Financial markets in the periphery (bonds, CDS and in some cases banks) have become rather volatile in recent days. This escalation follows several weeks of spread widening in the three smaller peripheral markets (notably Ireland and, to a lesser degree, Portugal and Greece). While each country has its own interesting idiosyncratic story, the danger of contagion has clearly increased. In this note, we discuss political deliberations on tapping the European Financial Stability Fund (EFSF). We outline how the process would work if it were to be activated and highlight the likely market reaction that could be expected on the announcement, based on what was observed in Greece. On balance, we believe that the sharp rise in market tensions over the last few weeks has increased the chances of the EFSF being tapped. This is in particular true for Ireland, in our view, where it seems increasingly difficult for the government to effectively backstop the problems in the banking system. We would stress, however, that no country will apply to the EFSF lightly or ‘just' to reduce debt-servicing costs: tapping the EFSF is a monumental decision that will shape economic, fiscal and financial policies in that country for many years to come. It is a step that a government would only take in case of no other viable alternative, in our view.
On balance, it looks increasingly likely to us that Ireland might not be able to avoid going to the EFSF eventually. That said, we would expect the Irish government to put up a ‘good fight'. As the Treasury is sitting on a comfortable cash buffer of €20 billion, it is fully funded until next summer. In addition, it can and has used the National Pension Fund Reserve (NPFR) to recapitalise the banks. A proposed change to the discount factor used in calculating the Fund's pension liabilities would free up additional cash reserves. Hence, contrary to Greece back in April, Ireland should be able to hold out for a while. In our view, the government will likely use this ‘borrowed time' boldly to restore market confidence. The four-year fiscal plan revealed this week is ambitious. It foresees budget cuts of €6 billion for 2011 alone (equivalent to 3.8% of GDP) and a total of €15 billion over the next four years. But because the real source of the Irish issue is not the budget deficit, but the problems in the banking system, fiscal austerity might not be sufficient to restore market calm. A key obstacle for the Irish government in the context of going to the EFSF is that other European governments will likely demand an overhaul of its highly competitive corporate tax system - which the Irish view as key to their attractiveness for inward foreign direct investment (FDI).
Portugal seems less in the market spotlight than Ireland - at least at this stage. This does not necessarily mean that the Lusitanian economy will be able to escape going to the EFSF. Rather, it is an observation that market dynamics look somewhat more benign for now. For example, 5y CDS spreads, at around 440bp, are about 150bp lower than in Ireland. And the Portuguese government is also sitting on a cash buffer (around €10 billion), this year's funding looks almost done and there's no need to worry about meaningful redemptions until next spring. What's more, from a fundamental perspective, Portugal has a productivity problem which causes it to be stuck in a low growth and poor competitiveness situation.
Hence, the genesis of Portugal's imbalances is different from that of the other EMU peripherals. While in Greece the key issue is fiscal indiscipline, in Spain a credit-fuelled housing boom-turned-bust, similarly in Ireland but coupled with an outsized banking sector, Portugal faces various structural deficiencies. The upshot is that the rebalancing in Portugal has an inherently structural nature and is unlikely to correct very easily and quickly, e.g., the current account deficit is still in double-digit territory. This is a reason for concern. So is the accumulation of external debt. The fiscal situation is only a by-product. Yet, while sudden shocks to the economy seem more unlikely in Portugal than elsewhere, the main risk is contagion. If markets behave in ‘systematic mode' and continue to associate the current difficulties in Portugal with, say, those in Ireland or Greece, access to funding might dry up to such an extent that Portugal too might not be able to avoid going to the EFSF eventually.
Trading Local Politics in Europe
Once a country has agreed on an adjustment programme with the IMF/EU authorities, the key issue is whether it can deliver the measures mapped out in the loan agreement. This largely comes down to its political stability and the parliamentary strength of the incumbent government. Markets often get spooked by reported shortfalls in revenues or adverse trends in economic activity. In our view, these concerns are only of secondary nature. If the country in question has implemented all it agreed to and the outcome still turns out to be falling short of (jointly generated) projections, a recalibration of the programme will likely follow. But non-compliance with the agreed action plan could immediately cause the next loan tranche to be withheld.
The IMF staff can tell many stories of situations in emerging markets where the Fund was ready to provide funding (against strong conditionality) but had difficulties in finding a political counterparty in the country that had a sufficiently strong mandate to negotiate an adjustment programme. While the IMF would typically wait patiently until the political situation has cleared up, Europe might not have the same luxury. Because of the close-knit nature of the EU and the EFSF, we believe that severe political instability in the periphery would increase the probability of a spillover of the crisis into the rest of the euro area, up to and including the core countries.
In Ireland, the financial crisis has indeed started to take a toll on the government's political support. Initially, Prime Minister Cowen's coalition started out with a clear majority in the Dail. But in the upcoming budget vote on December 7, it will have to rely on the support of the independent members of parliament. A total of four seats in the Dail are vacant at the moment and the High Court has recently ruled that by-elections need to be held as soon as possible. In reaction, the government has announced that it will hold one by-election in Donegal South-West on November 25, just a few days ahead of the presentation of the detailed 2011 budget on November 28. Depending on the outcome of this by-election, it might become more difficult to get the budget passed. In our view, markets will likely be nervously watching the political debate in Ireland in the run-up to the budget vote. On balance, we expect the budget to pass, but it could turn out to be a cliffhanger.
In Portugal, the minority government has already secured initial support from the opposition on the 2011 budget. The government has announced tougher and specific spending cuts. Overall, the belt-tightening measures represent savings of 3% of GDP, with the biggest focus on spending cuts (about two-thirds of the total). The goal is to bring the budget deficit down from an expected 7.3% of GDP this year to 4.6% in 2011. These targets seem broadly achievable, but only if all the measures are implemented swiftly and fully. The main risks relate to execution, especially in the context of a likely double-dip next year. After an expansion of 1.3% in 2010 on our projections, we expect growth to enter negative territory once again and foresee an economic contraction of the same order of magnitude in 2011.
While social cohesion is quite tight in Portugal and the number of workdays lost due to strikes compares favourably with other EMU countries - including some at the core of the euro area - investors will scrutinise any fiscal slippage, political tensions and social frictions to a great degree, in our view.
What Pushed Greece to Request Financial Support?
Not the Economy...
Deteriorating economic and fiscal fundamentals alone are not enough, we think. Indeed, when Greece applied for financial support in April, not even 1Q GDP growth was known to policymakers, market participants and economic agents. And, at that point in time, the magnitude of the latest reported economic contraction in 4Q09 (-0.8%Q) was far smaller than that in the euro area (-1.8%Q) - where the recession started earlier. Similarly, the public finance figures have been revised multiple times since, and do now show a much more worrying picture than initially reported.
This is not to say that there was no sign of impending deterioration in the outlook. Of course, more forward-looking (e.g., economic sentiment) or high-frequency (e.g., industrial production) indicators had already started to show some divergence between the growth path in Greece and the euro area as a whole. But, back then, the near-term trajectory did not seem to be that different in Greece relative to other EMU peripherals. What's more, Greece had announced just part of the belt-tightening it eventually ended up implementing.
Debt sustainability is not the issue either, in our opinion. Of course, Greece's long-term funding model was not viable, but the fact is that the market was willing to buy Greek bonds and provide funding even in the few months (and indeed weeks and days) before the crisis. While it was getting increasingly more expensive for Greece - but from a fairly low interest rate level - for quite a while there was a price the bond market was comfortable with.
Put differently, the issue is (the sudden drying up of) market funding. Basically, solvency and liquidity risks, which in theory are often portrayed as separate risks, are in practice deeply intertwined. For example, even if a government is solvent (i.e., its promise to repay the debt seems fully credible), market sentiment might change quickly and perhaps in a manner that has nothing to do with the economic fundamentals. Should that happen, a sovereign could face a credit event regardless of the viability of its long-term funding model.
Policymakers have some levers of defence that they could use, including inviting domestic financial institutions to participate more actively in the primary bond market and using existing cash reserves that the government might hold. The ECB's Security Markets Programme (SMP) also helps to stabilise bond market conditions somewhat, even though it is unlikely to backstop a full-blown buyer's strike. Only the EFSF can serve this purpose.
...but Broad Market Conditions...
Ultimately, a sovereign (or a corporate, for that matter) will need to apply for external emergency funding in order to prevent a default or restructuring when the market is no longer willing to extend credit. So, market conditions are the key tipping point that forces the decision to request financial support for any euro area country and beyond.
This is not to say that there is an element of pure automatism in an eventual default or restructuring - as governments, being able to impose taxes, privatise or sell assets, etc., effectively decide to restructure based on their own political considerations too.
Rather, it is an observation that broad market conditions - coupled with negative newsflow - can push a country towards a credit event (e.g., restructuring, default, etc.) regardless of whether economic conditions warrant such event.
...Led Greece to Seek Financial Help
Before the European governments and the IMF stepped in to save Greece back in April, Greek (and international) markets were severely disrupted. In particular, the short-end segments of sovereign credit curves (e.g., 2s5s) were inverted in virtually all EMU peripheral countries - as well as in countries ‘in between' such as Italy.
Put differently, the market was pricing - to various degrees - a high chance of default over the following several years pretty much across the board. In the case of Greece, that probability had risen to over four chances out of five at some point. And the market is much more concerned about the near term than it is about the medium-to-long-term - as signaled by the inversion of the credit curve.
Another sign of market stress - connected with the curve inversion mentioned above - was that interest rates rose rapidly over a very short timeframe. For example, the 10y yield on Greek bonds virtually doubled, rising from 5% to 10% within 48 hours. These dynamics were probably exacerbated by Greece's debt investor base, i.e., with two-thirds of the bonds held outside of the country, that resulted in a foreign buyer's strike, as overseas investors found it increasingly difficult to read the situation - the home bias works to the detriment of the country. Put differently, regardless of further falls in bond prices, making them attractively valued, there was no buyer around.
Similarly, the stock market was considerably under stress back in April. In Greece, both the general index and bank stocks were experiencing continued downward pressure and, at that point in time, it seemed that bank deposits were decreasing month after month - thought this trend turned out to be short-lived. Further, it was not clear, over last spring, whether social unrest would have ensued on a large scale (it didn't - at least so far).
Finally, there was scepticism on whether the government had the political will to address the various structural deficiencies, and uncertainty about the reliability of the public finance figures and the state of local governments (which, it turns out, account for just 1% of total debt in Greece - see The Local Finances: Do We Need to Worry, October 18, 2010). All this contributed to negative newsflow that fuelled market jitters.
What Happens to Credit Curves and Spreads?
More broadly, our interest rate strategists note that the chance of a default at a given point in time is conditional on not having defaulted already, so it initially rises and then falls as the probability of default rises (see Trading European Sovereign Spreads: A Tiered Approach, October 27, 2010). What's more, as spreads widen from low levels, the distribution probability of default becomes humped in the four-year area, before becoming increasingly skewed towards shorter timescales as spreads widen further. This explains why credit curves initially steepen as credit spreads widen and flatten when credit spreads widen significantly.
Typical sovereign credit curve behaviour: (i) the 2s10s credit curve stops steepening and starts flattening when the 5y credit spread reaches about 150bp, inverting at around 500bp; (ii) the 5s10s credit curve tends to stop steepening and start flattening when the 5y credit spread reaches about 150bp, inverting when it is around 300bp; (iii) the 2s5s credit curve tends to stop steepening when the 5y credit spread reaches 150bp and starts flattening when the 5y credit spread reaches 350bp, inverting when it is around 650bp; and (iv) 5y credit spreads tend to underperform 2y and 10y as spreads widen from 150bp to 350bp.
Is it Different Now?
Market reaction, when Greece's funding and solvency problems became more evident, was harsher than many had predicted. In other words, most of the above-mentioned dynamics took place over a very brief time span - as uncertainty over an outright default was considerable, political tensions and delays in the policy response were unsettling investor sentiment, and there was neither a historical precedent in Europe nor any plan to deal with such situations.
This time round we believe that there will be no uncertainty around what happens if another country is not able to access market funding, i.e., this country has the possibility to get financial support through the EFSF. Moreover, the ECB is now buying bonds. That may help to relieve some stress in the most dislocated markets and perhaps lead to a somewhat less chaotic market response.
Tapping the EFSF - How Does it Work?
Applying for financial help requires a number of steps, i.e., several weeks are likely to elapse between the formal request and an eventual disbursement of funds:
• Meeting the requirements of the EU Treaty, Article 122(2) - emergency assistance: The first hurdle for any euro area government planning to request financial support is to show that it is facing severe difficulties caused by exceptional circumstances beyond its control.
• Unanimous decision by all guarantors: In practice, all 16 euro area member states (apart from Greece, which is not a guarantor of the EFSF) have to accept the request for financial support. This is perhaps more likely to be the case if contagion worries become a key market driver once again.
• Negotiation of Memorandum of Understanding (MoU): The European Commission, in liaison with the ECB and the IMF, then needs to negotiate a MoU on specific economic policy conditionality and on several aspects related to the monitoring process.
• Signing of the MoU: The European Commission, on behalf of the euro area governments, has to sign the MoU - once it has been approved by the Eurogroup Working Group.
• Formulation of main terms of the loan agreement: The European Commission, in liaison with the ECB, then proposes the main features - but not all the various detailed terms - of the loan agreement to the Eurogroup Working Group.
• Negotiation of detailed terms of the loan agreement: The EFSF, in conjunction with the Eurogroup Working Group, negotiates the detailed terms of the loan agreement - which then needs to be approved by the member states.
• EFSF starts issuing debt: Before each disbursement, the European Commission, in liaison with the ECB, reports to the Eurogroup Working Group about compliance with the MoU. Then, the guarantors unanimously decide on whether to permit disbursement of the tranche of the loan.
The upshot is that the political hurdle to apply to the EFSF is quite high, in our view. This is not to say that no country will ever consider applying. Rather, it is an observation that the EFSF is likely to be the last resort for countries in funding difficulties when there is no other alternative left - not a facility that will be tapped ‘just' to save on funding costs.
How Will the EFSF Fund and Then Lend On?
The structure of EFSF debt issuance now put forward not only earns EFSF issues the AAA credit rating that was always planned; it also effectively removes the risk that different EFSF issues will be treated differently by the financial markets, depending on the composition of the guarantee of each issue. Even though the structure of the EFSF is a rather complex one, and despite the EFSF not being able to prefund, we don't expect the funding to stand in the way of a quick resolution of a bond market buyer's strike on the EMU periphery. The funding costs of the EFSF will be key in determining the costs of the emergency lending facility. At this stage, we would expect such a loan to carry an interest rate of 5-6.5% per annum (EUR asset swap - 1.5% 2y and 2% 5y - plus 300-400bp margin, plus a 50bp upfront service fee distributed across three to five years).
Our interest rate strategists expect that the market will be able to treat any eventual EFSF issues as ‘generic' credit, rather than as ‘heterogeneous' credit, varying between issues (see Clarification on the EFSF Programme, September 24, 2010).
On top of the two credit-enhancement mechanisms described in the EFSF framework agreement, the EFSF introduced an additional loan-specific cash buffer. This buffer is sized so that 100% of the bond coupon and principal is backed by cash (invested in short-term AAA securities) or by AAA rated governments. The three credit-enhancement mechanisms are:
• Over-guarantee: Euro area members will each guarantee 120% of their respective share in the issue. For example, Germany would guarantee 27.9%*1.2 = 33.5% of an issue at first. So, the issue will effectively enjoy a 220% guarantee (100% from the country seeking funding - since a default event of the issuing country needs to happen before any external guarantee is triggered - and an extra 120% from the remaining guarantors that have not stepped out yet).
• Cash reserve: The full 100% of the issue does not go to the borrower. A 50bp fee and a margin (paid upfront by the borrower and deducted from the cash amount remitted to the borrower) will be kept as a cash buffer. In the event of a delay or failure to pay by the borrower, and if the guarantor's payment doesn't fully cover the shortfall, the cash reserve will be used as first line of defence to cover such shortfalls. The cash reserve will be kept and managed by the EFSF. Using as a guideline the 300bp margin paid by Greece under the IMF package, we can assume that the cash buffer will be roughly 9% of the nominal value for a 3-year EFSF issue (NPV of 3*300bp + 50bp).
• Loan-specific cash buffer: In order to further enhance the quality of the bonds issued under the EFSF programme, a loan-specific cash buffer will be established. This will be sized so that all principal and interest payments are covered by AAA entities (with stable outlook) or by cash. This buffer will be taken from the proceeds of the bond issuance together with the cash reserve (thereby reducing the amount of cash disbursed to the borrower) and will be invested in high-quality liquid debt instruments.
Our interest rate strategists weigh a ‘theoretical valuation' of EFSF bonds against ‘practical observations' of where comparable issues trade. They find that the theoretical valuation puts EFSF bonds at tighter spreads over Germany than those observed in the marketplace for comparable credits (e.g., EIB, EC, KfW). In particular, our strategists compute the cash buffers and AAA guarantee allocation for EFSF bonds that would be issued in a hypothetical scenario in which Ireland applies to the EFSF for funding, leading the EFSF to issue €10 billion of 3y debt. The cash reserve would be the sum of 50bp (the service fee), and the running margin fee (paid upfront). Assuming a 300bp running margin fee, as indicated in the EFSF Q&A document, we get a cash reserve of roughly 9% for that issue (€0.9 billion). So, in the case of Ireland going to EFSF, AAA countries would be insuring around 73% of the issue (computed as the sum of percentage of AAA guarantors multiplied by 1.2). Put differently, there is an AAA rated sovereign guarantee on €7.3 billion of the issue principal and on 73% of the interest payments (non-AAA countries offer a further guarantee on 47% of the issue).
Read Full Article »