Conclusion: We’ve spent a lot of time on the road discussing RH over the past three weeks, and most specifically, our recent 45-page deep dive on RH’s real estate. The punchline of our analysis is that a) RH stores could (and probably should) get far bigger than even the RH bulls seem to think, b) Aside from reconfiguring 66 existing markets, there’s another 19 markets we identified where the spending rate on home furnishings by people making over $100k in income suggests that RH should expand to these markets with Design Galleries, and c) the availability and economics on large properties for all these markets are far better than people think. This analysis supports our $11 earnings power in five years (double the consensus), as well as our view that that this stock is headed well above $200. Here are some of the slides that we kept revisiting in our conversations.
1. Market Share Trumps Store Productivity: For the most, people underestimate the ramp in RH’s addressable market as the company continues to expand into new categories. Over the next five years, there should be $45bn upside in market opportunity for RH simply by expanding its presence into new categories at retail, including kitchens. An important note is that we analyzed every market of the US, and isolated only consumers making over $100k annually. The government’s aggregate numbers include every income level. But the fact of the matter is that the average American spends $857 annually on home furnishings, while those making over $100k spend $1,779.
2. Real Estate Methodology. In our analysis, we look at each market at a micro level. That is, we isolate the existing store, and then look at the demographics within a specific driving radius. We look for income levels, home values, and ultimately, how much money consumers at each income level spend on the categories where RH is expanding. This chart below shows Seattle, but we did this for every existing and potential RH market in the US and Canada.
3. Store can get MUCH bigger. The key to how we model RH is based entirely on market share. There are three factors that mater…store size, productivity, and market share. It’s absolutely impossible to pinpoint any one of those without knowing the other two. We think we can get pretty close. We already know that the highest productivity FLDGs are running at 8-9% share of their respective markets – and that’s before adding new categories like kitchens. Our model assumes that each FLDG hits 10% share in year five of our model, and generates $1,200 per foot (reasonable based on what we’re seeing today). That leaves us with an implied store size. In some markets like New York, it suggests that RH could have a store over 100,000 feet. Same for Houston (we think it expands its existing store). Most people we talk to cringe when we discuss anything bigger than the 20-25k box that we’ve seen built over the past two years. But this analysis suggests that RH could support 30 stores over 50,000 square feet, and all but 5 can support a 25,000 foot FLDG.
4. New Market Potential. In addition to existing markets, there’s another 19 markets where RH can, and should, build FLDGs. In 10 of the markets RH could add a store 45,000 feet or larger, and the biggest market – Montreal – could support an 80,000 sq. ft. store.
5. New Store Math is a Slam-Dunk. In this example (Cherry Creek) RH is taking over a Saks and is going from a 7,500 feet legacy store to a 56,000 foot FLDG. Implied market share at the property goes from 1.5% to 4.3% by our math, and despite the incremental $20mm in revenue, rent only goes from $1.3mm to $2.0mm. That takes occupancy costs from 12.6% to 6.5%, and likely lower as the store becomes more productive. But the key to this algorithm is that there’s $19.5mm in build-out costs, $15mm of which is being picked up by the landlord. Inventory costs in this business are minimal at the store level. So when you add up all the economics of the store, you get to a 6 month payback. It’s tough to find that elsewhere in retail. This leads us to think that our Gross Margin estimates (which don’t go above 39%) are potentially conservative.
6. Yes, there are more of these opportunities than most people think. There’s still a nice pipeline of free-standing locations – like what RH has in Greenwich, Houston, and Boston. But we’ll see more examples of the mall-anchor space as outlined above in Denver. Take for example the Cherry Hill mall in NJ. A high-end property with three anchor tenants – Nordstrom, Macy’s and JC Penney. Which one does not belong? JCP has less than half the productivity of Macy’s and Nordstrom, and arguably does not belong in any ‘A’ mall. That’s not where JCP’s customer shops. We think we’ll see more situations where the landlord buys out JCP, takes the space and carves it into 2 or 3 highly productive retailers – who will collectively transform that end of the mall. In this instance, we use RH, and arbitrarily pick CAKE and WFM. That would take annual REIT income from $10.1mm to $23.7mm (see second table below). That makes it pretty easy for the landlord to justify buying out JCP and building a couple million worth of walls, stair cases and escalators. As a point of reference, JCP has about 140 ‘A Mall’ properties. Yeah…big number.