After the dramatic rescue package for the euro area last spring, the governing council of the European Central Bank (ECB) decided on several measures in order to ensure an "orderly monetary policy transmission mechanism". Among them were secondary market purchases of private and European government bonds with the aims of propping up failing bond markets of governments in financial distress and allowing cash-starved commercial banks in those countries to sell them to the ECB in return for funding. In the days before announcing the package, the ECB council had surprisingly changed its previous traditional course of drawing a clear line of separation between responsibility for monetary policy and fiscal policy. By giving in to more than fifty European commercial banks and the European association of traders, the impression emerged that the ECB was not only driven by the markets but also by politics.
Since intervention has been limited to the secondary market (for the time being?1), the ECB has not literally violated the Treaty. However, it did infringe upon its spirit, as a central bank bail-out of government deficits is prohibited. The actors pointed to the extraordinary market tensions in an attempt to legitimise their move, accompanied by dramatic statements by A. Merkel and J.-C. Trichet, which unfortunately left the bank with no other alternative. While the German Chancellor expressed that "the euro is in danger", the ECB President spoke of the "most difficult situation since World War II if not since World War I" and noted that "we live in difficult times". It is true that in the days preceding the largest and most recent rescue package, markets for the sovereign bonds of the weakest and most indebted countries of the euro area nearly dried out. But markets in fact did not appear to be overly irrational.2 Their fear that they might not get their money back was generally realistic. What is more, economic science is not at all clear about whether huge spread increases and drying out markets are really "dysfunctional" or - in view of the fact that Greece's and Portugal's domestic savings are so small that they are no longer capable of keeping their capital stock constant and financing their public deficits - in fact "functional". While initial empirical evidence conveys the impression that, both in the past as well as during the crisis, sovereign bond spreads reacted in a systematic fashion to the fiscal policy stance (i.e. government debt levels and forecasts of future fiscal deficits) of individual countries and, hence, indicates that the aforementioned statements were potentially too alarmistic,3 more results are to be expected from further intensified research in this area.
Political Dependence of the ECB
Many observers have argued that, with the bond purchases, national fiscal policies could from now on dominate the common monetary policy. The ECB has obviously become more politically dependent. After all, in the first four weeks after its path-breaking decision, the ECB bought slightly more than EUR 40 billion in bonds which must be added to the much larger amount of Greek bonds that had already been offloaded onto the ECB's balance sheet via open market operations which applied lower haircuts than usual or no haircuts at all.4
After starting with purchases of €16.3 billion of bonds in the first week of the Securities Markets Programme (SMP), the ECB successively lowered the amounts in the following weeks (€10.4 billion, €8.8 billion and €4.9 billion). These numbers decreased in the following months, but this must not blur the clear impression that the ECB allows some countries and their banks to refinance at highly subsidised rates by offloading troubled assets in the amount of hundreds of billions of euros onto the ECB's balance sheet. But no official information has been released about the composition of these purchases in relation to whose debt was purchased or when it matures. Anecdotally, market voices speak of almost 75% Greek debt purchases, with Portugal and Ireland being the next biggest beneficiaries and some smaller Spain and Italy purchases thereafter. Without any doubt, it can be taken for granted that this portfolio would adjust based on the changing market perception of which country is weak. In any case, the basic principle followed by the ECB seems to be to reduce spreads and wipe out any shorting interest for a rather long time. The ECB makes use of the fact that there is no legal obstacle to targeted bond purchases. The ECB Council is not at all legally obligated to buy bonds, for instance, according to the economic weight of the issuing countries or their capital shares at the ECB.
In addition, targeted purchases of bonds issued by highly indebted euro area governments contain an element of subsidy which tends to severely weaken the governments' fiscal discipline: the interest rate premium on the bonds of fiscally weaker countries declines and the premium for stronger countries increases. Fiscally solid countries are punished and less solid ones, in turn, are rewarded for their lack of fiscal discipline and excess private and public consumption. The credit risk is thus simply rolled over from the bonds of the weaker countries to those of the stronger ones, with the liability held by the ECB. This risk transfer is not given a more favourable light by the recent revelation that the bond purchase programme was essentially an action of redistribution in favour of a Greek oligarch worth several million euros. Finally, not only targeted bond purchases but especially the lending of the ECB within its normal policy operations amounts to an implicit subsidy worth at least 2 to 3 per cent of GDP.5
These facts and circumstances, in particular the general perception that the ECB has become more dependent in political terms but enshrouds this via a huge lack of transparency, appear to be rather uncontroversial (in contrast to the assessment of whether this is welcome or not) and, thus, are not investigated further in this paper.
Financial Independence of the ECB and the ESCB?
Surprisingly less focus has been given to the aspect of the slowly vanishing financial independence of the ECB, which is at least equally as important. Who will actually have to pay the losses stemming from the purchased private and sovereign bonds if Greece and Portugal are unable to serve their debt in the end? Ultimately, the owners of the ECB would be asked to pay up, and by far the largest bill would be imposed on Germany. It is possible that the toxic bonds in the balance sheets of the ECB might eat up most of the Bank's reserves and equity capital if they were to fall in value by a sufficiently large amount - in the worst case, the amount could range into the hundred billions of euros. In this case, fewer central bank profits would be transferred to the accounts of the euro area governments, and with given public deficits and spending levels, taxes and duties would inevitably go up. In the extreme case, still more realistic for the Fed and the BoE than for the ECB,6 losses would exceed the entire equity capital of the European System of Central Banks (ESCB). Euro governments would have to prop up the ECB's capital in order to strengthen the reserves of the ECB or even prevent the ECB from having negative equity capital.
In order to re-establish the original structure of the central bank balance sheet, governments can recapitalise the central bank via newly printed government bonds indefinitely. One problem emerging in this context is that the governments' "deep long pockets" are rather deep or ultimately relatively empty in view of the reported public (let alone private) deficit and debt figures. With the bond purchases, ownership and the risk of the papers would be transferred to the national central banks (NCBs) in the euro area. Should a debt restructuring or debt waivers result, the central banks of the euro area would then be directly affected. Admittedly, the financial system would not be significantly shaken, since the commercial banks would not hold too many "Greek" bonds. But the potentially huge losses would ultimately be rolled over to the European tax payer - through direct taxation or indirect taxation via inflation (if governments are not willing to bear the income losses and a monetary overhang emerges). Does the ECB really want to follow in the steps of the Fed?
Currently, a mechanism for recapitalising the Eurosystem as a whole is not available. Instead, each national fiscal authority stands behind its own NCB financially. In view of current unorthodox monetary policies, this could be interpreted as a fiscal vacuum - a lack of "fiscal backing" of the ECB and the Eurosystem. This is why the ECB did not engage in outright bond purchases to a very large extent, except for the covered bond purchasing programme. Hence, in combination with the specific and carefully defined accounting principles of repo operations in the euro area (including a conservative imposition of haircuts) and the vast amount of reserve liabilities of the euro area national central banks, it was precisely this vacuum that had until recently prevented the ominous "tango" of euro area governments with the ECB. However, one should not forget that buying troubled assets via normal monetary policy operations does not present a solution for insolvent debtor banks and countries. It is self-enforcing, since approaching the ECB for quasi-free money is and will remain the least costly source of funds for commercial banks located in the euro periphery.
There is thus an inherent tendency for these clients to strive to continuously enhance their ECB funding. But in this case, the risk added to the ECB's balance sheet rises in the same proportion.7 Hence, ECB observers should not feel satisfied by any reference to the low scale of structural ECB bond purchases on secondary markets.
Sterilisation of ECB Bond Purchases
The ECB has decided to sterilise its bond purchases - offsetting those purchases through sales of other bonds or money market instruments to keep the overall money supply unaffected. This is to counter accusations that the ECB is monetising government debt. ECB President Trichet has been keen to point out that this cannot be regarded as "quantitative easing" because there are also some new operations to leave the supply of high powered money in the euro area constant.
Technically this works - and with the benefit of hindsight one can say that it did work - for instance through a tender of interest-bearing time deposits, via which banks deposit a certain amount of money for a limited time with the ECB. The ECB has actually already used this instrument "in order to signal to the markets its counter-inflationary stance". Although the bonds received have significantly longer maturities, the ECB has chosen only a one-week tender, indicating a greater emphasis on psychology than synchronous sterilisation. Another possibility explicitly mentioned by the ECB would be the issuance of debt certificates by the ECB itself.8
But can we really be sure that the Securities Markets Programme (SMP) does not contain important elements of quantitative easing? How airtight is the envisaged sterilisation of bond purchases as a protection against inflationary consequences? And does it target the main problem of unlimited bond purchases - the fact that confidence in the political and financial independence of the ECB is on the brink?
Note that the key aspect of quantitative easing in the UK and USA over the past year has been targeted interventions in bond markets to bring market interest rates down. This is exactly what the ECB is now doing, so it cannot be rejected that the SMP represents a clear variant of quantitative easing - let us therefore simply call it quasi-quantitative easing. Admittedly, the SMP differs from quantitative easing as practised in the UK and USA in terms of extra complications which were not encountered in the UK or US cases, i.e. that the programme lacks transparency and is politically highly controversial. For instance, we are not informed about the composition of the debt securities the ECB is buying (the suspicion is that the ECB is coy about identifying which debt is being bought because it is mostly Greek!), the criteria being used to select bonds to purchase, the ECB's bond purchase strategy during periods of primary issuance, how long the programme is going to last and how much may be spent.9 One quite obvious reason we do not have an answer to the last item is that there are clear internal disagreements within the ECB Governing Council on this issue. President Trichet conceded in an interview with Le Monde that the ECB decision to run the SMP was not made unanimously.10 He added that the ECB decision was made with "an overwhelming majority" - as opposed to the standard "unanimous decision" - which, taken by itself, is an important sign of internal divisions and a renationalisation of monetary policy.
While implementing its new "minimum price guarantees" for sovereign bonds, the ECB does not know exactly how many bonds it will have to buy in order to sustainably stabilise the prices of the jeopardised bonds. To be on the safe side, there might be a tendency to buy more than necessary, which would blow up the money supply to an unnecessarily large extent. In addition, the sterilisation measures are not mandatory but have been promised - except for the tenders of a time deposit just to put out a few feelers "to see how it will work". Conflicts of interest between the national central banks within the ECB are bound to occur. This is especially so since the process of consolidation in countries such as Greece or Portugal - if it is to be successful at all - is expected to last more than half a decade. What is more, the central bank in a country plagued by severe fiscal problems will consistently claim - generally with support from European commercial banks and traders - that the market for domestic bonds has dried out and is in dire need of support and that market movements are "dysfunctional".
But most importantly, the collection of deposits under any new programme appears to be rather irrelevant in the grand scheme of the overall ECB monetary policy stance. As of June 2010, the ECB was still offering unlimited loans within the framework of its refinancing programmes. Hence, the ECB's sterilisation programme does not appear especially relevant for assessing the current determinants of the money supply (chief among them is global excess liquidity, which is already circling the globe but has not unravelled due to small money multipliers). Indeed, in June of this year, the one-year tender through which the ECB has lent commercial banks huge amounts of money will run out in July. But at the same time, on May 10 the ECB announced new liquidity-enhancing operations with full allotment whose expansionary effect stands in strong contrast to the announced sterilisation.
It directly follows that the announced measures for sterilising this expansionary monetary policy are not fully credible. Issuing ECB debt certificates would make Greek and Portuguese bonds even less attractive. These countries would have to offer even higher returns in order to be able to place their issues. But this would clearly counteract the spirit of the most recent rescue package for the euro area. Apart from that, even a successful sterilisation would not smooth things over. If actually used for sterilisation purposes, the issuance of ECB debt certificates would contribute to a huge transfer of sovereign risk towards the ECB balance sheets - in the same way as is currently the case within the normal policy operations of the ECB when refinancing Greek and Irish banks. Offering time deposits to banks contributes to this kind of transfer. Through the bond purchases, the ECB acts like a fiscal agent by taxing other euro area creditors through higher bond rates in order to support a government which finds itself in a financial emergency. This same holds true when the ECB collects the money which was already spent for bond purchases. Other euro area creditors are put at a disadvantage because the ECB must offer higher interest rates in order to receive the money back, which in turn makes credit more expensive.
However, the claim that interventions through targeted bond purchases restore orderly markets does not seem compelling, especially if governments or the ECB decide which market movements are justified and which are not and, hence, which bonds to buy and which not to buy. For instance, a renationalisation of euro area monetary policy cannot be excluded insofar as the ECB might well feel inclined to buy Greek sovereign bonds in order to make it possible for French banks which are strongly engaged in Greece to reduce their holdings ("French bias") or to protect German banks with their strong stake in Spain from contagion effects ("German bias").11 While obviously affronting the general principle of a common monetary policy for the whole euro area, this potential outcome is even more critical because Greece is obviously not in need of additional access to the capital markets because it has already been bailed out and does not have to place its issues anymore. De facto, Greece does not even have to worry now about how and where its secondary market spreads trade, and as a result the discipline-enhancing bond spread mechanism is lacking. Instead, Greece will have to closely follow the "will of the Troika", as the Greek press calls the group consisting of the IMF, the ECB and the EU Commission. Unfortunately, this appears to be a case of throwing even more good money after bad.12
What is more, the risk premia in Southern European bond markets that are again rising clearly convey the impression that the assessment of the euro area by large investors has changed significantly since the adoption of the rescue package and the announcement of biting austerity programmes in Greece, Spain, Italy and Portugal. The saving requirements appear so drastic that their successful implementation appears to be nearly impossible and politically risky to say the least in the cases of Greece, Portugal and Spain. Hence, investors will have to be able to assess the euro area countries individually according to their country risk and not as members of a homogenous block. The main problem, currently increasing in importance, is that the ECB is curbing real returns on the bond markets through its bond purchases. Just for comparison: in an earlier debt crisis the real 10-year return of Spanish bonds rose to nearly 10 per cent. Currently, the real return of Spanish, Portuguese and Italian bonds only amounts to 3 to 3.5 per cent. This is almost certainly not enough to attract the private capital these countries are heavily dependent on. In Spain, the next performance test is expected to be at the end of July when the government has to raise more than 16 billion euro at one dash. Moreover, on June 18 Spain will need around 8 billion euro for repayment of Spanish bonds with shorter maturities.
"Sterilising" Monetary Policy Should Target the Asset Side of the Central Bank Balance Sheet
One problem inherent in both sterilisation approaches is that they reshuffle only the liability side of the ECB's balance sheet. Neither approach is well-suited to either diminish the bloated ECB balance sheet or to remove the (potentially) toxic covered bonds or sovereign bonds. In addition, the intake of potentially toxic assets both as collateral and by outright purchases in the central bank balance sheet artificially keeps the asset prices up. A credible strategy of sterilisation to deal with the consequences of the financial crisis should, thus, deal primarily with the asset side of the ECB balance sheet.
Efficiency of the ECB Bond Purchases
It did not come as a surprise that the ECB's bond purchases only affected markets on the first few days. For instance, around one week after the announcement of the bond purchases, the euro plummeted to a four-year low and bond spreads began to rise again, culminating in a scenario resembling the May 2010 situation at the start of November 2010. Other indicators of the degree of uncertainty in the markets also conveyed the impression that investors no longer believed in the sustainability of the "newly designed" euro area which was characterised by a daunting institutional failure to make sovereign default in EMU possible. Instead, markets perceived that "toxic" government bonds would ultimately be located on the ECB balance sheet, threatening the long-term stability of the euro. As a result, the European currency fell against most other currencies until attention shifted back to the massive credibility problems of the US macro policy mix.
Since the beginning, it appeared doubtful that the instantly lower bond spreads really signalled a gradual increase in confidence in bond markets.13 Much more likely, the activities of the ECB tended to bias the bond prices of peripheral euro area countries and fuel scepticism regarding whether and for how long the lower risk premia would hold. If central banks were to intervene in the market, i.e. against the not fundamentally calamitous insolvencies of Greece and Portugal, experience generally indicates that this would not go well. In this respect, bond purchases are akin to foreign exchange market interventions: the central bank intervenes in one asset class because price formation does not correspond to its view of what is justified and because this distortion threatens to spill over to other markets.14 The idea that spreads on certain financial instruments that are higher than the central bank would like should prompt an intervention has not, at least until recently, been a standard monetary policy tool. Consider two cases: either the ECB will hold the bonds to maturity (as indicated on some occasions) or will sell them earlier. In the first case, the ECB will effectively tax the private sector if the ECB really strives to shrink its balance sheet. It will have to sell non-sovereign bonds, which will lower their prices and increase the premia corporations will offer to pay for their bonds.
In the second case, imagine the intense political pressures on the ECB at any future point in time when the bonds have to be resold to the markets before their maturities, at which point it will become clear that adjustment will still take some time or that the core structural issues have not been dealt with sufficiently, leading the country-specific risk premia to skyrocket again quasi-automatically!
In both cases, it appears rather clear that the ECB will have to capitulate again. These considerations imply that we have definitely seen the persistent "exit from the exit" from ultra-loose monetary policies in the euro area. The danger has risen that the ECB will get caught up in the maelstrom of its role as the lender of last resort. The more troubled bonds the ECB buys, the more difficult it will be to deny further sovereign financing in the future, because doubts on the markets will prevail until an institutional solution of debt restructuring is installed as a fiscal agent to be financed by the governments themselves and not through the creation of money.
The ECB should therefore avoid an "anything goes" monetary policy under all circumstances and keep the deliberate secondary market bond purchases within the SMP as exceptional, brief, targeted and limited in volume as possible. Axel Weber and Mario Draghi, heads of the Bundesbank and the Banca d'Italia and potential candidates for the ECB Presidency, have already been publicly calling for a quick end to the bond purchase programme under the SMP.15 Draghi characterised the moment for withdrawal as when "the markets spontaneously resume trading of the securities of the countries involved." What kind of action Draghi had in mind if the markets are trading these bonds but only with an eye on the ECB staying in the secondary markets was left rather unclear. Even more worrying is that there is no clear indication of how the SMP is going to develop over the coming months, especially with respect to increasingly suffering Spain. Simply declaring that the SMP will be serving as a bridge until new state financing facilities agreed by the European Union can take over will clearly not be sufficient. Above all, public attention should be less focused on the ECB's direct purchases of distressed government bonds and more so on the much larger bail-out within the framework of the ECB's ordinary monetary policy operations.
Instead, keeping the deliberate bond purchases as brief as possible and preventing the ECB from becoming the bad bank of the euro area can best be achieved by immediately installing some institutionalised default mechanism like, for instance, a European Monetary Fund.16 This was well-recognised at the recent EU summit. The same mechanism could limit the damage in terms of credibility and reputation loss as well.
Though approached from another angle, the above analysis supports the view taken, for instance, by Gros and Mayer that without the immediate installation of a sovereign default mechanism such as a European Monetary Fund, the ECB risks turning into the "bad bank" of the euro area as timid investors offload sovereign bonds with uncertain repayment values onto the ECB's balance sheet. Although ever larger rescue packages have been prepared, investors clearly understand that some countries supported by the ECB's Securities Markets Programme (SMP) still have the potential to become insolvent. The increasing political and financial dependence of the ECB is the dire consequence. Accordingly, we observed the exchange rate of the euro decline proportionally to the deterioration of the ECB's balance sheet in the weeks following the introduction of the SMP. Since there are strong signs of a perpetuation of the exit from the exit from unorthodox monetary policies in the euro area right now, the internal value of the euro will follow and shrink very soon which - in turn - implies higher inflation in the long run.17
Recent events have vividly demonstrated that in the absence of a mechanism to manage an orderly sovereign default, adjustment programmes lack credibility and the balance sheet of the ECB is put at risk. Only sovereign funds tend to reveal the true opportunity costs to the initiators. However, by following the path of the ECB and the printing press, the opportunity costs of adjustment programmes wrongly appear to be close to zero.18 This is especially so if these programmes are not transparent (as in the current case of the SMP).
It has been shown above that the ECB will automatically transform into a quasi-fiscal agent of euro area governments in times of crises. If this happens in an environment of ultra-lax monetary policy and low transparency, this might damage the reputation and the credibility of the institution in the medium run already.
As a (maybe inferior) alternative to the immediate installation of a European Monetary Fund, the ECB could have contributed to sovereign debt consolidation by only accepting bonds issued by those countries which have introduced upper bounds to debt levels as collateral (after a transition period, of course).19 This proposal á la Martin Feldstein would be a welcome departure from the ECB's current practice of supporting commercial banks by accepting toxic assets as collateral and purchasing Greek and Portuguese bonds. This is especially so because imposing "debt brakes" and the resulting decrease in interest to be paid should be in the national self-interest of the respective countries.
Taken as a whole, the most worrisome aspect is that the euro area has stumbled into a perpetuation of unconventional monetary policies via the execution of the SMP and even more so by funding and heavily subsidising the assets of troubled banks and countries by accepting them within the framework of normal monetary policy operations. Of course, the immediate intentions are to bail out banks (primarily) and governments (to support issuance). What is difficult to see at the moment is how, once started, it can stop. We have already crashed into near-zero interest rates with no likelihood of escape in the near term (at least not without serious consequences). Hopefully, the ECB has not been checkmated by (a) the de facto abandoned Maastricht deficit and debt guidelines (apparently for all euro area countries) and the emerging illiquidity and insolvency risk, (b) giving in to apparently non-revertable government bond purchases under the pressure of powerful interest groups like European commercial banks and traders associations and (c) the huge degree of available global excess liquidity just waiting to enter the euro area through carry trades as soon as the ECB ventures to exit from its unconventional monetary policies. However, it looks a little bit like that.
In the light of my analysis, I have suggested to the members of the EP to ask Jean-Claude Trichet the following key questions in order to get an impression of the potentially huge opportunity costs of the SMP: What exactly is the composition of the sovereign bonds the ECB is buying? What criteria are applied to select which bonds to purchase? How can we characterise the ECB's bond purchase strategy in cases and periods of primary issuance? How long is the SMP going to last and what amounts may be spent? Finally, why does the ECB refuse to release internal documents which describe how Greece has concealed parts of its public debt? The ECB has been (too?) silent on these points, arguing that any publication on these issues would trigger the acute danger of financial market turbulences - just another facet of the "dysfunctional markets" argument. History repeats itself!
Ansgar Belke, DIW Berlin and University of Duisburg-Essen, Essen, Germany.
Revised version of a briefing paper prepared for presentation at the Committee on Economic and Monetary Affairs of the European Parliament for the quarterly dialogue with the President of the European Central Bank, Brussels, June 2010.