Everybody's buying Debenhams

If i'm not mistaken, your analysis results in an equity value of *minus* 240 million pounds.

It strikes me that the equity of a business that has been consistently able to generate free cash flow (to equity holders) in excess ot 100 million pounds per year through the crisis (with potential to go up as spending in the UK picks up), should be valued at less than zero ??

I'd certainly be happy to pay more than 0 for a business that's going to immediately begin returning cash to me. We are talking about actual cash flow available to equity holders, that has been used to distribute dividends, purchase shares *and* reduce the net debt at the same time. It's all real and tangible.

Perhaps your model assumes that the leases and the debt should cost much more than the're currently costing (debt costs just 4%... don't know about leases from the top of my head). And in that scenario interest costs would eat up all the profit. But is this assumption reasonable?

Banks and tenants seem to find resilience in this business are happy to lend at low rates. But is this a distortion caused by quantitative easing & low yields everywhere, and should we expect worse times to come for equity holders?

Javi Dieguez (twitter @javidieguez)

Hi Rich, thanks for your answer

First, the -£240m figure. To achieve the required return on capital of 10%, the business is worth £1,846m to you. Deducting all the components of enterprise value that are not equity (-£1,629m -£413m -£44m ), the equity is worth no more than -£240m. (please let me know if there's some flaw in this reasoning)

BTW I just noticed that the sign of "cash and cash equivalents" in your spreadsheet could be incorrect and the total enterprise value may possibly be £1,377m - £44m + £413m + £1,629m = £3,375m resulting in a slightly lower overvaluation according to your model (83%) and a slightly higher equity value of £1,846m - £1,629m -£413m + £44m = - £152m. Still, this appears as too low an estimation for equity value.

The discussion about the capitalization of rental payments is interesting, and i'd like to make some comments. Some businesses choose to own property, others choose to rent. And it makes sense to try to find metrics to make these businesses comparable. But IMHO capitalising rents may make the companies appear much more indebted than they actually are.

In my mind, paying for space is just an intrinsic cost of the retail business: you cannot get rid of it, like payrolls, or energy (but we never capitalize payrolls or energy!) . Bank debt is an optional cost. It's not intrinstic to the business. it can be repaid down to zero. In other words, renting is more akin to payrolls or energy than it is to bank debt. So we should be careful reclasifying rents.

We still face the problem of comparability. And to solve this, I'd rather apply a transformation to the companies that *own* property, than to the companies that rent it. When a retail business chooses to *own* property, in my opinion they are entering into another business: real state.

The "retail part of the business" still "rents" space from the real state part of the business. The "real state business" gets these rents as income, so they net to zero in the "consolidated" P&L. But what we would need to assess is: what is this property actually worth, and how does it compare to the amount of debt in this side of the business ? (ie, do these "real state" operations add, or deter, from value)?

Just my 0.02c.

red--

that's not the case for operating leases IMHO (as I think is the case for Debenhams). Tenants retain all the risks associated to the asset, and the asset and the lease-related debt are not in the retailers balance sheet.

Tenants will still want the relationship with the company to to be long lived, but so will the electric company and the workforce. I don't think that capitalizing the asset into the balance sheet, with ficticious debt is the best course of action.

This is specially true if we are then going to establish that a 10% return on that asset is required by the "providers" of that asset. Rents don't yield 10%, at least not in any normalized scenario.

I understand that we need comparability, but I think that retailers that own their retail space are embracing some risks not intrinsic to their business. Extra baggage, that I'd rather consider separately.

sorry, I meant landlords!! ;) (I've got to go and learn some more English!)

Thanks for the clarifications. If leases are non cancellable, It makes sense to include them in the balance sheet, and the new accounting obviously will be helpful to unveil this risk. Just some more food for thought:

- 8 times rentals could be too conservative (I've seen some operating leases being rolled over on a 5-year basis, for which assuming an average of 2.5 years until completion could be adequate). Debenhams terms should be investigated to have an accurate representation of the risk.

- requiring a return of 10% on these leases for valuation purposes could also be too conservative as rents usually go cheaper (as stated in my previous comment)

- True, if things went bad and Debenhams had to close *some* stores, it could have to end up paying some non-cancellable operating leases in their entirety, and it's sensible to take this risk into account. But putting them in the same category as bank debt is possibly overkill. The *landlord* ;) will be happy as long as someone is paying the rent. The possibility to resell or sub-let (even losing some money to get the deal done) lowers the risk. As does the fact that the risk is spread among >100 stores (it's not like failing a payment on a big syndicated loan, where creditors could take the business down the next day)

- Valuing a business based mostly on its balance sheet it's a bit like valuing as a liquidation play vs a going concern. If the business is strong (and Debenhams has demonstrated GREAT resilience through the crisis, not only NOT closing any space, but opening more and reducing debt at the same time) I think it's legitimate to give some importance to yearly cash available to the equity in a conservative scenario, than to focus on what could happen if Debenhams was to close all their stores tomorrow.

I'd also like to say that I find this discussion very interesting, and I'm enjoying the comments! :)

Red and Richard

Very interesting remarks and links, thanks!!. I can see some points, but I don't agree with everything ;)

Non-cancellable operating leases are indeed obligations (sometimes for just a few years, sometimes longer term). They should be taken into account as such. They can be a burden if things turn bad, and they should be included on the balance sheet.

But

- While adding them to the balance sheet would be fair and would make investors more aware of downside risks, it would not change economic reality substantially. Banks' perception of risk would not change (i'm sure they are well aware of operating leases) and they would keep charging 4% for their bank debt. Lessors would keep earning circa 4% pretax on their leases (rentals - depreciation charge). Equity holders would still get the same free cash flow available for them. Now, 4% is not too much to charge an overleveraged company for capital, is it? The only explanation I can see is that operating lease risk is lower than that of regular bank debt (be it for the possibility to sub-let the asset to somebody else, or whatever).

- I do think that the last line pointed out by the WSJ article is a valid point. Not every company uses operating leases to mislead investors. Debenhams sold all their freehold property in the 2003-2006 period. In that period, real state was expensive, and rental yields were low. It made more sense to get rid of property and use operating leases, as Baroness Retail private equity did with Debenhams. Of course it was a very profitable operation for equity holders at that time (specially because the property was most likely understated in the balance sheet...it was hidden value that they discovered). And now, of course Debenhams does not keep those proceeds (Templeman, Woodhouse, Lovering et al cashed them as dividend) and equity value is lower should it still keep the property or cash. But going out of property and into operating leases made sense. And if property values in the UK are still high (as they are), it probably still makes sense to prefer operating leases to property just to avoid downside risk in property values.

For example, in Spain, the Picasso Tower (a high quality asset) was sold at 50% of the "market value" it had a few years back (€400m vs €850m). If tenants owned the asset they would have been exposed to that risk. Operating lesees haven't been exposed. And if they have medium-term operating leases, they can even probably renegotiate rents a bit lower.

http://www.elconfidencial.com/economia/2012/05/11/ortega-marco-el-camino...

- My biggest concern is that adding the operating leases to the balance sheet as capital really distorts valuation. You can decompose the 2,375m of capital in two parts: the capital ex-leases, 747m earning 18% after tax, and the capitalized leases, 1629m earning 3% after tax. The weighted return is (18%*747 + 3%*1629) / 2375 = 8% (as in Richard's spreadsheet). The leases are really making total profitability look much lower, but it's fair: lessors will never earn a high yield.

Now, adding the two kinds of capital, and requiring 10% return, IMHO, distorts things, and any valuation resulting from this assumption, is therefore, distorted. THe distortion is not too different to the one that would result from adding receivables as a normal asset and payables as a normal liability (it would boost up capital, and reduce return on capital). Payables are indeed a liability (like operating leases, takes precedence over equity), and receivables are indeed an asset. But payables cost much less than the average cost of capital, and receivables earn much less than the average return on capital. So it makes sense to exclude both from the computations of return on capital (even if they shall still be taken into account for assessing risk, liquidation value, etc)

So, I don't think that 1,846 (that results from this 10% hurdle) is a good estimate of the "value" of the operating assets and I would argue that, even adding the tax shields, 23p falls quite short of the fair value of the common shares.

@javidieguez

Hi Red,
Thanks for your reply

When leverage is too high, trouble inevitably follows for equity holders. But this comes from the fact that the cost of debt usually goes over the roof (8, 9, 10%...) and it's very difficult to earn returns above that through the cycle (ie, some slight downturn, and equity is wiped out). Too much debt destroys value.

But in the Debenhams case (and most likely also other retailers), the (expanded) balance sheet may give an impression if higher risk than the economics suggest. With cost of non-equity capital running at 4%, any excess returns go straight to the shareholders. And 4% does not seem a tough hurdle to beat. Even when the high street is struggling, Debenhams is managing to achieve returns on capital of 8%, and is generating plenty of free cash flow for shareholders.

Yes, I indeed think the 10% is too high a WACC to value Debenhams as a business (ie to get an intrinsic "enterprise value"). This is abusiness that needs mostly space to operate. And this space can be obtained trhough low cost secured financing (either operational leases, of finance leases). Why use standard WACC of 10% ? Using a general valuation framework is fine to start with, but it must be adapted to the characteristics of each business.

Moreover, note that if 0,14 *ke + 0,86 * 4% = 10%, the implied cost of equity (ke) would be approximately 50% :-O I would not say that using a cost of equity of 50% understates risk for equity holders. ;)