Petr Grishin of VTB Capital -

"Honestly speaking, I don't know whether Brussels has its own formula. Or, to be direct, it will dictate to us, after we come to any kind of conclusion based on any kind of formula, that we need to lower the price by another 10%:" France Arhar, Supervisory Board Chairman of Nova Ljubljanska Banka and former Bank of Slovenia governor (1991-2001), on disagreements with the European Commission over the bank asset transfer price, in an interview to Delo (July 13, 2013)

We had several meetings in Ljubljana in July, looking to get a clearer picture about the reasons for, and the effects of, the delay in the banking system workout. The strongest feeling one gets these days talking to officials and banks is that there is a deep frustration with how various decision-making centres within the European Commission machinery are handling the situation.

While one could indeed debate the depth and quality of many of the inputs that Ljubljana sends to Brussels, it is evident that the reverse flow of recommendations is subject to similar doubts (in terms of consistency and coherence). Apparently, after Spain, Greece and Cyprus it is difficult to expect a particularly "benevolent" approach from the EC to anyone, and all the more so for Slovenia, a small country that has so far not allowed a single bank to close or licence to be revoked. Nor has it had enough courage to recognise the true magnitude of the capital gap all at once (opting for several piece-meal step-by-step repeated recaps).

Despite these hurdles, our takeaway was that the willingness to get the transfer process moving is strong and faces little opposition from within Slovenia. The authorities are aware that they have no bargaining power vis-à-vis the EC, and will have to accept the Brussels-determined ultimate cost of the banking system clean-up, whatever the number is. We continue to think that the delay is likely to be resolved in mid-autumn, with the first transfers taking place in the fourth quarter. To get this process moving is more important than where exactly the sovereign leverage will settle.

1. Slovenia's largest banks are under sustained pressure from growing non- performing loans (NPLs) - particularly in construction, real estate, and the financial holdings industry - and the respective recapitalisation needs.

At the end of the first quarter, more than 14% of total loans and 25% of loans to the corporate sector were overdue by 90 days or more. The construction sector is the most afflicted, with NPLs now exceeding 60%. Even these numbers could understate the magnitude of asset quality erosion, as indicated by the increased migration of loans into the highly impaired and crossover risk categories.

Banks tend to restructure a significant part of non-cash generating loans and delay the recognition of impairment, subject to the formal availability of collateral.

As credit quality deteriorated further in 2012 - an increase by one-third in loan impairment, driven by a 32% increase at Nova Ljubljanska Banka (NLB) and a 92% increase in Nova Kreditna Banka Maribor (NKBM) - the government responded with a range of tools including a Bank Asset Management Company (DUTB/BAMC) established to facilitate the clean-up of balance sheets and the further recapitalization of the banking sector. The BAMC law allows for up to €4bn in state-guaranteed bonds to fund asset transfers and an estimated €900m as a recapitalization target, based on a series of bottom up stress tests performed by the Bank of Slovenia and the IMF in 2012.

The government subsequently earmarked €1bn for recapitalisation, with asset work-out and transfer to BAMC as the core solution. NLB alone is supposed to transfer €1.9bn of gross loans to the BAMC.

2. BAMC is the central platform for bank restructuring but is still a work in progress.

According to the December 2012 Act on the Stability of Banks, BAMC remains fully responsible for asset workouts until 2017, when they revert to the Slovenian Sovereign Holding. A proposal to extend the lifespan of BAMC is under discussion; another suggestion is to lift the obligation for BAMC to sell at least 10% of assets annually and eventually to allow private co-investors in BAMC (like in Spain or Ireland).

As the first step - and in spite of the non-uniform nature of corporate NPLs - the largest banks' bad assets were supposed to be split into the three "packages": claims to companies already in bankruptcy are the first to be transferred to BAMC, followed by residential real estate assets, and then all the rest (including financial holding companies, where restructuring could be more complicated as it would require selling and liquidating multiple operating subsidiaries).

On the other hand, corporate borrowers in Slovenia tend to have loans from more than one bank, so in theory BAMC could benefit from its ability to consolidate the claims to a single borrower from individual participating banks. The by-laws adopted in March 2013 specify its operational asset-selection processes and valuation methodology, as well as some criteria for granting government guarantees. BAMC also needs to be adequately equipped for legal and workout processes as corporate restructuring procedures could eventually take the form of out-of-court restructuring.

3. Initially based on stress tests, the restructuring programme is now facing delays due to the EC's complicated review and approval procedures.

The Bank of Slovenia's stress tests have employed a series of assumptions to quantify potential transfer values: a total of €3.3bn was supposed to have been transferred to BAMC in 2013 at an average transfer value of 38% and total transfer cost of €1.1bn, potentially leading to the NPL ratio falling in the corporate segment from 25% to 9%. This also implied about 60% gross corporate NPLs at the three largest banks (NLB, NKBM and Abanka) would have been transferred to BAMC. The 38% number was accordingly presented to investors as the finance minister went on the road in late April.

In mid-May, as part of the 'European semester' policy coordination, the EC put forward new requirements, over and above what had initially been agreed between Slovenia and the ECB. They included system-wide asset quality review, and a new round of stress tests has been launched to ensure that the recapitalisation plans for each participating bank are consistent. The EC also requested that each tranche and its corresponding valuation be approved by the Commission. The EC's Directorate General for Competition (known in Eurocrat parlance as 'DG Comp') said it wanted to see the outcome of both the stress test and the asset quality review before giving its final approval to the restructuring programme (state aid case) for NLB and other banks. The government is now in talks with the EC to avoid individual tranche approvals.

4. Now stress-tests and asset quality review are more intimately intertwined. The spring BoS stress test was a typical top-down approach, with all the resulting "garbage in-garbage out" problems (we have discussed its assumptions in previous publications). The additional stress tests now required by the EC (Oliver Wyman has been retained for the job) will rely upon the results of the asset quality review (AQR), which was in turn assigned to Deloitte. Anecdotal evidence holds that around 50 Deloitte employees have been parachuted to NLB headquarters to do the sampling and individual analysis of selected exposures.

In this sense, the stress test will now have more of a bottom-up flavour than previously. Deloitte will not just review those NPLs "pre-packed" to be transferred to BAMC, but all corporate loans at the three largest banks, ie. the "good bank" part of their balance sheets as well. The review and stress testing are set to be completed in about three months (by the end-September for NLB, if all goes well).

One of the main stumbling blocks in the AQR is that, unlike in previous rounds when the ECB agreed quality groupings (and, hence, the valuations for the transfer) predominantly based on the value of the collateral, whatever this might mean in practical terms when pledged assets are highly specific and the market very shallow. Now, what Deloitte is asked to do is to put a greater emphasis on a borrower's standalone ability to repay (eg. its own cash flow generation capacity).

5. AQR aims to ensure no new requests for state aid, but it is highly discretionary.

According to EC documents, "the exercise should be system-wide to ensure lasting stability of the banking sector." This appears to us to be a valid reason, and a good rationale for a slower but more diligent progress, assuming there is any direct link between the speed and the quality of the outcome.

Yet it is clear that whatever comes out of it will be of little relevance for any purpose other than meeting the DG Comp's requirements. We are talking of a country where the corporate sector is notoriously overleveraged and the economy is in a deep recession. Barring refinancing, most borrowers will be deemed weak in any such cash-flow-focused exercise. AQR is therefore almost inevitably going to be highly judgmental, prone to model error and potentially outdated at the point of release.

This is what makes the whole exercise nearly pointless - DG Comp will apparently be in a position to claim, that 'to err on the side of caution, we think the Deloitte/Oliver Wyman numbers must be further stressed by x% to have a good safety margin, and then the average transfer price goes down by y%).

Forget the average 62% haircut, in short - it will have to be higher - simply because there is no objective way of knowing where it really should be, and the EC is not in the mood to make mild assumptions. The "costs of a banking crisis clean-up" are always, and unavoidably, underestimated.

6. While the terms of balance sheet clean-up are not finalised, banks are focused on maintaining adequate liquidity, which in turn creates a vicious loop for recovery. It is no news that Slovenian banks are increasingly dependent on the ECB and government funding. System-wide LDR at YE12 was 1.4-1.5x; central bank (read ECB's) liquidity was about 8.5% of total liabilities (vs. 4% at YE11).

In terms of immediately available collateral that banks can quickly deploy to get liquidity, ECB rules imply a 7% haircut for Slovenian government bonds (2nd risk category given the rating), but if downgraded even by one notch it goes up to 12% with a moderately negative impact on liquidity as those currently pledged are mostly state-guaranteed loans rather than bonds.

At the same time, state-owned banks have undergone the process of deleveraging as they have repaid almost all wholesale market funding (with the remainder being from development institutions). State deposits at NLB are at about €1bn, of which €400m is to be converted into equity.

Retail deposit pressure was significant during the Cyprus events, but mostly subdued with a manageable degree of conversion from contract regulated term deposits into current accounts.

7. Even if liquidity is adequate, in all situations like this it is difficult to find an investor for a bank with doubtful portfolio, uncertain recapitalisation prospects and stagnant new lending. So far, Slovenia has been providing support to the NLB and the NBKM to keep them just above the threshold. The former received €320m in capital from CoCos purchased by the state in 2012, subsequently triggered and converted into equity, and a further €400m recap is expected to be received from existing government deposits by the end of 2013. NKBM got €100m through a similar exercise in CoCos plus €400m planned over the next five years. Ultimately, and subject to a successful recapitalisation via the bad bank, the government would like to re-privatise the banks, or at least sell a stake to a strategic investor.

NKBM, shortlisted for privatisation among 15 other companies that are put up for sale (even before the State Asset Ownership Strategy is re-written and reconfirmed by the coalition), is reportedly in some form of due diligence. We remain sceptical as to the likelihood of the bank being sold.

Petr Grishin is Head of Macro Research at VTB Capital