Greece debt swap Questions and answers

27 July 2011

Greeces debt swap with private-sector participation

A position paper published by the International Institute of Finance (IIF), a global association of financial institutions, offers an approach to private-sector participation in Greek debt swap that has the explicit support of 30 financial institutions. This raises important questions, which we answer in the following:

How big is the private sectors participation?

The private sectors participation is calculated based on assumptions about the amount of Greek government bonds privately held (par, not market values) and about the rate of participation in the proposed mechanism. The IIF expects participation equal to 90% of the private sectors aggregate holdings. According to the institute, the given maturity profile results in a contribution to Greeces financing by the private sector of EUR 54 bn between mid 2011 and mid 2014 and of EUR 135 bn altogether in the period from mid 2011 to the end of 2020.

Composition of the new Greece package

180 160 140 120 100 80 60 40 20 0 Privat sector contribution via bond buyback programm Privat sector contribution via bond exchange/bond roll-over programm Credit enhancements for bond exchange/bond roll-over programm EFSF-loans to Greece for bond buybacks Recapitalization of Greek banks Direct loans to Greece
Source: Thomson Reuters, Raiffeisen RESEARCH

Contribution EFSF/IMF

Analysts Gottfried Steindl, CIIA Julia Neudorfer

gottfried.steindl@raiffeisenresearch.at julia.neudorfer@raiffeisenresearch.at

How does the debt swap work?

Participating investors would have the following four options of voluntary bond exchange or rollovers: 1. Bond exchange into new 30 year Greek bonds issued at par with step-up coupon payments (4.0% for the first 5 years, 4.5% for the 5 years thereafter, and 5.0% for the remaining period to maturity). The new 30 year instruments are collateralized

Jrn Lange, CEFA Peter Onofrej

joern.lange@raiffeisenresearch.at peter.onofrej@raiffeisenresearch.at

Matthias Reith

matthias.reith@raiffeisenresearch.at

Special

by triple-A rated zero-coupon bonds. Those bonds are purchased by Greece (like the Brady bond model) using funds provided by the EFSF. 2. Rollover of maturing Greek bonds into new 30year instruments issued at par on the same conditions as in option 1 and with the same system of collateralization. 3. Bond exchange at a discount (80% of the face value of the old bond) into 30-year instruments issued at par with higher coupon payments compared with options 1 and 2 (6.0% for the first 5 years, 6.5% for the 5 years thereafter, and 6.8% for the remaining period to maturity), but with the same system of collateralization. 4. Bond exchange at a discount (80% of the face value of the old bond) into 15-year instruments issued at par with coupon payments of 5.9%. On this model, there is only partial collateralization (80% of the loss, up to 40% of the notional value of the bond). This collateralization is performed with funds that Greece (again financed by the EFSF) puts in an escrow fund. The IIF assumes in its proposal that the four options will be chosen equally (each 25% of regis-tered par value).

Also on option 4, the par value of the old bonds is exchanged into new bonds at a discount of 20%. The NPV of the coupon payments of the new bonds set at a maturity of 15 years and discounted at a rate of 9% amounts to about 47.6 on 100. Collateralization is achieved up to a maximum of 40% of the instruments notional value (at the beginning 40% of 80% = 32%). The total of the NPVs from the (maximum) collateralization and the coupons here again yields an NPV of about 80% (47.6 + 32), or the cited loss of about 20%. Supposing that the assumed rates are the market rates at the time of debt swap / roll-over, the new bond will be traded at close to 80 in case of option 1 and 2 and at 100 in case of option 3 and 4.

What are the risks for investors during/after debt swap?

The collateral in options 1 through 4 relates to the principal, not to the coupon payments. So, whereas a triple-A risk may be assigned to the principal, the interest payments are subject to Greeces sovereign risk. In case Greece defaults during the maturity period, the investor loses the scheduled interest payments. In options 1 through 3, the difference between the NPV of the zero-coupon bonds and the par value of 100 may be added to that. The collateral for the principal must first be amortized over the maturity period; only after 30 years (upon final maturity) is the full amount guaranteed (provided the AAA-rated issuer is solvent). At the beginning, the substitution rate on the principal is thus about 33.6%, which gradually rises to 100% by the end of the maturity period of the zero-coupon bonds. When the bonds are rolled over (option 2), it is not clear whether investors will still bear Greeces full sovereign risk between the time of declaring their commitment and that of the rollover or will already have a claim to the principals collateral on the basis of the commitment undertaken. After all, many years may pass between the voluntary commitment to a rollover and the bonds maturing. It is also unclear whether the instrument will remain tradable during that time inclusive of or separated from the rollover commitment. In option 4, the collateral does not consist of zerocoupon bonds, but rather presumably of interest-bearing instruments with high credit ratings (probably also triple A). The advantage is that the collateral need not be amortized first. The disadvantage of this option is the lower substitution rate in case of a de-

How do the losses work out by the NPV method?

On options 1 and 2, the old bonds are exchanged or rolled over at the full par value into the new bonds. The net present value (NPV) of the coupon payments (average coupon of 4.5%) is discounted at a rate of 9%, which results in a price value of about 46.2 on 100. The zero-coupon bonds function as a kind of lump-sum repayment vehicle to which the investor has a claim in case Greece defaults. These bonds have a NPV of about 33.6% (100 discounted at a rate of 3.7%) at the beginning. The total of the NPVs (46.2 + 33.6) comes to just under 80%, which corresponds to the cited discount of about 20%. On option 3, the par value of the old bonds is exchanged into new bonds at a discount of 20%. The NPV of the coupon payments of the new bonds (average coupon of 6.42%) discounted at a rate of 9% amounts to about 66 on 100. If one takes the discount into account and adds the NPV of the collateral provided (zero bonds), the result is again an NPV for the new instruments of just under 80% (66 - 20 + 33.6), or the cited loss of about 20%.

Special

fault (at most 40% of the notional of the old bond which gives at most 32% of the face value of the old bond). Whether the new bonds will be tradable is not discussed in the position paper. We, however, see no obstacles to a secondary market for the new instruments.

selected option, past and future accounting practices, and the market environment at the time of the exchange and accounting. We currently assess the situation as follows. Options 1 and 2 represent an exchange of the old bonds into new bonds at 100% of par value. In theory, that should mean no need of writedowns. However, the conditions of the new bonds are significantly worse than interest usually paid on the market. Consequently, a significant price drop may be expected already on the first trading day. That will correspond roughly to the NPV loss of 21% under the assumptions in the IIF proposal. How this price loss is handled, however, depends on the accounting environment. If the new bonds are booked in their trading accounts, banks will have to adjust the book values quickly to market values and write off the difference through the profit-and-loss account (P&L). But if the new bonds are booked in the holdto-maturity (HTM) banking accounts, a valuation adjustment will only be necessary if a permanent value loss has occurred. However, we believe that is a difficult argument to make considering the collateralization of the bonds par value by a high-quality zero bond portfolio. If the new bonds are booked in the available-for-sale (AFS) banking accounts, no valuation adjustments will be necessary through P&L, but the difference between book and market values will have to be reflected in the revaluation reserve and hence in equity. In options 3 and 4, banks only receive new bonds equivalent to 80% of the old bonds par value. We therefore believe the difference relative to book value should be offset directly through P&L. If the old bonds were booked in the trading accounts, at current market prices this would probably mean writeups and hence a gain. Bonds in the HTM accounts, on the other hand, are likely to incur losses in most cases, while those in the AFS accounts should be capitalneutral. The losses recognized in P&L will replace the unrealized losses in the revaluation reserve in the statement of equity. Since the new bonds are likely to be much closer to the market in these options, we do not expect appreciable price losses for them in the near term under the IIF assumptions.

What rating will Greece get and will CDSs be triggered by the debt swap?
According to Fitch, the restructuring program will result in Greeces being rated in restricted default. The bonds affected by the exchange/rollover will be rated in default as soon as the offering period for the four proposed options ends. However, the default ratings are to remain in place only for a short time until the actual exchange process begins. The default will be cured with the issuance of the new debt instruments to the participating investors, and Fitch will give new ratings to both Greece and the outstanding bonds. According to available information, Fitch will classify Greece and its debt instruments in the low speculative category. According to initial statements by the International Swap and Derivative Association (ISDA), credit default swaps (CDSs) on Greece will not be triggered. The offered debt swap is no credit event. However, the relevant committee will not submit its final verdict until the offer becomes binding. According to current releases, the critical factor for the preliminary judgment is the voluntary status for participation in the debt swap. Moreover, from a legal standpoint, what happens is an exchange of old bonds for new ones without any change of the terms associated with the old bonds. A Change in terms is called restructuring which constitutes a credit event.

How do the losses calculated by the NPV method affect bank balance sheets?
The downgrade of Greece to default might necessitate an entry in the balance sheets for the old bonds. For example, the Institute of Public Auditors in Germany (IDW), according to a press release of 19 July 2001, considers taking an impairment charge necessary in the case of all the approaches discussed ahead of the summit meeting. The new bonds should, in our view, provide a reference point for the level of provisioning and writedowns. However, the precise amount of correction needed will depend on the

What does this mean for the banking sector?

The burdens arising from collateralization would be enormous for Greek and Cypriot banks, and recapitalization of those banks seems likely. For other European banks, the effects are less clear. Possible burdens arising from the banking accounts stand in contrast here to writeups from the trading accounts. 3

Special

The burdens are likely to predominate for the European banking sector. Set against that, however, are mostly profits from operating business and capital reserves. As long as a contagious spread to other periphery states can be avoided, this should not cause any major banking institution financial distress.

financing framework of EUR 109 bn. The target discount is therefore likely to be between 30% and 40%. Altogether, the sought debt reduction of EUR 26 bn offers some relief, but it is certainly not a big step toward making the Greek national debt sustainable (the debt level stands at about EUR 360 bn). That is supposed to (and must, in our opinion) take place by way of other measures. One thing to mention is cutting current deficits. Another is making government more efficient and industry more productive with extensive assistance from the EU (a long-term process), so the economy returns to a sustainable growth path. Finally, the high debt level must be made sustainable by means of a reduced interest burden.

What will happen if an investor does not participate in the voluntary pro-gram?
So far, the restructuring program has been presented to the financial sector as a voluntary offer, to which 30 well-known institutions have given their support. There is no plan for compulsory participation by all holders of government bonds. Nevertheless, the target participation rate is set very high at 90%. However, the free rider incentive is also very high, especially for holders of short-date Greek government bonds. If an investor allows an old bond simply to run out, he will receive, according to current information, the claims arising from the bonds in full (coupons and par value at 100%).

How will Greeces interest burden change?

How high is the interest load now?
Since most outstanding government bonds were already issued before the outbreak of the debt crisis last year, the coupon payments to be made on them by the Greek government appear very low compared with present yield levels. Of the debt instruments (bonds and notes) maturing by 2020, 33.4% have a coupon of at most 4.5%, which corresponds to the average interest rate of options 1 and 2 (excluding collateralization costs). Another 23.1% of the debt instruments maturing by 2020 have a coupon that is more than 4.5% but at most 5.9% (interest rate of option 4). Altogether, more than half (56.4%) of the debt instruments have a coupon that at most matches the average interest rate in option 4. Greece was thus able in the past to finance itself at relatively favorable conditions. On the other hand, only 12.5% of the securities to be serviced in the period to 2020 have a coupon that falls in the range between 5.9% and 6.42% (average interest rate in option 3).

How will Greeces debt profile change?

Based on the IIFs assumptions, the bond exchange will result reduce debt by EUR 13.5 bn (about 50% of the exchange is expected to occur via options 3 and 4 = EUR 67.5 bn; a 20% discount on that amount results in the above-mentioned EUR 13.5 bn). However, Greece is buying this debt reduction at the cost of a higher interest burden. Another debt reduction of EUR 12.6 bn is expected to result from a bond buyback facility. This facility will probably be financed by the EFSF; its aim will be to purchase outstanding bonds below par value. The problem is that the more prices on the Greek bond market advance, the more expensive this project becomes. Therefore, the bonds might be purchased to a lesser extent from private investors and to a greater from the European Central Bank (ECB). The ECB could directly determine the price discount with the EFSF, and this could ward off market speculation on price gains. The lower limit for the discount on par value may be put at about 20%; otherwise, these investors would be better off than they would be with an exchange/ rollover. But if one sets the price reduction on par value at 25%, this program will be very expensive at about EUR 50 bn and would probably not find a place in the second Greece packages announced

Coupon payments on Greek bonds

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% Share of EUR-denominated outstanding Greek bonds 1% < Coupon 4.5% 5.9% < Coupon 6.42% Others (Zero, Floating, etc.)
Source: Thomson Reuters, Raiffeisen RESEARCH
1% < Coupon 4.5% Others (Zero, Floating, etc.) Coupon > 6.42% 5.9% < Coupon 6.42% 4.5% < Coupon 5.9%

4.5% < Coupon 5.9% Coupon > 6.42%

Special

These favorable financing conditions are also reflected in the implicit interest rate (this years interest payments as a percentage of last years national debt), which the EU Commission estimates at only 4.5% in 2011.

What interest rates result from the four options of bond exchange/rollover?
Interest rates on EFSF loans are now at about 3.5%. For Greece, this is the cheapest source for funding. The interest rates offered in the bond exchange/rollover average 4.5% (options 1 and 2), 6.42% (option 3), and 5.9% (option 4). However, additional interest costs arise for Greece on options 1 through 3 due to the collateralization. Every EUR 100 from options 1 through 3 is to be collateralized with an investment of about EUR 33.6 in zero-coupon bonds. This amount must be additionally borrowed from the EFSF, for which an interest rate of currently 3.5% would apply, or EUR 1.18. The interest rates on options 1 through 3 thus increase for Greece due to collateralization costs by about 1.18 percentage points. Hence, applying the debt swap the interest load for Greece will actually increase, but the rates offered are way below market conditions.

At the same time, however, the current Greece package sets a precedent for debt restructuring within the European Monetary Union, although such a result was still vehemently ruled out by policymakers and the ECB just a few months ago. One therefore has to wonder how credible the latest political promises are and whether Greeces debt swap will actually remain an exception if the economic and budgetary situation in Portugal or Ireland develops worse than expected.

What does this solution of the Greek problem mean for other EU periphery states?
First, the easing of ESFS lending conditions (maturity lengthening and interest rate reduction) is worth mentioning because it benefits not only Greece, but also Ireland and Portugal. This step is especially welcome because of the critical influence of the interest rate level on a countrys long-term solvency. In particular, considerable interest relief of about 200bp accrues to Ireland, which has had to pay the highest interest rates on EU loans due to EU disputes over low Irish taxes on corporations. In this connection, rating agency Fitch has already emphasized the positive effects of the lower EFSF lending rates now also applicable to Ireland and Portugal on the debt dynamics of both countries. It is also positive for the future ratings of the two countries that EU policymakers are very concerned that the current debt swap of Greece is presented as a one-time special case within the euro area. The wording of the EU resolution (exceptional and unique solution) suggests that EU policymakers are not now considering using the Greece package as a blueprint for other periphery countries. The reception of the EUs declaration of intent by the rating agencies has thus been correspondingly positive.

Special

Special
Editor
Raiffeisen RESEARCH GmbH A-1030 Vienna, Am Stadtpark 9 Tel: +43 1 717 07-1521 Head: Peter Brezinschek (1517) Research Sales: Werner Weingraber (5975)

Acknowledgements

This report was completed on 29 July 2011

Economics, Fixed Income, FX: Valentin Hofsttter (Head, 1685), Jrg Angel (1687), Wolfgang Ernst (1500), Gunter Deuber (5707), Julia Neudorfer (5842), Matthias Reith (6741), Andreas Schwabe (1389), Gintaras Shlizhyus (1343), Gottfried Steindl (1523), Martin Stelzeneder (1614) Credit/Corporate Bonds: Christoph Klaper (Head, 1652), Christoph Ibser (5913), Igor Kovacic (6732), Martin Kutny (2013), Peter Onofrej (2049), Gleb Shpilevoy (1461), Alexander Sklemin (1212), Jrgen Walter (5932) Stocks: Helge Rechberger (Head, 1533), Aaron Alber (1513), Christian Hinterwallner (1633), Jrn Lange (5934), Hannes Loacker (1885), Richard Malzer (5935), Johannes Mattner (1463), Christine Nowak (1625), Leopold Salcher (2176), Andreas Schiller (1358), Connie Schmann (2178), Magdalena Wasowicz (2169) Quant Research/Emerging Markets: Veronika Lammer (Head, 3741), Mario Annau (1355), Lydia Kranner (1609), Nina Kukic (1635), Albert Moik (1593), Manuel Schuster (1529) Technical analysis: Stefan Memmer (1421), Robert Schittler (1537)

This document does not constitute an offer or invitation to subscribe for or purchase any securities and neither this document nor anything contained herein shall form the basis of any contract or commitment whatsoever. This document is being furnished to you solely for your information and may not be reproduced or redistributed to any other person. Any investment decision with respect to any securities of the respective company must be made on the basis of an offering circular or prospectus approved by such company and not on the basis of this document. RBI may have effected an own account transaction in any investment mentioned herein or related investments and or may have a position or holding in such investments as a result. RBI may have been, or might be, acting as a manager or co-manager of a public offering of any securities mentioned in this report or in any related security. Information contained herein is based on sources, including annual reports and other material which might have been made available by the entity which is the subject of this document. RBI believes all the information to be reliable, but no representations are made as to their accuracy and completeness. Unless otherwise stated, all views (including statements and forecasts) are solely those of RBI and are subject to change without notice. Investors in emerging markets need to be aware that settlement and custodial risk may be higher than in markets where there is a long established infrastructure and that stock liquidity may be impacted by the numbers of market makers which may therefore impact upon the reliability of any investments made as a result of acting upon information contained in this document. Special regulations for the Republic of Austria: This document does not constitute either a public offer in the meaning of the Kapitalmarktgesetz (KMG) nor a prospectus in the meaning of the KMG or of the Brsegesetz. Furthermore this document does not intend to recommend the purchase or the sale of securities or investments in the meaning of the Wertpapieraufsichtsgesetz. This document shall not replace the necessary advice concerning the purchase or the sale of securities or investments. For any advice concerning the purchase or the sale of securities or investments kindly contact your RAIFFEISENBANK. Special regulations for the United Kingdom of Great Britain, Northern Ireland (UK) and Jersey (Channel Islands): Raiffeisen Bank International AG. This publication has been either approved or issued by Raiffeisen Bank International AG (RBI) in order to promote its investment business. RBI London Branch is authorised by the Austrian Financial Market Authority (FMA) and subject to limited regulation by the Financial Services Authority (FSA). Details on the extent of the London branchs regulation by the Financial Services Authority are available on request. This publication is not intended for investors who are Retail Customers within the meaning of the FSA rules and should therefore not be distributed to them. Neither the information nor the opinions expressed herein constitute or are to be construed as an offer or solicitation of an offer to buy (or sell) investments. RBI may have effected an Own Account Transaction within the meaning of FSA rules in any investment mentioned herein or related investments and or may have a position or holding in such investments as a result. RBI may have been, or might be, acting as a manager or co-manager of a public offering of any securities mentioned in this report or in any related security. The RBI Jersey marketing representative office is not regulated by the Jersey Financial Services Commission as it does not perform any financial services activity in Jersey as defined by the Financial Services (Jersey) Law 1998 (FSJL).Special regulations for the United States of America (USA) and Canada: This document or any copy hereof may not be taken or transmitted or distributed, in the USA or Canada or their respective territories or possessions nor may it be distributed to any USA-person or person resident in Canada by any means other than via a US Broker Dealer. Any failure to comply with these restrictions may constitute a violation of USA or Canadian securities laws. Detailed disclaimer and disclosure as per Sect. 48f of the Stock Exchange Act: http://www.raiffeisenresearch.at -> Disclaimer