Like-for-like. Or not?

Bizarrely, like-for-like (LFL) sales - one of the most talked-about measures across the pub and restaurant industries - is not the subject of an industry standard. It is never, or extremely rarely, explained in the various companies' reports and accounts.

Why look at LFL sales growth?

When looking at LFL sales, or pretty much any other measure for that matter, it is as well to have some idea of what one is hoping to achieve by measuring them.

Is the measure being targeted for internal reasons, such as to motivate, incentivise, measure and reward staff, or are there external factors at work such as the desire to inform the markets as to progress and/or brag about performance?

Of course, both measures may be important but, while the latter is effectively information transmission, the former is more subtle in that any reward scheme should, at least in part anyway, target factors that could be affected by the staff being judged.

As sales growth can be impacted by external socio-economic factors, legislative changes and head office decisions, LFL sales growth is often something that staff have to live with rather than impact on directly, other than through their enthusiasm in serving customers who are already in the building.

Are they useful?

Although LFL targets may incentivise companies to do things that are not necessarily in the interests of the shareholders, we would suggest that they're not useless, so the answer is likely to be a guarded 'yes'. We would consider many, if not most, measures useful as they concentrate the mind - no small feat.

But in the case of LFLs, they do validate market trends, and if used consistently can, at least within the company in question, indicate to a company and to the wider market whether trading is improving or deteriorating.

They can highlight trends. For example, Mitchells & Butlers (M&B) has previously referred to a softening in its Vintage Inns operation. Covers and spend are clearly important here, but so are LFL sales - although they do not answer the 'why?', only the 'what?'

For an operator of a number of formats, such as M&B, which trades from Project S down to its Sizzling Pubs and Ember Inns, LFL sales highlight consumer shifts and can influence policy. That has to be, or at least should be, useful. What do LFLs measure - and what don't they?

LFL sales growth is a bald measure. They give you sales from a given footprint, but the things that they don't give you are legion. They don't measure return on capital, sales-mix changes, margin changes, the impact of competitors, external factors or the economy.

Nor do they differentiate between sales whipped and beaten out of a workforce for a short period - typically the few weeks, 'current trading', referred to after the end of a reporting period - and what represents true, sustainable growth.

While we've not had time to collate the information, it would be interesting to compare 'period LFLs' with 'current trading LFLs'. We would suggest that the betting would and should be on the latter. And nor do LFLs comment - or at least they shouldn't - on the weather, sporting events, the blend between new units and mature units, the quality of sales or their sustainability.

How are LFLs measured - and what are the tricks?

This sounds easy, but a definition is elusive. Consistency over time within the same company may be more important than absolute consistency across an industry, or particularly between industries. But there are a large number of potential issues that need to be ruled upon.

For example, which units should be included in the survey? Companies would clearly like to include the good and leave out the bad. More often than is desirable, this is exactly what happens. A few of the questions that need to be imposed include:

Should units be owned for both 12-month periods? Well, yes, preferably, but what about dire units that you shut for a day, can they be left out? And what about non-core units? If they're bad - and they're rarely good - can they be excluded too?

How should one treat units held for disposal? Unless the crown jewels are being sold (and here the company in question could be facing real problems), these are rarely the best units and it is therefore helpful that they are excluded. After all, good units to be sold may not be 'deemed for disposal' until after the period end.

What about managed-to-tenancy shifts? Again, this depends upon the policy for disposals and/or acquisitions.

And what about invested units? Well, define investment. Is it a lick of paint, eg, a judicious sparkle, or is it an extension/virtual rebuild? Invested units may well be left out, commented upon separately or left in, so there's plenty of variety in treatment. If they're left in then one could argue that closed units should be left in too.

How should promotions be treated? Should honeymoon units be excluded (or at least referred to)? What about trying to exclude the World Cup, the poor summer, or foot-and-mouth? Why are only bad things usually excluded? Does the decision to write off capex as repairs through the P&L flatter 'uninvested' LFLs? What about costly promos such as DJs, shows, etc?

Should price inflation be stripped out? Well, no, but what about deflation? More questions than answers, I'd agree, but one issue that would be relatively easy to clarify would be "what proportion of your units, or better still sales, are included within the LFL figures that you've just given?" If bad units are ruthlessly excluded, this may be a small percentage. It is possible for LFLs to be positive, but for group sales to decline.

Some busted companies never reported LFL sales declines. It was what they left out that sank them.

A like-for-like conclusion

LFL sales growth is a useful measure. However, if manipulated, poorly explained and/or inconsistently used, it can be dangerous. The following, however, could be useful:

Sales growth could be split into its two main constituents: price and volume. The one may be better than the other.

Price rises could be accidental (for example, inflation) and are arguably finite, while volume growth may be accommodated by putting on more capacity.

Sustainable return on capital is critical. Did the company in question make and will it continue to make, comfortably in excess of your cost of capital on your investment? A CROCE/WACC (cash return on capital employed /weighted average cost of capital) calculation will then suggest whether or not the company's market capitalisation reflects its earnings abilities.

This sounds more complicated than it really is. Having spoken to a number of people on this subject, one concludes that LFLs are useful as a measure but should not be relied upon to the exclusion of all else.

What should be included in the figures remains a subject of hot debate.

Mark Brumby is an analyst with Blue Oar Securities​