Letter number 107 of October 2017

  • NEW

News : Why have the main European banking and insurance groups listed – or why are they planning to list – some of their subsidiaries?

Early 2018 if all goes well, each of the four major French financial groups will have IPOed one of its subsidiaries: Amundi for Crédit Agricole in late 2015, First Hawaiian Bank for BNP Paribas in the summer of 2016, ALD for Société Générale in June, and Axa Financial via AXA America Holdings for Axa in the first half of 2018.

Although we will focus here on French examples, this is not a French phenomenon per se: between 2014 and 2015, RBS sold its US retail banking subsidiary, Citizens Financial Group, on the market; ING sold its North American and European insurance subsidiaries on the market; Santander IPOed its Mexican, Brazilian and US subsidiaries; Unicredit listed its online bank FinecoBank; Crédit Suisse spent a long time thinking about IPOing its Swiss retail banking activity and Deutsche Bank announced that it would soon be opening up the capital of its asset management subsidiary.

This is more than a coincidence and this movement is a result of a combination of several factors, including good stock market performances.

Although it is rare to see a single reason behind this sort of structural movement, we have identified one that they all have in common, at least for the banking groups: the improvement of prudential ratios, (Basle III). All of the subsidiaries IPOed (or to be IPOed) have been recording good performances (if this was not the case, they wouldn’t be candidates for an IPO): their return on equity is higher than their cost of equity. The logical consequence of this is that the value of the equity of these subsidiaries is higher than the book value of this equity[1]. Selling shares while retaining control makes it possible to generate capital gains, which although not recorded on the income statement given that under IFRS a fully consolidated entity remains consolidated, they are directly incorporated into equity which they will thus increase. Moreover, the minority interests created by opening up the capital (but which replace equity capital, group share) are recognised prudentially[2], following corrections that we won’t burden you with here. Because there is no reason why the amount of the weighted commitments on the asset side of the bank’s balance sheet should be modified by the sale of a minority stake, the prudential ratio, the ratio between equity capital and commitments, is thus increased.

So, the recent sale of 23% of ALD enabled Société Générale to improve its CET 1[3] prudential ratio by 13 base points; for BNP Paribas this figure was 15 points thanks to the IPO of First Hawaiian Bank, followed by a subsequent sale on the market, reducing its stake to 62%. Finally, the Amundi capital increase, to refinance part of the Pioneer acquisition, which was only 40% followed by Crédit Agricole which held a 73.9% stake, enabled it to improve its solvency ratio by 9 points while retaining a stake of 70%.

This is probably not the main reason, because compared with CET 1 ratios which are currently 11-12 %, the improvement is marginal; but it’s a welcome side effect that improves prudential ratios that are rising as a result of regulation.

Sometimes, an IPO is an alternative to a disposal that is impossible, given the lack of a buyer (this is how Coface found itself being IPOed by Natixis). First Hawaiian Bank (retail bank in Hawaii, with a 44 % market share) and ALD (leading European player in long-term car rental) are both on niche markets on which there are very few buyers and additionally, their leading positions would make it difficult for a competitor to acquire them.

Compared with an ordinary sale, an IPO, followed by partial sales, enable the seller to spread its capital gains and its cash which it receives over time. If it received them all in one go, it would be difficult for it to re-employ them immediately and would expose it to requests to return this idle cash to shareholders. The listing of the life insurance and asset management activities of AXA in the USA was positioned as a way of reducing the group’s exposure to this major asset (estimated at around $12.5bn) and as a money maker. Once the subsidiary is listed, it will be easier to sell a new tranche of its shares to finance the acquisition of non-life insurance assets, AXA’s stated aim, for unit amounts of €2 to 3bn. 

By listing a subsidiary, some companies are seeking to attract the attention of investors to an asset that would otherwise be lost in the haystack of the group, and that has attractive features in terms of growth or risks. For example, it is unlikely that investors would have noticed ALD, with its 10% growth rate, within the Société Générale group. They have certainly noticed it now that the subsidiary has been listed as the IPO process resulted in an increase in the residual stake held by SG. The size must be substantial relative to the group: at least 10% (ALD capitalises €5.5bn compared with €40bn for SG). Will this mean that the Société Générale share will be more highly valued? Perhaps. What we do know though is that hope is life!

The IPO of a company is one of the standard liquidity mechanisms used in joint venture shareholder agreements (as are drag / tag along clauses). The main reason for IPOing Amundi was to enable the exit of a minority shareholder (Société Générale, with 20%) that Crédit Agricole did not wish to buy out (negative impact on prudential ratios), and at the same time to replace it with the market, in order to finance subsequent acquisitions through capital increases (like Pioneer) or dilution in the case of payment in shares of the listed subsidiary.

The regulators now require financial institutions to justify the fungibility of their subsidiaries’ capital (disposal is an option) and their recovery plans in the event of major problems, for which the disposal of entities is the first option generally used in order to justify their ability to raise capital. One of the key questions that regulators have when reviewing these plans relates to the ability to carry out these sales quickly. So a subsidiary that is already listed is interesting from this point of view, even if this alone is not reason enough for IPOing it. 

When only one minor subsidiary is listed, the holding discount is not a problem. Things would be different if several major subsidiaries ended up being listed.

             *             *             *

Finally, we should also note that nothing is irreversible: BNP Paribas had bought up the whole of the free float of First Hawaiian Bank in 2001 and AXA did the same with AXA Financial. As the saying goes, ‘only fools and dead men don’t change their minds’, but normal people do and so company strategies, just like the world around us, will also be subject to change.


[1] If you’re still grappling with this concept, check out chapter 26 of the Vernimmen.

[2] In Europe if the subsidiary is regulated as a bank

[3] CET 1: Common Equity Tier One


Statistics : Yield curves in some major countries

12 years. That’s the maturity period for the first Swiss government loan with a positive interest rate. For other shorter periods, there are only negative interest rates. For Germany, this period is 7 years, its 6 years for France, 4 years for Spain and 3 years for Italy.

0.49 %: That’s the cost of 10-year debt for Germany and about what a 30-year loan would earn for you in Switzerland. The figure is 1.85% for France at 30 years, compared with 1.9% in the UK and 3% in the USA. So there are some good things in the euro zone.

The slope of the interest rate curves is not steep. So investors aren’t expecting a very large increase in interest rates very soon. They may be right, they may be wrong.

Research : IPOs: the virtue of the greenshoe clause

With Simon Gueguen, Lecturer at the University of Cergy-Pontoise

The article that we discuss this month[1] looks at the greenshoe clause, an option used during IPOs. The term greenshoe comes from the eponymous children’s shoe company (now trading under the name Stride Rite Corporation) which used this technique for the first time in 1960. The technique involves a bank borrowing shares in an IPOed company from a shareholder (adopting a short position), generally around 15% of the amount of the IPO, and selling them in the IPO process together with the other shares sold. If the share price drops after the IPO, the bank buys shares on the market and closes its short position using those shares. If the share price rises, the bank exercises its greenshoe option (also called an overallotment) and the additional shares are issued or sold and acquired by the bank, which then closes its short position. All in all, the buyer or seller behaviour of the bank tends to stabilise the share price, and the additional shares are issued or sold if the market gives the IPO a warm reception.[2]

Greenshoe option have become commonplace over recent years. But do they really benefit the issuer? There are two possible theoretical responses.

The first notes that the greenshoe option could encourage the bank to undervalue the IPO price, in order to place more shares and thus receive a larger total fee. The greater complexity of the transaction is also advantageous to the bank which will be able to use its technical expertise to the detriment of the company.

The second suggests that this technique, by authorising an adjustment of quantities, encourages a more accurate and more stable IPO price. It is thus possible that the presence of a greenshoe option is a good thing for both the bank and the company.

Seeking to measure the consequences of the greenshoe option for the company’s existing shareholders, Jiao et al use a data base of IPOs on the Japanese market between 1997 and 2011. What makes this data base interesting is that greenshoe options have only been authorised in Japan since 2002, becoming systematic only from March 2010. It is thus possible to factor into the study the endogenous nature of the decision to include a greenshoe option[3].

The inclusion of a greenshoe option is more frequent when the amount of shares listed is high, when it represents a large part of the company’s capital and when the transaction includes a large portion of existing shares. These are situations in which existing shareholders are seeking liquidity. Another advantage of the greenshoe is that the additional shares listed in the event of exercise may come from existing shareholders, even if the latter are subject to a lock-up clause (prevented from selling during the period that follows the IPO). In other words, the greenshoe partially cancels out the lock-up clause and brings additional liquidity. This is why greenshoe options are more frequent in cases where there is a lock-up clause.

With regard to the consequences of the greenshoe option, Jiao et al measure whether the inclusion thereof, all other things being equal, lead to an increase in banks’ fees. The effect is negligible economically (less than 1% of the total). However, the inclusion of this option results in a very clear reduction in the undervaluation of the share being IPOed[4]. Here, the effect is around 25% and is must larger than the fees effect. Moreover, the share’s medium-term performance (6 months after the IPO) tends to be better, and volatility is lower, in the case of a greenshoe. The article thus concludes that the greenshoe option is a good thing for the issuer.

This article shows that the systematic inclusion of a greenshoe option in IPOs today is not to the detriment of issuers. On the contrary, it seems that this price stabilising mechanism reduces the total cost of the IPO. So this goes some way towards responding to the polemic sometimes raised by the exercise of greenshoes which are seen as (too) favourable for the bank that is carrying out the transaction.


[1] Y.Jiao, K.Kutsuna et R.Smith (2017), Why do IPO issuers grant overallotment options to underwriters?, Journal of Corporate Finance, vol.44, pages 34 to 47.

[2] For more on greenshoes, see chapter 25 of the Vernimmen.

[3] In the USA, greenshoe options have been used systematically for a very long time, which explains the choice of the Japanese market for the study.

[4] This undervaluation is measured by the share’s performance on the day of the IPO.


Q&A : What is structural subordination?

Subordination can be contractual when it is included in a contract which provides that a given debt may not be repaid before another given debt has been repaid beforehand. It can also be structural when it is the result, not of a contract, but of the group’s structure.

So, in cases where subsidiaries have large debts with third parties, lenders at the level of the parent company are exposed a special risk that will be taken into account by the rating agencies. 

The subsidiaries’ lenders will have direct access to the assets in the event of liquidation and the parent company’s lenders will be mechanically subordinated. They will in fact only be able to recover part of the value of these loans after the subsidiaries’ lenders have recovered what is due to them. This is called structural subordination which becomes more serious very rapidly when the risk of bankruptcy increases.

New : Comments posted on Facebook

Regularly on the Vernimmen.com Facebook page[1] we publish comments on financial news that we deem to be of interest. Here are some of the comments published over the last month.

How long has PwC audited the accounts of Goldman Sachs?

For 91 years, since 1926, sic. Certainly it can be assumed that it is no longer with the same members of PwC. . . As for Wells Fargo, it has been 86 years since it was faithful to KPMG (since 1931), a recent auditor of Citigroup, for only 48 years (1969). Deloitte is a bit new to Morgan Stanley (20 years since 1997).
This could change under influence. . . of European regulations which impose a change every 20 years and to open the auditors' mandate to competition at least every 10 years.
This is rather healthy, at least in principle, to have an outside look that changes regularly and a competition that can play truly. The most virtuous countries in governance are not always those who claim to be.

Report sine die of the Aramco IPO?

This is what the Financial Times thinks and it is not a huge surprise. Which investor, in a pure financial logic, would like to become shareholder of a group with a small free float (5 to 10%), alongside an undemocratic state with decision-making processes at least opaque and with a monstrous conflict of interest because of the oil tax, by far the first resource of Arabia? Not to mention a sector which, due to the energy transition, may have to leave a significant part of its assets in the ground forever. Last but not least, the difficulty of carrying out a process of making Aramco a truly independent public company offering normal guarantees of internal governance and reporting at a point in the stock market cycle of which we take no great risk to say that it is closer to its end than its beginning.

Richard Thaler, 2017 Nobel Prize for Economics

Behavioral finance is thus rewarded through one of its pioneers. Even if its applications in corporate finance are limited to this day, behavioral finance has figured prominently in the Vernimmen for years (chapter 15 for those who would have forgotten).
It will be noted that he teaches in Chicago like Eugene Fama, designer of the theory of efficient markets and Nobel Prize in 2013 for views very different from those of Richard Thaler. A demonstration of how academic freedom within the same institution is a real source of wealth. Michael Jensen has not (yet) received the Nobel Prize (for his work on agency theory and governance) among the founding fathers of today's finance.

So typically American!

Facebook had plans to distribute new non-voting shares to its shareholders allowing Mr. Zuckerberg to continue to sell Facebook shares without falling below 50% of the voting rights; a level of votes he gets thanks to a third category of shares with 10 voting rights per share. The country that prides itself on being the champion of good governance is far from putting it into practice at home (one seventh of the S & P 500 members have several classes of shares with your different rights).

Shareholders took the case to court and Facebook has just withdrawn its project. If in France, everything ends with songs according to the Canadian proverb, in the United States, it is with a trial that everything ends!

To get the week off to a good start, remember that since July 2017 the SP500 index manager no longer accepts multi-class stock companies in its index; even if those already in the index can keep them in place.

Have a nice day (with songs and without being sued if possible ...).

Toshiba sells its memory chip division to a consortium led by Bain Capital for $ 19 billion

LBOs of this size have not been seen for a long time, the big LBOs in recent years have been around $ 5 billion. But we are very far from historical records, since for example the 10th largest LBO in history was $ 27 billion in size.

At 3.5 times EBITDA, the price reflects both the situation of a seller forced to sell to escape the worst (due to its losses in its nuclear division) and a cyclical sector with heavy investments that makes that only a fraction of EBITDA is available to service debt. Its listed Asian competitor, SK Hynix, is currently valued 3 times EBITDA.