The hidden debt of lease obligations

Unless you’re an accountant or an experienced and well-read investor, chances are you either haven’t heard of capitalising lease obligations or, if you have, it’s something you don’t do because it can be a lot of work.

Until recently I fell into the second group. I knew what lease capitalisation was (don’t worry if you don’t; I shall explain all shortly) but I didn’t do it because a) none of my investments had problems with crippling lease obligations and b) it was a lot of work.

However, the argument (or excuse) that it’s all too much work is about to become null and void, thanks to a new accounting standard known as IFRS 16: Leases.

This new standard is about to shine a great deal of sunlight onto what was previously a dark and largely hidden debt, so now seemed like a good time to review the basics of lease obligations and how much of what is generally a good thing can be too much.

After much pondering, frowning and chin-scratching I came to the (long overdue) conclusion that lease obligations are basically the same as debt obligations.

You can learn a bit about lease obligations and how I’m going to factor them into my investment process in this month’s Master Investor magazine (below).

I’ve also included Burberry, The Restaurant Group, Next (all of which I own) and Marks & Spencer (which I don’t own) as examples of how the hidden debt of lease obligations can seriously distort the accounts of retailers and other property-based businesses.

Capitalising lease obligations

Looking at total future lease obligations

In the article above I said, “I’m going to start using a multiplier of six-times minimum lease payments to calculate ROLACE (return on lease-adjusted capital employed) for lease-heavy stocks on my stock screen.

First of all, I should have mentioned where you can get the rental figures. You can get them from SharePad, and possibly other data providers.

You can also find them in the notes to the accounts. For example here are the rent figures for The Restaurant Group:

Lease payments for The Restaurant Group

Contingent rents are rents that are based on a performance metric such as revenues. These are not considered leases by IFRS 16 as the amount owed varies depending on revenues, so they’re not a fixed obligation like debt.

So If I was going to calculate the future lease obligations for The Restaurant Group, I would only use the “minimum lease payments” figure in the rental multiplier calculation. And here is that calculation:

Future lease obligations = £78.2m * 6 = £469.1m

That’s a reasonable estimate, but I now think a better (or at least more conservative) approach is to just use the total lease obligation figure quoted in the notes to the accounts. Here are those figures for The Restaurant Group:

Minimum future lease obligations for The Restaurant Group

I’ve ignored the first table because it shows finance leases rather than operating leases.

With finance leases, virtually all of the risks and rewards of ownership are transferred to the lessee, such as with a hire purchase agreement. Under current accounting rules, these are already recorded on the balance sheet, so we don’t need to capitalise them (i.e. turn them into a capital asset and associated liability on the balance sheet).

The second table relates to operating leases, which are leases for stores, factories and so on. This is what we need to capitalise.

In the Restaurant Group example, it has total lease obligations (net of lease receivables) of £1.1 billion.

That’s a long way north of the £0.5 billion estimate via the six-times multiplier. The reason for the difference is that the six-times multiplier assumes that the average lifetime of a fixed lease is six years.

In The Restaurant Group’s case, it has more than half its leases extending beyond five years, and many of them seem to stretch beyond 10 or even 20 years (it’s hard to tell for sure because the company doesn’t give that much detail; I wonder why).

IFRS 16 will discount those distant future payments (i.e. reduce them by a reasonable interest rate such as the rate on long-term bonds) and so the IFRS 16 figure may end up below £1.1 billion. However, I’m happy to NOT discount the lease obligations figures because a) it’s easier and b) it will make the lease obligations look worse for companies with very long leases, and that’s precisely what I want to do.

Being extra cautious with long leases

I want long leases to look ugly because they lock companies (like The Restaurant Group) into fixed commitments long into an uncertain future.

If that future turns out to be worse than the company expected (perhaps because every Tom, Dick and Harriet opens up a restaurant) then the company is obligated to pay rent for unprofitable sites for the next 10 or 20 years.

Let’s look at this another way. Let’s assume The Restaurant Group decided to borrow £1.1 billion to pay all those rents up front (in reality, landlords would accept a lower amount if paid up front, but I’m going to ignore that detail). If it did that, the company’s borrowings would go from their current £360 million to about £1,460 million.

Is that a sensible level of debt? Let’s have a look:

Over the last ten years, The Restaurant Group produced profits of about £50 million per year if we ignore the losses of 2016 (I’m being very generous). It also made average lease payments of £60 million, which it would no longer have to pay if it paid them all up front with a loan. For the sake of simplicity, I’ll assume that the payments on its new £1.1 billion loan equal its old rent payments.

With profits of £50 million and borrowings of £1,460 million, the company would have a debt-to-average-profit ratio of 13.3.

To put that in context, my upper limit for that ratio for cyclical companies (which includes restaurants) is 4.0, so by that measure, The Restaurant Group is indebted (to landlords) to an insane degree.

Adding lease obligations to the balance sheet also has a seriously negative impact on the company’s profitability metrics, primarily its return on (lease-adjusted) capital employed. With leases accounted for, The Restaurant Group managed to generate returns on lease-adjusted capital employed (ROLACE) of just 5.6% over the last decade, and that’s excluding the losses of 2016.

As an investor, I’m targeting average returns of at least 10% per year. Whether or not I’ll get them I do not know, but what I do know is that it will be harder to achieve if I invest in companies producing returns of less than 6% on their capital.

Now, there are various reasons why this isn’t an entirely fair assessment of The Restaurant Group as it stands today (primarily because of its recent acquisition of Wagamama), but it’s enough to make my point:

Fixed lease obligations are basically the same as debt obligations, so it makes sense to look at a company’s total minimum future lease obligations, either directly or by using a rent multiplier.

If you do that then it becomes clear that companies like Burberry or Next have much higher returns and much lower risk estates than companies like The Restaurant Group or Marks and Spencer.

And if you want to do these calculations yourself, I’ve added lease obligations to my Company Review Spreadsheet which you can access from the Free Resources page.

Author: John Kingham

I cover both the theory and practice of investing in high-quality UK dividend stocks for long-term income and growth.

15 thoughts on “The hidden debt of lease obligations”

  1. Very good article John.
    It is indeed going to be interesting to see what impact the revised figures, once reported and adjusted, will have on share prices — if any.

    Did you run the numbers for our disasterous foray into Ted Baker?

    Regards LR

    1. Hi LR, yes I ran the numbers for Ted Baker (although I wouldn’t call it disastrous just yet!).

      Here are a few points that spring to mind:

      Substantial but not excessive leases

      Ted Baker has fairly high but not exceptionally high lease obligations relative to profits. It’s 10yr average net profits are £32m and average lease obligations are £194m, giving a 10yr lease to profit ratio of 6.

      In comparison, M&S has a ratio of 9 (excluding the terrible results of the last three years).

      Burberry has a 10yr lease/profit ratio of 2.7 and Next has a ratio of 3.5.

      So Ted Baker’s lease obligations are relatively higher than Burberry or Next, but not in the same league as M&S.

      Above average lease flexibility

      The good news is that Ted Bakers lease obligations are relatively flexible. About half of its rent is variable because it’s a heavy user of concessions, e.g. an area within a department store that exclusively sells Ted Baker items and is Ted Baker branded. These rents vary with revenue to an extent, so if revenues go down then so does rent, to a degree.

      Ted Baker’s lease obligations are also relatively short, with most having exit dates in less than five years, giving the company room to move or close stores at fairly short notice.

      This is in stark contrast to M&S where most leases are >25 years.

      Weak profitability

      Once lease obligations are factored in, Ted Baker has generated a return on lease-adjusted capital employed (ROLACE) of about 8% over the last decade, with only one year exceeding 10%.

      10% is my minimum cut-off for ROLACE, unless there is a short-term cause for a lower figure, which doesn’t seem to be the case with Ted Baker.

      The problem here is that it’s impossible to know if Ted Baker’s returns are weak because a) it’s deliberately keeping prices low in order to maximise growth, or b) it doesn’t have pricing power and can’t raise prices high enough to produce a decent return.

      The truth is probably a bit of both, but the recent collapse in profits because of pricing pressure from heavily discounting competitors suggests that option (b) is the major factor. It will be interesting to see what Burberry says about the impact of heavily discounting fashion brands. If Burberry shrugs it off then that’s a clear suggestion that Burberry has pricing power (as a luxury brand the whole point of Burberry’s products is that they’re expensive and exclusive) and Ted Baker doesn’t.

      I haven’t completely changed my mind on Ted Baker yet, but recent events and this consistently weak profitability do undermine its credentials.

      But I’m not going to throw it in the bin just yet. Instead, I’m going to wait to see what happens over the next year or three.

  2. Hi John

    Yes, a very interesting and extremely worrying article.

    I fall into the category of investors that you highlight that had heard of lease obligations but didn’t understand the impact of them. Armed with you article I’ll be interested to revisit my Marston’s shares when they are due to be reviewed and see what impact their inns and pubs leases have on ROCE (or ROLACE as you quite rightly say). It could well be the reason why they’ve never performed as a business or an investment as the accounts and standard metrics would suggest.



    1. Hi Steve, I definitely think ROLACE is a major improvement over ROCE.

      Just had a quick look at Marstons and back of the envelope version:

      Borrowings around £1.5bn
      Equity around £1bn
      Lease obligations around £0.5bn

      So capital employed goes from £2.5bn to £3bn with leases included. Not a huge difference but not trivial either.

      Average profits are around £100m (that’s slightly optimistic) then ROCE is around 4% and ROLACE is around 3.3%. Both of those are pretty terrible, so unless I’ve missed something (quite possible) then this certainly isn’t a company I’d invest in. That doesn’t mean it’s a bad company, just that (perhaps) it has no pricing power.

      Also, its leases are quite long with the vast majority being more than five years to exit, which I would normally call risky but pubs might be a special case, seeing as they’re often in the same place for centuries.

    2. Steve, Phil Oakley just did 3 issues of Investors Chronicle covering P&L, Balance Sheet and Cash Flow statements in the 3 issues in turn. Marston’s is used as the example, so could be interesting for your review.
      I’m of the same view as John, in not wanting to consider Marstons although I did own it years ago. My most recent foray into pubs was Fullers which I sold at a small margin having watched it for 2 years on a very wide bid – offer spread because the shares are largely family held, and Greene King which now has been bought by Li King for asset stripping on the freehold property.
      To be frank I guess that was my last venture into the pub trade. There’s no money in it because they have to keep spending to refresh their facades and unless they are really top notch at food they don’t make enough margin in the beer and spirits sales. It’s very hard to keep a pub popular for a long time – they come and go as far as fashionable ventures are concerned. Wetherspoons does well though.
      They also don’t have much of a level playing field in that there is a wide margin between pub taxes and supermarket prices on drink sales – the government refuse to deal with it that’s why, together with business rates, pubs close at 15-20 a week.


  3. A very good and useful article. I also really like the writing style—very clear and well-explained.

  4. Do you know when IFRS:16 comes in whether the calculation of ROLACE will just be the same as ROCE, that is EBITDA/(Total assets-current liabilities), where total assets now includes right of use assets?

    1. Hi Andrew

      There are various definitions of ROCE but the one I start with is:

      EBIT / (NA + B)

      where EBIT = earnings before interest and tax, NA = net assets (shareholder equity) and B = borrowings

      Actually, I use net ROCE, which uses EAT (earnings after tax) because that tends to look worse for heavily indebted companies because it’s net of debt interest payments.

      Anyway, post-IFRS 16 you’ll still have to include lease liabilities to calculate ROLACE as these won’t be included in the standard ROCE ratio.

      For example, if a company has an IFRS 16 right of use asset of £1m and a lease liability of £1.5m (for technical reasons the asset and liability are not always exactly equal), then the net effect is a reduction in equity (net assets) and therefore capital employed of £0.5m.

      We then need to add the lease liability to capital employed (net assets + borrowings) to lease-adjust capital employed, which will add £1.5m to the existing capital employed figure.

      So the net impact of IFRS 16 in this example to the company’s LACE (lease-adjusted capital employed) is -£0.5m +£1.5m, which is £1m.

      You would get the same result if you calculated LACE as total assets minus current liabilities, because total assets would have increased by the £1m right of use asset.

      Hopefully that makes sense? Basically you still need to lease-adjust capital employed by taking account of the right of use asset and/or (depending on how you calculate capital employed) the lease liability. It’s just easier because those figures are on the balance sheet rather than in the notes to the accounts.

  5. Good article !!

    When you add the lease obligation to the debt (transforming lease into debt) shouldn’t you as well increase the operating earnings (pre-financing) because they no longer have the lease payments included?

    1. Hi Asis, that’s a good point.

      This can all get a bit complicated, so I’ll lay out some background for those who aren’t familiar with the mechanics of return on capital employed (ROCE):

      In the standard ROCE calculation, returns are measured as EBIT (earnings before interest and tax) because the capital employed figure is equity plus borrowings, so ROCE is effectively answering the question “what would this company’s return on equity (ROE) be if we raised additional capital through a rights issue to pay off all debt”.

      If a company did that then its debt interest payments would disappear, so its returns (pre-tax) would equal EBIT.

      With ROLACE we’re answering the question “what would this company’s ROE be if we raised enough equity from shareholders to pay down all debts and pay for all lease agreements in advance”.

      So yes, in that case both interest and annual lease payments would disappear, so returns would be EBITR (earnings before interest, tax and rent). So the text-book version of ROLACE would exclude rent payments as you said.

      The text-book version works like that because it allows investors to compare the performance of a company’s productive assets (fixed capital and working capital) regardless of how those assets are funded. So if a company has a widget factory, the idea is that how that widget factory was funded (via equity, borrowings or a lease agreement) doesn’t affect the underlying performance of that factory.

      However, while differences in the funding mechanism (equity, debt, lease) don’t affect the underlying performance of a company’s productive assets, they do affect the riskiness of the business.

      For example, a company that funds the purchase of a factory with equity has no obligation to pay out a fixed amount each year. Yes, there are dividends, but these are optional. But when debt or leases are used to fund assets, there are obligatory fixed payments to make, often for many years. This fixed expense, combined with the inherent variability of a company’s income, creates financial risk.

      And here’s the point: I want my profitability metric to be risk-adjusted so that riskier returns have a negative impact on the ratio. For example:

      If a company takes on more debt, this will typically INCREASE ROE, making the company look more attractive by that metric. I don’t want more indebted companies to look more attractive, so I don’t use ROE.

      If a company takes on more debt then this has NO IMPACT on the text-book version of ROCE (using EBIT). This is better than ROE, but still doesn’t reflect the increase in risk, which is why I don’t use the standard version of ROCE.

      If a company takes on more lease obligations this will typically IMPROVE ROCE, so this is why I use ROLACE, because it takes account of the increase in risk from lease liabilities.

      My version of ROCE and ROLACE (which I try to remember to call net ROCE or net ROLACE) usies EAT (earnings after tax) which is net of both debt interest and rental payments. By using earnings net of all expenses:

      Net ROLACE will go down as a company uses more debt because EAT is reduced by the interest payments.
      Net ROLACE will go down as a company uses more leases because EAT is reduced by the rental payments.

      Of course, you could just use the standard ROCE or ROLACE calculations which use EBIT or EBITR, and then use a separate ratio to measure liability-related risk such as interest cover, debt/earnings ratios, debt/equity ratios etc.

      I do this as well, comparing debt to 10yr average earnings. But I still don’t want my profitability metric to increase as risk increases. I want it to reflect risk, i.e. be risk-adjusted, and I think an easy way to approximately risk adjust profitability is by using net earnings after all (financial risk-related) expenses.

  6. Hi John,

    How is it you calculate ROLACE in sharepad because there is no lease liabilities data other than when selecting capital employed and ticking lease adjusted?

    Thank you

    1. Hi Liam

      Unfortunately SharePad doesn’t have total lease liability data.

      You can either:

      a) get the data from company annual reports (search for “operating leases” and its in the notes to the accounts at the back) or

      b) use current lease expenses (which SharePad does have) and a multiplier, such as 6x or 7x, to “capitalise” the lease expense into a lease liability.

      Hope that makes sense.

      1. Thanks again John.

        Would I be correct in saying that the data you share in your free Company review sheet is the data you download from SharePad. For instance, you calculate capital employed manually by adding together NAV, Total Borrowings and Lease Liabilities (calculated as lease payments * multiplier)?

      2. Hi Liam, almost. All of the data does come from SharePad except lease liabilities, which I’ve manually extracted from company reports and put into a spreadsheet. I add one new company each day, so fairly soon my entire stock screen will use ROLACE.

        But if you didn’t want to do that then the lease expense * multiplier route is a reasonable (but slightly less accurate) proxy.

Comments are closed.

%d bloggers like this: