A number of methods have been proposed to define the value posted in the balance sheet when a brand is part of the assets of an acquired company, or any other instance when this valuation is needed. They can be positioned on a two-dimensional mapping. The horizontal axis refers to time (but do we base the analysis on the past, the present or the future?).
This axis discriminates between valuations based on historical costs (those that helped build the brand), valuations based on present earnings, on market price, and those which rely on a business plan: that is to say, a forecast. The vertical axis is a real/virtual dimension. Some analysts rely on hard facts (historical accounts are facts, as well as present earnings). However, some methods rely more on estimates about the present (the replacement method), or about the future (the discounted cashflow method). We now analyse these methods in turn.
Positioning brand valuation methods
Valuation by historical costs
The brand is an asset whose value comes from investments over a period of time (even though accountants do not strictly regard this as a true form of investment). The logical approach would therefore be to add together all the costs associated with a particular period: development costs, marketing costs, advertising and communication costs, etc. These costs can be determined objectively, and will have been in past income statements.
As we can see, this approach allows us to overcome the tricky problem of separability, by isolating the direct costs associated with the brand and also by attributing to it the indirect costs such as the sales force and general expenses. Even though this method is simple and logical, it nevertheless raises the following practical difficulties, which reintroduce a certain subjectivity:Over what period should costs be accounted for? Numerous brands are very old as we have seen: Coca-Cola dates back to 1887, Danone to 1919, Lacoste to 1933, Yves Saint Laurent to 1958, Dim to 1965. Should we include costs right from their beginnings? Everyone knows of old brands that no longer exist. Companies must go back in time and ask themselves if past advertising still has an effect today.Which costs should be taken into account? Investment in advertising has a dual marketing role: one part generates extra sales, which can be measured immediately, while the other part builds brand awareness and image which facilitates future sales. The practical difficulty is in estimating year by year the weight that should be attributed to each part. Also, how far ahead are we looking when talking about future sales? On top of this we have to look at the advertising wear-out curves over a given time period. If, as has been shown in studies on the persistence of attitude changes, such effects decrease in a linear manner over, for example, five years, it may be that expenses arising over this period, including only 20 per cent of those for year n –5, can be posted.It is not simply a question of adding up the costs, you also have to take into account an appropriate discount rate which has to be calculated. On top of the subjective nature of the answers to the above questions, valuation by costs causes several basic problems which are linked directly to a partial understanding of the brand:When creating a brand, a large part of the long-term investment does not involve a cash outlay, and therefore cannot be posted to the accounts. These include stringent quality controls, accumulated know-how, specific expertise, involvement of personnel, etc. All of these are essential for encouraging repurchase, for the brand’s long-term reputation and for word-ofmouth. There would be no trace in the accounts of brands like Rolls-Royce because there were no advertisements for it.One of the major strategies to create a strong brand consists of choosing a competitive launch price, which may be the same as that of competitors’ even though the product is upgraded. Swatch is an ideal example of this. They could have opted for a slight price differential, or a price premium, to cover the costs of innovation and of upgrading the product. They decided, however, to set an aggressive price that was equal to that of their competitors, thus maximising the brand’s price/quality ratio and enhancing its attractiveness. This is one of its key success factors. Unfortunately, this non-cash investment would not appear in a system where only cash expenditures are registered.The method therefore favours brands whose value only comes from advertising and marketing and which have a significant price premium. It would not apply to brands such as Rolls-Royce or St Michael (Marks & Spencer’s brand) which advertise very little. It could also be said that past expenditure is not a guarantee of present value. There are several brands that are heavily advertised but of little value and are coming to the end of their life.This method is favourable to recent brands and a fortiori to internal brands that are in the process of being created, as we have already seen.
Valuation by replacement costs
To overcome the difficulties arising from the historical costs approach, it might be better to place oneself in the present and to confront the problem by resorting to the classic alternative – as we cannot buy this brand, how much would it cost to recreate it? By taking its various characteristics into account (awareness, percentage of trial purchases and repurchases, absolute and relative market share, distribution network, image, leadership, quality of the legal deposition and presence in how many countries), how much would we have to spend, and over what period, in order to create an equivalent brand?
Is it possible to remake Coca-Cola, Schweppes, Mars, Buitoni or Martell? Probably not. How about Benetton, Bang & Olufsen, Saab or Epson? More than likely. For a certain number of brands, the question no longer arises since it is impossible to recreate them. The context has changed too much:They were created in an era when advertising expenditure was negligible and the brand was nurtured over time by word-ofmouth. Today, it costs so much for a 1 percent share of voice that it has become impossible to create a leading brand through unaided awareness. In any case unaided awareness is a restricted area and to gain access a competing brand must leave. This is because of memory blocks. There is no reason why today’s well-known brands should allow themselves to be thrown out.It is difficult to imitate the performance level of brand leaders. Backed by research and development and an intangible but very real know-how, they enjoy a longlasting competitive advantage and a resulting image of stability. Any challenger is taking a risk. Unless they have access to the necessary technology, their chances of encouraging repurchasing and loyalty are virtually zero.Major retailers have now become exacting gatekeepers. They give pride of place to their own brands, only selling one or two national brands that tomorrow will be international.Finally, considering the high failure rate of new product launches, it is easy to understand the uncertainty of the return on the large amount of money that has to be invested in the long term. If you are going to pay a lot you might as well buy certainty. Hence, the clutter of takeover bids, raids, mergers and acquisitions of firms with strong brands that are already market leaders.
On the other hand, when these factors which hinder market entry are no longer present, the market is more accessible. The possibility of creating tomorrow’s brand leaders from scratch ceases to be theoretical, even though uncertainty and the necessary time element may still exist. Therefore, future Benettons will probably be created. Franchising allows wider market penetration without admitting defeat at the hands of major retailers.
What is more, the fashion industry is open to new ideas. In this domain, style is more important than technology. Computer services and the high-tech world in general are also open to innovation. Generally speaking, the future will see the emergence of new international brands, each positioned in its own particular niche. They will thus no longer seek global awareness but will aspire to be leaders in particular market segments.
Brand valuation by replacement costs nevertheless remains very subjective. It requires the combined opinions of experts and ambiguous procedures. On top of this it should be remembered that the aim of the valuation process is not, in itself, to arrive at a value but to get an idea of the economic value of the asset in question – in this case the brand. Cost methods focus on the inputs, whereas the economic value is based on the outputs – what the brand produces and not what it consumes. Profit is not generated through investments but through market domination and leadership.
Valuation by market price
When valuing a brand why not start with thevalue of similar brands on the market? This is how property or second-hand cars are valued. Each apartment or car is inspected and given a price that is above, equal to or below the average market price of similar goods.
Even though this method is very appealing, it raises two major problems when applied to brands. First, the market doesn’t exist. Although such transactions are often cited in the financial pages, acquisitions and brand sales are relatively few. Brands are not bought to be sold again. In spite of this, we can get an idea of the multiples applicable to each sector of activity (from 25 to 30) thanks to the number of transactions that have taken place since 1983. Thus, such an approach could tempt some wishing to value a brand.
However, there is a major difference between the real estate market and the market for brands, which is relatively small. On the real estate market the buyer is a price-taker, that is, the price is fixed by the market. Irrespective of the use that he or she will make of the property, the price remains the same. For brands, the buyer is a price-setter, that is, he or she sets the price of the brand. Each buyer bases his/her valuation on his/her own views, on potential synergies and on his/her future strategy.
Why did Unilever pay s100 million for Boursin, the well-known brand of cheese? It can be explained by the pressing need of this group to acquire shelf space in major supermarkets in which it had previously been absent. Having at its disposal a compulsory brand, they saw a way of opening the door to other speciality products. In April 1990 Jean-Louis Sherrer was bought for three times less than the price that Mr Chevalier paid for Balmain two months earlier. For Mr Chevalier, Balmain was a means of entry – or rather re-entry – into the luxury market. Hermès, which was already present on this market, didn’t need to pay this price (Melin, 1990).
In abstract terms the purchase price is not the price paid for the brand but is the interaction between brand and purchaser. To use the price paid for a similar brand as a reference, without knowing the specific reasons behind that brand’s purchase, ignores the fact that an essential part of the price probably included the synergies and the specific objectives of the buyer in question. Each buyer has his/her own intentions and ideas. The value cannot be determined by proxy.
This is what distinguishes fundamentally the market for brands from that for real estate, or for example for advertising agencies. In the case of the latter, norms and standards exist that are not dependent on the buyers’ intentions (50 to 70 per cent of the gross margin on top of the net assets). Despite this, valuations in the luxury market frequently take into account recent transactions and use a multiple of the sales (1.5 for Yves Saint Laurent, 2 for Lanvin and for Balmain, 2.9 for Martell, 2 for Bénédictine).
Considering the difficulties which are inherent in the cost-based methods or in the referential methods on a hypothetical market, prospective buyers tend rather to look at the expected profits from brand ownership. Since the third type of approach relies on two major philosophies, we are devoting a special section to it.
Valuation by royalties
What annual royalties could the company hope to receive if it licensed the rights to use the brand? The answer to this question would form a means of directly measuring the brand’s financial contribution and would also solve the problem of separability. The figure obtained could subsequently be used to calculate the discounted cashflows over several years. The difficulty is that this is not a very common practice in most markets. They are found in the luxury and textile markets.
From a conceptual point of view, it is not certain that this method properly separates just the value of the brand (Barwise, 1989). In fact, companies often use licences to reach countries where their brand is not present. However, the royalty fee does not include solely the use of the brand. The brand owner also undertakes to supply a package of basic materials, know-how and services, which allow the licensee to maintain the brand’s appropriate quality level.
Valuation by future earnings
Since the brand aspires to become an asset, it is best to begin by a reminder of what an asset is. It is an element which will generate future profits with reasonable certainty. Valuation methods have been developed on the basis of expected returns of brand ownership. Naturally, these tie in fully with the purchaser’s intentions. If he/she wishes to internationalise the brand, it will be of more value to him/her than to a buyer wishing to keep it as a local brand.
The value measured by expected profits cannot be separated from the characteristics of the future buyer and from his/her strategies for the brand. This explains why the stock market value compared to a predator’s value of a branded company will always be structurally lower. The former valuation is related to the existing business, taking into account current facts and figures provided by the firm. The latter comes from the overvaluation created by the prospect of synergies, complementary marketing processes and the attainment of strategic market positions.
The process of valuing the expected profits of the brand can be divided into three independent stages (see Figure below):The first step involves separating and isolating the net income associated with the brand (and not with the company for example).The second step is to estimate the future cashflows. This requires a strategic analysis of the brand in its market or markets.The third step involves choosing, by using a classic financial method, a discount rate and period.
A multi-step approach to brand valuation
This is the classic method of valuing all investments, whether tangible or intangible. The analyst calculates the anticipated annual income attributable to the brand over a 5- or 10- year period. The discount rate used is the weighted average cost of capital, which if necessary is increased to take account of the risks arising from a weak brand (that is to reduce the weight of future revenues in the calculation of the present value). Beyond this period, the residual value is calculated by assuming that the income is constant or growing at a constant rate for infinity (Nussenbaum, 1990). The following formula is used:
This is the classic model for valuation by the discounted cashflow method, even though analysts offer numerous variations of it (Mauguère, 1990; Melin, 1990). This method was used to value Cognac Hennessy at 6.9 billion francs, based on a capitalisation of its net revenue over 25 years at a rate of 6.5 per cent (Blanc and Hoffstetter, 1990).
This method was also used to value the Candia milk brand as part of a restructuring programme. The final figure, which was around 1.8 billion francs, was the result of a business plan within which two questions were discussed:Knowing that milk is a commodity, what percentage of Candia’s future sales will be generated by products which are heavily marketed, differentiated and have a strong identity which justifies a price premium?At how much do we estimate the price premium that Candia can demand over more ordinary products? In such markets, even a tiny difference may amount to huge profits.
Sceptics of this method (Murphy, 1990; Ward, 1989) object to its three sources of uncertainty: the anticipation of cashflows, the choice of period and the discount rate:By definition any forecast is uncertain. This does not apply only to brands, but to any investment evaluation – tangible or intangible – which is calculated by the above method. For brands, cashflow forecasts could be ruined if a competitor launched a superior product which was not accounted for in the calculations. This argument overlooks the fact that these forecasts were made after an in-depth analysis of the brand’s strengths and weaknesses (on the basis of the criteria presented earlier). It can be assumed that these were included when the anticipated cashflows were calculated. In any case the discounting rate takes into account the anticipated risk factor.A second criticism lies in the subjective nature of the choice of a discounting rate. However, on the one hand analysts test the sensitivity of their findings against variations in this rate, and on the other hand, this rate is fixed by taking into account stable company data, such as its average cost of capital. The only subjective factors are the risk premium and the future rate of inflation. Furthermore, very often the risk is zero from the purchaser’s point of view as he or she feels that success is a certainty.Finally, there are those who criticise the choice of period for calculating cashflows. Why 10 years and not 15? What is the value of forecasts made so far ahead? On the one hand, the brand may disappear after only a few years and on the other, in certain volatile sectors three years is already a long time (eg laptop computers). This is where certain valuations come from: brand value should be based on that which is certain, ie the net income of the brand at the moment. This is the basis of the multiple method. Brand value is calculated by applying a multiple to the current profits of the brand, measured over three years (t–1, t+1). This approach does not need internal data.
Valuation by present earnings
Who can predict the future? How can one be sure that the forecasts of a business plan will be matched? In fact, one of the reasons so many internet brands have been heavily overvalued is that they made no profit whatsoever (eBay excepted).
The brand valuation process relied exclusively on forecasts and business plans which were created just to attract new investors, so the founders could resell before the collapse of the illusion. Interbrand, a major brand valuation company, has promoted a specific approach to circumvent this problem. No business, no brand.
Interbrand valuations rely exclusively on three years: last year, this year and next year. After partitioning each year’s revenue to pay for the invested capital which made the business possible and other direct intangible assets, one is left with a global residue, made of a weighted average of the residues of each of these three years.
This residue should be then multiplied by a figure called ‘the multiple’, hence the name of the Interbrand proprietary method: the multiple method. Although Interbrand seems to have moved now to the most orthodox method (discounted cashflow), we analyse this former approach on which many brand valuations have been based.
In the financial valuation of companies, it is typical to examine what is known as the price/ earnings ratio (P/E). This ratio links the market capitalisation of a firm to its net profits. A high ratio is a signal of high investor confidence and optimism in the growth of future profits. Even though the brand is not the company, the same reasoning can be applied:
The only difference lies in the fact that for a brand there are no data on its market capitalization because it doesn’t exist, therefore it is this that we are trying to calculate. This notional market value of equity is the price to be paid for the brand (before the effect of overbidding). In order to calculate this, it is necessary to determine M, the multiple which is equivalent to the P/E ratio specific to the brand.
There are four stages to this method:Calculating the applicable net profit.
Interbrand used the profits for the last three years (t–2, t–1, t), thus avoiding a possibly atypical evaluation based upon a single year. These profits were discounted to take account of inflation. A weighted average of these three figures was calculated in accordance with what we consider to be the most and least important years. This weighted average after-tax net profit which is attributable to the brand forms the basis of all calculations.Assessing the brand’s strength.
This method uses a set of marketing and strategic criteria to give the brand an overall mark. Interbrand uses only seven of these factors and takes a weighted sum of the individual marks for each factor in order to calculate the overall mark, as can be seen in Table below.
A method of valuing brand strength
Estimating the multiple.
A relationship necessarily exists between the multiple (an indicator of confidence about the future) and this score for brand strength. If this relationship was known precisely, the multiple would then be predicted by the brand strength score. For this, Interbrand developed a model known as the ‘S-curve’ which plots the multiple against brand strength.
The model is based on Interbrand’s examination of the multiples involved in numerous brand negotiations over recent periods – in sectors close to the one being studied. The P/E of the companies with the closest comparable brands are used. Interbrand then reconstructed the company’s profile and brand strength. Plotting the multiples (P/E) against the reconstructed scores results in an S-shaped curve (see Figure below).
The Interbrand S-curve – relation between brand strength and multipleCalculating brand value. This is calculated by multiplying the applicable net brand profit by the relevant multiple. We can illustrate this method by an actual case. In 1988 Reckitt & Colman valued its brands in this way. They valued household and hygienic goods where they were market leaders, as well as food products (condiments) where they were also a leader, and finally pharmaceutical goods where they had an average position. The specific situation enjoyed by those brands in the first group is as follows:world leadership;growing markets, with few new entrants except for distributors’ own-brands;unaided brand awareness (eg Airwick) high in the UK and in Anglo-Saxon countries but less so in France;customers’ brand loyalty;strong brand image and assurance of quality;for each of its brands, little possibility for diversification.
Reckitt & Colman estimated that 5 per cent of profits on these brands came from sales under distributors’ own-brands. Interbrand considered that the remaining 95 per cent was the brand’s gross profit. The income generated by the brand can be calculated by subtracting the expected return on investment from net assets. The net revenue was weighted according to the importance of each brand and discounted for the previous three years. The following results were obtained for each category:household and hygienic products: £53.8 million;food products: £24.7 million;pharmaceutical goods: £17.1 million.
What multiple should be applied? For the first group, the multiple used by Reckitt & Colman in 1985 when buying Airwick was applied. A multiple of 17 was used for food products and was based on recent transactions in the sector during the last few years, for example the BSN–Nabisco takeover bid. Finally, a multiple of 20 was used for the pharmaceutical group. In fact, recent transactions in the pharmaceutical industry had been using multiples which were closer to 30. A lower multiple was chosen in this case because of Reckitt & Colman’s relatively weak position in the sector. By applying these figures to the net revenue in each category, the following brand values were estimated:household and hygienic products: 53.8 × 20 = £1,076 million;food products: 24.7 × 17 = £420 million;pharmaceutical goods: 17.1 × 20 = £342 million.
Comparison of the cash flow and multiple method
The multiple method, which was developed in the UK, is becoming a classic. It was, in fact, used by such companies as Rank Hovis McDougall and Grand Metropolitan whose decisions to post brand values to their balance sheets caused a controversy which is still not settled. It is also the method which communicates the most through books, articles and seminars. The simplicity of the method used is such that it is uncharacteristic of the stringent world of financial analysis. All this said, is it valid?
First, the multiple method is not all that different from the classic method of discounted cashflow. It is a particular example of it. When a constant and infinite annual cashflow is expected, the present value of the brand is defined thus:
As we can see, the multiple is none other than the inverse of the cost of capital adjusted for risk (1/r). If a constant growth rate (g) of annual income is expected, the multiple is:
Equations aside, the point to remember is that we cannot reproach the method of discounted cashflows for making certain hypotheses, since the multiple approach is itself a particular hypothesis, which is equally as questionable but not explicit. It draws its apparent validity from the fact that all its calculations are based upon:net known profits attributable to the brand over the previous three years;marketing data and the subjective opinions of managers regarding brand strength;multiples based on recent transactions by similar companies;an S-curve, using information from a database to plot these multiples (or P/E ratios) against brand strength scores.
However, face validity (or appearance) does not mean validity per se. In its present form, Interbrand’s method poses various problems:Market multiples, which were used as parameters for the S-curve, are not valid indicators of the strength of the brands even though they were the mainstay of these transactions. In fact the final transaction price includes both the estimated value of the brand and a certain amount which is due to overbidding. For example, in the fight between Jacob Suchard and Nestlé, the initial bid was 630 pence and the final bid was, 1,075 pence! Market prices include the effect of this overbidding and thus overvalue the brand. It is therefore rather curious that we are trying to link market multiples to a value for brand strength as this value ignores the effect of overbidding. For this reason a certain doubt arises about the applicability of this method to value and post to the balance sheet unacquired, internally created brands. The value attributed to the asset will be greater than the value of the brand as it will include an unspecified amount Which is a result of overbidding! The fact that companies may nevertheless have used this method to represent their brands as assets in no way validates this approach.Even in a market where there is no overbidding, the stated multiple measures the value of the brand from the point of view of the potential buyer. It expresses his vision, his strategies and any synergies that he may expect. The fact that in 1985 BSN did not buy Buitoni despite it being reasonably priced does not mean that Buitoni was worth less but means that it was worth less in the eyes of BSN. In 1988 Nestlé valued it at several billion Swiss francs. It again seems strange to try to relate market multiples, which are closely linked to the buyer, to the scores for brand strength, which are calculated by an outsider and do not include the synergistic benefits. This poses a problem when internally created brands are posted to the balance sheet. They are valued in the context of a ‘going concern’ according to their current benefit to the companies who own them. On the other hand, multiples supplied by the market are calculated with the idea of using them for a totally different reason.For the moment, no illustrations of the Scurve showing the variance around the curve have been published. This variance is a measure of the quality of the empirical relationship between the two variables. As it is, the curve would have us believe that there is zero variance, which is impossible. A single brand strength score probably corresponds to several multiples or at least to a range of values (within which the S-curve is found). Such uncertainty causes problems as in reality the financial value of a brand is very sensitive to even a slight change in the multiple. Going back to the Reckitt & Colman’s household and hygiene brands, we see that a one point variation in the multiple results in either a £53.8 million increase or decrease in the value of the brand. This is a far cry from the principles of prudence, reliability and rational certainty which govern accounting practice and information.The very validity of the S-curve is questionable. Interbrand uses the following argument: a new brand grows slowly during its early stages. Then, once it moves from being a national brand to being an international one, its growth is exponential. Finally, as it moves from the international to the worldwide arena, its growth slows once more. For example, the difference between Buitoni’s purchase and resale price signalled the transition of a national brand to a European wide one. Experience shows that brands are susceptible to large threshold effects. Their strength with customers and retailers is developed in stages. Thus, today, a moderately known brand may be worth virtually the same as a little known one. However, beyond a certain threshold, it grows in value. Research on brand awareness has shown that, in markets with intensive communication, it is only once a brand has reached a certain level of aided awareness that its unaided awareness will start to increase. This is due to a memory block. Likewise, major retailers are replacing middle-of-the-range brands with their own products. These brands rely more on supply than on demand and they would cease to be sold if the retailers replaced them with their own brands. Thus their future is very unstable. This would lead us to believe that the relationship between brand strength and the multiple – provided that both are assessed by the same potential buyer – is better illustrated by a stepped graph (See Figure below).
Stepped graph showing relationship between brand strength and multiple
In conclusion, the widespread use of the multiple method is not proof of validity, as we have just seen, but testifies to its simplicity and handiness for non-specialists, and therefore its internal educational value. A small variation in the chosen multiple leads to important differences in the value of the brand. The present method of choosing the multiple is unsatisfactory from the point of view of reference multiples and of the brand strength scores.
What can we make of a total score which is obtained after subjective weightings of factors which are sometimes redundant or in any case correlated? This wish for simplicity is to the detriment of the method’s validity. Despite its claim to be accurate, the multiple method in its present form is just as subjective as that of discounted cashflows. To use a hundred or so criteria instead of seven would change nothing. By doing this, we introduce a certain amount of redundancy between the criteria, which results in more weight being given to some factors.
As long as the method is subjective, it should remain transparent. The multi-criteria method gains nothing from being summarised in a single score since there are many implicit hypotheses in the weightings. The brand profile should be used instead to make a realistic, valid business plan, materialising in discounted cashflows.
Last but not least, the multiple method is too sensitive to small variations of the multiple itself. Multiplying 800 million by seven or eight makes a lot of a difference. Such sensitivity is at odds with the principle of prudence. Brand valuation is not an exact science. It is not acceptable to obtain outputs that can vary by millions of pounds just by changing the multiple by 1 unit. This is probably why recently Interbrand moved unobtrusively towards the classic financial methodology, the discounted cashflow approach (Table below).
Another estimate of the financial value of brands (2007)