What's Woolworths worth? - Part 3

Charlie Munger once said that for all Warren Buffett's talk of discounted cash flow valuations, he'd never actually seen him do one, to which Buffett replied that there are certain things a gentleman only does in private.

Well, gentlemen can evidently afford certain luxuries that aren't available to us. Quite reasonably you expect us to explain how we reach our valuations and sometimes that means baring our spreadsheets.

Our valuation of Woolworths isn't technically a discounted cash flow, but it's similar and we're very aware that it's just as dependent on the assumptions that go into it. That's why we've spent parts one and two of this series providing some background to the most important ones (the growth and margins in the Australian supermarkets business). We're also providing you with our valuation spreadsheet, and we encourage you to take it somewhere private to test your own assumptions.

Woolworths is a good example of why many prefer to keep these things close to their chest. One specific problem is that while we've penciled in 8% operating or EBIT (earnings before interest and tax) margins as our 'bullish' scenario for 2019 – unchanged from 2014 – we'd probably be disappointed if it happened, because it would mean that management was still failing to face up to the challenge from Coles.

At the other end of the spectrum, while a fall in the food and liquor operating margin from 8% to 5.8% would on the face of it be disappointing, it would more than likely be accompanied by improved long-term prospects.

Give and take

In other words, while we've combined a low margin with a low growth rate in our bearish scenario and vice versa for the bullish case – as we need to in order to place some outer limits on our valuation – we think that such combinations are relatively unlikely in practice. In reality there's likely to be some give and take between margin and growth, leading to outcomes closer to the middle, but how the company ends up there will make a difference as to how it's placed beyond 2019.

On the whole, if some medicine needs to be taken (and we think it does), we think it's probably better done sooner rather than later. The valuation problem could be overcome by taking the forecasts further into the future, but even we are not prepared to play with specific longer-term forecasts in public.

It's also worth stressing that our bearish and bullish cases don't reflect the worst and best possible outcomes. It's possible that the margin could fall to 4% or rise to 10%, but the chances are remote and it doesn't help frame a picture of the valuation. It makes more sense to stay realistic but acknowledge the additional risks.

Picking the numbers

The spreadsheet (you can download it here) is split into sections for each of Woolworths' operating businesses. There are yellow boxes in each for you to enter an annual revenue growth rate for 2014-19 and EBIT margins for 2019, for bearish, neutral and bullish scenarios.

We've further refined the inputs for the Australian food and liquor business, because this is the one that matters most, contributing 87% of underlying EBIT in 2014. For this division, you can break down the margin between the gross margin (what's left after paying the direct costs of each product – most of all the cost of buying or making the product) and the cost of doing business (the EBIT margin amounts to the gross margin minus the cost of doing business).

By changing the assumptions you can come up with different EBIT outcomes for each division. At the bottom, you can then enter estimates for central overheads, the net financing costs (for Woolworths' net debt which amounted to $3.4bn at the end of June 2014), tax and the profit deducted for minority or 'non-controlling' interests.

Finally, you can enter assumptions for shares on issue to give earnings per share figures for 2019, payout ratios to give dividends per share figures and price-earnings multiples to give implied future share prices and your total return given a specified buy price.

So how should you go about picking the numbers? Well, the obvious first point to make is that we're looking into the future, so you need to be careful to avoid getting 'anchored' to past performance. But the past can also inform about what's likely to happen in future (indeed it's all we've actually got) so it would be wrong to discount past results entirely.

The other fundamental point is to know when to use help from external sources. We could spend weeks trying to estimate Australia's future population growth rate, but we'd still probably come up with a worse number than the Australian Bureau of Statistics and the time would be better spent trying to understand the threats and opportunities facing Woolworths' businesses. Make sure, though, that you consider whether data providers might suffer from any bias (forecasts provided by an industry group, for example, may be optimistic).

Revenue assumptions

With all that in mind, let's consider our revenue assumptions for Woolworths. The basic factors affecting Woolworths' revenue are how many people there are, how much food and groceries they need or want, how much it all costs and how much of it they get from Woolworths.

'How many people there are' is relatively straightforward: the ABS's central case is for population growth to slip from 1.7% in 2012 to 1.0% in 2045, so for 2014–19 something around 1.6% seems fair and matches predictions made by IBISWorld (see Chart 1).

What about the need and desire for food and groceries, and how much it costs? There are two elements to the first part of this: the needs and the wants. The needs are fairly constant, but the wants – whether you buy the gourmet sausages or the homebrand – will fluctuate with consumer sentiment. On top of this, though, you need to consider how much people are eating out: reduced sentiment means fewer meals in restaurants and more demand for supermarkets, so to some extent the economic sentiment balances out.

So as a rough guide, at least over the short term, you'd expect people to continue to spend the same proportion of average wages on groceries, and that's expected to rise by around 3% a year (see Chart 2) – but over longer time scales you need to consider that we're getting more efficient at producing food, leaving more room for other areas of expenditure.

They placed a greater emphasis on the daily feed than their entertainment budget in the Stone Age, for example, and the trend still holds for the 1980s. According to the ABS, from 1986 to 2006 household spending on 'communications services' rose 8% a year and spending on 'goods for recreation and culture' rose by 7% a year, while spending on food only rose by 0.4% a year. All of which is to say that over time we'd expect spending on food and groceries to lag economic growth and the increase in average wages.

As you can see from Chart 3, 'Food and non-alcoholic beverages' inflation has averaged a little over 2% over the past few years (according to the ABS), although it's been over 10%, following the floods in 2011, and as low as minus 3%. Woolworths' average shelf prices before the cost of promotional activity have followed the trend, though in muted fashion, as they have after the effects of increasing promotional activity, although in this case they've tended to be several percentage points lower. Since 2012, food price inflation has recovered back to around the average and we'd expect that to continue, but we'd also expect Woolworths' price growth to remain some way below this, after promotional activity, partly funded by gross margin reductions (as we'll come to shortly).

Falling market share

As you can see, from the relative certainty of population growth, the waters are getting murkier and it gets worse still with 'how much of it they get from Woolworths', otherwise known as market share.

This will be directly affected by how much Woolworths is prepared to compete on price, although there will be a lagged effect. It takes time for perceptions to change around 'price leadership' and the longer Woolworths lets Coles get ahead on this, the harder it will be to turn it around.

Even with heavy promotional activity, we'd expect Woolworths to continue to lose market share at least for the next few years, which is to say that its comparable store sales growth will continue to lag Coles (see Chart 4) until perhaps recovering towards the end of the five-year period if it takes the necessary action.

With population growth of 1.5% and falling market share, it's hard to see how Woolworths will meet its internal targets for a 3% annual increase in selling space without gross margin reductions – and that, in turn, will keep a lid on price inflation and sales per square metre, which has averaged growth of just 2% a year over the past 15 years and has barely budged for four (see Chart 5). This is the practical expression of food price deflation and we think it's unlikely to change.

As you'd expect, growth in total supermarket sales approximate to growth in selling area (square metres) plus growth in sales per square metre (see Chart 5). As a result, our growth assumptions centre on the 3% target for growth in selling area, with 2% in the bearish case and 4% in the bullish, but as already mentioned we think it'll be hard to meet these assumptions without reducing the gross margin.

For all this hard work, then, we come up with a revenue growth range of 2–4% a year, which again emphasises Woolworths' defensive nature.

Margin pressure

The bigger impact will be felt in the margin assumptions. This is where it gets really murky, because it's impossible to say how much you need to reduce prices in order to change perceptions and get people to come into your store. You can fast-track the process by shouting out price-matching guarantees but, as they've found in the UK, this can be a dangerous approach. And of course, it'll only do you any good in the long run if quality is maintained while making any price cuts.

It can make things easier to break the EBIT margin down into the gross margin and cost of doing business (CODB), because the latter is harder to shift for a company that's already operating efficiently. Technology may provide some gains in the supply chain through the use of things like radio frequency pallet tagging, and there may be further gains to be had from things like self-service checkouts. In our neutral case, though, we've stuck with the current CODB, and edged 0.1% higher in our bear case and 0.1% lower in the bull case.

That takes us to the gross margin, where the real battle will be fought. We outlined in part two of this series why we think Woolworths will need to be more proactive on pricing, but part of this can be achieved through increased private label offerings and/or lower supplier costs (as explained in part one).

In our bullish scenario this is enough and we therefore assume no change in the gross margin, but in the neutral case we take 1% off, and in the bear case we deduct 2%. These approximate to fully self-funded price cuts of a similar magnitude and, combined with private label and reduced supply costs, they would give Woolworths considerable ammunition for a gentle price skirmish – in the bearish scenario particularly. This in turn would almost certainly lead to improved growth rates although, as we've noted, they may take a little while to appear.

The resulting EBIT margin is 5.8% in the bear case, which is still higher than Coles (which managed 5.3% in 2014), much higher than the 4% average for the world's largest 30 food retailers (according to CLSA) and much much higher than the 'no more than' 2.3% guidance for the year to February 2015 given yesterday by Tesco.

Other assumptions

We go through a similar process for each of the other operating divisions, but we won't spell out the detail here, because the valuation is overwhelmingly influenced by the assumptions for Australian Food and Liquor and we don't want to confuse matters. Further thoughts on Masters can be found in Woolworths still learning with Masters from 13 Aug 14 (Buy – $35.74).

Notable features are that we assume slightly better growth for the NZ business, some fairly disastrous results for Big W and a break-even result for Masters in 2019 even in our bullish scenario.

In the bear case we've assumed that Lowe's has finally exercised its put option on the Masters joint venture – adding about $50m to the annual interest bill for the cost of buying it out and increasing minorities because Lowe's is no longer shouldering some of the losses.

The minority interests represent the profit share on the 25% of ALH that Woolworths doesn't own (assumed to be $50m), offset by Lowe's share of losses from Masters (which is assumed to be nothing in the bull case due to it breaking even, nothing in the bear case due to it being bought out and a $25 post-tax loss in the neutral case).

In addition to the possible increase in debt to buy out Masters, we've assumed net debt rises by up to about $1bn from the current $3.5bn (adding around $50m to finance costs), reflecting the fact that Woolworths' free cash flow has been lagging its targeted dividend payout ratio of 70% (see Chart 6).

Total returns

This leaves the share price, which you can enter in the green highlighted cell. At the current price the stock is on a price-earnings ratio (PER) of about 15 reflecting the fact that we're not the only ones who think the 8% food and liquor margin is at risk. By putting in PERs for 2019, the spreadsheet will give you an approximate total return for the 2014–19 period (approximate because, to keep things simple, it doesn't discount the dividend payments individually according to when they are received). We've opted for 2019 PERs of 14, 17 and 20 for the three scenarios, which would mean total annual returns ranging from minus 4% to plus 13% at the current price, and minus 2% to plus 16% at our proposed new Buy price of $27.

As regards Australia's other listed food retailers, we'd continue to Avoid Metcash for the reasons given in Metcash model is fundamentally flawed from 21 Mar 14 (Avoid – $2.85). Wesfarmers is more interesting, but the situation is confused by its higher price (a PER of 19) and other businesses, some of which aren't going so well. We're planning to have a good look at it in the new year, but for the time being it remains a Hold.

And that's going to be Woolworths' new recommendation as well. When we upgraded the stock a couple of years ago in Put some Woolies in your basket on 23 Oct 12 (Long Term Buy – $29.20), we were anticipating growth in the mid-single digits with a reasonable degree of confidence. Not only does that outcome now look optimistic, but the risks have also increased, particularly in light of the worsening sales performance over the past two quarters.

We've been slow to adapt our thinking, but we're not going to compound that error by clinging to our past valuation when the facts appear to have changed. Given the weaker outlook and some concerns over management's response we're reducing our Buy price from $36 to $27 and our Sell price from $50 to $40. We're also reducing our recommended maximum portfolio weighting from 10% to 8%. HOLD.

Note: Our model Growth and Income portfolios own shares in Woolworths.