Letter number 118 of January 2019
- QUESTIONS & COMMENTS
Based on our practical experience of IFRS as teachers or investors, we believe that there are a number of areas that would benefit from a reform of existing provisions, even if only to comply with the true and fair view accounting principle. This is in fact a paradox as principles are intended to guide standards and the true and fair view principle is one, if not, the most important of all accounting principles.
1 – Increasing wealth just by buying at the right price
In 2017, the British tobacco group, BAT, the fifth largest market capitalisation on the London Stock Exchange, (£61bn), recorded sales of £20.3bn and net earnings (group share) of £37.7bn.
Yes, your eyes are not deceiving you, £37.7bn for only £20.3bn in sales. You knew that tobacco had become very profitable, but you hadn’t realised by just how much!
The first of the two of us who saw these figures secretly wondered whether a junior employed by a financial data provider that compiles annual reports might have misplaced the decimal, that their internal checking systems in place had failed, and that the correct figure was obviously £3.77bn. But no! BAT’s 2017 net earnings are indeed £37.7bn.
The other thought that BAT must have sold a very large subsidiary, for example its international business, generating large exceptional capital gains which logically were reported at the bottom of the income statement. But this was not the case either. BAT made no significant disposals in 2017 that would have resulted in capital gains that could have boosted its net earnings by this much.
However, in 2017, although it had not sold any substantial assets, BAT had bought up the balance of shares that it did not hold in RAI, in which it previously held a 42.2% stake.
And so what?
And so, according to IFRS, in such cases, the 42.2% stake initially held is revalued at the price paid for the balance of the shares, almost as though BAT had for a fraction of a second sold its 42.2% stake before buying it back as part of a bid for 100% of the shares in its subsidiary. And the higher the control premium, the greater the increase in shareholder wealth (on paper), as in any event, goodwill is recorded the first year at market value and will not be depreciated for some time, and possibly quite a long time. This is somewhat reminiscent of a previous IFRS rule according to which when a company’s solvency deteriorated and the market value of its debt declined as a result thereof, the company could record a gain equal to this decrease in value. Here, the increase in wealth was the result of the company’s decline.
We leave you to decide on the appropriateness of such an approach which swells BAT’s book equity by 62 % and massively increases deferred tax liabilities by £27.1bn (out of a total opening balance sheet of £40bn) compared to £0.7bn initially, when too many readers of the accounts still believe that these are outstanding debts, which is not the case. So how do you calculate a return on capital employed or on equity that corresponds to reality, i.e. to what was invested by investors and not to the effects of revaluations?
Good luck with that!
2 – Losing without losses
A group acquires a young company, as happens every day, in order to acquire a new technology that appears to be particularly promising. In this case, goodwill represents virtually the whole of the acquisition price, which is hardly surprising in an economy in which intangible assets are increasingly important.
Unfortunately, a few years later, competing technologies turn out to be more efficient. The group then decides to sell the company for a song.
What happens to the goodwill generated by the initial acquisition? Logically, it should be depreciated in full. But no, only the percentage that the assets’ sale price represents in terms of the value of the assets of the cash generating unit to which it belongs, is depreciated. In other words, peanuts. It will remain on the balance sheet while the company behind this goodwill is no longer part of the group. Strange, isn’t it? True and fair view, really?
3 – Generating earnings through share buybacks
A large group buys back shares and its broker delivers the shares in blocks from the start of the buyback programme which is spread over several weeks.
Well, the auditors of this large group insisted that the change in value over the duration of the share buyback programme be recorded on the income statement, in order to cancel the shares, while this is not required under IFRS.
One would like to discredit the P&L account, one wouldn’t do things any differently.
4 – Recording debt as equity
This is what happens now with so-called perpetual hybrid debts, where form takes precedence over substance. There only needs to be a clause covering an increase in the interest rate paid by the issuer if it fails to exercise its early redemption option (most often after 5 years) for this debt product to be recorded as equity under IFRS.
However, we remind our readers of the accounting principle that reality prevails over appearance, which the IASB has often rightly highlighted to demonstrate the quality and superiority of its standards.
Given that equity is the cornerstone of corporate finance, we really should call a spade a spade, and debt should be called debt. So-called perpetual hybrid debts are subscribed by investors seeking to invest in debt, they are arranged and placed by banks’ debt capital market (DCM) teams, and in the majority of cases, companies exercise their early redemption options at the first call debt (usually after 5 or 6 years). So, this is debt and not equity, regardless of what the IASB currently says.
For example, at the close of 2017, Volkswagen was able to include hybrid debt with its equity thanks to clauses providing for 0.25% and 0.75% increases in interest paid if the early redemption options were not exercised, which would certainly encourage it, when the time came, to opt for the early redemption of these products that would have become too costly.
The IASB is currently considering modifying IAS 32, the standard which allows this treatment, and which is what we recommend. If it does so, the way in which hybrid debt is booked could effectively change by around 2025-2028. There’s still time for us to raise this issue again!
5 – Recording equity as debt
This is the case for bonds redeemable in a variable number of shares for which interest and capital are recorded as debt. This goes against the most basic common sense given that the capital will never be redeemed in any form other than shares. The fact that it could be redeemed in a variable number of shares makes absolutely no difference.
Issuers do win in the end if they have an option to convert these bonds into a fixed number of shares, as they can then book them as equity. Too bad for careless issuers or those that have received poor advice! Although the IASB is also considering this issue, it does not appear that it is keen to change its opinion for the moment.
6– Confusing exceptional and recurring items
IFRS has got rid of the notion of exceptional earnings on the pretext that it depends on a judgement, which is true of course, but is there really anything under IFRS, US standards, that doesn’t depend on a judgement?
So, how does a company communicate on its annual results when they include exceptional earnings, and where it appears logical to highlight net current earnings? The exercise is complicated by the intervention of the ESMA in Europe which does not allow companies to focus on indicators that are not drawn from the financial statements in their financial communication.
This rule is based on a good intention, that of preventing companies from focussing on ad hoc, contingent and special indicators, which put a gloss on the facts and mislead those reading financial statements.
As a result of this abdication, companies have increased the amount of data they provide by 50%, as illustrated by the 2017 annual report of the UK advertising agency WPP:
It’s then up to the reader to work out what’s what. But who should be believed? The IFRS accounts which show a 31.9% increase in EPS or the company which reports a simple and modest +6.4%, qualified as ‘headline’? And EBIT? Did it grow by 2.8% as stated by WPP or drop by 7.5% according to IFRS? And what does the 4.7% increase in EBITDA mean when only the company provides this information, since under IFRS, there is no definition of EBITDA?
Probably unconsciously, the IASB is detracting from the credibility of the accounting standards that it has established by pushing companies to publish corrected results, with the blessing of the European stock market regulator, which, just like Pontius Pilate, does not want to judge and requires everything to be put on the same level.
Let’s hope that the IASB, which has announced that it wants to take another look at accounts presentation, backtracks on this issue. Is it that hard to ask managers to distinguish between recurring and non-recurring items and to explain them in notes to the accounts and to ask auditors to audit them like they did a few years ago? Meanwhile, a lot of time has been wasted and readers have been unnecessarily misled!
* * *
We’d like to invite our readers who have come across other inconsistencies, anomalies or mishaps associated with IFRS to use the letterbox on the Vernimmen.com website by clicking here, so that we can add them to an updated version of this article when the time comes.
 And if you’re one of them, we recommend that you read sections 7.6 of the Vernimmen where you’ll see that this is not the case.
 “On a balance sheet, only the date does not involve a judgement”. Roman Weil.
Asset betas, also known as unleveraged beta, or betas of the operating assets, correspond to the equity betas once you have neutralized the impact of the financial structure. Here is a list of 28 sectors with their asset beta:
As most companies are into financial or banking debts, the asset beta is lower that their equity betas. Hence an average asset beta lower (0.79) than the average equity beta which is one.
Asset betas are useful to compute the cost of capital using the direct approach.
With Simon Gueguen, Senior Lecturer at the University of Cergy-Pontoise
The most frequently used models for valuing capital assets are constructed as extensions of the capital asset pricing model (CAPM), according to which the expected return on an asset depends on the share of non-diversifiable risk. These models include premiums based on factors that have been identified as factors that will increase expected returns on assets. More specifically, in 1993, Fama and French suggested that shares should be valued using a three-factor model, the two additional factors (on top of non-diversifiable risk) are:
- a size factor (SMB for small minus big) according to which small capitalisations have higher expected returns than large ones;
- a valuation factor (HML for high minus low) according to which value stocks (those with a high book-to-market ratio) have higher returns than growth stocks.
Sometimes other factors are included, such as momentum.
The size factor is often criticised. Given that it is an empirical measurement, results are often unconvincing so its impact on expected returns, once risk has been taken into account, is low. Moreover, there is no real theoretical explanation underlying the size factor. In a standard frictionless market model, a premium based on size would appear to be an anomaly. The article that we look at this month rehabilitates the size factor and sheds light on its possible sources.
The idea behind this article is to measure the importance of the size factory by controlling the impact of other factors. Asness et al identify a strong correlation between the size and “quality” of the share. Quality refers to a set of the features of the underlying company, in particular its growth, earnings, dividend pay-out ratio, credit rating, etc. Asness et al recalculated the importance of the size factor once the quality factor had been taken into account, over a very long period (US stocks between 1926 and 2012) and on a large number of markets (data for 23 other markets in developed countries between 1983 and 2012). The results are very convincing.
By ignoring the quality factor (but controlling the value and momentum effects), the size factor results in an alpha of 14 basis points. This means that, all other things being equal, the shares corresponding to low capitalisations bring an expected monthly return that is 14 basis points higher than shares corresponding to high capitalisations. Once the quality factor has been taken into account, the same alpha rises from 14 to 49 basis points! The impact becomes significant and the size factor can now no longer be ignored.
In addition to rehabilitating the size factor, this study sheds light on the source thereof. The shares of small capitalisations are generally less liquid than those of large capitalisations. Once the factor associated with the share’s liquidity is taken into account, the alpha of the size factor becomes low again (although still positive). So, it seems that the higher return on small capitalisations is largely explained by a premium on the share’s illiquidity risk.
These results are important for practitioners from at least two points of view. In terms of market finance, it is said that the size factor should not be neglected in valuation models, and that there is still works to be done on the study of its sources. And in terms of corporate finance, a premium associated with small size means a higher cost of capital for the company. This could help to provide justification for mergers.
 C. ASNESS, A. FRAZZINI, R. ISRAEL, T.J. MOSKOWITZ and L.H.PEDERSEN (2018), Size matters, if you control your junk, Journal of Financial Economics, vol.129, pages 479 to 509
In the consumer goods sector where, given the size of margins which are currently at an historical high and limited investments due to a slowdown in growth, many groups (Unilever, Nestlé, L’Oréal, Diageo, etc.) have been buying back their shares.
Heineken, for its part, has stated that it does not wish to buy back its shares, except in order to return income from the sale of significant assets as it did in 2015 for €365m. Is it right?
As we explain in chapter 37 of the Vernimmen, a share buyback only makes financial sense when:
- The price of the buyback is lower than the value of the share.
- The increase in the company’s debt ratio will lead to better performance by managers.
- The funds returned to shareholders could not be put to work in the company in investments that could earn at least their cost of capital.
With consensus of the 27 analysts who follow Heineken of a target share price of €92 compared with a current share price of €80, the Heineken share price does not appear to be particularly undervalued when we consider the behavioural propensity of financial analysts for optimism. A 2018 P/E ratio of 21.3x, EBIT multiple of 17.7 and P/B ratio of over 3 do not spontaneously appear as low for a profitable company whose EBIT is expected to grow by 4.8% and its net earnings by 8.2% by 2020.
With a 2017 net debt/EBITDA ratio of 2.4, Heineken is clearly leaving itself room for manoeuvre so that if the opportunity arises, it can make acquisitions without having to call on its shareholders. In as far as the Heineken family hold 50% of the share capital through a holding company, the family is the one to decide whether or not it is comfortable with this capital structure, and it seems that this is the case given the stability of this ratio since 2011. Our second point is thus a relatively theoretical matter in this case.
Since 2011, Heineken’s average ROCE has been 10.6%, so above the cost of capital which is just under 7%. Marginal ROCE, i.e. return on new investments, is 10.8%. This makes Heineken a company that is creating value, as evidenced by its long-term share price which has risen by 117% since January 2007, compared with 98% for Molson Coors, 65% for Carlsberg, 27% for AB InBev and 134% for AmBev.
The steady rise in the dividend payout ratio, from 33% in 2011 to 43% in 2017, shows that surplus cash flows are returned to shareholders through dividends.
It also seems to us that as long as Heineken increases its volumes of beers sold (by 4.8% per year since 2011), and while it can still identify investments that earn at least their cost of capital, then the answer to the question is yes.
Regularly on the Vernimmen.com Facebook page we publish comments on financial news that we deem to be of interest. Here are some of the comments published over the last month.
Index funds with management costs of 0%?
This was launched during the summer by Fidelity, known for its active management and not its passive management skills, to limit its losses of assets under management moving to index funds and ETFs. The success of this first product on the market with management fees of 0% is undeniable: its US equity fund and its non-US equity fund have raised so far more than $2bn in 6 months, a quarter of which for the international fund and 3/4 for the US equity fund, illustrating the domestic bias of investors.
It's definitely two loss-leaders for Fidelity. It is necessary to open an account at Fidelity to be able to subscribe, which allows the asset manager to offer other products more remunerative for it to its new customers, or de-incentives the current ones from moving to the competition. In addition to the volume effect necessary to reduce the costs invoiced to zero, Fidelity relies on indices that it has developed, which avoids having to pay royalties to the owners of indices like Dow Jones or MSCI; and practices securities lending on the shares held by these funds. Securities lending allows Fidelity to receive fees from short-sellers who needs to borrow shares to short-sell them. In a nutshell, it allows other investors to bet down securities prices held by Fidelity on behalf of its customers. . . This is the price to pay for having zero fees. As we say in the United States, there is no free lunch.