Letter number 69 of September 2012


News : A brief look at "Other comprehensive income"

For a few years now, US GAAP and IFRS have required the booking of changes in the fair value of assets and liability, either as net income/loss for the financial year or as shareholders’ equity.  In the latter case, this entry is made either in a separate table that reflects this “Other comprehensive income”, or in a single table that groups it together on the normal income statement.  It is then referred to as the “Statement of comprehensive income”.  The second option is not used very frequently, and in our mind, rightly so, since the information content of these other items is insignificant and they tend to pollute the reading of the income statement. 

So what is “Other comprehensive income” ?

Under IFRS, it is described in six points, and under US GAAP, the last four on our list:

1/ Changes in fair value of tangible fixed assets

Under IFRS, firms are entitled to value their fixed assets in fair value on the assets side of their balance sheets.  Latent capital gains will then form part of “Other comprehensive income” and shareholders’ equity.  Subsequently, in the event of a drop in fair value, this drop is first set off against “Other comprehensive income” until it has been depleted in shareholders’ equity, then, if applicable, entered directly on the income statement as it should be.  

Very few groups have opted for this system given the complexity of implementing it.  Portugal Telecom is an exception.  This option should be distinguished from the option allowed at the time of the shift to IFRS, where, for example, Publicis revalued its building at the top of the Champs-Elysées (13,500 sqm of offices and shops), from €5m to €164m in 2004, resulting in a latent gross capital gain of €159m, which was booked under shareholders’ equity.  The building has now been amortised on this basis, as the historical value was simply revalued in 2004 and has not been revalued since. 

We believe that a group that resorts to this option sets itself apart by sending out a signal or confirming the fact that its financial situation is less than stellar as it has had to externalise latent capital gains. 

2/ Changes in fair value of intangible fixed assets

The same principle applies as for tangible fixed assets.

3/ Changes in the value of pension fund commitments resulting from changes in assumptions

Currently, under IFRS, pension fund commitments can be entered on the income statement or under “Other comprehensive income”, or they can be booked using the corridor method(1). With the new version of IFRS, which will apply from 2013, all changes in value resulting from a change in assumption with regard to mortality, turnover rates, retirement dates, and increases in salaries or pensions, must be booked as "Other comprehensive income”.  However, differences between the expected rates of return on pension fund assets and the rates actually achieved every year, should be booked on the income statement itself.

What is the informative content of these change for the analyst?  Insignificant in our view.  At the very most, they will draw the analyst’s attention to the assumptions that were made and that are no longer valid, possibly indicating a form of accounting aggressiveness on the part of management.

4/ Currency translation differences resulting from the consolidation of subsidiaries that present their financial statements in a currency other than that of its parent company

On this very standard point, the same principle has applied for a long time under French GAAP.  The informative content of these items is insignificant for the analyst since these currency translation differences only trigger a change in the book amount of the share in shareholders’ equity of the consolidated subsidiaries that is owing to the parent company as a result of variations in foreign exchange.

5/ Changes in fair value of financial assets classified as available for sale

They are revalued on the balance sheet, the unrealised capital gains are not externalised on the income statement but booked as "Other comprehensive income", and accordingly as shareholders' equity.   On the day on which the asset is sold, the additional value obtained compared with the last revalued value on the balance sheet, plus all past capital gains and "Other comprehensive income”, will be entered on the income statement to make up the pre-tax capital gains, the difference between the sale price and the historical cost price. 

When a loss in value has to be booked as an impairment of the available-for-sale financial asset, it is booked on the income statement. 

6/ The efficient part of a financial risk hedge

Remember that as it currently stands, the standard covering financial instruments (IAS 39), provides that variations in fair value of the part of the cash flow hedge deemed to be inefficient, are booked on the income statement.  Variations in fair value of the part of the hedge deemed to be efficient are booked as “Other comprehensive income" and as shareholders’ equity.  More specifically, this only applies to cash flow hedges (for example, interest on a debt at a variable rate) and not to fair value hedges (for example, the value of a debt at a fixed rate). 

When the hedged transaction takes place, the items previously booked as “Other comprehensive income” are then transferred to the income statement in order to neutralise the gain or the loss in value on the hedged item, since the hedging had been deemed to be efficient.

* * *

Accordingly, “Other comprehensive income” corresponds to all of the items that change shareholders' equity, with the exception of net loss/income, transactions involving shareholders (dividends, share issues, share cancellations), and the effects of corrections of errors and changes in accounting methods. 

Both IFRS and US GAAP stipulate that firms should classify “Other comprehensive income”, either as:

• items that could subsequently be transferred to the income statement;
• items that will not subsequently be transferred to the income statement.

Among the latter, we draw attention mainly to the changes in the fair value of assets that were depreciated in the past.  We could also add the currency translation differences involving subsidiaries that will never be sold, in as far as we can use the term “never” in this area.  But since that is not the case, they are booked with the first group.

* * *

A study of 41 European groups in the banking, mass market and telecoms sector show that items seem to be arbitrarily allocated to “Other comprehensive income”, with each sector having its own specificities.  Banks show a predominance of changes in fair value on available-for-sale financial assets, with more foreign exchange effects for industrial groups.
OCI as a % of Net income + OCI                             
Breakdown of OCI
The more in-depth study over a longer period of time of the results of Carrefour and Vodafone does not, for these two groups, reveal that “Other comprehensive income” shows a systematic negative or positive bias, something which would have troubled the analyst.

Accordingly, “Other comprehensive income” seems to be merely a receptacle for volatility, whether it be volatility that is experienced (currency translation differences, changes in pension calculation assumptions) or that has been created by introducing fair value into the balance sheet (change in the fair value of assets, booking of hedges).

 Thanks to Strahinja Ninic and Oussama Lemsyeh for their work on this article.


  (1) For more information, see chapter 7 of the Vernimmen.

Statistics : Legal provisions of M&A transactions

For four years now, the law firm CMS has published an annual study on the legal provisions of M&A transactions in Europe, in which they acted as advisors to the seller or the buyer in mainly mid market deals (with a value below $500m). Comparisons are available for the USA on deals completed in 2010.

The full study can be downloaded here.

Key take-aways are:

Frequency 2011 Average 2007-2010 USA
Purchase price adjustment 47% 48% 82%
Working capital and net debt are the most popular forms of purchase price adjustments, featuring in respectively 26% and 22% of deals with a purchase price adjustment. Equity/net worth is the third most popular item but present in only 12% of the transactions under review.
Locked box mechanisms (1) 38% 29%
They made a stellar performance in German speaking countries where they are now present in 58% of the deals without a purchase price adjustment ( versus 14% in 2009).

Earn-out (2) 14% 17% 38%

mainly based on EBIT or EBITDA (44%) and turnover (25%).
De minimis (3) 62% 52%
Very popular in the UK (found in around 80% of deals), less so in Central and Eastern Europe (51%) and in Southern Europe (44%).

Basket (4) 65% 49% 95%

In the UK, the Benelux and German speaking countries, once the basket threshold is exceeded the buyer is entitled to make the entire claim, or just the excess of the claim over the basket as in France (66% of the deals) or in the USA. In Europe, 45% of deals under review had a basket threshold above 1% of the deal value versus only 12% in the USA.

Liability caps 82% 78% 86%

Limiting the aggregate liability of the seller under the sale agreement is standard: in 24% of deals the limit is the purchase price and in 23% of deals it is between 10 to 25% of this price. In the USA the liability cap is significantly lower.

Warranty and indemnity insurance (5) 11% 11%

They are mainly used when the seller is a private equity fund.

Security for warranty claims 37% 42%

Putting part of the purchase price in an escrow account is the top solution (50%) followed by retention of part of the price (37%) and a bank guarantee (28%).

Material Adverse Clause (MAC) 16% 17% 93%

This is where there is a big cultural difference between Europe and the USA.

(1) For more on this item see the Vernimmen.com newsletter number 64 dated January 2012.
(2) For more on earn-out, see chapter 43 of the Vernimmen, 2011.
(3) An individual claim is only considered if it is in excess of a minimum value.
(4) The sum of individual claims exceeding the de minimis are only considered if they exceed the basket threshold.
(5) For more on this topic see the Vernimmen.com newsletter number 59 dated July 2011.

Research : Friends with money

with the help of Simon Gueguen Lecturer-reseacher at the Paris Dauphine University

In comparison with disintermediated financing, bank loans are characterised by strong institutional ties between the bank and the firm. A lot of studies have been carried out on these ties and on their consequences in terms of financing. The article we look at this month(1), examines a particular type of link – interpersonal relationships. 

When the lender and the borrower have known each other for a long time, as a result of having studied together or worked for the same firm, the terms negotiated for the loan are optimal.

The analysis covers a very large sample of 19,554 loans, granted by US banks, mainly unlisted banks) to listed firms (5,057 different firms) between 2000 and 2007. Of this sample, Engelberg et al obtained (2) a list of the organisations in which the 65,074 managers concerned worked or studied. 

They considered that there was a personal link between a bank manager and a manager at a firm when one of these two conditions was met:

• they graduated from the same school/university less than two years apart;
• they worked at the same time for the same company, or were both members of the same board of directors. 

This condition must be met at least five years before the date of the loan and involve a different company than the bank and the firm taking out the loan(3). According to these criteria, there is at lease one interpersonal link in 29% of the loans in the sample. 

An econometric analysis shows that, for equivalent loan terms (4), the spread  (difference between the interest rate and the LIBOR, taken as a benchmark) invoiced by the bank, is 28 base points (0.28) lower if there is an interpersonal connection. This is highly significant since the average spread in the sample amounts to 88 base points.  The gain is even larger for firms with a low rating – eight base points for loans with high ratings (between AAA and A), and up to 51 base points for those with low ratings (CCC and below). Engelberg et al checked that these advantageous terms were not set off by other contractual clauses.

The question that is raised is this: Are these advantageous terms justified? It might be suggested that there is connivance which results in a “gift” to the borrower, at the expense of the lender’s shareholders. 

The article shows that this is not the case at all. 

Firstly, the rating of firms improves following the granting of a loan when an interpersonal relationship exists. Next, these firms outperform firms which have contracted loans without any interpersonal link on the stock exchange, by more than 5% per year (over a 3-year period).
Engelberg et al conclude that these links in reality facilitate the selection and monitoring of projects by the bank. They help to reduce the usual phenomena of moral uncertainty and adverse selection, which are particularly prevalent when ratings are low. The apparently advantageous terms are thus justified by the future performance of the borrowers. 

So who said that there was no price on friendship?

(1) J.ENGELBERG, P.GAO and C.A.PARSONS (2012), Friends with money, Journal of Financial Economics, vol.103, pages 169-188.
(2) They used the data supplier, BoardEx, which specialises in interpersonal relationships.
(3) These conditions are set to ensure that the relationship goes back a long way before the loan. Engelberg et al sought to eliminate the possibility of an inverse causality (a seat on a board given in exchange for favourable terms, for example).
(4) The criteria used as control variables are rating, maturity, past institutional relationships and the number of lenders. 

Q&A : What is a unitranche loan?

Unitranche loans are loans that we have seen more frequently since the 2007 financial crisis in the financing of small-to-medium-sized LBOs, where a single lender grants the senior debt which is subordinated in the form of a 5- to 8-year bond, reimbursable at maturity.  The unit amounts are a maximum of a few tens of millions of euros.

In other words, and unlike a senior loan, borrowers are not required to make any repayments during the life of their loans, which reduces the pressure on them during LBOs and increases their margin for manoeuvre when making investments in organic or external growth.

In France, the lender is very often a debt fund or a mezzanine fund which, as it does not have the status of a lending institution, cannot grant loans as a result of the banking monopoly, but can subscribe bonds (a form of shadow banking!).  

Part of the interest is paid in cash during the life of the bonds, and another part is capitalised and payable in a bullet payment when the bonds are repaid (we talk about PIK or payment in kind).  An additional remuneration, in the form of warrants, is sometimes included like in a mezzanine financing (equity kicker).

The yield to maturity of a unitranche loan is half-way between that of a senior debt and a mezzanine financing, i.e., from 11 to 13%, which is in line with the logical expectations for this product. 

Granted by a single investor, it has the dual advantage of being quick to put in place, and eliminating any conflict of interest among several lenders, in particular if the debt has to be renegotiated if the company gets into difficulties.  Since one can’t have one’s cake and eat it, in the event of problems, the private equity fund that owns the asset runs the risk of losing it very quickly to the mezzanine lender who, being the sole lender, shouldn't have too much trouble reaching agreement with itself and presenting a united front to the shareholder!

We leave it to academics in a few years time, to reach a conclusion on whether firms under LBOs carried out using unitranche financing create as much value as those using the more traditional LBO financing.  It’s possible that this may not be the case, since a poorer performance may result from a smaller degree of control over the use of cash flows by the lenders, given that this product is reimbursed at maturity.

In the mean time, the regular problems that arise in the syndication of LBO bank loans and the succession of phases of opening and closing of the high yield bond market, mean that the unitranche loan is likely to be around for a while yet.