European Bond Documentation and Ineffective Negative Pledges
Negative Pledge provisions are designed to prohibit or restrict the amount of secured debt the bond issuer and/or the corporate’s group can incur. In bond documentation, an effective Negative Pledge clause provides some protection for the unsecured bondholder against event risk. However, as the current recessionary environment has highlighted, when distressed credits have required immediate liquidity, companies have accessed funds from (often) relationship banks which are willing to lend on a secured basis. Where such lending has been implemented within the intended limits of negative pledge provisions, this ‘event risk’ may still affect the credit rating to a measured extent. Where such lending is beyond intended limits, this may have stark implications for the ultimate recovery prospects of unsecured bondholders. Similarly, fresh lending at a level within the group structure which subordinates the existing unsecured bondholders (or ‘structural subordination’) can also lead to a “ratings cliff”, where unsecured bond ratings can be driven down multiple notches. At this point, seemingly protective documentation provisions have been exposed as ineffective. The rating then becomes more reflective of rate of recovery analysis coupled with any weak bond provisions.
Within Fitch’s European corporate universe, dramatic recent examples include Royal Ahold (‘BB-‘), British Energy (‘D’), and TXU Europe (‘D’). An ineffective Negative Pledge provided little protection for Capital Shopping Centres’ unsecured credit rating whereas Green Property’s heavily-covenanted bond documentation captured the new owner’s secured borrowing within the group thereby triggering the bondholders’ put option mechanisms.
Since many ineffective clauses still exist in corporate bonds and MTN programmes, rating cliffs will continue to compound negative rating actions for distressed companies. As a whole, solid investment grade companies do not intend to exploit ineffective Negative Pledge clauses, and often cite them as the reason why their current funding should not be secured, but many an MTN programme or set of bond documentation allows secured funding to happen, when required. Investors are tolerant of loose covenant language when the issuer has a solid investment grade rating, and when liquidity is strong (like now), but this tolerance can be costly particularly when corporate activities (acquisitions, change of ownership or strategy) cause downgrades. At a time when sterling investors are requesting more covenant protection, this report highlights the weakness of certain Negative Pledge clauses and clauses intended to prevent structural subordination both in the UK and in the rest of Europe and clarifies Fitch’s remit in relation to documentation issues. Overall, continental European bond issuers often offer a much weaker covenant package than their UK counterparts.
Rating agencies typically do not incorporate the weakness or strength of covenant packages within the ratings assigned to investment-grade unsecured corporate debt. That weakness or strength may come to the fore when the company is, or progresses towards, non-investment grade territory and/or its capital / funding structure begins to change. The assignment of the issuer’s senior unsecured rating to its senior unsecured bonds is conceptually based on the capacity of the issuer to service its outstanding senior financial obligations.
There are exceptions, most commonly, where the bond documentation contains explicit subordination (where the bonds would typically be identified as such and rated at a level below that of the issuer’s other unsecured debt).
Typically, after having reviewed the bond documentation, Fitch applies the senior unsecured debt rating of a company to its individual senior-ranking unsecured bond issues. Fitch has also added value to investors by frequently commenting on certain provisions in bond documentation (including change of control, rating triggers, meaningful financial covenants, disposal of assets restrictions, etc.) as per Green Property, Taylor Woodrow, and Sistema. In particular, where documentation has changed compared with a previous issue, Fitch has flagged meaningful changes (Casino, Guichard- Perrachon). The recent press release on Spanish infrastructure provider Abertis S.A. highlighted the potential weakness of the domestic debt programme’s structural subordination-related provisions compared with the current group structure, a factor currently mitigated by management’s public statements on financing policy.
When a corporate event takes place which can cause a prepayment of certain debt, where Fitch is aware of these provisions, the rating may be affected in so far as the provision imposes a strain on the company’s financial or operational flexibility, in attempting to meet the potential, perhaps unexpected, prepayment obligation (e.g. the demerger of Kesa Electrical from Kingfisher, the sale of Porterbrook by Stagecoach Group plc). Very rarely, bonds of the same ranking from the same issuer may be rated differently – as was the case with bonds issued by Prosiebensat1 Media AG. Here, following the downgrade of the company’s ratings from investment grade to subinvestment grade, change of control provisions in a new bond, which were absent in the issuer’s existing bond, were deemed to be of material benefit to the new bondholders. This notching differential reflected the greater protection afforded to the new bondholders by bond repurchase mechanisms, particularly when assessed against the backdrop of a deteriorating credit story compounded by the likelihood of an imminent change of ownership.
|
Documentation provisions which have affected the relevant bond’s rating compared with the senior unsecured rating of the corporate include:
There are also provisions which are designed to reduce ‘event risk’ but do not generally impact ratings of comfortably investment grade companies:
|
Fitch is obliged to accept the opinion of legal counsel and the relevant trustee (and occasionally a covenant-required “independent financial adviser”) as to whether or not a trigger event has occurred, although Fitch will often instigate a Rating Watch designation, together with an indication of the likely alternative ratings, should a potential event be material in scope.
Where bank funding includes other additional provisions, whether as a potential enhancement or a potential detriment relative to the senior unsecured rating, Fitch registers such mechanisms in its rating assessment but sometimes it can not publicly disclose the details of such provisions due to the private nature of these documents. It is Fitch’s view that transparency can be improved by publication of such data within the company’s financial information.
Covenant protection may be reflected in a corporate’s rating insofar as its affects all unsecured creditors. Such protection is more meaningful if reflected in long-term public bonds, rather than easily-refinanceable private bank debt, though even here, investors are reminded that a covenant to achieve a certain ratio is at best an indication of company policy, rather than a self-fulfilling prophecy. Obviously, if a breach, or a potential breach, of covenant results in an Event of Default, this does affect the rating. Fitch is aware of the benefit that a known (identifiable, usually relationship) syndicate of banks can have in waiving potential Events of Default versus a faceless, arguably, un-‘coordinatable’ group of bondholders.
Investors may think that financial covenants are set at levels to protect the rating – this is often not the case (as seen in the disclaimers within the reports on Anglian Water and Southern Water). Such financial covenants are usually set, with varying degrees of headroom, for other trigger or early-warning mechanisms.
Fitch can not second-guess whether loop-holes in Negative Pledge, Change of Business, or Disposal of Assets provisions will be utilised, or whether previously adopted group funding arrangements will be subverted through subsidiary-level funding. If Fitch did second-guess such ploys many of its ratings would be lower than the business and/or financial profiles justify. The agency can only warn investors of mechanisms that a currently comfortably investment grade company could exploit. Nor can Fitch assume in its ratings that the bond with the most restrictive covenants is bought back so that onerous provisions no longer apply to the whole group. The exposure of investors to the potential “ratings cliff” which can be thrown up by these issues is an element of risk in fixed income investment. In reality, it is almost impossible for documentation alone to eradicate the ‘event risk’ that is inherent in corporate creditworthiness. Where Fitch believes event risk on identifiable grounds is elevated (e.g. a management team is acknowledged to be aggressive in its expansion plans), management’s likely behaviour under such scenarios is reflected in individual company’s ratings.
Increasingly bonds have bondholder put options should a ‘change of ownership’, or ‘restructuring event’ (howsoever defined), and a ratings downgrade to non-investment grade occur as a result of the former event. Fitch is very aware that the issuer’s documentation has required the agencies to be clear, within a stated time-horizon, if the downgrade is related to the trigger event. There is a danger that only selected information may be provided as to, say, the acquirer’s intentions, thereby creating minimal or lots of uncertainty, which could result in a rating representative of a scenario that may or may not go ahead. Under these circumstances, Fitch’s remit is to remain objective, assess the information at hand, remain transparent as to its rating rationale, and stand apart from the potential gamesmanship of bondholders that do or do not want to negotiate something more than the contractual put of the bonds at par or other conflicting vested interests.
In relation to these clauses that use credit agency ratings for certain triggers, regardless of whether Fitch is specifically mentioned in the relevant clause, where Fitch rates the company it will continue to make its credit opinion known. Indeed, in the case of TXU Europe where Fitch downgraded the company a week before the other agencies, the Financial Times stated “Now the decision on whether or not a put option in the bonds should be triggered by the downgrade of Fitch . . . lies with an independent financial adviser”. The article, entitled TXU Shows Pitfalls of Covenants, then went on to quote a major UK sterling investor saying “A downgrade to junk by any of the agencies should be considered prejudicial given the likely effect on liquidity at TXU Europe”.
The remainder of this Special Report focuses specifically on the effectiveness of Negative Pledge and structural subordination-related provisions in documentation.
Ineffective Negative Pledge clauses can include the following weaknesses:
definitions of “relevant indebtedness” may incorporate other detrimental exclusions/ inclusions.
Structural subordination may be made easier if:
Where documentation tries to protect against structural subordination there are often provisions for principal subsidiaries to provide upstream guarantees. However, in relation to this there are a few issues:
Fitch has also noted that certain financing structures have achieved a similar net effect as secured funding to the detriment of the unsecured creditor, often working around conventional Negative Pledge provisions.
As an introduction, this drafting is representative of the typical parameters of a Negative Pledge.
“Cl.3. (a) So long as any of the Bonds remains outstanding… the Company will ensure that no Relevant Indebtedness of the Company or any of its Subsidiaries and no guarantee by the Company or any of its Subsidiaries of any Relevant Indebtedness of any person will be secured by any mortgage, charge, lien, pledge or other security interest (each a “Security Interest”) upon, or with respect to, any of the present or future business, undertaking, asset or revenues (including uncalled capital) of the Company or any of its Subsidiaries unless the Company shall, before or at the same time as the creation of the Security Interest, take any and all action necessary to ensure that:
i. all amounts payable by it under the Bonds, the Coupons and the Trust Deed are secured equally and rateably with the Relevant Indebtedness or guarantee, as the case may be, by the Security Interest to the satisfaction of the Trustee; or
ii. such other Security Interest or guarantee or other arrangement (whether or not including the giving of a Security Interest) is provided either (A) as the Trustee shall in its absolute discretion deem not materially less beneficial to the interests of the Bondholders or (B) as shall be approved by an Extraordinary Resolution…of the Bondholders,
save that the Company or any of its Subsidiaries may create or have outstanding a Security Interest in respect of Relevant Indebtedness and/or guarantees by the Company or any of its Subsidiaries in respect of Relevant Indebtedness of any person… where such Security Interest is in respect of a company becoming a Subsidiary of the Company after [the date of the bond issuance] and where such Security Interest exists at the time that company becomes a Subsidiary of the Company… and any Security Interest thereafter created in substitution for such original Security Interest, by the same guarantor as of such original Security Interest, which is created over or secured on assets whose value immediately prior to such substitution, in the opinion of the Trustee, does not materially exceed the then current value of the assets subject to such original Security Interest.”
However, the definition of “Relevant Indebtedness” allowed a carve-out of secured debt in the Domestic Currency:
“Relevant Indebtedness” means any present or future indebtedness… in the form of or represented by notes, bonds, debentures, debenture stock, loan stock or other securities offered, issued or distributed by way of public offer, private placing, acquisition consideration or otherwise and whether issued for cash or in whole or in part for a consideration other than cash, and which (i) with the agreement of the person issuing the same, are quoted, listed or ordinarily dealt in on any stock exchange or recognised over-the-counter or other securities market and (ii) are denominated or confer a right to payment in, or by reference to, any currency other than the Domestic Currency of the relevant issuer, but shall in any event not include Project Finance Indebtedness.;
“Domestic Currency” in respect of an issuer means the lawful currency for the time being of the country of incorporation of such issuer;
Consequently, this UK property company, with only UK assets, can raise secured listed debt in sterling, but not in a foreign currency. The exclusion of “Project Finance Indebtedness” also allows certain types of secured lending within the group.
When this company changed its funding strategy, and previous commitments to unsecured bondholders were said to be no longer applicable, increased levels of secured debt and fewer unencumbered assets led to the downgrade of CSC’s unsecured credit rating to non-investment grade territory. This subject then broadens out into concepts covered in Fitch’s “UK Property Companies’ Secured Debt & Joint Venture Funding”, dated January 2003.
Particularly in Continental European bonds, the raising of secured debt in the domestic currency is excluded, and/or access to secured bank debt is thought to be excluded as this type of funding is not traded or listed on a recognised exchange, and is not an over-the-counter instrument.
Domestic Currency exclusion: The following example is representative of many such clauses:- Carlsberg Finans A/S: “Cl.4… provided that the above restrictions shall not apply in respect of an issue of bonds, notes, debentures or other securities which (i) are denominated in Danish Kroner and (ii) are initially offered by or on behalf of the Issuer, the Guarantor or, as the case may be, any Subsidiary primarily to persons resident within the Kingdom of Denmark.”
Capital Markets Debt: In August 2002, Fitch highlighted to investors the three word change
(“whether or not”) in Casino, Guichard-Perrachon’s EUR4.0bn EMTN programme which weakened the Negative Pledge clause to exclude capital markets debt. The Relevant Indebtedness clause originally read:
“indebtedness for borrowed money whether or not represented by notes or other securities which are for the time being, or are capable of being, quoted, listed or ordinary dealt with on any stock exchange, over-the-counter-market or other securities market…”
This was changed to exclude bank debt:
“indebtedness for borrowed money represented by notes or other securities which are for the time being, or are capable of being, quoted, listed or ordinary dealt with on any stock exchange, over-thecounter- market or other securities market…”
Similarly, the debut Eurobond of Russian oil company OAO Tatneft in 1997 contained language which included only capital markets debt in its negative pledge language. Consequently, the company was able to raise USD500m in bank debt secured on export receivables. At the height of the 1998 Russian crisis, cash flow from these receivables was blocked by several of the lending banks, to the detriment of unsecured bondholders. In this case, exercise of the bank’s security directly impeded the timeliness of payments to bondholders, while payments still flowed to certain secured bank lenders.
When British Energy (“BE”), a UK nuclear power generation group, fell into financial difficulties in September 2002 and immediate liquidity was required, the UK government provided funding which progressed in size and status to secured lending.
By mid-September 2002 GBP650m of new funding was cross-guaranteed by principal group entities (which had been the case with existing bonds), but also secured by a pledge over materially all unpledged assets of the BE group (which had not been the case previously). Bondholders benefited in as far as the refusal of the government to provide support would have most likely seen BE enter administration, and the crystallisation of substantial long-term non-financial liabilities. However, the additional liquidity was not intended to repay nearterm debt but to support immediate incremental liquidity needs, largely for trading collateral.
Unsecured creditors found themselves with a stillsolvent issuer, but subordinated and facing low recoveries (albeit higher than under administration), estimated publicly by Fitch at below 50%.
BE’s Negative Pledge clause, in documentation common across all three bonds, was typical in many respects with two exceptions. One was very visible (billed as an “Exception to Negative Pledge”), which carved out any secured indebtedness not exceeding GBP250m, and with a maturity of not less than 15 years. Less visibly, and ultimately of more significance, the definition of Relevant Indebtedness contained implicit exclusions as follows:
“Relevant Indebtedness” means any present or future indebtedness… in the form of or represented by notes, bonds, debenture stock or other debt securities (whether issued in cash or in whole or in part for a consideration other than cash) which are transferable without any need to obtain the consent of the issuer thereof, but shall in any event not include Project Finance Indebtedness.”
Common to many Negative Pledge clauses, this clause includes “notes, bonds, debenture stock or other debt securities”, but does not specifically prohibit bank loans, or loans from the government or another entity. Even if this secured financing was captured under the listed instruments definition (which is itself open to wide interpretation), indebtedness that can be transferable with the need to obtain the consent of the issuer was excluded from BE’s definition of Relevant Indebtedness. In the event, the government’s loan did not qualify as relevant indebtedness and neither triggered the negative pledge, nor required consent.
In February 2003, when this company announced overstated profits in its US Food Services division and other concerns came to light, that week’s subsequent announcements referred to secured lending by Ahold’s relationship banks for the group’s near-term liquidity requirements.
The secured funding is now publicly disclosed as EUR600m and USD1.3bn of bank facilities “collateralised by a pledge of shares of Ahold’s significant Dutch and US subsidiaries. Ahold may use borrowings under the 2003 Credit Facility to refinance intercompany indebtedness, fund intercompany loans, provide for working capital and for general corporate purposes, including the issuance of letters of credit”.
Of publicly discloseable documentation Ahold is working within two restrictions. One set of documentation restricts bank secured debt (defined as Private Debt of the Issuer or the Guarantor) to 30% of the group’s total consolidated fixed assets. In Ahold’s EUR3bn EMTN Programme, the Negative Pledge reads thus:
“Cl.3 So long as any of the Notes or any relative Receipts or Coupons remain outstanding, neither Koninklijke Ahold N.V. nor any Subsidiary will secure any Public Debt or Private Debt, then or thereafter existing, by any lien, pledge or other charge upon any of its present or future assets or revenue unless the Notes and any relative Receipts and Coupons shall be secured by such lien, pledge, or other charge in the same manner. The foregoing shall not apply to (i) any security arising solely by mandatory operation of law, (ii) any security over assets existing at the time of acquisition thereof, (iii) any security comprised within the assets of any company acquired by or merged with the Issuer or the Guarantor or any Subsidiary where such security is created prior to the date of such merger or acquisition, (iv) any security over assets pursuant to the general terms and conditions of a bank … (v) any security approved by an Extraordinary Resolution of Noteholders, and (vi) any security arising out of contractual obligations entered into prior to [the date of the programme issuance].”
“Public Debt” means any loan, debt, guarantee or other obligation of the Issuer or the Guarantor represented by or securing bonds, notes, debenture or other publicly issued debt securities which are, or are capable of being, traded or listed on any stock exchange or other organised financial market.
“Private Debt” means loans, debts, guarantees and/or other obligations of the Issuer or the Guarantor in excess of 30 per cent. of the total consolidated fixed assets of Koninklijke Ahold N.V. and its Subsidiaries, not being Public Debt.
The outlined secured funding is allowed because the above Private Debt carve-out only applies to the Issuer (Koninklijke Ahold N.V, Ahold Finance B.V. Ahold Finance Europe B.V. and Ahold Finance USA Inc (each, an Issuer) and Guarantor (Koninklijke Ahold N.V in relation to the other Issuers).
In another set of bond documentation, for the US debt (issued by Ahold Finance U.S.A. Inc) whose issue is guaranteed by Koninklijke Ahold N.V. (Royal Ahold), the group is working within the Limitation on Liens applicable to Senior Debt Securities which stipulates that:
“Royal Ahold will not, and will not permit any subsidiary to, incur, issue, assume or guarantee any indebtedness for money borrowed or any other indebtedness evidenced by notes, bonds, debentures or other similar evidences of indebtedness for money borrowed (referred to in this summary as “debt”) if such debt is secured by pledge of, or mortgage, deed of trust or other lien on any part of its or any such subsidiary’s undertakings, assets or revenues (…”mortgages”) without effectively providing that the senior debt securities of all series issued under the indenture (and if Royal Ahold so determines, any other debt of Royal Ahold or such subsidiary then existing or thereafter created which is not subordinated to the senior debt securities) will be secured equally and ratably with (or prior to) such secured debt so long as such secured debt shall be so secured.
This restriction, however, will not apply if… all such secured debt which would otherwise be prohibited, plus all attributable debt of Royal Ahold and its subsidiaries in respect of “sale and leaseback transactions… would not exceed the greater of (1) USD750m and (2) 15% of consolidated net tangible assets.”
This clause appears to prohibit secured debt which falls within the group’s USD750m / 15% parameters. Presumably, Ahold has fitted its new secured funding into these provisions.
In the case of Ahold’s rating, a combination of the declining operating performance, uncertainties over the US Foodservices division (potentially fraud), a resultant management-less company, the appearance of secured debt and uncertainty as to future funding availability has caused the rating to fall from low ‘BBB’ territory to low ‘BB’ territory. This rating was underpinned by Fitch’s view that the group had a sound operating basis, but it has near-term refinancing issues, which management is seeking to address.
These latter mechanisms of lending at the OpCo level opens up the complementary, or alternative avenue to a negative pledge, namely structural subordination.
Just as secured debt subordinates the unsecured creditor, structural subordination has the same effect. “Structural subordination” takes place when, for example, lenders to the holding company (“HoldCo”) find themselves subordinated by external lending taking place at the operating company level (“OpCo”). The operating company lenders (whether secured or not) have a more direct claim on its assets; the holding company’s creditors (probably only as shareholders in the operating company) have a secondary claim.
Sometimes it is a group’s treasury policy that all lending is undertaken at one level – usually the HoldCo, or ‘parent/mother’ company, or TopCo level. In this case TopCo acts as a holding company and its main assets are mainly shares held in the OpCos. This is illustrated in Diagram A. Structural subordination can be limited if the following type of wording occurs within bond covenants:
“The Company will procure that so long as any of the Notes remain outstanding… the aggregate principal amount… for the time being outstanding of (a) all Secured Borrowings of the Company and the Subsidiaries and (b) all Borrowings which are not Secured Borrowings of Subsidiaries which are not or do not become Guarantor Subsidiaries… shall not exceed a sum equal to 50 per cent. of the [consolidated] Adjusted Capital and Reserves. Subject to the Trustee being satisfied as to the legal validity of the guarantee of the Notes given by such Subsidiary, the Company can procure, at any time, that any Subsidiary becomes Guarantor Subsidiary without the consent of the Noteholders or Couponholders.”
It is hoped that Secured Borrowings (as defined in an undisclosed Trust Deed) include bank lending in order to be effective. The ‘inner limit’ would be calculated as per Diagram B.
In Diagram B, even if OpCo 1 gave a guarantee to HoldCo (which its bondholders would benefit fro