Turkey’s abandonment of its quasi-currency board regime does not yet mean the end of the IMF programme by a long shot, but the odds of its collapse from within have clearly risen sharply. In the aftermath, the central bank’s decision to provide liquidity to the financial system has come as little surprise given the extent to which the payments system nearly ground to a halt, but would also signal further sharp declines ahead for the lira. The dilemma in which policymakers find themselves now (as opposed to November) is that in having abandoned the pegged FX regime, the banking sector is set to suffer whether the central bank decides to provide more liquidity or not. The triage exercise to follow is likely to follow the lines of what will hurt the economy least, and as in the past, preservation of the banking sector is likely to come before preservation of the value of the exchange rate and inflation. Restoration of banking sector stability in some form is as crucial to the government’s own ability to meet financing needs as it is to the corporate sector.
In November, the liquidity squeeze claimed one major bank, and while the sector was clearly weakened to an extent that made it extremely vulnerable to a shock such as last month’s, it survived in part due to the fact that the central bank’s decision to provide liquidity above and beyond its IMF-mandated Net Domestic Asset target was accompanied by a maintenance of the FX regime. Banks took a hit on their treasury bond holdings, but not (at least to any similar degree) on their sizeable net open FX positions. This time around, the float of the lira in the face of sky-high interest rates implies a hit to both fixed rate treasury bond holdings and net open FX positions. A decision to sustain liquidity injections to the banks will potentially help ease losses on bond holdings, but exacerbate losses on open FX positions that have yet to be closed due, simply, to the lack of foreign exchange available. Any attempt to tighten monetary policy as a means to stabilise the exchange rate, however, is likely to exacerbate the squeeze on domestic liquidity. Due to the Treasury’s overarching need to roll over what was the largest-ever maturity of treasury bonds last month, the decision to cut maturities to one month effectively imposes a deadline of 20th March on the authorities to stabilise the situation.
PRIORITY 1: SAVING THE BANKS
The authorities appear to have concluded that providing liquidity directly to the banking sector offers it the best chance of survival, even if it means a sharply weaker lira in the short term. Liquidity injections are likely to find their way into the FX market in relatively short order as banks seek to close their open FX positions, which suggests that barring massive injections, domestic lira rates will remain extremely high in real terms. That said, any measure of real rates will be extremely difficult to ascertain over the next several weeks as the devaluation works its way into inflation. While it is arguable that the Turkish banking sector was already in crisis prior to the past week’s developments, the scope of the crisis has now broadened to even the private banks deemed healthy until now. The takeover of a small bank less than a week into the FX float is not likely to be the last. Failure to continue liquidity injections at this stage would almost certainly result in multiple bank failures (as indeed two major state banks appeared to do last week. On the other hand, as the exchange rate spirals weaker and domestic rates remain prohibitive, the probability of a flight-to-quality type run on the banking sector, favouring the larger private banks which were deemed healthy until recently, recedes in favour of a generalised run.
The likely path forward should be along the following lines:
1. Liquidity support – 100% deposit guarantees have already been implemented. The cost-benefit analysis that has likely taken place through the weekend will be along the lines of determining whether there is enough value left in the system to justify defending it against further loss. Moreover, the fact that many banks are already in a state of deep insolvency exposes the government to massive payments to cover deposit guarantees. These factors would argue against providing liquidity support to the sector. One obvious risk in this regard is that political and economic costs are muddled. The choice amounts to one of whether to allow depositors rather than taxpayers to assume the cost of the banking crisis. The growth of the banking sector during the past five years, and its importance to the corporate sector as a source of credit, however, provides a powerful incentive to extend liquidity support in spite of the damage already caused. For this reason, and the importance of a stable payments system for the government’s own financing, the government is likely to move forward with direct liquidity support.
2. How aggressive will the government be? – the government’s debt/GDP ratio prior to the crisis was not overwhelming at roughly 50% of GDP, but clearly, the cost of resolving the banking sector and the sharp rise in real financing costs risks a massive jump in debt/GDP over the relatively short term. Beyond political constraints, failure to push for more aggressive restructuring of the banking sector up to now may have reflected the desire to avoid large outlays of public funds or assumption of bad banking sector debts in the early stages of the IMF programme. The costs of this less aggressive approach are now coming home to roost, and the government is now forced into the position of assuming large private sector liabilities during a period in which its own access to liquidity is coming under pressure.
3. Rather than closing failed private banks, nationalisation is even more likely now than in the past given the desire to prevent a further deterioration in confidence and an all-out bank run. Resolution of the failed banks will be delayed, pending stabilization in confidence and markets. Recapitalisation efforts will take place once liquidity injections stabilise the financial system.
The result is likely to be a sharp rise in government indebtedness as it is forced not only to bear the higher funding costs associated with the crisis, but is likely to be forced into assuming much of the bad debt generated by the crisis. Based on past systemic banking sector restructurings globally, a cost to the government of 12-20% of GDP would typically be expected, though gauging the costs in Turkey, where the degree of cross party lending and FX exposure may remain uncertain even to the authorities at this stage, will take several months, if not years.
EXTERNAL DEBT RISKS
According to the latest joint IMF-BIS statistics on external debt, Turkey had $25.2 billion in liabilities outstanding to foreign banks due within one year of 30 June 2000. Non-bank trade credits of one year or less were $2.1 billion as of end-September and foreign debt securities due within one year were $1.1 billion.
For the authorities, the next redemption of external bonded debt comes only in mid-April, with the maturity of a $650 million equivalent Samurai bond. Another $1.5 billion in maturities come only in November and December. We assume that FX reserves, together with IMF balance of payments support, will be sufficient to stave off a sovereign external debt default.Though this crisis is arguably more severe than those suffered previously, Turkey has a strong history of having avoided external debt defaults during periods of macro distress. The risks are much higher, however, for private sector debtors as credit lines are scaled back and domestic sources of credit disappear as well. As with domestic bank liabilities that the public sector may be forced to assume, the market will need to attempt to assess the extent to which the government will be forced to take on external debt liabilities as well. We voiced this fear several months ago as one in which the official decision to allow banks under state administration to run net open FX positions well in excess of prudential limits risked blowing up in the government’s face. Even if private banks have, as the government has promised, stuck to rules limiting net open FX exposure to 20% of capital, the resulting losses and inability to close these positions should be recognised as potentially increasing the government’s short-term FX liabilities above and beyond what is recognised in the official external debt statistics.
IMPLICATIONS FOR MARKETS
There is still much to be uncovered as to the rot in the financial system and the fallout on the government’s balance sheet. The government appears to have little choice but to provide liquidity in the short term, and even then, the probability that this move stabilises the payments system cannot be taken for granted. The immediate victim of this strategy will be the lira, given the still sizeable net open FX liabilities, the likely cut in credit lines, and the general public demand for US dollars given the parlous state of the banking sector and the expected surge in inflation. As noted last week, gains in the equity market related to simple asset price inflation (in line with higher CPI) will come as little surprise, but the risks to the corporate sector are arguably as severe as those to the banking sector, given the importance of the latter in providing credit to the former and, indeed, the prevalence of cross party lending within some of Turkey’s largest industrial conglomerates.
Exporters are the obvious safe haven, and given the overvaluation of the lira until last week, there should be plenty of idle capacity with which to boost output.The key for this sector is that it can expand production without recourse to bank credit, which is likely to be lacking even to this sector. Finally, for fixed income, the backdrop has not changed, and the risks are for substantial further downside as it becomes clear that the government’s debt burden is set to increase sharply. Perhaps most disconcerting is the lack of any signs from the government that it is willing to recognise its culpability for the crisis. Failure to do so is likely to hasten the loss of public support for the government and weaken the government’s response to the crisis. The difference between 2001 and 1994 is arguably that the private sector believed much more in the current programme, and as such was less hedged against its failure. If the government decides to follow the old policy prescription to such crises – central bank provision of liquidity and a restoration of the depreciation-inflation spiral, it is in for yet another rude awakening.
Gene Frieda is head of operations at Forecast Pte, Singapore, a subsidiary of 4Cast.