The inconsistent and incomplete recognition of intangible assets in financial statements distorts performance metrics. Invested capital and profit are understated – to what extent depends on the business dynamics and nature and source of investment in intangibles. The combined effect is generally to overstate return on capital.
With the ever-increasing importance of intangible assets, few companies are unaffected by this accounting problem. We suggest adjustments to help your analysis, provide an interactive model to illustrate, and calculate an intangible asset adjusted return for Amazon.
We have discussed the inconsistencies in intangible asset accounting under both IFRS and US GAAP in other articles. Essentially, intangibles that are acquired separately, or together with other assets in a business combination, are recognised as assets in the balance sheet and amortised over their useful life. However, the expenditure that produces most (although not all) internally generated intangibles is immediately expensed.1There are some internally generated intangibles that are recognised (capitalised). Under both IFRS and US GAAP expenditure on software would always be capitalised and, under IFRS only and subject to some fairly strict criteria, certain development projects also result in intangible assets being recognised. However, most of the expenditure that creates or enhances internally generated intangible assets is likely to be immediately expensed. This is very different from tangible asset accounting, where both acquired and internally constructed assets are recognised in the balance sheet.
For more detail about intangible asset accounting see our article Intangible asset accounting and the ‘value’ false negative.
The consequence of the incomplete and inconsistent recognition of intangibles is that you will either need to adjust, or otherwise allow for the distortions, when interpreting profit and measures of return on capital.
Return on equity and return on invested capital are not only commonly used as measures of performance and a point of comparison between companies, but can also be a key input in valuation models. We have used ROIC (strictly a forward-looking incremental return) as an input in various enterprise DCF and target valuation multiple models featured in our articles. Understanding the impact of intangible accounting on return measures is vital to obtain realistic inputs.
Comprehensive recognition of the cost of intangibles may be subjective but is still useful for analysis
While the accounting for intangibles causes analytical challenges, we are not necessarily saying that current accounting practices are wrong. This is a difficult issue that raises questions of relevance versus the reliability of financial statements. Estimates of internal intangible asset ‘investment’ and the useful life and recoverability of the resulting assets are all difficult. Permitting or requiring capitalisation would lead to significant subjectivity in financial statements and the potential for manipulation by unscrupulous management. However, we believe that investors need more information about the differing nature of operating expenses, and the extent to which these are ‘future-oriented’ and result in value creation and therefore intangible assets. Even if intangible asset investment is not capitalised in the primary financial statements there should, at the minimum, be better information provided in the footnotes.
Impact on profit and returns
The lack of recognition of internally generated intangibles almost always results in a higher return on invested capital. This can easily be seen by observing the returns for different sectors with higher (pharmaceutical and consumer goods) or lower (utilities and banks) intangible asset investment. Clearly there are many factors that affect return on capital. However, the fact that sectors where intangibles dominate often tend to have higher observed returns may have more to do with the accounting than underlying economics.
The ROIC impact is a function of two factors:
- Lower operating profit: For a growing business the intangible expenditure that is expensed would be greater than the amortisation of past intangible expenditure had it been capitalised.
- Lower invested capital: The lack of intangible asset recognition in the balance sheet reduces invested capital.
The net impact of these two factors is generally to produce a higher reported ROIC (and forward-looking expected incremental ROIC) for companies which have higher (expensed) intangible investment. This reduces inter-sector comparability and means that accounting returns are a poor indicator of the underlying economic returns. When a reported ROIC is used as a DCF or target valuation multiple model input, you will need to apply a higher figure in cases where intangibles dominate and one that is structurally above the cost of capital.
Consider adjusting accounting data to capitalise estimated expensed intangible asset investment
A better solution, and the subject of this article, may be to adjust your data. In our view, capitalising the estimated amount of intangible asset investment that has been expensed, and amortising this over the estimated life, provides more relevant measures of return on capital. It also facilitates better judgement regarding the returns implicit in DCF cash flow forecasts.
Clearly, free cash flow is unaffected by the accounting treatment of intangible asset investment. However, when forecasting free cash flow, accounting data regarding profitability and returns is a vital component. Although adjustments to capitalise intangibles may be highly approximate, we think that it is a worthwhile exercise.
In the model below we demonstrate how capitalising intangible investment affects profit and returns. We initially begin with an assumed reported (after expensed intangible investment) profit and ROIC. Inputs of the amount of operating expenses that are ‘future oriented’ and the average useful life of the resulting intangibles are used to derive the profit and invested capital adjustments. We generate historical intangible investment through an assumed historical growth rate. A more explicit estimate for prior periods may be preferable in practice, which we apply below in the case of Amazon.
Interactive model: Impact of intangible asset capitalisation on ROIC
— iphone and ipad users: This model formats best if viewed in Google Chrome —
Please enter your email address to receive an excel version of this model
The second calculation in the model is the reverse adjustment. Input an estimate of a return that reflects full recognition of intangible assets and the equivalent ‘intangibles-expensed’ ROIC is presented. The prior inputs related to intangibles also apply to this calculation. Notice that higher intangible investment produces a higher ‘intangible-expensed’ ROIC. Because the ‘full intangible recognition’ ROIC can be a good approximation of an underlying economic return, this calculation may enable better estimation of inputs for valuation models based on reported ROIC. An estimate of underlying economic returns may be easier because, for many valuations, these would likely be equal to, or at least not greatly in excess of, the cost of capital.
The impact of business combinations
Intangibles that are purchased, rather than internally generated, are always recognised as assets in financial statements. This primarily arises due to business combinations, where all intangible assets of the acquired company are recognised at the acquisition date. The price paid in a business combination applies to the business rather than individual tangible and intangible assets. The cost of intangibles therefore has to be approximated by measuring their fair value on the acquisition date.
In subsequent periods these intangibles are amortised; however, any expenditure that enhances or replaces the assets is subject to the normal capitalisation rules and is therefore generally expensed. This creates a form of double-counting where current profit is reduced by the amortisation of past intangible purchases and, in addition, by the expensed cost of similar replacement assets.
For example, if a valuable brand name is acquired in a business combination the (estimated) value of that brand at that time is reported as an asset because it has been purchased, albeit in combination with other assets. However, subsequent expenditure to enhance that asset or create new brands will almost certainly be expensed, and not capitalised, under current accounting rules.
Full recognition of intangibles removes the need for non-GAAP adjustments
The inconsistent accounting for purchased and internally generated intangibles causes problems in performance measurement and leads to frequent adjustments to remove the amortisation expense in non-IFRS and non-GAAP metrics. We explain more about non-GAAP measures and the problem of intangible amortisation in our article Should you ignore intangible amortisation?
However, the problem of double counting does not apply if you make adjustments to capitalise all intangibles in the manner we suggest. Due to the prospective consolidation of acquired companies, the capitalisation of future-oriented operating expenses will automatically recognise the intangibles that gradually replace those arising from the business combination. Of course, the intangibles arising from a business combination will be measured at their fair value at the time of the transaction, whereas other intangibles would be measured based on capitalised expenses. However, both represent the cost to the business of investment in intangibles. If (estimated) intangible assets are comprehensively capitalised then it would not be appropriate to add back any intangible amortisation.
Intangible adjusted return for Amazon
The model above is intended to provide an indication of the difference between a reported and an intangible adjusted return, and to highlight the key drivers of that difference. The application of such intangible asset adjustments in practice will likely require a more granular approach, where the annual expenditure in prior periods is identified and individually capitalised and amortised. You may even be able to differentiate between different types of expenditure, the different intangibles they would generate, and be able to apply different amortisation periods. However, this all depends on the granularity of disclosures provided in the financial statement footnotes. Unfortunately, for many companies, this is not much.
Amazon disaggregation of operating expenses helps identify those that are ‘future-orientated’
Amazon provides a useful disaggregation of operating expenses that helps investors. It differentiates spending on Technology and Content and Marketing from other operating expenses, both of which are clearly future oriented and likely to result in intangible assets. Amazon future-oriented expenditure is significant; in 2020 it was over 2.8x reported operating profit and nearly 17% of revenue. Growth in this ‘investment’ has also been high at about an average 37%p.a. for the last 10 years, which is higher than the growth in revenue.
Amazon disaggregation of ‘future-oriented’ operating expenses
Amazon 2020 financial statements
Of course, ‘future-oriented’ does not mean it automatically results in intangible assets and, conversely, other expenses may also have a forward element to them. In addition, estimating the useful life of the resulting intangibles is difficult and it is impossible to identify whether assets arising from past investment has subsequently been impaired. Nevertheless, even if the outcome is approximate, we believe the exercise to capitalise unrecognised intangibles to be valuable.
In the model below we present calculations of the reported and intangible asset adjusted ROIC for Amazon for 2020. The model is interactive and you can change the inputs for the percentage of each expense item that is assumed to be forward-looking and the useful life of the resulting intangibles. The calculated invested capital and profit adjustments are based on the expenses reported from 2000 to 2020.
Interactive model: Amazon intangible adjusted ROIC
— iphone and ipad users: This model formats best if viewed in Google Chrome —
Note: For this illustration, we define invested capital as 2020 equity shareholders’ funds plus gross debt and lease liabilities less cash and marketable securities. We have not used the average invested capital for the year, nor have we made any adjustments on account of the US GAAP treatment of operating leases. The useful life input is limited to a maximum of 20 years simply because we have only loaded the last 20 years of Amazon data into the model.
The assumed intangible useful lives in the model are our estimates. However, there is some information in Amazon’s financial statements that helps. The intangibles we are attempting to capitalise are the same as those Amazon has recognised in its financial statements as a result of past business combinations. The footnote extract below shows those intangibles and the related amortisation.
Amazon intangible assets acquired through business combinations
Amazon 2020 financial statements
Amazon says that the useful life of acquired intangibles ranges from 1 to 25 years and that the average remaining life is 14.4 years. We find the last figure somewhat confusing since the disclosed future amortisation for 2021 compared with the gross prior asset cost would seem to indicate a shorter period of around 9 years. However, Amazon also specifies an average useful life for each intangible asset category which clearly shows a much shorter useful life for technology and content.
For Amazon, the effect of our intangible asset adjustments (using the base-case assumptions that appear when this page is first loaded) is very significant. We estimate that operating profit would increase by 121% and invested capital by 146%. The increase in invested capital is a function of the amount of intangible investment relative to existing tangible assets – the technology and content expense alone exceeds reported tangible asset capital expenditure. The increase in operating profit is driven by the past rapid growth in the Amazon business combined with the significant intangible ‘investment’. This results in a far greater add back of the investment expense compared with the amortisation of intangible assets acquired in prior years.
The combined effect of the changes in operating profit and invested capital is to reduce ROIC from 24.5% to 22.0%. In this case the reduction is mitigated by the past high growth; for most companies with high expensed intangibles the profit impact would be lower than the effect on invested capital, resulting in a larger reduction in ROIC.
Intangible adjusted ROIC provides a more comparable measure of return …
We think that the adjusted metrics are a better basis to assess performance and to make comparisons with other companies. This is particularly the case where companies have a different balance between expansion by acquisition and organic growth. For Amazon most growth has been organic, as indicated by the relatively small overall contribution of acquired intangibles and goodwill to reported invested capital. This means that reported ROIC is higher than for comparable companies where expansion has, to a greater extent, been driven by acquisitions.
… and perhaps more realistic valuation multiples
In addition to affecting performance and returns metrics, the capitalisation of internally generated intangibles also affects valuation multiples. The price or enterprise value component of multiples is unaffected because the full (estimated) contribution of intangibles should2At least in theory the capitalisation of intangibles should not affect market prices. Whether, in practice, this would provide new information to the market is an interesting question. already be incorporated in market prices. However, as we demonstrate for high growth companies such as Amazon, the existing accounting of expensing most internal investment in intangibles creates a drag on profit and consequently overstates valuation multiples. For Amazon, based on the assumptions in our model above, the capitalisation of intangibles would reduce EV/EBIT from 72x to 32x.3The EV/EBIT multiples is current EV, which we have approximated by taking current (September 21, 2021) market capitalisation plus 2020 balance sheet net debt, divided by 2020 reported, and our adjusted, operating profit respectively. and have a similar effect on other valuation multiples, including the price earnings ratio.
Intangible asset cost versus fair value
Our analysis uses historical cost measurement for the capitalised intangibles. This is a good basis on which to assess the amount of investment and allows for the better matching of the inputs and outputs of a business when assessing performance. However, the cost of intangibles is likely to be very different from their fair value. Most intangibles are unique assets where value can be gained or lost very quickly. This may also be the case for some tangible assets, such as certain properties and mineral reserves but, generally, tangible assets are more likely to be, in effect, fungible, with value not deviating greatly from depreciated cost.
Historical cost of intangibles is still relevant, even if fair value can be very different
Some argue that the unique nature of intangibles, and the potential for value to deviate significantly from cost, means that a cost-based measurement is of little use. We disagree. We think that the investment in intangibles and the resulting return on capital is still useful. Value is reflected in achieving a return premium and it is for investors to assess rather than for management to provide.
In certain circumstances management assessment of intangible (or indeed tangible) asset fair value, notwithstanding the subjectivity, would be highly useful for investors, but we think this should be provided in the footnotes, in addition to the cost-based measurement in the balance sheet.
Accounting disclosures need to improve
It has long been debated about how best to report on intangibles in financial statements. As we have previously explained, the current inconsistent and limited recognition of intangible assets causes analytical challenges for investors. In addition, there is often limited disclosures in both financial statements and management commentary related to intangible investment. We think that investors would greatly benefit from improvements to both the narrative reporting and financial statement data regarding intangibles and we welcome the attention this area is getting from standard setters.
Our concern is that any changes may only relate to narrative reporting. While we think that disclosures about the types of intangibles held by companies and the strategies for their management are important, attention should also be paid to recognition and measurement. We have demonstrated how profit and return on capital are significantly impacted by the lack of intangible asset recognition. Investors need not only narrative but also numbers to help them estimate this impact.
Insights for investors
- The lack of intangible asset recognition in financial statements results in an understatement of invested capital and, for growing companies, an understatement of profit.
- For most companies ROIC calculated from published financial statements is overstated due to unrecognised intangibles and is a poor indicator of economic returns.
- Capitalising ‘future-oriented’ operating expenses as intangible assets may be subjective, but it results in more realistic measures of performance and returns.
- Use our intangible adjusted returns model to better understand how unrecognised intangibles affect profit and return on capital.
- Intangibles purchased as part of a business combination are recognised as assets but the profit and return on capital impact is not comparable because it depends on the relative balance of organic and acquisition-based growth.
- Combining an estimate of capitalised internally generated intangibles with acquired intangibles that are already recognised in financial statements may produce more relevant and comparable metrics.
Don’t miss new articles published by The Footnotes Analyst – subscribe here:
No advertising, no spam, just new articles. Unsubscribe anytime.
Send us a question or comment about this article:
For more articles select subject tags:
Or see suggested related articles below: