Copying and distributing are prohibited without permission of the publisher.


Covered bonds through Covid: the French issuers’ perspective

By Bill Thornhill
10 Jul 2020

Funding officials from Caffil, Credit Mutuel Arkéa and La Banque Postale gathered in early July to take part in a roundtable to discuss how they navigated their way through the peak of the crisis when covered bond market volatility was at its highest, and how they expect to manage funding in its aftermath, as the pernicious impact of the pandemic takes its toll.

French banks have been among the most prolific, issuing as many as 10 covered bonds since March. The trio of issuers on this roundtable were fortunate enough to have pre-funded ahead of the crisis and were therefore able to avoid issuing when spreads were at their highest. 

With the European Central Bank’s Targeted Longer‑Term Refinancing Operation expected to form an important part of funding for the foreseeable future, there is no immediate pressure to source liquidity. However, since the TLTRO’s short three year term is not suitable for funding banks’ long-term assets, syndicated issuance is likely to remain a strategic imperative. 

Given the uncertain economic outlook, issuers such as La Banque Postale are keen to preserve as much flexibility as possible by attaching a call option to their retained covered bonds. In this way they can quickly free up collateral and potentially issue in the market, should they wish.

Even so, after a decade of central bank reliance, covered bond liquidity has been badly affected, and, in contrast to the global financial crisis, when covered bonds were the first sector to reopen capital markets, covered bonds were the laggard on this occasion, with the corporate sector spearheading the return of liquidity back to normality. 

Apart from liquidity woes, the scale of retained issuance could lead to renewed concerns over asset encumbrance, particularly where borrowers have limited excess collateral to support their deals. But in this regard investors are less worried about French banks, whose covered bonds benefit from high and stable ratings and who have plenty of spare collateral to ensure cover pool quality remains pristine. 

As for the public sector, the financial situation of French local authorities is likely to be manageable, partly because the government has provided support, but also because the impact is likely to be spread over more than one year. Even so, French hospitals have massive investment plans and, as their biggest lender, Caffil was perfectly placed to respond with its Covid-19 response covered bond issued at the end of April. But with French SMEs hit hard, Arkéa was also able to respond with a Covid-19 related transaction issued under its social bond framework. 

As far as the housing market is concerned, the French government decided against declaring a national mortgage moratorium and instead permitted banks to deal with prospective payment holidays on a case-by-case basis. As a result, payment holidays in France have been a fraction of those granted in the UK and Italy — regions where borrowers could have taken mortgage payment holidays for opportunistic reasons, rather than out of necessity. 

Although the European Banking Authority has provided helpful guidance, it is impossible for issuers to really know the extent to which payment holidays will migrate to delinquencies and eventually to defaults. So, for the time being, affected loans will be considered as performing and will count towards a covered bond’s asset coverage test. However, a second spike in the Covid-19 infection rate, leading to renewed lockdowns, could result in stubbornly high unemployment, potentially leading to further mortgage market stress. 

Despite uncertainty over how loans affected by payment holidays will eventually perform, the European Covered Bond Council has been lauded for its proactive decision to adapt its harmonised transparency template by including new data that will give investors a better idea of the broad scope of affected loans. In any case, with sovereign and issuer credit risk playing a more important role in covered bond ratings than cover pool quality, it’s questionable whether the large regional variances in payment holidays will translate to much regional spread differentiation. 

Apart from the fact that central bank action will exert a sustained flattening influence on credit curves, mortgage covered bond ratings are well protected from issuer downgrades. But as a public sector issuer, Caffil’s covered bond ratings are more closely linked to the French Republic, more so with Standard & Poor’s than Moody’s.

Participants in the roundtable were:

Daniel Rauch, covered bond portfolio manager, Union Investment

Dominique Heckel, head of long-term funding, La Banque Postale

Laurent Gestin, head of investor relations, Credit Mutuel Arkéa 

Matthias Ebert, head of covered bonds, DZ Bank

Ralf Berninger, head of investor relations at Caffil’s parent, SFIL

Moderator: Bill ThornhillGlobalCapital

GlobalCapital: How have French covered bond spreads and interest rates evolved since the start of coronavirus and how do you expect the market to evolve? 

Matthias Ebert, DZ Bank: We have had a rollercoaster ride this year, with 10 year euro swaps falling to minus 0.33% in the run up to the coronavirus, this being close to the historic lows from August 2019. This was then followed by a sharp reversal to plus 0.16% by March as economic concerns weighed in. But since then the market has trended lower and 10 year swaps yields are currently at minus 0.15%. At the same time, French covered bond spreads widened five-fold from about 7bp over swaps in early February to 36bp by mid-April, before re-tightening by up to 30bp through to mid-June — which took that market back to levels that are only slightly higher than they were pre-crisis. 

Scope for further spread tightening from this point on would seem to be rather limited as reoffer yields will probably remain negative and Covid-19 risks are still weighing upon the market. However, some further spread convergence between core and peripheral eurozone covered bonds is still possible.

GlobalCapital: Six French issuers tapped the euro covered bond market in March and early April at spreads as wide as 40bp, but Caffil and Arkéa were more patient to make their return. Why was that?

Ralf Berninger, Caffil: We issued a €750m 20 year in early February at 5bp over mid-swaps. We took some time to look at the market for Covid-related transactions as the market was developing very quickly. We have a social programme to finance investments by public hospitals. A lot of the investments this year are expected to be Covid related, so issuing a Covid-related transaction made a lot of sense for us. At the end of April we issued the first Covid covered bond in the form of a €1bn five year, which was priced at 22bp over. We returned to the market in mid-June to issue a €1bn 10 year at 7bp over. 

Laurent Gestin, Crédit Mutuel Arkéa: We had managed some pre-funding in 2019, so we were not in any hurry to return to the market and we were happy to wait for markets to return to more interesting spread levels for issuers. We closely monitored opportunities to enter the market and when we felt that spreads were at a decent enough level we pressed the button at the end of May and issued a €1bn 10 year that was priced at 15bp over. 

GlobalCapital: How has the distribution of deals evolved through the crisis in terms of investor types and order book granularity? Where do you see the market heading from here and why? 

Ebert, DZ Bank: In the early days of the crisis, in mid-March, spread widening fears in combination with liquidity constraints limited investors’ appetite for euro covered bonds. The allocation of bonds to real money investors collapsed in percentage terms from the mid-to-high 20 area to a high single-digit area. But after the ECB announced various stimulus measures, from March 10 through to April 30, we observed an improvement in liquidity, and the widening in credit spreads meant that euro covered bonds suddenly offered a very compelling investment case. 

The investor base quickly rose beyond the traditional buyers of covered bonds, and order books increased from more than €1bn in March to €6.5bn for BPCE’s €1.25bn 10 year green covered bond issued in mid‑May. The real money bid returned, with the share in this deal rising to roughly 30%. 

After a significant tightening in credit spreads between April and May, the market seems to have bottomed out in June at levels that are slightly higher than pre-crisis. But euro covered bonds have now limited spread tightening potential. This is reflected by the number of investors in primary euro deals dropping from as high as 190 in BPCE’s deal to around 60, with an increasing share of bonds allocated to bank treasuries and official institutions relative to real money buyers. 

Some financial institutions have recently had to rely again on the European Central Bank’s covered bond purchase programme to achieve their target deal size. 

GlobalCapital: Where do you see the market heading from here?

Ebert, DZ Bank: For the second half of this year we believe that most primary covered bond reoffer yields will remain negative and supply will be skewed to the long end of the curve. Over-subscription ratios are likely to be low but sufficient, with bank treasuries and official institutions dominating order books. Price sensitivity is likely to increase moderately. However, there is still a lot of uncertainty at this stage of the crisis. 

GlobalCapital: Why has LBP not yet returned to the euro covered bond market — did the European Central Bank’s Targeted Longer-Term Refinancing Operation provide a more convincing funding alternative? 

Dominique Heckel, La Banque Postale: We have not been in a hurry to return to the market because we managed to raise funding early in the year before the crisis set in, with a €750m 15 year that was launched one day after Caffil’s 20 year in early February at 7bp over. 

When the crisis initially struck, the commercial paper market was closed, so our treasury team decided to participate in the TLTRO and our contribution was to issue retained covered bonds to provide us with collateral that we can pledge to the ECB. This was a very tactical and cost-efficient decision because we can access a very low cost of TLTRO funding, which offers a discount of up to minus 1% for one year. 

The three year term helps fund part of the balance sheet and makes a lot of sense for shorter assets such as consumer loans. Also, if you look at the loan guarantee scheme, these are also expected to be shorter. So I would say that, through the TLTRO, you can fund part of your balance sheet and you don’t necessarily have to do very long dated issuance — at least for part of your funding. 

Berninger, Caffil: For us it is very simple. We only look a capital market funding. TLTRO is not something that fits with our lending activity.

Gestin, CM Arkéa: We think the TLTRO is very attractive and we will take a significant part. The TLTRO reduces market funding needs by generating LCR liquidity without issuing any unsecured funding. However, it does not contribute to maintaining overall liquidity reserves, and for that purpose we still need to do some unsecured funding which would be most pertinent in the short and longer maturities. 

GlobalCapital: As an investor, how do you see the recent developments in the euro covered bond market? 

Daniel Rauch, Union Invest: Compared with the global financial crisis, the awareness of the central banks to react quickly and in massive size was a good sign. As a result, we saw that the spread widening was very much on a more muted scale in comparison with 12 years ago, so that’s the good news. 

But there are always pros and cons. One of the more negative aspects is the large production in retained covered bond volumes. We understand that is an option that banks usually take advantage of this format, compared with market financing. This was most important for the southern European countries but particularly now also for northern European countries. While that’s helpful for issuers, it means that from an investors’ point of view you are missing this primary supply. 

The primary market plays an essential price discovery role. The Canadian and French issuers that came into the market in March and April did quite an important job. But if you miss this function in other markets and jurisdictions you are a little less certain on where the real value is and that is a negative side-effect of the growth in the retained volumes. 

Additionally, retained issuance should not play a central role in the financing of a bank as three year — or four years in the case of older TLTROs — does not fit to the natural tenor of the asset side and therefore you have some mismatch. I would really like to see banks, and especially those that have a stable deposit base, not forget that they have clients waiting for material to invest in. 

GlobalCapital: Can a retained 10 year covered bond that is pledged to the ECB under its three year TLTRO facility be considered as 10 year funding or three year funding?

Heckel, LBP: It’s true that a 10 year retained covered bond provides the covered bond programme with 10 year funding, but when you look at the group level funding between the SFH and the sponsor bank it is neutralised, so that means it doesn’t give you long‑term funding. The maturity depends on how you’re going to deploy your bond. If you are going to pledge it to the ECB for three month repo funding, you’re just getting three month funding. It’s not possible to create long‑term funding just by issuing a long dated retained covered bond, because in the end the maturity is determined depending on how it is used. 

Gestin, CM Arkéa: I agree with Dominique. The critical question is around the collateral and what you expect to do with this excess collateral — do you use it to pledge it, or do you keep it in the liquidity reserves? 

Rauch, Union Invest: But from my perspective, the more you take collateral and pledge it to the ECB — and are not going to the wholesale market — the worse liquidity is likely to become. Liquidity at the peak of the crisis in March was particularly acute, as banks and brokers were stuck with trading inventory that they couldn’t sell. This overhang of inventory began to fall over the course of time, when they were able to offload bonds to the ECB. 

This process was quite different from the global financial crisis, where we saw the covered bond market was one of the most important tools to reopen the capital markets. Of course, we were talking about a banking crisis at that time and now we are not; it’s more of a real economic crisis. 

But even so I was astonished so see that liquidity in the corporate bond market was quicker to return than in the covered bond market. I put this down partly to the fact that banks have been much more easily able to issue retained deals for central bank purposes, rather than publicly syndicating deals as corporations are obliged to do. As a consequence, liquidity has worsened in the covered bond market, which is not beneficial.

Heckel, LBP: It’s very important to issue benchmark deals regularly in the market. The practical choice around use of the TLTRO and retained covered bonds is something that can change. From our perspective, we have decided to issue callable retained covered bonds. So, if the situation changes, or if we want to free up the collateral that is used for retained covered bonds, then we can call our bonds and potentially issue in the market. The idea behind this was, faced with this situation, to maximise our flexibility. Again, we made the decision at the time because we saw that the commercial paper market was completely shut down, so it was critical to take action to prevent a potential liquidity crisis by issuing retained bonds for the TLTRO. 

GlobalCapital: How has the French regional public sector managed to deal with the Covid-19 crisis so far? 

Berninger, Caffil: The French government has just commissioned a report on the financial impact of the pandemic on the financial situation for local authorities. Overall, a large part of this year’s tax revenues are based on the economic activity of previous years, so the impact of the Covid-19 pandemic is likely to be spread out over more than one year. For this year the government-commissioned report expects a reduction in local government revenues of around €7bn, out of total revenues of €230bn. To a large extent, these losses in tax revenues are due to a slowdown in real estate transactions. To compensate for the losses, the government has announced measures of €4.5bn that will support the local authorities that have been particularly impacted. 

When we put the figures into perspective, the financial impact for local authorities will be manageable. For public sector hospitals, it’s not a question of revenues but of additional investments. We expect a big increase in investment needs, based on the lessons we learnt from the pandemic. And there’s one important measure for hospitals that has already been announced before the pandemic. The central government said it will take more than €10bn in debt from public hospitals over the next two years, which will create some financial flexibility for future investments. 

There will be a massive investment plan for public hospitals to make sure that the sector is prepared for any potential future pandemic. Caffil is the biggest lender to public hospitals in France, so we expect to play a big role in financing these investments. 

One of the lessons we learned from the pandemic is that, when you look at public hospital investments, you don’t just need investments in specialised medical equipment — you also have to adapt many of the buildings. When you deal with a contagious disease, it’s important to have separate entrances, separate elevator systems, or specific air-conditioning systems. It’s necessary to ensure patients that come to the hospital for other treatments don’t get infected with the Covid-19 virus. When we look at our social bond framework, it has been set up to finance investment by the public hospital sector, so it’s very well adapted to finance these additional investments. 

GlobalCapital: Will Covid-19 use of proceeds work only with a social component? 

Gestin, CM Arkéa: ICMA has given some guidance to address social bonds related to Covid-19. A social bond is probably the most suitable type of bond to mitigate the pandemic’s social consequences, by financing SMEs and healthcare, for example — which are eligible assets categories included within Arkéa’s sustainable bond framework. So, taking into account this guidance, we believe that the social component is quite an important message that you want to talk to investors about if you are looking at issuing a social bond with a Covid-19 focus — so that’s what we did at Arkéa. 

We were able to draw on the fact that Arkéa has a strong presence in financing SMEs, especially within our core territories of Brittany and southwest France, but also everywhere in France. We believe that we could contribute to this fight against the economic and social consequences of this pandemic by issuing this type of social bond. The deal is really targeting classes of eligible assets that are already included within our social bond framework and are considered by ICMA as the most suitable class of asset. 

Berninger, Caffil: I totally agree with Laurent. Coping with the pandemic is really a social issue so social bonds are very well adapted. We also have some issuers with sustainability bond frameworks which could also be used for Covid response bonds. But in that case, the focus will be on the social component under the sustainability framework. 

GlobalCapital: How will the pandemic affect the credit quality of the French public sector? 

Berninger, Caffil: When we look at the operating side of the French local authorities budgets, before investment expenditures, they had a surplus last year of close to €40bn. The impact of the Covid pandemic on local government revenues is expected to be around €7bn, so it’s manageable — even though there are likely to be some isolated cases with specific problems.

GlobalCapital: How did the French government help the housing market? 

Heckel, LBP: The government’s measures were aimed more at helping businesses and households, rather than directly affecting the housing market. Individual lenders have been able to extend loan deferrals on a case-by-case basis, as opposed to a government-sponsored mortgage moratorium, and the French government has set up a loan guarantee scheme to help businesses. 

GlobalCapital: What is the estimated take-up of payment holidays on mortgage loans in France and how do you think this will compare internationally? 

Heckel, LBP: We only have half year results, so we will need to wait for the precise numbers, but we think mortgage deferrals could amount to around 5% of outstanding mortgage loans. 

GlobalCapital: The take-up of mortgage loan deferral schemes was much lower in France, Spain and the Netherlands compared with the UK and Italy, for example, where the declaration of a national mortgage moratoriums boosted take-up rates to as high as 20%. Do you think there may be other factors to explain these differences? 

Heckel, LBP: The variance between European countries may depend on the make-up of their respective mortgage markets. For example, you may see a higher take-up of mortgage deferral schemes in the buy-to-let sector compared with the owner-occupied sector. French unemployment protection is also fairly generous, which perhaps explains the relatively low take-up of mortgage deferrals — but if we face another lockdown in the autumn, we may well see this figure rise from 5%. 

GlobalCapital: When you consider the French mortgage market and the low rate of mortgage deferrals, how do you think it will fare compared with other countries?

Rauch, Union Invest: We’ve already seen some feedback from the rating agencies which shows there is a huge differential between different regions, with France at the bottom end of the range, whereas the UK is reporting much higher numbers. 

The good news here is that the European Covered Bond Council has also been quick to react by introducing a new section on the covered bond label’s Harmonised Transparency Template that specifically tracks mortgage payment deferrals. Some issuers have begun to fill this field in, but as of next quarter I think we will start to see much better data. 

I think the real importance here will be to understand more about the nature of affected loans — in terms of the duration of payment deferral, and whether the loans remain in the pool to be included in the asset coverage test. We need to get a common understanding about the treatment of those loans. But if loan deferrals range around 5%-7% I wouldn’t be too worried, because there are other forms of protection to overcome credit deterioration in the cover pool. The big question for issuers is the extent to which rating agencies raise required over-collateralisation will evolve over time. 

GlobalCapital: What does the EBA’s guidance on payment holidays look like and what is the French regulator’s advice regarding renegotiated loans?

Heckel, LBP: The choice of whether to keep affected loans in the cover pool will be made at the bank level. The EBA has provided guidance on how to treat loans without misleading investors on their credit deterioration. It’s important to avoid having these loans classified as being under forbearance — but at the same time it’s quite difficult to give a clear indication. We don’t really know whether the payment holiday is going to last a few months or whether the client is likely to be in more serious trouble. It’s quite difficult to draw a clear distinction between these two groups. The guidelines are helpful, but they are not that easy to implement. 

Gestin, CM Arkéa: As Dominique said, it remains at the bank’s discretion to decide whether to include or exclude these loans that have benefited from payment holidays. We have EBA guidance, but there’s no clear regulation to take any specific course of action and classify these loans as non-performing. 

So, unless these loans have already been restructured, or have registered a credit event during the payment holiday, they are likely to be eligible for inclusion in the cover pool of the home loan SFH product. 

GlobalCapital: What does your prospectus documentation say about loans that are more than 90 days overdue — is this aligned with EBA/domestic regulatory guidance? 

Gestin, CM Arkéa: All loans subject to payment incidents are systematically excluded from the cover pool of our home loan covered bond programme. Allowing a payment holiday on a home loan is, however, different from a payment incident. First, the borrower may, in some cases, have applied for a payment holiday, only from an opportunistic approach and not because of payment difficulties. Second, the approval of a payment holiday is at the bank’s discretion: we are in a different situation from one when a client is not paying because he can’t. At this stage, these loans are thus not considered as non-performing but they are clearly monitored to address any potential problems.

GlobalCapital: Do you think there might be a cliff effect where, at some point in time when it becomes clear how borrowers’ circumstances have changed, payment holidays rapidly migrate to delinquencies?

Gestin, CM Arkéa: Currently, payment holidays represent a small percentage of total outstanding home loans at Arkéa. If these loans were to comply with SFH eligibility criteria and were also in trouble, the credit quality of the programme’s cover pool would not be jeopardised and we have considerable capacity to replace these loans with performing ones, if needs be. 

Heckel, LBP: I think it’s possible that many clients are opportunistically taking payment holidays. As long as we are in a situation where the client is in employment or furlough and the situation does not last too long, I think the cover cool can be managed perfectly well. La Banque Postale has already filtered those loans that are affected by payment holidays. 

With respect to programme documentation, the decision on how to classify loans and include them in the pool is being made on a case-by-case basis by each issuer and its sponsor bank. If they opt to keep the loans in the cover pool, there will need to be a decision on how to account for their value. And then, of course, we need to be transparent with our reporting. 

GlobalCapital: Will you be using the updated Harmonised Transparency Template to provide transparency for investors on loans affected by payment holidays?

Heckel, LBP: We will, not least because it ties in directly with the EBA’s guidance on transparency and reporting. The updated HTT will definitely help issuers provide information on the extent of affected loans and what they are doing with them, in terms of keeping them in the pool or not. Tracking this data is really something that’s important for everybody.

Gestin, CM Arkéa: We will also be using the updated template at Arkéa from the next quarter onwards and reporting regularly as usual via our covered bond investor webpage. The covered bond label is the only organisation that has expressed a specific demand for public transparency on this subject. If other institutions, like rating agencies or the local regulator, were to request more transparency on this, then we would also report as we already have the information at hand. 

GlobalCapital: As an investor, what do you think about the transparency around payment holidays? 

Rauch, Union Invest: I think it was a really great idea from the ECBC to be the first mover by creating a higher degree of certainty for investors. We always highly welcome improvements in data quality. But given the high degree of interpretation that’s required by issuers on how to account for such loans, it would probably be helpful to have a common definition. 

For example, it might be helpful to know whether the payment holiday was granted for three months or six months. I think it would also be useful to understand more about how these holidays are granted. From what I understand, it has been relatively easy in the UK to get a payment deferral, whereas in Spain it has been rather difficult. In this context, it would also be helpful to know the proportion of clients that genuinely need a payment holiday compared with those that are opportunistically taking one. 

In any case, it should not be forgotten that the primary driver of spread is not really the credit quality of the pool, it’s much more linked to the country rating and the unsecured credit quality of the bank. As of now, the credit quality of the cover pool ranks behind the issuer and sovereign rating. But if we were to see a dramatic deterioration in cover pool quality, then it’s likely spread performance drivers would change. But, as of today, I would say the potential deterioration in pool quality is likely to be rather limited. 

GlobalCapital: Do you see scope for spread differentiation between different regions based on how renegotiated loans are treated and the extent to which these loans have affected cover pools? 

Ebert, DZ Bank: After the widening in April, spread differentials have since become quite small. From what I understand, all the major rating agencies say that mortgage loans with payment holidays do not present a major driver for covered bond ratings; country and issuer ratings are more important. 

However, rating agencies are still analysing how loans with payment holidays will perform, so it’s probably too early to draw a final conclusion. The take‑up of payment holidays will vary over time and this is likely to be shaped by the economic development, which is also very difficult to predict. 

There is still uncertainty over whether we will see ‘V’, ‘U’ or ‘L’-shaped recovery. This factor will probably have a significant impact on the rating agencies’ views on the performance of the mortgage loans. Hence, a high level of transparency around the treatment and magnitude of renegotiated loans in cover pools is probably key in this extraordinary situation. And we think that credit differentiation in covered bonds due to payment holidays could be possible but will probably be very limited. 

GlobalCapital: Will mortgage loans with payment holidays and a surge in retained covered bond issuance have an impact on your ratings?

Heckel, LBP: The covered bond ratings could withstand a downgrade of the French state by two notches with S&P, though overall our covered bonds have four notches of unused rating uplift compared with our senior unsecured rating. All things being equal, that means the bank’s unsecured rating would need to fall by a total of five notches before the triple-A covered bond rating is lost. As long as payment holidays remain in their current low range, I do not expect any impact on the covered bond rating. The rating agency transaction update doesn’t mention the scale of retained covered bond issuance that would have an impact, nor even the fact it could have an impact on the rating.

Gestin, CM Arkéa: I agree with Dominique. There have been no notifications from rating agencies that payment holidays will have an impact on the covered bond programme or sponsor bank’s ratings. And if payment holidays stay as low as they are today, I expect ratings to remain stable. 

Ebert, DZ Bank: If more and more issuers use retained covered bonds for ECB purposes, it will increase encumbered assets and that should have an impact on the issuer rating, shouldn’t it? 

Heckel, LBP: The impact will vary between banks, depending on the amount of covered bonds pledged for repo and the make-up of the banks’ balance sheets with respect to the overall proportion of encumbered to unencumbered assets. For us at LBP, it’s a tactical decision to have access to the TLTRO. This is not a facility that is supposed to last for ever. At some point in the long run there is a definite need to match long‑term mortgage assets, which have an average maturity of eight to 10 years, with your liabilities.

Gestin, CM Arkéa: At Arkéa we took the decision not to issue any retained covered bonds and keep our mortgage collateral in reserve. So this situation does not apply to Arkéa. As far as TLTRO is concerned, its use has an impact on asset encumbrance, the level of which is closely monitored.

Rauch, Union Invest: The topic of asset encumbrance is not new. Asset encumbrance can also be caused by assets other than mortgages that are being pledged for ECB repo purposes, as well as the high quality collateral used in swap derivatives. 

There is, however, a potential problem with refinancing this TLTRO debt, especially when you consider the varying extent to which it has been used in different countries. The market is flooded with liquidity at this point in time, but there is a question over what might happen in three or four years’ time, when this funding needs to be refinanced, as we could by then expect to see a large wave of public issuance. 

There is also a problem with transparency, as it’s not always clear to get hold of data on the volume of retained issuance that has been pledged for repo. And if retained covered bonds volumes have been high, you need to ask how much over-collateralisation is left in the covered bonds that were actually placed in the public market with real investors.

If, for example, a borrower with limited excess collateral were to issue a large amount of retained covered bonds for ECB purposes, then I would really start to question whether I am better or worse off as a result. But, I am less worried about French issuers than those in other jurisdictions. 

Ebert, DZ Bank: We know that French covered bond ratings are well protected, but this is not necessarily the case with all peripheral European covered bonds. If this crisis proves to be more severe or longer than expected, how do you expect to trade peripheral risk?

Rauch, Union Invest: We do a lot of number-crunching and are also quite selective with the covered bonds that are part of our universe and those that are not. Before the crisis began, we would not necessarily have bought covered bonds where the rating buffer between the issuer and covered bond was very limited or zero. 

But at the end of the day it’s a question of spread. Would we require more compensation for taking the additional risk? Yes, of course! We are in a low yield environment and covered bonds have been badly affected by this. Real money investors have shifted to other asset classes in search of more yield, so if there is an additional pick-up, it could of course help. 

GlobalCapital: How will sovereign and issuer ratings evolve and what impact will this have will on covered bond ratings? How many notches does your covered bond rating have in terms of a buffer? 

Berninger, Caffil: As a public development bank, we benefit from a letter of comfort from the French Republic. When we look at our lending activities, we are executing public policy missions for the French government, so our issuer ratings at the group level are very closely linked to the rating of the French Republic. 

When it comes to Caffil’s covered bond ratings, it varies very much from one rating agency to another. Under S&P’s methodology there is a close link between the issuer and covered bond rating, but for Moody’s there’s a lot more flexibility. 

GlobalCapital: How will the French housing market likely progress over the next year or two? 

Heckel, LBP: France is progressively coming out of lockdown period. It’s a bit early to say what we think will happen, but fundamentals are still quite strong and rates remain at low levels. We still have some dynamic regions in France where the housing market is strong. Ultimately the rate of unemployment will be really important to monitor. Then we also have some factors that are likely to have changed as a result of the crisis, such as home working, that could potentially support demand in the regions. So, if we do not face another lockdown period and we are on track for a gentle economic recovery, we expect house prices to fall by up to 5% over one year, from where they had been in March. 

GlobalCapital: How will Covid-19 and cheap ECB money impact your covered bond funding strategy for the rest of this year? 

Berninger, Caffil: TLTRO is not a topic for us. And when you look at what we have issued this year including the Covid-19 transaction, we have already issued three deals so we are very well advanced for the year. For the second half we would normally expect to do at least one additional benchmark. 

Gestin, CM Arkéa: Our public issuance programme hasn’t changed since the Covid-19 crisis. We had a need to issue covered bonds this year, especially given high redemptions in 2021 and the strong loan production we experienced in 2019. So we decided to remain active in the covered bond segment. We issued a €1bn 10 year covered bond in May, a higher size than achieved in previous issues. Unless spreads come back to levels that would allow us to do some pre‑funding for 2021, we won’t have any further need for public covered bond issuance in 2020. 

The Covid-19 aspect of our senior non-preferred issued under our social framework was something we hadn’t obviously planned, so this was an additional topic. 

Over the past three years we have had strong funding needs for regulatory and rating purposes and have regularly issued Tier two, senior non-preferred and senior-preferred debt to build the right layers of debt. This has been part of the funding strategy in 2020, irrespective of Covid-19. The funding programme is quite ambitious and in line with what we did last year. 

Heckel, LBP: We have also been active in senior non-preferred. This year we had a very pragmatic approach to our 2020 funding, which involved taking part in the TLTRO. 

There is also a question over the outlook for commercial activity in terms of the new loan production — especially in home loans, where we’ve already missed three months of production — and whether we will see a strong pick-up in new loan production over the remainder of the year. It’s possible we may adjust our plans once we have more clarity on the development of assets on our balance sheet, with a view to potential issuance before year end, but at this point it’s a bit early to say.

By Bill Thornhill
10 Jul 2020