It is rare and quite ironic, to say the least, that Lenin's words are used to address events in financial markets. His pithy phrase, however, undeniably captures the exhausting speed at which events have been unfolding around us over the past several years. These unprecedented events have elicited an unprecedented policy response. Rather unfortunately, not all policy actions in these exigent circumstances have been of the ‘right' kind. In this note, I argue that a rather long list of the ‘wrong' kind of policy decisions, including the recent bond swap by the ECB, have retroactively changed the rules of the game and introduced regulatory risk into financial markets. By subverting the rules and regulations that they themselves put in place, policy-makers have ensured that investors will demand to be paid yet another kind of risk premium in future crises, a premium for bearing the risk of retroactive changes in regulations.
Why hasn't such a risk premium asserted itself already? Because the ‘right' kind of policy response - QE and the related strategy of financial repression - has nullified the risk premia. QE, for example, has helped to push risky asset prices higher and bond yields lower, thereby acting in exactly the opposite direction to risk premia. The fact that the use of unconventional policies has been so aggressive has meant that the net impact on financial markets has been a benign one. Financial repression - a set of policies designed to keep bond yields or currency values lower than unconstrained market pricing would have generated - is part of the standard policy toolkit for extreme scenarios. QE or regulatory policies that push bond yields lower are one way that policy-makers can financially repress investors. If one argues that financial repression does change the rules of the game, then the same argument would have to apply to QE. This is clearly a grey area, but on balance, both QE and financial repression are best viewed as aggressive extensions of standard policy tools employed under extreme economic conditions.
Policy-makers may well use a similar combination of QE and financial repression in future crises and override a prospective rise in the risk premium by stimulating markets vigorously. However, if the willingness or the ability of policy-makers to be über-aggressive wanes in future crises, perhaps because of policy-maker concern over forfeiting public confidence through issuing too many central bank liabilities (see The Global Monetary Analyst: QE Coming of Age, February 1, 2012), the risk premium could well dominate asset markets.
What are the implications? The implications are profound and manifold, but I point out only two of the important ones here. First, peripheral Europe and EM economies know full well that a rise in yields during downturns or crises (which would happen if the risk premium was the dominant force) inhibits the rolling over of liabilities, public and private. In turn, this dampens the economic activity that was being financed by these liabilities.
Second, there are repercussions, away from economics, for the first principles of finance. If the market price in future crises also includes a premium for regulatory risk, it will give only a murky signal about how much of the macro and market uncertainty is in the price. Through their decisions, policy-makers would have made markets less efficient. This would hurt not just markets but policy effectiveness too, given how much more reliant policy has become on markets and investor sentiment.
To better understand the interplay between the benign and malign effect of policy decisions on asset prices, it is necessary to first understand which policy actions themselves are of a benign or malign nature and why.
Two kinds of policy actions - the ‘right' and ‘wrong' kind: In order to avoid grossly understating the grey area between what is ‘right' and ‘wrong', I prefer to make the distinction between the two using concrete examples. As a simple framework for classification, the ‘right' kind of policy measure is one that extends (even aggressively) the existing realm of policy without retroactively bending or changing the rules that govern financial markets. The ‘wrong' kind of policy violates this principle.
It is easy to see why such a broad distinction will not do. Take the example of financial repression. It is easy to argue that distorting asset prices and almost forcing investors and economic agents to expand their activities is simply an aggressive extension of the standard tool of policy rate cuts. However, my colleagues remind me that keeping official institutions senior to private investors is also an integral part of financial repression. Arguably, this is what the ECB has done. How then do I classify financial repression? Precisely in order to avoid broad, sweeping statements, a nuanced discussion follows.
A Laundry List of What I See as the ‘Wrong' Kind of Policy Decisions
• The ECB's recent refusal to accept losses on its purchases of Greek government bonds via the SMP programme: The main objection to a ‘wrong' tag is that official institutions, particularly those acting to safeguard the system, are always senior to private bond holders. The issue, however, is the timing and manner in which this seniority has been asserted. Had the ECB announced at the very outset of its SMP purchases that it would not accept any losses on its purchases in the event of a restructuring, investors returning to peripheral bonds would have been fully informed and could have decided to participate, keeping in mind that the bonds they acquire or hold would be subordinated to those purchased by the ECB. By announcing and exercising its seniority only after investors came back to peripheral bond markets, the ECB has effectively changed the rules in a way that I believe violates the spirit of the law.
• The introduction of the PSI: Deciding to use the PSI in July 2011 broke a prior promise to not involve the private sector before 2013. It sparked a major wave of contagion into other peripheral bond markets.
• Emphasising the ‘voluntary' route for the PSI was an attempt to avoid triggering sovereign CDS contracts: Success in restructuring Greece's debt without triggering the CDS would amount to the bypassing of an important instrument designed to insure investors against exactly such a risk.
• Efforts to restrict sovereign CDS trading to holders of sovereign debt: While such a policy would have been in line with the intent behind the initial design of sovereign CDS contracts, imposing such a restriction on investors retroactively would have disenfranchised a significant proportion of investors in the market.
• Bans on short-selling: Even though short positions were driving markets into a tailspin faster, policy-makers would have been better off addressing the root of the problem, weak balance sheets of private financial institutions.
Policy measures have been of the ‘right' kind as well: Not all uncertainty is bad. The use of unprecedented measures like QE and capital injections should make investors think twice before they test the willingness and ability of central banks to act. The fact that the zero lower bound is no longer a restriction and that the design of QE can be adjusted in size and scope to address the problem at hand provide central banks some relief by introducing the ‘right' kind of uncertainty among investors. How? The threat of aggressive QE helps to keep bond yields low, spreads tight and risky assets buoyant. In other words, the possibility that central banks re-enact their aggressive easing in future crises could well serve to quell risk premia then.
Similarly, the use of capital injections by governments to bolster balance sheets of private financial institutions was the right kind of policy response, in my view. By providing explicit and tangible assistance to financial institutions, policy-makers addressed the root cause of the lack of confidence in these institutions. The injections did not circumvent the rules of the game. Capital injections are and will likely continue to be an important tool in the crisis and an essential part of maintaining financial stability.
Going beyond the past sins of emerging markets: On balance, DM policy-makers have gone a step further than their EM counterparts in the EM crises of the past through their actions. Why? Because EM economies could be chastised or avoided by global investors thanks to the presence of a safe haven in DM markets. By ensuring that there is no safe haven, DM policy-makers have troubled investors globally by giving them no place to hide. EM policy-makers were not able to use financial repression very effectively, though they may have wanted to. DM policy-makers, however, have willingly wielded their ability to financially repress investors with the skill of Parthian archers.
Summary: Unless policy-makers aggressively use the ‘right' kind of policies to support asset markets in the next crisis, the risk premium from the legacy of the ‘wrong' kind of policies from this crisis could become the dominant force. Asset prices could stay subdued, credit spreads could widen and bond yields could stay high if this turns out to be the case. Aggressive action by policy-makers in the next crisis would thus become even more of a necessity. I believe that policy-makers have put themselves on a very slippery slope in the future.
Of the importance of unintended consequences. Our colleague Joachim Fels once described the current stage of the crisis in Europe with the acronym CCC, standing for a crisis of confidence, competency and credibility. One could add a fourth ‘C' to this list, for consequences. A recurrent feature of the global crisis is the unintended consequences of policy actions. Perhaps the two most important milestones of the past four years were policy decisions, whose unanticipated consequences caused in each case a considerable degradation of the situation and extended the crisis both in scope and length. The first of these events was the decision by US authorities to let Lehman Brothers fail in September 2008, now widely acknowledged to have been a major policy error. The second was the decision by European governments to initiate restructuring of public debt in Greece without having first put in place a robust safety net for solvent governments. This is also now widely regarded as a policy error. In each case, we note that many in the market initially applauded these decisions (when not actively calling for them beforehand) without seemingly fully appreciating the consequences.
A euro exit by Greece would be no small matter. We are struck by the apparent complacency with which many in the market and more worryingly in policy circles speak of the possibility that Greece exits the euro area. The PSI agreement on February 20 has postponed the issue, but we do not think it has resolved the Greek situation once and for all. Debt sustainability remains problematic. An economic recovery in Greece remains remote (on these points, we refer readers to upcoming publications by our colleagues Daniele Antonucci and Paolo Batori). Political tensions across Europe have not been put to rest. Questions that have been postponed will likely resurface. Far from considering a Greek exit from the euro a benign event, we think that should such an exit materialise, the unintended consequences would be unfathomable for Europe and the world, and would be immensely more difficult to counter than those of either the failure of Lehman Brothers or a sovereign default. We explain why below.
The Lehman Brothers collapse caused a fundamental change in the perception of bank credit. The consequences were bad enough but could be mitigated to some extent because public authorities deployed two lines of defence. They substituted the (then) superior credit of governments for that of banks by extending funding guarantees to financial institutions or forcefully recapitalising them. They also deployed the full range of central bank instruments to support the economy.
The Greek PSI in turn caused a fundamental change in the perception of sovereign credit in Europe, leaving only one line of defence, the central bank. This line of defence has proved resilient, indeed more resilient than we anticipated. Resolute action from the ECB to shelter banks and governments from liquidity risk has proved relatively effective at stabilising the market, so far.
Should Greece leave the euro, what would change is the nature of money itself, potentially disabling even the central bank as a stabilising force. Who then would be able to step in to mitigate the consequences is, to say the least, unclear. As for what these consequences would be, this is clearer in our view: taking this route would likely trigger a wide-scale bank run and threaten the existence of the euro as a whole, and with it that of the European Union. We justify this assertion below.
What is also clear is that the consequences would reach beyond Europe. The European Union, even with little growth, remains the largest economy in the world, accounting for 26% of global GDP. If it gets further destabilised, so would the global recovery.
So bad it cannot happen? Prior to exploring the unintended consequences of a Greek exit from the euro area, we underline that what we are providing here is a demonstration ad absurdum. We do not expect Greece to leave the euro, precisely because the consequences would be so damaging. It is however impossible to quantify objectively the risk of such an event. If it happens at all, we think it will happen because a subset of policy-makers has not appreciated the consequences. For our expectation to prove founded, therefore, the unintended consequences of a Greece exit have to be anticipated, meaning understood by all parties.
Why would a Greek exit from the euro area be so damaging?
The euro is irreversible. Is it? The starting point of the analysis is the simple yet immensely important observation that if Greece were to leave the euro, this would imply that the euro is reversible.
Note that this is not currently the case. The adoption of the euro by a country is, according to the terms of the European Treaties, irrevocable. There exists no institutional mechanism for a country to exit the euro area, nor was such an exit ever intended to take place. The reasons for this are straightforward:
Breaking down the fungibility of money. If the euro is reversible, it is not reversible in only one country. It is reversible in all. If Greece can leave the euro, then other countries may also leave the euro at a subsequent date. The implication is what we described in an earlier piece as a breakdown in the fungibility of money.
To explain this point, one must consider that in a fiat money system, money is the liability of a bank. It is the liability of the central bank in the case of central money, i.e., banknotes and bank reserves held at the central bank. It is the liability of commercial banks for commercial money, i.e., deposits.
As long as the euro and the Eurosystem exist in an irreversible form, these different forms of money are effectively fungible. Should the euro be reversible, however, these different forms of money are no longer completely fungible.
Federal money versus national money. Which is which? The euro may be a federal currency, but a deposit in a bank is effectively national money. If a country were to leave the euro, it would most likely be redenominated in that country's new currency (‘most likely' as opposed to ‘definitely' as there is no precedent and currently no legal reference that can be applied to such a scenario).
Strong money versus weak money. Which is which? As long as the euro is irrevocable, the distinction made above between federal and national money is irrelevant. By contrast, if the euro were to become reversible, this distinction matters. A euro held in a country more likely to abandon the euro becomes a weaker form of money than a euro held in a country more likely to keep it, with banknotes the strongest form of money.
Under the more extreme scenario where the euro breaks up altogether, the weaker forms of money would be deposits in countries more likely to see their new currency depreciate after break-up, the stronger forms of money being deposits in countries whose new currency would more probably appreciate. Where banknotes rank exactly under such a scenario is less clear (they are the liability of an institution - the Eurosystem - that would no longer exist as such).
How a run on money would unfold. In a situation where different forms of money are no longer entirely fungible, what one ought to expect is a run away from the weaker forms of money towards the stronger forms. More explicitly, this would take the form of:
• Savers withdrawing deposits from the banks of countries deemed weaker, preferring to hold banknotes instead; and/or
• Savers withdrawing deposits from the banks of countries deemed weaker and transferring them to the banks of countries deemed stronger.
The erosion of deposits in Greece and Ireland provides but a glimpse of what would happen elsewhere if Greece exited the euro. This erosion of deposits has, it will be argued, already taken place partially in Greece, where banks have lost a quarter of their deposit base since the end of 2009. It has also already taken place partially in Ireland, where banks have equally lost approximately a quarter of their domestic deposits and a larger proportion of their foreign deposits. In both cases, what has taken place is consistent with a migration of money away from weaker forms towards stronger forms, meaning to deposits in German or Dutch banks. This is reflected in turn in the build-up of the infamous intra-Eurosystem claims.
Our point, however, is not that an exit from the euro area by Greece would threaten the deposit base of Greek banks, which is a given. It is that such an exit would threaten the deposit base of most banks across Europe, at least the banks of any country for which, as a consequence of a Greek exit, the probability of leaving the euro and devaluing its currency becomes meaningfully different from zero. More explicitly, the partial run observed in Greece or Ireland would both assume larger proportions and more importantly occur across a larger number of countries.
Contagion ought to be a familiar pattern by now. To illustrate this point, it suffices to recall that the failure of Lehman Brothers closed access to capital market and interbank funding to all banks, regardless of whether they were solvent or not. Similarly, the initiation of PSI for euro area sovereigns caused contagion not just to Ireland and Portugal but also to Spain, Italy, Cyprus, etc., and eventually France and Austria, regardless of whether the governments of these countries were objectively solvent or not.
No line of defence? The precedent of deposit haemorrhage in Greece and Ireland is nonetheless relevant in that it highlights the tools necessary to counter a bank run. Paradoxically, the determined action of the Eurosystem in Greece and Ireland to counter this haemorrhage illustrates why a run of broader magnitude would be very difficult to control.
In a situation of a bank run, what prevents catastrophic outcomes (widespread bank failure) is the ability and willingness of the central bank to replace deposits and fund banks directly itself. This is what happened in Greece and Ireland, where the Eurosystem increased its lending to domestic banks.
Limits to the ability of the Eurosystem to intervene... The Eurosystem is however constitutionally constrained to only lending to banks against ‘adequate' collateral. The meaning of ‘adequate' is elastic, as evidenced by recurrent loosening of Eurosystem-wide eligibility criteria through the crisis and by the acceptance of weaker collateral in the context of Emergency Liquidity Assistance (ELA) in both Greece and Ireland. It remains the case that the central bank cannot fund banks unless they have unencumbered assets to provide as collateral. A first threat to the ability of the Eurosystem to replace deposits on a wide scale is the availability on the balance sheet of banks of sufficient amounts of unencumbered assets.
...and perhaps limits to its willingness. More fundamentally, if the euro were to become reversible, the nature of the risk taken by the central bank by replacing deposits would change substantially. It is one thing for the central bank to fund banks that are and will always remain in the euro area. These banks are within its jurisdiction and are its natural counterparties. By contrast, if a country can leave the euro, its banks may no longer be ECB counterparties in the future, are regulated by authorities whose interests may no longer be aligned with those of the ECB, and are active in a market that may no longer use the euro as a currency.
If the ECB is sensitive to the market and credit risk that it takes in its operations (which it is, judging by its refusal to take losses on its holdings of Greek bonds purchased in the context of the Securities Markets Programme), then it might become more reluctant to perform its role of lender of last resort without reservation if the euro were made reversible.
If not the ECB itself, at least some components of the Eurosystem may become (even) more reluctant to support the banks subject to a run. The corollary of the migration of deposits referred to above is the rise in intra-Eurosystem claims, which are claims of the national central banks of the ‘stronger' countries on the ECB, matched by claims of the ECB on the central banks of the ‘weaker' countries. As long as the euro exists and is irrevocable, these claims have very little relevance, in our view. They are an accounting mirage, resulting from the disaggregation of the balance sheet of the central bank of the euro area (the Eurosystem) into 18 sub-balance sheets (that of the ECB and of the 17 national central banks). As long as the euro exists, the exposure that central banks have to each other within the system imply no risk, in our view. Should the euro become reversible, however, then these intra-Eurosystem claims would suddenly imply genuine risk, as the national central bank of a country exiting the euro might realistically default on its obligations towards the remainder of the Eurosystem.
And the credit crunch would likely come back. Arguably, the points we have listed here are of secondary importance. More important is the fact that if banks lost their deposits, they are very unlikely to extend credit to the economy. This would push any country affected by the run back towards the credit crunch, which the ECB painstakingly managed to avoid and to turn into a less damaging credit squeeze through its long-term refinancing operations.
Conclusion: if one country goes, it is monetary union that goes. Our conclusion at this stage is that Greece leaving the euro would not be a local event. It would change the nature of money everywhere across the euro area by making the euro reversible. It would turn most of money supply (commercial money) back into national money. It would hinder the ability (and possibly willingness) of the Eurosystem to act as a lender of last resort for the entire euro area banking system. For all practical purposes, it would be the end of the euro as a genuine single currency. It would also likely trigger an unstoppable run on banks, which would push large parts of the continent onto a depressionary and politically painful path.
To preserve the euro if Greece left would require total federalism in the rest of the area. Is there a way to avoid the conclusion we outlined, meaning to have one country leaving the euro area without triggering a catastrophic run in other countries? We think the answer is yes, but only under conditions that are unrealistic to us in the near term.
For a Greece exit from the euro to happen without triggering the chain of events we have described, it would need to be ensured that the euro remains irreversible for all countries other than Greece, and perceived as such. In other words, it would need to be credibly ensured that Greece is the only country that will ever leave the euro.
Needless to say, in a context where policy credibility has been a major casualty of the crisis (as reflected in Joachim Fels' CCC acronym) the only way to assure that no other country would ever leave the euro would not be policy statements but institutional and constitutional reforms that bind countries together forever. We believe that this would need to include at least: