Turkey’s Controversial Banking Sector Package
Augusto Lopez-Claros Tolga Ediz
Investors have been concerned for some time that Turkey faces a systemic banking sector problem, which is impeding financial intermediation and adversely affecting growth prospects. Both bank and corporate balance sheets took a major hit fr om the fall in the lira and from the sharp rise in real interest rates precipitated by consecutive financial crises. The result was a sharp fall in credit extension and the emergence of a rapidly growing non-performing loan (NPL) portfolio last year.
Turkey’s case highlights the close links between the state of the banking system and the health of the corporate sector. Such links call for a comprehensive solution, with the problems in the banking and corporate sectors ideally tackled simultaneously. So far, the authorities have been overwhelmingly concerned with the banking sector, dealing with banks’ foreign exchange liabilities, the closure of a number of private banks and the restructuring of state banks. There has been no equivalent attempt to address related problems in the corporate sector and, in particular, the implications for it of a growing debt overhang, the counterpart of the banks’ NPL problem.
The authorities, with the Banking Regulatory and Supervisory Agency (BRSA) taking the lead, have finally come up with a comprehensive plan. Before turning to an assessment of this plan, however, we review briefly the broad consensus of “best practice” that has been established in the regulatory and academic community on how a banking sector restructuring package should be put together. A credible approach involves three elements:
Recognising and allocating losses, including a transparent assessment of the “true” magnitude of the non-performing loan portfolio. In order to minimise moral hazard, it is important that shareholders bear the brunt of the losses.
Restructuring financial claims, including recapitalising viable banks, resolving non-performing loans, and closing unviable financial institutions. This will often involve the setting up of asset management companies (AMCs) to deal with NPLs, and a restructuring of corporate debt with the public sector sometimes acting as an agent of change and as a co-financer of the restructuring project.  The public sector’s involvement should be limited, however. Recapitalisation should involve injection of tradable instruments, allowing the public sector to gain a stake in the banks so that the fiscal accounts benefit from any upside to the resolution. Legal changes are usually needed to ease bankruptcy procedures. 
Operational restructuring, to restore the profitability of banks once they have been recapitalised and NPLs have been removed from their balance sheets. This often involves a change in senior management, an improvement in systems and controls and major downsizing of the bank in order to ensure that it is put on a sustainable growth path. Otherwise it is likely that the bank will soon become insolvent, with all the loss of public funds this implies. Past experience has shown that it is this part of the restructuring process that is the most difficult.
Turkey’s plan: just a start
The highlights of the BRSA’s banking sector package are as follows:
Banks with risk-adjusted capital ratios of less than 5%, and accounting for more than 1% of total banking sector assets, will receive a direct capital injection from the Deposit Insurance Fund (DIF), conditional on shareholders injecting an equivalent amount.
Banks with capital ratios of more than 5% but less than 8% will receive cash credits from the DIF, with the latter taking seven-year maturity convertible bonds as collateral. There is to be no requirement on banks’ asset size or bank owners to inject capital.
Banks’ accounts will be prepared by independent auditors (contracted to the bank) which will then be checked for “sound auditing practices” by another firm appointed by the BRSA. The BRSA will then verify the findings with its own assessment.
Banks are to be required to lend 60% of the amount injected by the public sector to the corporate sector. In addition, public sector banks will receive an injection of around $1.1bn to lend to the private sector.
The authorities estimate that the operation will cost around $4bn, although it is not clear whether this includes the money to be injected to state banks. It is the authorities’ expectation that they can use IMF or additional World Bank funds to finance the package. The second stage of the banking sector restructuring will involve the setting up of Asset Management Companies (AMCs) that will be in charge of resolving non-performing loans and restructuring existing corporate loans. The details for this phase of the process (see below) are yet to be worked out, although the law allows the DIF to take a 20% stake in AMCs.
We have argued before that a comprehensive approach to addressing the non-performing loan problem of the banking sector was an important missing element of Turkey’s IMF programme. The package outlined so far seems incomplete although the BRSA is expected to publish an “action plan” that should provide detail on the implementation of the package, including measures on restructuring NPLs and setting up of AMCs. To the extent that the recapitalisation plan constitutes the first stage of a comprehensive solution, it is a welcome addition to Turkey’s efforts to rehabilitate the banking system. However, it matters how such an operation is carried out. Even at this early stage, we have several concerns.
First, the package has the flavour of a wholesale bail-out of the banking system and introduces a strong element of moral hazard. Ten banks, accounting for 97% of total assets, qualify for direct capital injections from the DIF. Prudent bank owners that had forgone the possibility of high returns, and therefore are not now in need of a bail-out, are penalised. We would have preferred a more selective approach, coming on top of the closure of a significant number of undercapitalised institutions.
The authorities argue that moral hazard concerns are overdone given that bank owners have to put up capital of their own and that the treasury will gain a share of the bank. In their view, prudent bank owners are not penalised, as they will not have to relinquish shares in their company or receive expensive loans from the treasury. In our view, this does not alter the “heads I win, tails you lose” incentive for bank owners. The public sector’s option to retain a share of the bank is essentially meaningless, given that it will receive 100% of the bank in the event of insolvency. And the true reward for prudent bank owners would have been an increased market share from the exit of under-capitalised and less well-run institutions.
Second, the recapitalisation operation, by itself, might not significantly reduce the perception of risk in the banking sector. Banks’ capital ratios should vary across banks according to the regulator’s assessment of asset quality and management skill. In many cases, this means that banks’ capital ratios should exceed 8% by a reasonable margin.  The current plan seems to class banks with capital ratios of above 8% as adequately capitalised without exception.
Third, past experience suggests that bank recapitalisation should occur in conjunction with operational restructuring. At a minimum, this requires that senior management should change or be under very close regulatory scrutiny. It is unclear whether capital injections will be made conditional on specific management actions and, apart from one or two exceptions, there is little evidence that private sector banks are willing to undergo painful restructuring. It is indeed possible that the BRSA’s “action plan” will address these concerns. Maybe the BRSA will be tougher on poor management practices and give specific targets to banks on operational profitability. But it is also the case that a credible action plan would require tackling a deep-seated corporate governance problem in Turkish banks, which are often run by family-owned conglomerates with close links to the political establishment and with bank management often beholden to the interests of the largest shareholders, reinforcing perceptions of “crony capitalism”.
Fourth, we are worried that the banking sector package will reduce the likelihood of foreign entry to the banking system in the short-term. Although foreign banks will welcome the prospect of merging with better capitalised firms, domestic bank owners are likely to be more reluctant to sell and capital injections will complicate banks’ valuations. The recent collapse of the BNP-Paribas/Finansbank merger is a bad omen in this respect.
Fifth, we are puzzled by the requirement that 60% of public sector capital injections should be on-lent to the private sector. A quantitative credit target may not be consistent with operational restructuring which might require a restructuring of banks’ balance sheets in favour of liquid instruments. (Although it may be true that this is a populist measure to appease the corporate sector and would in any case be impossible to monitor; this is hardly a credible way to get the restructuring process started).
The way ahead: hard work starts now
These concerns not withstanding, it is the next phase of the restructuring package that will determine the success of the overall plan. Hence, a final verdict can only be reached after the publication of the BRSA’s “action plan”. The authorities need to make sure that a “true” picture of the size of NPLs emerges as quickly as possible. Past experience suggests that the precise nature of the asset resolution process, in particular the set-up of AMCs, wh ere the authorities will face difficult trade-offs, is often crucial for success. And the authorities should prepare for the eventuality that much larger volumes of public sector funds could be needed in this regard. Further legislative changes, for example on the bankruptcy code, may have to be implemented in due course. In short, there is a long way to go. In the meantime, we fear that the authorities will come to regret such a broad recapitalisation package in the period ahead.
Selected banking sector indicators: All data as of 2001Q3
 Part of the difficulty is that co-ordination issues arise when the problem is of a systemic nature. See “Financial Restructuring in Banking and Corporate Sector Crises. What Policies to Pursue?” by Claesens, et al, World Bank WP, April 2001.
 See “Recapitalizing Banking Systems: Implications for Incentives and Fiscal and Monetary Policy,” Patrick Honohan, World Bank WP, 2001
 See “Bank Capital and Regulatory Intervention” by Tolga Ediz, William Perraudin etal FRBNY Economic Review 1998.