Introduction
The issue of the Non-Performing Loans (NPLs, from now on) has been long debated over the last few weeks and a new system has been designed with the purpose of creating a market for these troubled assets, which at the moment are still occupying the balance sheets of most Italian banks. After the agreement to create a bad bank, many Italian banks and other financial institutions have agreed to contribute money for the creation of a “rescue” fund, which will be named “Atlante”. Atlante is a € 4.25bn fund that will participate to the subscription of the upcoming capital increases of Italian banks seeking recapitalizations and that will take actions to revive the NPL market in Italy.
The structure
At first glance, Atlante looks very much like a private equity fund. Italian Minister of Finance Mr. Padoan has referred to it as “an entirely private operation”, which it is indeed, even though it is clearly satisfying the interests of its investor in an unusual way, i.e. not by providing strong returns but instead by attempting to restore stability and confidence in the Italian banking system. It is enough to take a look at how Atlante is going to operate in order to find support for this claim, in fact: (1) it will be supported by the Italian government with ad-hoc legislation (e.g. so far it has been exempted from the requirement of launching a full takeover offer after passing the 30% ownership threshold); (2) it will purchase the securitized NPLs at non-market price, probably somewhere between what the market is offering (around 20% of nominal value) and the book value of these loans (around 40% of nominal value); (3) it will support upcoming recapitalizations, namely Banca Popolare di Vicenza (see below) and Veneto Banca.
The fund has been promoted by Mr. Penati, President of Quaestio Capital Management SGR, with the blessings of the Italian government and of EU Authorities, which will monitor the operation to make sure it does not turn into mere State Aid. With respect to this last point, it should be stressed that, in fact, Cassa Depositi e Prestiti (a formally private institution but de facto under the control of the Italian Ministry of Finance) will participate to the fund with only a rather small stake, and the only “real” aid from the State has and will come under the form of ad-hoc legislation. Thus, there appear to be no grounds for classifying the operation under the State Aid umbrella. Moreover, while the government has indeed a big interest in fixing the Italian banking situation, its objectives are undoubtedly aligned with those of the largest players in the Italian financial services industry, which suffer from the current instability and lack of trust.
A number of Banking Foundations and Insurance companies have contributed money to the fund, together with small but healthier Italian “popolari” banks such as Banca Popolare di Milano and Banca Popolare dell’ Emilia Romagna. Cassa Depositi e Prestiti appears to have contributed around €500m, while Unicredit and Intesa will have the largest individual stakes in the fund, with €1bn each.
As noted above, the fund has raised a total of €4.25bn. This amount could potentially but unlikely be more than doubled, as Atlante will be able to lever itself up to 110% (but at what cost?). According to Quaestio SGR, at least 30% of the fund will be used to buy the more junior tranches of securitized NPLs, while up to 70% will be devoted to investments in banks (i.e. to back recapitalizations). The fund will have a 5-year horizon, with the possibility of extending to 6, 7, or 8 on an annual basis, and is offering a 6% return per annum. In terms of governance, Quaestio SGR has made clear that: (1) the SGR will vote in the elections of board members of participated banks “respecting stringent independence requirements”; (2) the SGR will not exercise management powers over controlled entities (with the Italian meaning of “direzione e coordinamento”, i.e. it will not use elected board members as means to exercise control over the bank(s) in question); (3) the SGR will not interfere with the management of day to day activities of controlled banks.
Purchasing NPLs
As explained above, one of the two primary functions of Atlante will be to act as a buyer of the more junior tranches of securitized NPLs, requiring “an IRR lower than traditionally demanded by specialized investors”. In fact, senior tranches will carry the guarantee of the Italian government (part of the Bad Bank proposal, as discussed in a previous article), which will make them more appealing to investors. The simple dynamics are shown in the picture below.
Depending on the outcome of the operation and on market demand, Atlante may turn out to buy also the equity tranche, which in alternative would be retained by the originating banks.
Mr. Penati has repeatedly insisted on Atlante playing a role to reduce the Bid-Ask spread in the Italian NPL market, that is, stimulate competition and reduce information asymmetries with the purpose of pushing up the current value attached by the market to these securities. In fact, as of today, the NPLs are valued at around half their book value (e.g. recently Carige has refused an offer from Apollo to buy all of its NPLs at 17% of their nominal value, less than half the value at which these loans are currently booked). How and whether this spread reduction will be achieved is hard to predict, however Quaestio has done it for us, and their predictions (quite optimistic in our view) are shown in the picture below.
Without going into detail, the graph above shows the potential improvement in sale price (expressed as a percentage of nominal value) of a portfolio of NPLs, starting from the assumption of a straight sale of the portfolio to a specialized investor (20.7%) and arriving to a final price that takes into account the securitization, the government guarantee, and the effect of Atlante, as well as other factors. Specifically, the value should be enhanced by the securitization process itself (the benefits of splitting the risks), by the State guarantee (the “GACS” above, i.e. “Garanzia sulla Cartolarizzazione delle Sofferenze”), and by the Atlante effect, which should create value not only by buying the junior tranches but also by stimulating competition. With respect to this last point, Atlante will stand on the buy side offering a higher price for bundles of NPLs, thus potentially increasing competition in the market and leading other players to bid higher. The issue however, as will be discussed in the final section, is whether it will be able to do so in a credible manner, especially considering its little endowment (a little more than €8bn compared to about €350bn of NPLs in total) and the fact that the market is far from being fragmented, with only few (mostly foreign) specialized investors with lots of money and bargaining power. The other factors include reductions in servicing costs (almost irrelevant) and time to recovery, as well as an increase in recovery rate. We believe that the latter is very much dependent upon an overall recovery of the economy, and Quaestio’s forecast is indeed rather optimistic. We are also doubtful on the possibility of reducing time to recovery, in spite of the fact that the government will pass some ad-hoc decrees. In fact, we are convinced that an important reason of the strong devaluation of NPLs, especially by foreign investors, is the fact that going through a recovery procedure in Italy implies having to deal with Italian bureaucracy, and a recovery process can take up to 6 or 7 years. It is hard to believe that, even with the right legislation, we could see a considerable reduction in these figures.
Subscribing Capital Increases
The issue of NPLs has generated, for many small Italian banks, the need to recapitalize in order to comply with Basel requirements. However, it is clear that, with the NPL issue yet to be solved, troubled banks seeking for capital are very much unlikely to find support from the market, and even from their own investors.
Thus, Atlante will provide capital to make sure that these banks are able to raise the money they need. One of these recapitalizations has already taken place, it is the case of Banca Popolare di Vicenza, discussed in the following paragraph. Atlante is expecting to commit another important portion of its endowment over the summer, when Veneto Banca, another institution plagued with NPLs, will carry out a €1bn capital increase. The operation has been underwritten by Banca IMI but, as we shall see in the next paragraph, this is far from being reassuring. Finally, Banco Popolare is expecting to raise €1bn within the next 18 months, and in any case prior to the merger with Banca Popolare di Milano. In our opinion, this last case is indeed less alarming, and should encounter fewer issues along the process. This is because the capital increase is finalized to the merger with Banca Popolare di Milano, which is in a much better shape than many other “popolari” banks, and this should make investors feel a bit more confident.
The case of Banca Popolare di Vicenza
As already mentioned, Atlante is already active, and it kicked in for the first of the three capital increases planned for the upcoming months. In April 2015 Banca Popolare Vicenza seemed to be in good shape, share price was around €40 and the bank did not seem have particular problems. However, the Asset Quality Review (the process by which EU authorities assess whether the book value of banks’ assets is appropriate), revealed that the valuation of some of the NPLs was too optimistic, so that the bank had to revise down their value, wiping out most of the equity capital. Their Common Equity Tier 1 (one of the major ratios used to identify the stability of a financial institution) suddenly fell to 6.81%, therefore the bank needed to raise capital in order to meet the Basel Capital Requirement of c.a. 11% (specific for unlisted banks). BPVI planned to raise € 1.5bn and then list the shares in the Milan stock exchange, in order to give liquidity to shareholders. Unfortunately, the market and the former investors only subscribed 8% of the total amount offered, and Atlante had to intervene subscribing all the remaining shares (Unicredit acted as underwriter, but Atlante stepped in as sub-underwriter a few days before the end of the subscription period). At the end of the process, Atlante had bought 15bn shares at € 0.10 each (against a share price of €40 just one year before), ending up with 99.33% of the bank. In fact, Atlante took care also of the 8% subscribed by other investors, as these orders were conditional on the execution of the IPO, which Borsa Italiana denied in the aftermath of the transaction due to lack of sufficient floating shares to allow for regular trading (the minimum requirement to list in the Milan Stock Exchange is at least 25% of floating shares).
One of the major question marks relates to the kind of approach that the fund will take on BPVI’s governance: Atlante’s articles of association clearly state, as outlined above, that “The SGR will not exercise management powers”, i.e. it will not use the elected board members as means to exercise control over the banks in question. However, Mr. Penati seems to be oriented towards a restructuring of the bank, to carry out “within 18 months”, according to his own statements. Another possibility that Mr. Penati is considering would be to sell the bank to a specialized PE fund, also in view of the future commitments of the fund.
Will it work?
What is sure is that the operation will inject fresh capital in the accounts of many troubled banks that would otherwise not be able to raise capital in the market, thus avoiding external bailout. On the contrary, there are still many doubts in assessing whether Atlante will be able to play an important role in the market for NPLs. This will mainly depend on two variables:
- How many capital increases Atlante will have to subscribe. Apart from Banca Popolare di Vicenza, of which Atlante is already the majority shareholder, there are two other right issues to be completed, (Banco Popolare and Veneto Banca) for a total amount of €2bn. If both of them will need help, Atlante will step in, thus necessarily devoting less resources to the purchase of NPLs.
- Private investors’ interests for the senior tranche of the NPL-backed securities issued by the Bad Bank. If institutional investors participate heavily, the magnitude of the Atlante effect will be much greater.
Atlante has a maximum leverage of 110%, i.e. around € 8.9bn of total available funds. In the worst-case scenario, Atlante will have to subscribe also the other two capital increases in full, thus ending up with around € 5bn of residual resources, which will be used to buy both the mezzanine (30%) and the equity tranches (20%), assuming the latter will not be retained by the banks. Assuming also that the government will guarantee only 50% of the Bad Bank securities (i.e. the senior tranche will only be about 50% of the total securitization) and that the market will be there to buy the senior tranche in full (which seems a reasonable assumption) the total amount of funds raised by the Bad Bank should be around €10bn (actually more, as senior tranches should be more expensive, but we will disregard this to make it simple). Further assuming that Atlante’s effect will be poor and that the Bad Bank faces costs leading it to buy NPLs from banks at 60% of book value, the total securitized NPLs should amount to around €16.7bn, with a €6.7bn overall write-off. Pretty bad and, in magnitude, quite limited compared to the €350bn total NPLs. In the best-case scenario, assuming same leverage and same proportions of the different tranches, Atlante will only have to subscribe the capital increase of BPVI, having in this case around € 7.5bn left to purchase NPLs. Assuming that in this case the equity tranche will be retained by the originating bank, Atlante will invest all of its resources in the mezzanine tranche (still 30% of the issue). In this case the Bad Bank should have around €25bn of funds (five of which are coming from the originating bank itself, again we are disregarding different costs). Further hypothesizing that Atlante has yielded the desired effects (being this the best case scenario) and that the Bad Bank is able to purchase NPLs at 90% of book value, the total securitized NPLs should amount to €27.8bn, with a €2.8bn overall write-off.
Of course, these calculations are sloppy, and their purpose is certainly not that of showing the actual (in numbers) impact of Atlante on the NPL market, also because this would be a time consuming and poorly rewarding task, as there are many uncertainties and dilemmas yet to be addressed. What the examples above want to show, instead, is how poor, in relative terms (i.e. compared to the €350bn total NPLs), could be the impact of Atlante be if one considers only its ability to actually buy the securities. In fact, we hope, the effect of Atlante will span beyond the simple purchase of NPL-backed securities to encompass, as in part discussed above, the competitive landscape in the Italian NPL market. This obviously depends on how the government will act, especially in terms of passing legislation that will speed up recovery timings, as well as on the real appetite of specialized investors for these securities. Standing as a “biased” buyer, Atlante may push other investors to bid higher, thus potentially generating a reviving effect for the market. The main concerns, however, are represented by its small endowment and by the concentration of specialized players, smart investors that will certainly make things more complicated. All is left to do is just wait and see, exciting developments are on their way.
[edmc id= 3946]Download as PDF[/edmc]
0 Comments