Tuesday 13 October 2015

The Greek NPL issue and a possible resolution path

 Δημοσιεύτηκε στο Greek Economists for Reform
It is argued that the creation of an Asset Management Company or bad bank that would acquire most if not all of the 100bln NPL would be advantageous to the Greek economy the Greek banking system and the society, provided, it is structured so as to minimise political influences and hazards. The AMC would operate for 15-20 years, allowing for a smooth work out of the loans and maximizing recovery rates. The cleaned banks would be recapitalised from the private sector. The proposal would cost less to European and Greek taxpayers and can be an engine of economic growth. Similar solutions have been applied to many countries the most recent in Italy[1] and can be adapted to Greece.
Η δημιουργία μιας «κακής τράπεζας»  (εταιρεία διαχείρισης ενεργητικού ή δεινοτράπεζα) η οποία θα αποκτήσει την πλειονότητα των Μη Εξυπηρετούμενων Δανείων (ΜΕΔ) των τραπεζών που ανέρχονται σε πλέον των 100δις θα έχει σημαντικά πλεονεκτήματα για την Ελληνική οικονομία, αρκεί να ελαχιστοποιηθεί η πολιτική επιρροή και κίνδυνοι. Η δεινοτράπεζα θα έχει διάρκεια ζωής 15-20 χρόνια, επιτρέποντας την ομαλή αναδιάρθρωση αλλά και την μεγιστοποίηση των ποσοστών ανάκτησης. Οι καθαρές από ΜΕΔ τράπεζες θα ανακεφαλαιοποιηθούν από τον ιδιωτικό τομέα. Η πρόταση αυτή θα κοστίσει λιγότερα χρήματα στον Ευρωπαίο αλλά και Έλληνα φορολογούμενο και μπορεί να γίνει η ατμομηχανή της ανάπτυξης στην Ελλάδα. Παρόμοια λύσεις έχουν επιτραπεί να εφαρμοστούν και σε άλλες χώρες με πρόσφατο παράδειγμα την Ιταλία.
The greatest challenge that Greece and the Greek economy is facing is not the debt but the current state of the Banking system. Although the total government debt is excessive, the cash flow maturity profile is favourable for Greece. A further re-profiling of the Greek debt although desirable will not have any big immediate impact on the economy of Greece unless there is a functioning banking system.
The government debt is a long term problem contrary to the Banking problem which is current and urgent. The four systemic banks in Greece have mostly stopped functioning as credit institutions and operate, for all practical purposes, as payment systems. There is no credit into the economy and the balance sheet is mired with Non-Performing-Loans (NPL).

In this note, we describe the nature of the Greek NPL issue and lay down the principles that need to be adhered to, in order to have a successful resolution for the Greek economy, the banking system and the tax payers of Greece and Europe.
Finally, a sketch of a possible resolution path is given. Given the size of the NPLs one can easily understand that resolving the banking NPL and recapitalization crisis would define whether Greece goes back on the path of growth and prosperity or stagnate for many decades.
Facts and Figures

Greek banks have accumulated over the years an unprecedented high number of Non-Performing-Loans (NPL). The last official figure of Dec 2014 had placed the number close to 39% of all loans, but since then, the situation has deteriorated. Already the Bank of Greece report[2] on Monetary Policy talks of a total of 100bln. The situation is even worse if one tries to estimate the total Non-Performing-Exposure (NPE) which is probably well over the 100bln mark with no sign of abetting.
This is a staggering amount, in percentage terms, the highest in the world. It represents 55% of GDP. For comparison, Cyprus is second with NPL’s accounting for 45% of GDP[3].
A further complication comes from the high degree of recidivism or relapse of loans that have been already restructured. According to data[4] becoming non-performing again, approaches 70% for the first 12 months after restructuring the loan. This is due to the practice of restructuring loans for window-dressing the balance sheet or for funding reasons.
Clearly, the different types of loans have different delinquency/default rates and recovery rates, with Consumer loans having the lowest recovery, highest delinquency and default rate and expected loss given default. Details can be found in the Blackrock reports (project Solar, project Aura and associated AQR tests). For example out of the 64bln of residential mortgages 20% has defaulted (360+DPD, Denounced) but when you include the restructured the percentage doubles to 39.9% or 25.5bln.
The majority of the loans are indeed collateralised. Most of the consumer loans come with personal guarantees and the same goes for Small Business Loans or for loans to professionals. Effectively, all these guarantees boil down to real estate. Most of the wealth of the Greek household is in the form of real estate which for years was the main savings and pension alternative (also tax efficient, legal and illegal).
This means that dealing with the Greek NPL problem is a huge exercise in the real estate markets.
System Nature of Crisis
Looking at the magnitude of the NPL it is clear that this is a systemic and a national problem and not just a bank with a bad loan book. All four systemic banks in Greece are more or less in the same position. They are exposed to the same economic environment and the same customers.
Moreover, any attempt to implement a partial solution affects the totality of NPL and the recovery rates. For example, imagine that Bank X, attempts to recover some of the losses by repossessing the collateral and selling it. In the current market this would depress the value of real estate even more and Bank X would be forced to mark the rest of the portfolio at even lower prices forcing them to raise more capital. This may force other banks to do the same, creating a down spiral.
In addition, the large numbers of borrowers that have defaulted infect the rest of the loan book. Namely, the large number of defaults is encouraging strategic defaulters. It is interesting to note that according to Blackrock, in the case of residential mortgages, unemployment was the third explanatory variable with LTV and Coupon being first and second. There was also a much higher default rate for high value loans.
Strategic defaulters also pose a huge threat to the incentives of the economy. A company that continues to service its debts find itself at a market disadvantage and survival forces it to become a strategic defaulter too.  
Perhaps the biggest problem is the cultivation of a non-payment culture, which was championed by political parties and politicians. Many borrowers are hoping to enter into increasingly good restructuring terms, even total write down and thus procrastinate or chose to default even if they can service the loan.
A final problem comes from the possibility of using the NPL’s for clientelistic reasons. Removing the loans from the banks strips away levers of influence and in some cases the temptation of personal favours or enrichment.
Roots of the problem.
The roots of all credit crises are the same. It comes down to cheap and abundant credit that makes banks and borrowers complacent. All of the Greek loans are of the same “vintage” 2003-2009. These are the bubble years of the Greek economy fuelled mainly by rampant state borrowing.
When markets stopped refinancing the state, the rapid and forceful fiscal adjustment of the program gave rise to a deep recession, wiping out more than 25% of GDP in five years. Real wages and salaries suffered and real estate prices collapsed (40% drop between 2007-2015[5]) pushing most mortgage holders into negative equity position and encouraging strategic default.
Thus the high degree of NPL can be attributed to three factors:
a.       Bad bank policies. Imprudent lending with high leverage (high debt to income ratios) and dubious if not illegal practices of lending on undeclared income (Blackrock report), high LTV ratios or political influences. The Bank of Greece supervision on these practices and also on credit growth seems to have been very soft.
b.      Severity of recession that wiped out 25% of GDP
c.       Over borrowing by consumers and businesses
d.      Culture of non-payment encouraged by certain political parties. Laws that were meant to protect (foreclosure moratorium) the most vulnerable members of the society were abused by many. Also cumbersome and time-consuming legal procedures that allowed many strategic defaulters to escape or postpone payment.

Principles of resolution
1)      It is clear from the above that any solution must be holistic and not piecemeal fudge. In other words, most if not all of the loans, despite their inhomogeneity must be dealt at once.
2)      NPL’s cannot remain inside the banks as they infect good ones, encourage strategic defaults and most importantly of all, makes recapitalization through private means extremely difficult due to the non-transparency of losses. New shareholders must be responsible and accountable for new NPL not old ones. Many investors are under the false impression that owning a Greek bank would allow them to cherry pick certain high real estate assets and thus propose solutions that are not optimal.
3)      In order to increase the recovery rates one must have time to work them out. The sheer number and systemic importance of the loan portfolio means that at least 15 to 20 years will be needed for a total wind down.
4)      In order to have a just resolution, political interference must be avoided at all costs.
5)      Recapitalizing banks with no NPLs would much easier. Experience has shown that prospective investors require a backstop on NPLs in order to invest. If the banks are clean this is not an issue.
Structure of Solution
The proposed solution is based on the bad bank model. At the last count, around 12 countries in Europe have used a variant of the Bad Bank or Asset Management Company (AMC). The main differences are on the ownership structure and the loss allocation. Some are funded totally by the state (e.g. German FMS) others have a private-public mix (Spain’s SAREB).
The nature of the Greek banking NPL problem begs for a similar solution. In the following we give an outline of how bad bank might be structured, adapted on the Greek idiosyncrasies.  
1)      Creation of an Asset Management Company (AMC) otherwise known as bad bank that is going to acquire most if not all of the 100bln NPL from the four systemic banks. Needs work on legal framework.
2)      The majority of the shareholders (>55-60%) of this AMC must come from the private sector (including Banks) and if possible, ESM should participate. The total equity needed is no more than 5-7bln. Having private majority is one of the key demands of Eurostat in order to avoid consolidation with national accounts.
3)      The AMC receives around 60bln in total of guarantees from the Greek State (AMC leverage 8-9). These should be structured in an appropriate way as to minimise losses to the taxpayer. More importantly these guarantees if structured efficiently would not count towards the Greek debt (like in Spain’s Sareb)
4)      Currently, Greek banks have accumulated provisions of around 40bln against the 100bln of NPLs. Part of the state guarantees would be used as payment for the 100bln worth of loan notional. Thus, the AMC would buy the loans at a discount of minimizing the future losses to the equity holders and the guarantor.  Again, this is another key prerequisite of Eurostat.
5)      There is a cut-off date for bad loans (say 31-12-14). Only loans that were “red” before that date would be transferred to the AMC. This is to avoid acceleration of defaults. 
6)      The AMC would have a life span of 15-20 years. This is adequate time to work out without market disruption the volume of loans and increase the recovery value.
7)      The profit and loss and credit structure of the AMC can be adjusted to reflect the risk sharing.
8)      The charter and governance structure and principles of the AMC must ensure no political influences and the just application of the social principles that would govern any restructuring. An international board of directors along with the Greek one should ensure this.
Advantages of the scheme
·         Much needed time is given to work out the NPL’s minimizing market disruption and maximizing recovery rates and losses to the taxpayer.
·         AMC can be used as an engine for growth by involving private equity firms for companies that are in trouble but have sound prospects.
·         Banks are clean of NPLs and can be easily raise capital from the markets. Moreover, they can now be the engines of growth for the Greek economy rather than chasing defaulters.
·         Concentration in one vehicle of all loans that a borrower might have at different banks. This would facilitate restructuring and reveal any strategic defaulters as everything would be consolidated.
·         The structure of the AMC guarantees independence and transparency.
·         The social criteria are applied with no political interference.
·         Avoid the need for a wide bail-in of shareholders and creditors
·         Recapitalising the Banks using funds from the private sector would be much easier as the balance sheet would be more transparent. Also the reduced size of the balance sheet would require much less equity capital.

Possible hurdles of the Scheme
·         Needs a whole new legal framework urgently.
·         Many classes of loans make the scheme tedious.
·         Resistance from vested interests both political and banking since they would lose leverage.
AMC vehicles have been used with success in many countries. The nature and magnitude of the Greek NPL problem can be addressed successfully only if Banks are cleaned of all bad Loans in one go and are allowed to function as banks. This can be achieved by an AMC that has transparency and independence. 
Greek Loan Composition and approximate Non-Performance-Exposure as of June 2013 (Blackrock)

Size (funded)
NP Exposure
Consumer Loans
Small Business Prof




State related


[1] 6 Sep 2015, http://www.bloomberg.com/news/articles/2015-09-06/italy-in-final-talks-with-eu-commission-on-bad-bank-padoan-says
[2] Bank of Greece report on Monetary Policy, June 2015 (in Greek)
[3] Data from World Bank, 2014.Greece has gone up since the reported 34%
[4] Blackrock report, Project Solar
[5] Bank of Greece housing data, Link