The term hybrid product has been bandied about so much that it can be difficult to determine what it really means. As a generic term, it is used to define financings that do not easily fall under equity or debt. This hold-all category is much used in finance textbooks (including the Vernimmen[1]). So it covers convertible bonds, bonds with warrants, deeply subordinated debts[2], or their successors, exchangeable bonds (which is quite wrong as an exchangeable bond is clearly a debt), bonds redeemable in shares, etc.

What is quite obvious is that little by little, the distance between what happens in practice and what is referred to in academic terminology has grown. So today, if you talk to investors or issuers about hybrids, they are certain to understand “hybrid bonds” and will automatically think of some of the standard features of this product.

That being the case, let’s restrict ourselves in our analysis of hybrids as this semantic shift surely reveals what is really happening on the market. It does seem that over the past few years we have seen the emergence a type of bond issue that complies with certain standards (mostly set by rating agencies) which enable companies to improve their credit ratings. Common features for these bond issues are:

• subordination (in the event of liquidation they are repaid after other debts but before equity);
• very long maturity, even perpetual;
• a coupon (fixed then at a floating rate) that may be deferred in the event of dividends not being paid to shareholders;
• undertaking by the company in the event of early repayment to replace this issue with the issue of products of the same type or of shares;
• increasing remuneration over time with step-ups at certain dates.

Let’s illustrate this with a concrete example of the issues made by Total in 2015:

The bond pays a coupon initially fixed at 2.25%. Total has a first repayment option after 6 years, i.e. in 2021. If the bond is not repaid, the coupon becomes floating (risk-free rate + margin of 1.861%). A second option arises 5 years later, if it has still not been exercised, the margin increases by 0.25%. And again 15 years later with an increase of 0.75%. This issue does not make provision for an mandatory payment date for Total. 

Given their features which are substantially riskier than plain vanilla bonds, agencies rate these bonds 2 to 3 notches below senior debt. The corollary of this risk (highlighted by the rating) is that the yield offered is significantly higher than that of plain vanilla bonds.

Such issues provide a form of security for the company’s other creditors as hybrids have a very long maturity. They will in fact only be repaid after the other debts. Accordingly, rating agencies reclassify them partially as equity when calculating their ratios (mostly 50% debt – 50% equity). Getting hold of this equity credit is indeed the main motivation of issuers.

But issuing hybrids also makes it possible to appeal to specific, specialised investors, which would not have purchased standard bonds. Such investors are looking for high yields and are prepared to take on a certain amount of risk.

Volumes of hybrid issues have increased significantly in Europe since the beginning of the 2010s:

The obligation (or rather the undertaking given to rating agencies) to replace existing issues with new issues of the same type suggests a bright future for the hybrids market, which is becoming, de facto, a recurrent market – issuers who have issued such products once are highly incentivised to issue again every 6-7 years. For volumes observed since 2013, we see a mix of replacement issues and new issuers. Equity credit has become a sort of mildly addictive drug for these issuers!

However, the product is based largely on a sort of misunderstanding.

For rating agencies, this product has features that are clearly half-way between debt and equity, especially with their very long maturity periods. Contractually, it provides comfort to senior lenders by introducing an additional buffer.

On the other hand, investors buy hybrids on the understanding that they will be reimbursed on the exercise of the first call option.  There is a tacit undertaking on the part of the company that it will exercise its call option. In practice, what happens if it doesn’t do so? Nothing immediately, but it is likely that investors will feel as if they’ve been deceived and the company will be cutting itself off from this market. Investors will also feel comforted by the probable reimbursement based on the technical motivations that the issuer has to reimburse:

• the step-up which makes the issue worth more;
• the fact that issues are structured so that the rating agencies cancel the benefit of the equity content on the date of the first call option (considering that the residual economic life then becomes too short, i.e. less than 20 years until the major step-up).

In practice we see that companies exercise their call options, even when economically this is not always optimal, but there are exceptions. Südzucker, for example, made the choice of deferring reimbursement as it was contractually entitled to do. This decision had the advantage of reassuring the rating agencies that their approach was correct. 

As the rationale is largely owed to handling by rating agencies (but also tax treatment), the fear of a change in the dogma or the tax treatment represents a deterrent for certain issuers. Issue related documentation also contains a right for the company to reimburse the hybrids in the event of a change in terms of accounting, tax treatment or handling by rating agencies.

This is not totally theoretical. For example in 2013, Moody’s changed its approach by removing equity credit from non-investment grade issuers[3]. This was a double penalty for companies that lost their investment grade status because they also lost part of their equity in the view of Moody’s (this was the case for ArcelorMittal for example which chose to reimburse its existing issue).

Hybrids present an attractive opportunity for investors. In a world of low interest rates and narrow credit spreads, they generate a relatively high yield that is relatively high while investing in blue chip companies. So for EDF, the investor will receive 2.625% on the hybrid to be reimbursed in 10 years, compared with 0.99% on the conventional bond with a maturity fixed at 10 years. Accordingly, this is a good opportunity compared with high yield bonds, the issuers of which have a much lower credit rating.

So hybrids rely on a subtle balance between constraints imposed by rating agencies and the comfort provided to investors on the probability of reimbursement. But, we’re talking about real debts, which are risky, but still debts, and for the purposes of valuation, all hybrids should be included under net debt. 

There are some products that may create the illusion of fitting the mould that we have described for hybrids – a subordination, a long maturity, a step-up. But this is just an illusion. For example “ORDINANEs “: obligations à durée indéterminée à option de remboursement en numéraire et/ou en actions nouvelles et/ou existantes (perpetual bonds with an option for reimbursement in cash and/or new and/or existing shares), of which we’ve seen a few issues in France, resemble hybrids. The main differences are that they can be converted before the call option and that the step-up is nothing like that of hybrids (up to 8%) In reality, these products are really only convertible bonds with an added gimmick (the apparently limitless maturity) just so that they can be booked under equity in IFRS.  This amounts to intellectual imposture since the step-up is done so that the convertible bond is reimbursed in cash if it is not converted into shares. The rating agencies will not allow themselves to be duped and these instruments are reserved for unrated issuers.

So, real hybrids have become a separate class of assets with its own investor base and specific features. But their future lies in the hands of the rating agencies. If they could stabilise their analysis, this market could become a very significant compartment for bonds, as is already the case for banks and financial institutions.

On its way up for issuers rated by Moody’s: 1.66%; still a far cry from levels last seen in 2008 (2.5%) and 2009 (5%), but the collapse of oil prices is hitting hard some US issuers in the shale gas and oil sector

With Simon Gueguen, lecturer-researcher at the University of Paris-Dauphine

Empirical studies on capital structure often yield results that are difficult to interpret[1]. The different variables tested relating to the company (size, sector, valuation ratios, etc.) do not provide much of an explanation. The results of this study show the strong presence of “fixed effects”, which indicate that factors specific to each company explain the heterogeneity of capital structures observed. A frequent conclusion is that the capital structure of each company tends to remain stable[2]. The study whose results we present here[3] shows, on the contrary, that over the long term, the capital structure of each company is highly unstable.

What is measured in this study is the stability of the position relative to the capital structure (i.e. financial leverage) of each company. Do companies whose financial leverage is the highest on a given date have a greater chance of displaying even higher leverage in later years? DeAngelo and Roll’s answer to this question is “no”.  Stability of financial leverage of every company is only observed over the short term.  The study mainly covers US firms (non-financial firms) between 1950 and 2008. A number of tests and different ways of measuring the stability of capital structure are carried out.  A noteworthy result is the result obtained for companies that could be observed for at least 20 years. If we divide these companies into quartiles in accordance with their financial leverage, 69.5% of them are present in at least 3 different quartiles over the period observed. In other words, over 20 years, stability is the exception and instability the rule. 

DeAngelo and Roll show that there are periods during which companies may display a capital structure that is relatively stable. These periods are short and are generally characterised by low levels of debt. Frequently, the end of these periods of stability is marked by sharp growth. The typical example cited by DeAngelo and Roll is that of the 1950s, when many companies abandoned a conservative approach to capital structure and took out debt in order to finance growth.

From an academic point of view, it is interesting to ask what models are the most appropriate for explaining the results of this study. Those favoured by DeAngelo and Roll are models in which companies have a target debt ratio, but which varies over time. On the other hand, two types of models are rejected:

• models with target debt zones that are fixed in time and that cannot explain instability over the long term;
• models without any target and with a random development of debt, which do not explain the (short) periods of stability.

There are of course many questions on capital structure to which the article fails to provide answers. The main one is probably the following: what are the determinants of a target debt level? Empirical study shows that variations in a company’s debt level after a period of stability are greater than variations in target debt based on sectorial medians. Comparables have an influence on financial policy (and capital structure)[4], but do not explain the extent of the instability. There is still a lot to be learnt on this topic!

When a company wants to buy back its shares[1], it can:

- Buy shares on the market

- Buy shares from a given shareholder

- Make a public offering to buy back shares open to all shareholders

- Distribute put warrants to all of its shareholders

1) Buy shares on the market

Listed companies can buy shares on the market. Depending on the country, buy-backs have to be authorised by shareholders and may be limited in volume (for example, a maximum of 10% of the shares every year or 18 months) and in price (a maximum share buy-back price is set). Furthermore, in most countries, they cannot represent more than a certain percentage of the average trading volume. Generally, the shares bought back will be cancelled but they can also be kept by the company (as treasury stocks) to be handed over in the case of an acquisition, for the exercise of stock options or for the conversion of convertible bonds. Treasury shares lose their voting right and their right to a dividend. They can also be used to enhance liquidity through a liquidity program implemented by a broker, or be resold after a buyback program.

To implement this program, the company can simply ask a broker to buy a certain number of shares per day (to the extent permitted). It may also give a mandate to a broker to buy a certain volume at a certain time. The broker will then look to buy in the best interests of the company without taking a commitment to deliver the securities exactly to the average volume weighted for the period.

The broker may also take a firmer commitment on the purchase price. Certain purchase algorithms (modulating the amount of securities purchased each day depending on the performance of the title) and a knowledge of the market of this share (about when shopping in the day) allow brokers to guarantee the company an average purchase price of shares less than the weighted average of the daily volume for the period. Brokers talk about buybacks "VWAP minus". This "bonus" is actually achieved mainly by allowing the broker flexibility on the duration of the repurchase program which allows it to determine its effective maturity.

Similarly, some companies offer to share the performance in order to align the interests of the broker and the issuer. For example if the weighted price for the period is 100 and the broker managed to acquire the shares at an average of 95, performance (100-95 = 5 here) will be distributed between the issuer (eg 85% or 4.25, the effective purchase price will be 100 - 4.25 = 95.75) and the broker  (in our example 15% or 0.75).

2) Purchase from a specific shareholder

A company may repurchase shares from one of its largest shareholders. This allows the issuer to organize the disposal or partial disposal of the shareholder. It is an alternative to a sale on the market of a block which could weigh on the price of the share. Thus in 2014, L'Oreal bought an 8% stake from Nestlé. Note that as this operation is not egalitarian, it must respect certain price constraints (market prices) to ensure fairness.

3) Public offer

In practice, the Board of Directors, using authority which must be given by the EGM, offers to buy all or part of the shares of all shareholders at a certain price and for a certain period (a few days to a few weeks). If too many shares are presented to the offer, the company will reduce redemption requests and cancel the repurchased shares. For example, in 2016, Alstom launched a share buyback offer on 29.5% of its capital: only 39.1% of the shares tendered were actually bought back because the amounts provided far exceeded the size of the offer. If, however, an insufficient number of shares is presented, the issuer acquires and then cancels all proposed shares.

Unlike the two types of transactions presented above which are made at the market price (or marginally below), Public offers are offered with a premium over the market price, most of the time between 10 to 15%.

4) Put warrant

In some countries, a share buy-back can be accomplished by issuing put warrants to each shareholder, several warrants giving the holder the right to sell one share to the company at a specified price. Such a warrant is a put option issued by the company.

Of the four tools described, this one is clearly used the least.

[1] On the why a company decides to buyback its shares, see Chapter 36 of the Vernimmen.