Sold: Dunelm’s Share Price Is Too High For My Liking

Consistent double digit return on capital is impressive for a retailer

YES Dunelm has produced very impressive post-tax returns on lease-adjusted capital for a store-based retailer.

Typically, store-based retailers produce fairly weak returns on capital because stores are expensive capital assets, regardless of whether they’re purchased or leased.

However, Dunelm’s policy of operating almost purely from out-of-town superstores has allowed it to produce impressive profits per square metre of rented space, and per pound invested in stores, fixtures and stock.

More specifically, Dunelm’s return on capital has averaged more than 15% over the last decade, and never once fell below 10%. This is far above average and extremely impressive.

Dunelm’s return on sales (profit margin) is also impressive for a low cost volume retailer, averaging 10% over the last decade. In comparison, Tesco’s margins have almost never exceeded 5%.

YES The vast majority of Dunelm’s business is focused on homewares, from soft furnishings such as curtains and cushions to rugs, lights and more recently a fast-growing selection of furniture from beds to desks and sofas.

Traditionally the company focused on out-of-town superstores, but this has now evolved into a “total retail” system blending in-store and online ordering with click and collect and home delivery.

YES Dunelm has been focused on homewares since 1979 when it started out selling seconds such as curtains that Marks & Spencer and other big retailers didn’t want.

YES When it joined the Million Pound Portfolio in 2016, Dunelm didn’t have a clear goal. That has since been remedied and now the company’s purpose is “to help everyone create a home they love”.

The company’s overall strategy has remained largely unchanged for many years.

Essentially the strategy is to provide a wide range of value-for-money homeware products across a wide range of price points. These are sold through large out-of-town stores and now a very fast and functionally modern website, with customers able to pick up in-store or at home regardless of how they order.

This gives the customer convenience, with Dunelm effectively being a one-stop-shop for homewares for a large portion of the UK population (my wife was very excited when Dunelm opened its first store in our hometown).

YES There have been no attempts to “transform” Dunelm at any point, and that’s a good thing. Instead, the company is continually evolving by reinvesting earnings to adapt to the ever-changing retail landscape.

YES Dunelm has grown at a healthy pace of around 5% to 7% over the last decade. This is ahead of inflation and has required the roll out almost 70 additional stores, but this is in no way excessive.

Single digit growth can typically be absorbed with ease, as it doesn’t involve a rapid increase in new stores, new processes, new staff and so on. And the cost of fitting out new stores and buying in additional stock has been comfortably covered by the company’s high returns from capital employed.

YES Dunelm only made one meaningful acquisition during the last ten years. It acquired the loss-making Worldstores business out of administration for around £10 million, but eventually incurred costs of £30 million to pay off suppliers and tie up other loose ends.

At the time the media seemed to be quite negative about this acquisition. Worldstores was the UK’s largest online furniture catalogue, but it was a failed business which dragged on Dunelm’s returns for a year or two and was eventually broken up and sold off or closed down.

But the whole point was to acquire Worldstores’ technology platform, furniture supplier relationships and two-man delivery infrastructure. And from that point of view it was very successful, giving Dunelm exceptionally good online ordering and delivery capabilities.

NO Dunelm’s products do not become better just because more people use them.

YES Dunelm’s most valuable assets are its brand and its culture as a long-time family business.

Although people won’t willingly pay more to buy Dunelm branded pillows, the brand is useful as a way to lower search costs for customers.

In other words, if someone’s thinking about buying curtains, they might just do a web search for “Dunelm near me” rather than “curtain shop near me”. So being synonymous with your product is incredibly valuable, as companies like Domino’s and Victrex (LON:VCTX) know only so well.

As for Dunelm’s family business culture, the company was managed by the founding couple, Bill and Jean Adderley, from 1979 to 1996. After that their son, Will, who grew up working in the business, took over. Will was CEO until 2016, giving the company a near 40-year period as a family-run business.

Will is currently on the board as deputy chairman and, when combined with his mother’s shares, the Adderley family still own more than 50% of the company.

This history and ongoing founding family involvement can have various benefits. The primary benefit is that founders often have a longer time horizon than hired managers.

Founders often think about handing their business onto the next generation, which invariably involves thinking about how to build the company over decades. This typically precludes high risk strategies, such as using debt to fund excessive growth or acquisitions.

Another frequent benefit of long-time family-run businesses is a desire to have the right people in the business and to treat them as more than just transactional cogs in a corporate machine.

With Dunelm, this showed up in the short-lived tenure of Will Adderly’s replacement, who became CEO in 2016 and left for “personal reasons” in 2017.

In reality, the new CEO was generally thought to be a bad fit with Dunelm’s culture, and it’s hard for a CEO to win an argument with a deputy chairman who owns 44% of the company’s shares. It also shows up in Dunelm’s employees, many of whom have been with the company for decades.

Having said all that, a strong brand and family-like culture focused on the long-term are not golden tickets to a bright future, as John Lewis has found out in recent years. Dunelm still has to outwork and outthink the competition, but if it can do that then the odds are at least skewed in its favour.

YES It’s the leading homewares retailer having overtaken John Lewis in 2012. This should give Dunelm some economies of scale advantages, and being #1 is a social proof which helps increase trust in the business.

NO Customers can easily buy curtains or cushions from another retailer.

YES Dunelm currently has outstanding debts of almost £320 million, the vast majority of which are future store lease liabilities. That’s about four-times its average earnings over the last ten years of £84 million, which is a very low level of leverage for a store-based retailer where lease liabilities are often double-digit multiples of average earnings.

The reason for Dunelm’s low leverage are twofold: First, it doesn’t borrow very much from banks or other lenders. Second, its store lease agreements are very short.

Almost two-thirds of Dunelm’s store lease agreements end within the next five years, whereas some retailers and restaurant chains frequently sign leases lasting ten or twenty years (and in the case of M&S, it has one store lease extending more than 200 years into the future).

These short leases give Dunelm much greater flexibility with its store estate, which is incredibly useful and a material advantage in a world where online shopping has already become the default for most people.

In other words, Dunelm can close or move stores relatively quickly with relative ease, whereas retailers with 20-year or 200-year leases will have to pay a king’s ransom to get out of those lease agreements early.

NO Although the homewares market has performed exceptionally well during the pandemic, that’s down to luck rather than any intrinsic defensive or counter-cyclical qualities of the market itself. So in a normal recession, people will typically put off buying curtains and cushions until the economy improves.

For example, Dunelm’s like-for-like sales (which strips out the effect of new store openings) declined by more than 8% during the 2005/2006 slowdown and by almost 10% during the 2009 financial crisis.

However, during that time the company was still increasing its store count, and that more than offset sales declines in individual stores, leading to an overall increase in revenues. Even so, the underlying homewares market is still cyclical.

YES Dunelm’s core homewares and furniture markets are both very mature and are expected to grow more or less in line with inflation over the next five years or so. This ignores 2020 and 2021, which are unusual because of the pandemic.

The market is expected to have returned to its pre-pandemic size and growth trend by 2022.

YES No abrupt regulatory changes are expected in either homewares, furniture or retail in general.

NO Retail is currently going through a period of disruption thanks to the World Wide Web and, more recently, smart phones.

However, this transition is now largely over, with almost all competitive retailers having sophisticated online capabilities, from websites and apps to the infrastructure necessary to handle deliveries, returns and click and collect (for those retailers that still have stores).

I think this disruptive period for retail will be over quite soon, but for now I think the pace of change is still fast enough to call it disruptive.

YES Dunelm isn’t overly dependent on any one customer, employee or supplier.

YES Dunelm doesn’t rely on patents to any meaningful extent and it isn’t dependent on just a handful of key products. And products which do need to be updated tend to have minor updates rather than ground-up redesigns (it seems unlikely that there will be any great need to “reinvent” the curtain anytime soon).

YES Although commodities are used in the production of Dunelm’s products, they don’t materially impact selling prices from one year to the next.

YES The pandemic is the obvious problem which is having a seriously damaging effect on most retailers. And perhaps over the next year or two it will begin to negatively impact Dunelm seriously as well.

For now though, Dunelm has performed surprisingly well during the pandemic, with revenues from July, August and September up an incredible 37% over last year. Although Dunelm is a quality business and was performing well before lockdown, its success through the pandemic is almost entirely down to luck.

With people spending much more time at home, it was inevitable that many of them would want to improve the environment in which they were suddenly spending the vast majority of their time. And for those working from home, the need for a decent desk and chair (at the very least) would be a worthwhile investment.

Of course, there’s a chance that the pandemic is just pulling demand for homewares from the future into the present, and that Dunelm will see weak sales in 2021 and perhaps 2022. But either way, I don’t see the pandemic or anything else as an obvious risk to the company’s longer-term prospects.

YES It seems extremely likely that people will still want to buy cushions, clocks and curtains 20 years from now, in approximately the same volume per person as they do today.

On that basis, and as I’ve already said, it seems reasonable to expect Dunelm’s underlying markets to grow in line with inflation, or around 2% per year if the Bank of England can keep inflation near their target. On that basis I think growth of 2% per year for Dunelm is a reasonable low level expectation.

In addition, Dunelm has said for a long time that it thinks it can operate 200 superstores in the UK without excessively cannibalising sales from other nearby Dunelm stores. I expect Dunelm to reach 200 stores within 10 years from its current count of 173. That’s an increase of 16% or just under 2% per year.

So if we combine like-for-like growth of 2% with store rollout growth of 2%, then perhaps Dunelm can manage growth in the region of 4% per year for the next decade.

On top of that, the company wants to get people to shop more often and put more things in their basket when they do. So perhaps Dunelm can take add another 1% or so to its overall growth rate (a very ballpark guess).

On that basis, I think it’s reasonable to assume that Dunelm can grow by around 5% per year on average over the next ten years. For comparison, that isn’t far off the 7% growth rate it achieved over the last decade (ignoring the impact of the 2020 dividend suspension) when its store rollout program was far more aggressive than it will be over the next decade.

A 5% growth rate over 10 years will produce total growth of almost 65%. That would take the dividend from a pre-pandemic high of 28p to almost 46p, assuming similar growth across revenues, earnings and dividends.

Over the ten years from 2030 to 2040 I have even less idea what might happen. Perhaps Dunelm will expand internationally, but who knows?

If Dunelm sticks to the UK, it seems reasonable that Dunelm might grow in line with inflation, plus perhaps another 1% per year for population and disposable income growth, giving it a growth rate beyond the next ten years of 3% (again, obviously a ballpark figure).

NO I like Dunelm. I think it’s a quality business which is likely to continue growing for many years to come. However, at its current price I don’t think it’s likely to outperform the market over the next ten years, and its dividend yield is a little too low for my liking. More specifically:

Dunelm has one of the lowest stock screen ranks of any of the model portfolio’s holdings. It’s ranked 79th out of more than 200 companies, which isn’t bad, but it’s 28th out of 34 holdings in the portfolio.

This low-ish rank is mostly due to Dunelm’s high PE10 and PD10 ratios. These are 33 and 71 respectively, which are above my preferred maximums of 30 and 60. If the dividend hadn’t been suspended then these ratios would still be relatively high for a company with somewhat pedestrian growth prospects.

The shares also seem to offer relatively weak value from a dividend yield plus dividend growth point of view. Here’s why:

If Dunelm reinstated its dividend at the 2019 level of 28p, then the shares would have a dividend yield of just 2% at their current price of 1,342p. A 2% yield is a little low for my liking, but a 2% yield might be okay for a company that can grow at 8% or more per year over the next twenty years.

But, as I’ve already said, my ballpark estimate for Dunelm’s growth rate is 5% over the next decade and 3% beyond that.

A 2% yield plus a 5% growth rate gives an expected total return of 7%, which is a bit weak and no better than the UK market’s long run rate of return.

Even worse, if Dunelm grows its dividend by 5% annually for ten years then its dividend will reach 46p in 2030. If the shares stayed exactly where they are today, then in that 2030 scenario they’d have a dividend yield of 3.4%.

A dividend yield of 3.4% is nothing special, especially if Dunelm’s growth rate has slowed to 3% by that point.

So for me, there’s a real risk that even if Dunelm performs well as a company over the next decade, its share price might not increase significantly or even at all from where it is today.

This is similar to the situation from mid-2013 to mid-2016, when Dunelm’s share price was sometimes above 1,000p. Back in 2013 Dunelm’s dividend was 16p, giving the company a dividend yield of just 1.6%.

Investors were clearly optimistic about Dunelm’s future, but they were too optimistic and had to wait almost seven years to see any capital gains at all.

In contrast, Dunelm’s shares fell as low as 500p in 2018 as the company worked through the Worldstores acquisition and other headwinds. At that point, the dividend yield was more than 5%, a far more attractive yield especially when combined with a high quality business.

And let’s not forget, from 500p Dunelm’s investors have seen an almost 200% capital gain. So as always, valuation matters, even for high quality businesses.

NO Dunelm is a high quality business which I’m happy to own. However, at more than 1,300p I don’t think the shares are good value for money.

On that basis, and somewhat reluctantly, I removed Dunelm from the model portfolio and my personal portfolio a few days ago.

The proceeds will, as usual, be reinvested into a new holding next month.

Original Post

Written By: John Kingham