In this article I calculate a Buy Price and Sell Price for Dunelm’s shares based on a review of the business and its potential future dividends.
Dunelm has a long history of dividend growth
Dunelm Group PLC is a leading (actually the leading) homewares retailer in the UK, selling all manner of soft and hard home furnishings from cushions to dinner sets and sofas through a nationwide network of superstores and its Dunelm.com website.
Dunelm’s long-established strategy of rolling out more and more out-of-town superstores has seen it grow from a single store in 1984 to more than 170 stores today.
As you might expect, this led to a massive increase in revenues, earnings and dividends over the years. Looking back at just the last ten years, revenues have doubled from £540m to £1,100m, earnings went from £60m to £100m and the base dividend went from 11p to 28p per share.
This track record of consistent dividend growth (ignoring the almost inevitable dividend cut during the pandemic) is exactly what I’m looking for in a company, so its worth pulling up a few more figures:
- Revenue per share growth of around 8%
- Earnings per share growth of around 3%
- Dividend growth of around 12%
- Capital employed growth of around 4%
We can also put some numbers on the company’s growth consistency:
- Revenue per share increased eight times in nine years
- Earnings per share increased seven times
- Dividends per share increased eight times
- Capital employed per share increased seven times
That’s very consistent growth and few companies, especially retailers, can sustain that level of consistency for so long.
Using SharePad to check the figures all the way back to 2004 shows the same pattern of consistency holds, so the last ten years of relatively consistent growth wasn’t a fluke.
However, while growth is nice it doesn’t mean a thing if it’s unlikely to be sustained long into the future. For that to happen the company has to have some sort of durable competitive advantage, and a good place to find evidence of competitive advantages is in a company’s profitability ratios.
Profitability has been consistently high
The chart above shows Dunelm’s results for my two favourite profitability ratios: Net return on capital employed (including lease liabilities) and net return on sales (profit margin).
Given that Dunelm is still very much a store-based retailer with lease liabilities of several hundred million pounds, the fact that it has consistently generated returns on lease adjusted capital of more than 10% is very impressive. Not many store-based retailers can manage that and those that do are usually very high quality businesses.
Having a consistently high return on capital is especially important because it allows a company to grow organically and at a decent pace by reinvesting some of its earnings.
So in Dunelm’s case it produces returns on capital of around 15%, most of which it pays out to shareholders as dividends, special dividends and share buybacks. The rest is reinvested into expanding its asset base (e.g. inventory, store fit-outs, machinery, tech infrastructure) which then forms the foundation of future growth.
Debt isn’t high for a store-based retailer
As I mentioned above, Dunelm is a store-based retailer with a network of more than 170 stores, the majority of which are leased. Leases are effectively a form of debt (the store is borrowed from the landlord in exchange for rent payments) so it’s important for store-based retailers to keep their non-lease liabilities (i.e. borrowings) in check.
In Dunelm’s case it has combined borrowings and leases of £359m (as at its 2020 annual results), which sounds like a lot. But over the last decade Dunelm earned an average of £84m per year, so its debts are just over four-times its average borrowings, and that’s not bad for a store-based retailer.
Generally I’ll avoid a property-heavy business if its borrowings and leases total more than five-times its average earnings, so on that basis Dunelm has acceptable levels of debt (to be honest most retailers fail this test, typically by a country mile).
Note: The decline in borrowings in the chart above reflects a change in accounting policy (IFRS 16) where leases used to be reported in absolute terms but are now discounted (typically) by the company’s weighted average cost of debt.
Dunelm may have valuable but hard to replicate competitive advantages
So far we’ve seen that Dunelm has been very successful for a very long time, producing consistently high profits without excessive debt.
However, in a relatively free market high profits will attract competition and if a company cannot fend off competitors then its past success will not be replicated in the future. The only way to defend against competitors over the longer-term is to have some form of durable competitive advantage, otherwise known as an economic moat.
Some companies have obvious competitive advantages such as a brand name that allows the company to charge abnormally high prices (e.g. Ferrari), or network effects that make it almost impossible to compete against the market leader (e.g. Rightmove).
But Dunelm has no obvious competitive advantages and exactly how Dunelm has been able to maintain its above average profitability decade after decade is a bit of a mystery (at least to me).
After an extended period of head-scratching my view is that Dunelm’s main competitive advantages are deep competence, stakeholder goodwill and true long-term focus, built up over more than 40-years as a family-run business looking after its customers, employees and suppliers.
This is good because “soft” advantages like deep competence, stakeholder goodwill and true long-term focus are hard to replicate. But it can also be bad, because those advantages can be destroyed by a dangerously incompetent CEO.
Having the founders and their son as near-majority shareholders reduces this risk, as does having the son as Deputy Chairman. And historically it looks as if they’ve been willing to nudge out unsuitable CEOs when necessary.
Dunelm has performed very well through the pandemic
When the pandemic initially struck in early 2020 Dunelm was hit hard, as were most retailers, and in the first UK lockdown its stores were closed for many weeks. Even its online operations were closed for a short time as social distancing measures were introduced to warehouses and other non-store operations.
As Dunelm’s online operations re-opened there was a significant increase in demand, albeit from a very low base. But by June 2020 the company’s total sales were actually up 20% relative to June 2019 and online sales were up more than 120%.
By October 2020 Dunelm’s sales were up 37% year-on-year and its 2021 interim results (covering the period to December 2020) announced an earnings per share increase of 32%.
That all sounds fantastic and it is a very commendable result, but it isn’t an isolated case. Sales at Next’s branded homewares business were up 55% in the 12 months to Jan 2021 and N Brown’s Home Essentials business grew from 30% of overall revenues to more than 40%. So pretty much any homewares business did well.
The reasons for this boom in homewares are fairly obvious:
- Lockdowns: During multiple lockdowns people were literally ordered to stay at home. They couldn’t go out and spend money so instead they spent it doing up their home (where they were spending all their time).
- Homeworking: Millions of people switched to homeworking and suddenly required a home office, desk, filing cabinet, laptop stand and who knows what else, which Dunelm was happy to provide.
This is all good for Dunelm because it enabled the company to re-start its suspended dividend earlier than might have been possible otherwise, but I don’t expect last year’s impressive growth rate to be sustained for any meaningful period of time.
Having said that, I’m not super-pessimistic about Dunelm either. Looking beyond the pandemic the company has benefitted over the last several years from a broad increase in homewares spending.
This has been driven largely by social media, with people taking selfies and posting them online. As a result, their “friends” can see what the inside of their home looks like and therefore having nice interior décor is more important to more people than it was five or ten years ago (gotta keep up with the Joneses, right?)
I expect this trend to continue, perhaps for several decades, as the social media-loving millennials and Gen Zs replace the less technologically inclined Gen Xers and Baby Boomers.
Overall then, I like Dunelm. I think it’s a high quality business operating in a market with long-term growth prospects, and I think it probably has valuable and hard to replicate competitive advantages which make long-term growth more likely.
I would be happy to invest in Dunelm at the right price, so let’s look at one way to calculate a target Buy Price and Sell Price for this company.
Calculating a Buy Price and Sell Price based on Dunelm’s future dividends
A company’s intrinsic value is equal to the present value of all its future dividends.
Historically Dunelm also returned some cash to shareholders through buybacks, but we can can take account of that by assuming we’re going to buy the whole company.
If you own the whole company then it makes no sense to do buybacks because you’d be buying back shares from yourself, so if we assume we’re going to buy the whole company then we can also assume that any cash historically used for buybacks will instead be used to pay dividends.
To value a company we’ll need to build a dividend model, and to do that we need to think about a variety of factors that could drive or constrain future dividends.
For Dunelm, some of the main factors are:
Return on capital employed and retained earnings:
If we conservatively assume that Dunelm will fund future growth via retained earnings rather than additional debt, then return on capital will be a constraint on that growth.
Think of it like a savings account. A savings account paying 10% interest can only grow by 10% per year even if all of that interest is reinvested. So the higher the return on capital, the higher the potential growth rate (or alternatively, the higher the potential dividend payout).
Increasing number of stores:
Dunelm is a UK-only business so for now it’s limited by how many people there are in the UK and how much they want to spend on homewares. This translates roughly into a maximum number of stores and the company has long had the goal of reaching 200 superstores in the UK.
So far, unlike many other areas of retail, there doesn’t seem to be a huge shift away from store-based purchases. I assume people still like to physically browse, touch and feel homeware products like cushions, curtains, ornaments etc, so for now I will assume that Dunelm’s 200 store goal won’t change for at least the next decade.
Per store revenue growth:
A combination of inflation, real price increases and online growth has driven consistent revenue per store growth in the region of 3-4% per year over many years.
Although Dunelm’s online revenues have increased dramatically over the last decade (going from almost nothing in 2011 to £200m in 2020) I don’t think this dramatically changes the company’s overall growth prospects. Instead, I think most of this growth comes from sales that have shifted from instore to online rather than being in addition to store sales (although as I said above, this shift is not on the scale seen by clothing retailers for example).
This revenue cannibalisation isn’t a bad thing though. Dunelm’s stores would see sales shift online regardless of whose website they went to. So it’s better for Dunelm if it can retain those sales via its own website rather than seeming them disappear into the pockets of a competitor.
So those are the major factors, and here are some related assumptions which will underpin the dividend model:
- Return on capital employed: This increases to 21% in 2021 and beyond, which is slightly above the historic norm (which has been trending upwards in recent years)
- Increasing stores: The total number of stores increases by a historically normal 2% per year, growing the total from 175 in 2021 to the 200-store target in 2028
- Increasing revenue per store: Revenue and profit per store increase by the historically normal rate of 3.5% per year largely as a result of inflation and online sales growth
- Long-term growth: Beyond 2030 Dunelm’s growth stabilises at 3-4% per year due to a combination of inflation and mild expansion into new markets
Remember, these are just assumptions and they will of course be wrong. But the point is not to be right; the point is to be both realistic and conservative and I think these assumptions are both.
Feeding those assumptions into my investment spreadsheet gives the following dividend model (see the description below if you prefer words to a wall of numbers):
Translated into English the above table says:
- Stores increase from 175 today to 200 by 2028
- Revenues almost double by 2030 and EPS sustainably reaches 80p with average growth running at 4.8%
- Dividends grow from 42p in 2021 to 61p in 2030 (these dividends are much higher than past dividends because historically Dunelm returned cash to shareholders as a mix of base dividends, special dividends and buybacks, whereas this model assumes those are all lumped together into a higher base dividend)
- Long-term growth slows to 3.5% as Dunelm reaches saturation in the UK
There are two main outputs from this model:
- Buy Price: £6.64
- Sell Price: £12.51
The Buy Price is the price at which those estimated future dividends would produce an annualised return of 10%, which is the target rate of return for my portfolio. So if Dunelm’s share price was equal to the Buy Price of £6.64, the estimated 42p total dividend in 2021 would give a dividend yield of 6.3% and of course that’s quite attractive.
The Sell Price is the price at which those same dividends would give a 7% annualised return, which is the long-term average return from the UK stock market. In other words, if a company’s shares aren’t expected to produce a better long-term return than the market then I would be looking to sell them.
Note that despite the names, the Buy Price and Sell Price are not hard commands to buy or sell at those prices. Instead they are landmarks which can be used to help you navigate your investment journey, but exactly which way you choose to go is up to you.
As I write, Dunelm’s share price today is £15.15, so it’s comfortably above the Sell Price. This means its estimated future returns are going to be below the market average of 7%, assuming of course that the above dividend model is indeed conservative.
The fact that Dunelm’s share price is above the Sell Price feeds into another metric which I call Margin of Safety.
This basically tells you where the actual share price is compared to the Buy and Sell Price. For example:
- If the share price equals the Buy Price then Margin of Safety is 100%
- If the share price is halfway between the Buy and Sell Price then Margin of Safety is 50%
- If the share price equals the Sell Price then Margin of Safety is zero
Given that Dunelm’s share price is above its Sell Price it has a negative margin of safety. In other words, if things turn out as I assume in the model then Dunelm’s total shareholder return will underperform the market.
- Dunelm’s Margin of Safety at £15.15 = -44%
However, this is based on share price, so if Dunelm’s share price halved and if its future dividend prospects remained the same, then its Margin of Safety would improve dramatically.
Calculating a target position size for Dunelm
One final point which isn’t included in the dividend model above is Dunelm’s target position size. Given that different stocks have different expected returns, it makes sense to weight a portfolio towards its most attractive holdings, and you can do that if you have an active position sizing policy.
In my case, I have a default target size of 3% for cyclical companies like Dunelm and 4% for more defensive companies (I’m a defensive value investor so I favour defensive holdings). I then use a simple formula which adjusts that target size based on the margin of safety. For example:
- If Margin of Safety is 100% then I increase the target size by 50% (4.5% in Dunelm’s case)
- If Margin of Safety is zero I decrease the target size by 50% (1.5% in Dunelm’s case)
The target position size is adjusted on a sliding scale (my spreadsheet will calculate this for you), so with a very negative Margin of Safety Dunelm’s target position size ends up being very small:
- Dunelm’s Target Position Size: 0.2%
This tiny target position size reflects the fact that, based on the above dividend model, I’m not interested in owning Dunelm at its current price.
However, I like Dunelm and I think it’s a quality business, so rather than owning Dunelm it will instead go onto my Watchlist.
If at some point investors become disappointed in Dunelm and the price moves below say £8.00 within the next year then I might be interested, as long as Dunelm’s expected future dividends remained largely unchanged.
In the row “discount factor (target)” the 2021 column is 91%. Does that mean the discount factor is 9% per year (since 100% – 91% = 9%)?
I was expecting that you would discount the future dividends at a rate that resembles loss of purchasing power, which might be around 3% per year.
John Kingham says
Hi Ken, that’s what it looks like but it’s actually a slightly different calculation.
The discount rate is effectively the inverse of the rate of return.
So for example let’s say I’ll give you £100 one year from now if you give me some money today. How much would you pay me today to get £100 a year from now?
If you want a 10% return on your money and if we call the amount you’ll give me X, then we know that X * 110% should equal £100.
We can solve that equation by switching it around:
X * 100% = £100, which can be re-written as
X = £100 / 110%, or alternatively
X = £100 * (1 / 110%), which simplifies to:
X = £100 * 91% (rounded)
So the 91% effectively discounts the £100 by the inverse of a 10% rate of return, if that makes sense.
I use 10% as that’s my target rate of return. Different people will use different values; Buffett uses the risk-free rate based on long-term government bonds and then he tries to buy at much lower price than that. Personally I’d rather use my target rate of return, but there’s no one “right” way to do it.
Bob Barnacle says
For year ending June 2020, notice debt jumping
from £44m (14% gearing), to £311m (155% gearing),
with corresponding increase in Property, Plant and Equipment.
Is that to do with with lease accounting changes or something else ?
John Kingham says
Hi Bob, yes that’s the IFRS 16 lease accounting change. Lease liabilities now show up on various financial data websites as borrowings so you now have to search the annual results to pick apart how much debt is owed to banks or landlords. For Dunelm, actual borrowings are the same at £45m, lease liabilities make up the rest.
The biggest threat to Dunelm business model is AR. When AR technology catches up Dunelm might end up like Blockbuster. Since thanks to globalisation its easy to make things abroad at a fraction of cost and selling online via AR and centralised distribution centres will make a competitor even more efficient. In a situation like this any advantage Dunelm has will become net zero. There is a possibility Dunelm adapting to new technology but long established retail business are the worst adaptors of modern trends in retailing. This is why I favour business like tobacco, confectionary products and household detergents. For such business growth is minimal and also lack of innovation in product. As a result existing companies earn a high ROIC but face few competitors.
John Kingham says
By AR I assume you mean Augmented Reality? I guess there’s a risk that people will be happy to view new cushions or curtains in their home using AR. But how hard would it be for Dunelm to use AR technology as well? And it already has the UK’s largest (most efficient?) distribution infrastructure for homewares, so a competitor would have to replicate that. Obviously that isn’t impossible and there are pure-play online competitors already, but not all of them are successful. I think what is likely to happen over the next decade or so is that Dunelm starts to reduce its superstore estate as people become more comfortable buying online.
Dunelm also has the potential to follow Next’s lead (disclosure: I own shares in Next). Next has developed their Total Platform as a way to become a platform for smaller clothing retailers and brands to piggyback on, using Next’s larger and more cost effective infrastructure rather than developing their own. I think Dunelm could probably do something similar as it’s already the #1 go-to brand for homewares. And it has the store network which can be used for click and collect or returns.
So yes there are risks of course, but my guess is that Dunelm will be bigger in ten years than it is today, while also being significantly more tilted towards the online world.
As for companies behind branded tobacco, detergents etc, I agree and I own some of those too!
I cant fault your argument! Dunelm may well be able to migrate towards Augmented Reality. Interestingly some of the things you have added in your comment I was oblivious too! Perhaps its a good think for me to stick to FMCG sector because if you don’t know the business well enough its probably not a good area to invest in!
The only kind of retail business I would actually consider investing in would include LVMH, Hermes, Kering or Games Workshop. I think these companies have a very wide moat partly because the products are all ‘old school’. I think you should do a piece on Games Workshop very good company but far too expensive! However its a good illustration of a robust retail business.
John Kingham says
I think a Games Workshop article would be interest. I haven’t looked at it in detail although I used to visit some of its stores in the 80s and 90s.
My guess is I’ll think it’s expensive, but we’ll see. I’ll probably be able to do a review sometime in June.
For me Dunelm is one of the Corona stocks, i.e. stocks which did well last year due to pandemic: these include some customer staples like Clorox, Zoom, Netflix, but also some consumer discretionary like Nike etc.
In my opinion all of them are trading now above their intrinsic value, so no point to invest at this time.
John Kingham says
Hi Eugen, yes there were lots of companies with very abnormal results last year. I think Zoom is very interesting. Personally I don’t see how it does anything that Microsoft can’t copy with about ten minutes of coding and a few more servers. I think most of its success is that the name Zoom sounds “cool” and people think they’re on the cutting edge when they say “I’ve been on oh so many Zoom calls”. Microsoft just needs a catchy name for its video conferencing product.