural advantage QVC has over traditional retailers is the company’s up to the second information about what is selling and more importantly what isn’t selling.  The traditional retail business model involves a company guessing what will sell (to be fair they call it “merchandising”), buying it in advance, filling up a store with it, then reconciling everything at the end of the week or the month.  What sells gets ordered again (if they can get it in time) and what doesn’t sell gets discounted.

QVC knows what is selling every second of every day, and can be more nimble with what products are bought and put on air.  It is hard to process if you haven’t actually seen it, but the model is VERY powerful.  QVC keeps a schedule for what will go on air and when, but the products can be moved around (or removed, or added to) based on customer preference.  On TV this is done by producers who track the sales performance of items shown on TV, following both phone and online sales, and helping the show hosts tailor their message to improve the sales performance.  Producers can then take that information and adjust future show product to reflect their learnings from the recent performance – if, for example, blue shoes were selling great but brown shoes were “out”, future shows could include a second line of blue shoes and remove the brown shoes altogether.

You can see this phenomenon by taking a tour of the QVC facilities.  Below is a blog post talking about this from someone who took the tour last month:

http://www.mediapost.com/publications/article/256724/a-tour-of-qvc-inside-tvs-mecca-of-selling.html

The practical benefit to this is that QVC is able to smooth out sales and offer more protection against a “fashion miss” or just a generally weaker sales environment.  You see this in their comp sales and in the customer loyalty stats, QVC knows what their customer wants and how to present it to them.

A logical question here would be “what distinguishes that customer information and sales knowledge from any other ecommerce first retailer”?  I believe QVC is advantaged because of their content.  Unlike traditional eCommerce which leans heavily on transactions (a need driven decision), QVC produces copious amounts of original content which entertain and inform – ultimately that content leads to the purchase, and the experience leads to trust, which then leads to the loyalty / retention mentioned above.

If you’re skeptical, then maybe this will help.  For every 1 customer watching QVC at any one point in time, there are 9 others watching who aren’t yet customers.  That’s another fact that blew me away.

QVC the Compounder

QVC is a classic compounder.  Predictable (though slightly cyclical) top line growth, some fixed cost leverage / sales mix benefits to margin, consistent financial leverage, and share repurchases drive mid-teens compounding over time.

Here’s how the algorithm works:

• Net active customer file grows 1-2% per year in mature markets, >3% in developing markets

• Spend per customer increases another 2-3% per year

• Total revenue grows 4-5% on average

• Fixed cost leverage and the continuing shift to eCommerce drives ~20bps of margin per year

• This pushes EBITDA growth to 5-6% on average

• QVC maintains 2.5x of leverage, pushing Free Cash Flow growth to >6%

• The company puts 100% of FCF into share repurchases, which in 2015 is $1.1bn on a current market cap of $12.9bn (8.5% yld)

• Taken together, per share cash flow grows roughly 15% per year

There are caveats to this.  The first is that the company does business in Europe and Japan, which exposes them to FX fluctuations.  Both revenues and costs are primarily local which means margins don’t move around much, if at all, so changes are easily modelled.

The second is that the price of the stock does affect the rate of compounding given ½ of growth per share comes from share count reduction.  The higher the share price, the lower the rate of growth, the lower the price, the higher the rate of growth.  Said another way, if the share price doesn’t change over the next twelve months my expected rate of compounding would increase to 17%+, then 20%+ the following year on the same assumptions.  The share price going lower is accretive to earnings.

A look at the revenue history:

There are a few things that should be taken away from the revenue history

• You can see the cyclicality in the organic growth #’s, but the consolidated revenue impact is muted

• Organic growth rates are remarkably stable

• The US is 70% of the revenue (more on EBITDA) and is primary economic engine

• Recent consolidated revenue growth has been negatively impacted by FX changes

Underlying profitability shows a similar trend:

As you can see above, there is margin expansion in the business as sales rise (as seen in the US & UK), but in cyclical downturns, some of the margin is given back (Japan and Germany).  Over time, as revenue grows, margins grow too.

And finally, share count over time and leverage:

You’ll notice that in 2014 the leverage jumped and the share repurchase slowed.  Until October of 2014 the company owned a collection of Internet companies which were “sold” to Liberty Ventures (LVNTA – I wrote this up a few weeks ago on VIC) along with $1.0bn in cash in exchange for LVNTA stock, which was then given to QVC shareholders as a dividend (equivalent to ~$4.50 / share in value).

That had the effect of stopping share repurchases for six months and increasing leverage to a level above their stated target (though no one complained, you basically got a quick 20% on the stock through the dividend).

Post the “re-attribution” as it was called, the share repurchases have resumed and leverage is working its way back down to ~2.5x (through EBITDA growth, not debt repayment).  

QVC buys zulily for $2.0bn

Long time QVC followers got a little surprise two weeks ago when the company announced plans to buy zulily (ZU) for $2bn in a 50/50 mix of cash and newly issued QVC shares.  This was the first major acquisition we’ve seen from QVC – and likely disappointed some investors who were hoping for consolidation with HSN (HSNI) over farther afield eCommerce acquisitions.

Net net, I think the acquisition is neutral to positive.  In the downside case QVC suffers some near term per share FCF dilution versus what they could’ve produced by simply buying back stock.  In the upside case, QVC is able to both improve the profitability / stability of ZU and more importantly use ZU as a customer acquisition vehicle for QVC.

Some facts:

• The acquisition price of $2.0bn is 1.35x forward revenue, 23x forward EBITDA and $415 per active customer (on 2016 consensus)

• QVC is using $900m of revolving debt (at ~2% pretax) and issuing 39m shares to fund the deal

• The deals is structured as a tender offer, which I believe will be complete by Oct 1st

• Cost synergies are expected to be in the “tens of millions”, I calculate $30m

• Revenue synergies were not quantified, but ZUs proxy from earlier this week (found here) suggests management thinks they could be as high as $740m

• From that same proxy, ZU management believes the stand alone ZU business does $200m of EBITDA in 2017, and adds $193m of EBITDA synergies to QVC, suggesting the actual multiple QVC paid could be as low as ~5x EBITDA.

Below are management’s estimates for the ZU business and the synergy estimates:

Management is meaningfully more optimistic about their business than Wall St is, below is a comparison of management estimates versus consensus.

Reading through ZU’s IPO prospectus and most recent annual report it is clear that zulily and QVC have some philosophical overlap in terms of the customer they are trying to serve and how they are trying to present her with a shopping / entertainment experience versus simply trying to find incremental transactions.  QVC brings products to the television, and highlights a “today’s special value” while zulily relies on emails and flash sales.  Both companies focus on exclusive or limited products as a way to differentiate.  Both companies focus on customer lifetime value and relationship building.  That said, QVC has been at it successfully for almost 30 years, zulily is just 5 years old.   

Financially, I frame it this way:

• What would QVC have produced in FCF per share as a stand alone company?

• What does the combined company look like if consensus is right and there are only $30m in synergies?

• What does the combined company look like if management is right on the zulily projections, and there are $30m of cost synergies but no revenue synergies?

• What does the company look like in the ‘blue sky’ scenario where management estimates are right, and they get both the costs and the revenue synergies?

Here’s what I’ve come up with (there is much more financial detail at the end of the report):

Going through this exercise showed me a few things.  First, the deal is dilutive unless the revenue synergies are real, but it’s definitely NOT a “bet the company” transaction.  Second, if the revenue synergies are real (or even 25% real), QVC management will have hit a home run.  In that case, the per share growth rate in earnings skyrockets, creating significantly more value over time than would occur through repurchasing stock.

In very real sense, this transaction looks very much like a “heads I lose a little, tails I win big”.  

QVC and HSN

HSN (Home Shopping Network) is QVCs primarily TV shopping competitor in the US.  HSN was founded 1981 in Florida and reached national distribution by 1985.  Despite being older than QVC, HSN is roughly half the size of QVC in the US (HSN does not have an International presence).

I’ve largely skipped over the history of QVC and how Liberty came to be the owner of the company.  Much of that story overlaps with the history of HSN.  If you’re interested and have time, read the book “Cable Cowboy” (Robichaux, 2002) – there you can get color on how the ownership structure and distribution of the businesses have developed over time.

The punch line, though,  is that John Malone controls QVC (through Liberty Interactive); and after a long period of being owned by Barry Diller through IACI, HSN is now a stand-alone company publicly traded under the ticker “HSNI”.  Through its involvement in IACI, QVC is the largest single owner of HSN with ~38% of the shares outstanding (worth about $2.50 per QVC share today).

Because of the long history between the two companies (and because John Malone has said this at least once), many believe QVC will eventually own HSN.  I am firmly in that camp.  I think the biggest barrier to getting that done is valuation – HSN trades for 17x+ forward FCF versus QVC at 10x; and presumably QVC would have to pay a premium to acquire HSN.  Management at QVC has always maintained that the synergies from the deal don’t justify the premium, and as long as QVC owns ~40% of HSN, they effectively have a ‘blocking position’ on anyone else buying them.

My hope is that the ZU acquisition works, leads to faster growth at QVC, which then leads to a higher multiple, which then leads to a deal with HSN.  These businesses probably should be together.

Below is a look at QVC US vs HSN:

QVC Stand Alone Estimates

Figured it made sense to lay these out first.  Below are my organic growth and EBITDA margin assumptions by geography:

Roll all that up, and it looks something like this:

In the ZU proxy referenced above, QVC management laid out estimates through 2019, below are those estimates and then a comparison vs my estimates:

As you can see, its pretty close.  One thing I would note is that I DO NOT include startup costs for France (the newest QVC market) in my estimates, either costs or EBITDA (or capex).  Management estimates do include those costs (and subsequent benefits presumably).

Now, reconciling EBITDA to cash flow has a few moving pieces.  There is an accounting for the cash owed to our business partner in Japan (Mitsui), the dividend we get from HSN, Capital Expenditures, interest and taxes (for which I assume 32%).  Here are my #’s:

I do not include any drag from working capital, but many do (as they grow, they finance some sales through “ez pay”, which requires some working capital).  I also don’t assume any incremental capex for France (as mentioned above).  I don’t include losses from China (it is not consolidated).  I run interest expense through assuming flat debt levels and bring taxes up with earnings.  Taxable earnings are calculated as EBITDA less D&A (not including acquisition amortization of $323m) less interest.

I then repurchase shares with FCF, which looks something like this:

Dividing the FCF by the updated share count (in this case I like to use end of period “actual” given magnitude of the repurchase activity, as opposed to “average”).

Adding ZU to the mix under the “Consensus estimates plus ~$30m in synergies” assumption, looks something like this:

In the zulily section, all I did to get to the incremental FCF per share from higher ZU accretion was adjust this model up with managements estimates (with and without revenue synergies).

What is QVC worth?

As mentioned above, I see three outcomes post ZU:

1. ZU is a bust, newco QVC generates $2.60 / share in FCF next year

2. ZU works,  but the revenue synergies aren’t there; newco QVC does $2.70 / share

3. ZU is a homerun, newco QVC does $2.80 growing 20%+

As I look across retail, I see forward multiples as low as 12x forward earnings and as high as 19x.  I see QVCs primary competitor trading for ~17x, and the market broadly trading for 16x (QVC is DEFINITELY a market plus business).

Conservatively I’d put FV at 15x FCF.  That still leaves room to retire ~7% of the shares every year and still reach a 13-14% annual rate of compound growth.  In the “ZU bust” scenario, that implies $39 / share in value for the business (+40% upside), not including the $3.00 / share in value we have for HSNI which is under-earning on our BS.

In the “ZU homerun” scenario I would argue the business deserves a retail premium multiple.  At HSN’s multiple the business is worth $47.50 / share (+73% upside), again, not including the $3.00 / share of HSN.  

Either way, QVC is trading today for somewhere between 10.6x in the “ZU bust” scenario and 9.8x in the “ZU homerun” scenario (not accounting for HSN).  Way too cheap for a business of this quality with these growth prospects.

PLEASE hit me up on the message board with any questions or pushback.  I LOVE this idea and would love to get as many of you involved as possible.

Catalyst

- Closing ZU at the end of Sept 2015

- Earnings in late October early November

- November 12th Liberty Analyst day in NYC