€ liquidity is getting more valuable: what can be done about it?

It is a recurrent phenomenon that liquidity gets more valuable at the end of the year: this is seen in the fact that, notwithstanding the higher amount outstanding, the overnight money market rate (EONIA) increases in the last weeks of the year. While the phenomenon was suppressed during the crisis by the overabundant liquidity provision, before the crisis, the average increase between the beginning of December and the end of the year, was of the order of 36 basis points, concentrated in the last few days of the month (see table 1). 
The explanation for this phenomenon is that banks want to window-dress their balance sheet at the end of each year, by showing higher liquidity buffers. 
Table 1. Changes in money market rates 


Source: ECB, Bloomberg, Author’s calculations ; Notes: Difference between beginning of December and end of year/ end of first three weeks of December



This year the phenomenon appeared earlier and was visible also in maturities longer than overnight, differently from previous years, because the balance sheet of banks at the end of the year will be considered for the Asset Quality Review (AQR) of the European Central Bank and there is thus a stronger incentive for window dressing as banks want to make a “Bella Figura” when examined by the ECB (table 1 and chart 1). 
Chart 1. Euribor curve at various points in time

 Source: Bloomberg, Author’s calculations
Obviously, given that, since October of 2008, the amount of central bank liquidity outstanding is not decided by the central bank but by the aggregate behaviour of the banking system, the increased demand of liquidity is not only apparent in its higher price (the overnight money market rate) but also in a higher quantity: the amount of excess liquidity outstanding has increased over the last week by some 40 billion €, breaking the decreasing trend (Chart 2).
Chart 2: Euro area excess liquidity

Source: Bloomberg, Author’s calculations
The source of this increase is also interesting: most of it derived from the fact that banks have underbid by some 30 billion € the operations aiming at sterilising the impact of the Securities Market Program (SMP) purchases (some 180 billion €). There are two reasons for this: first, with money market rates approaching the rate on Main Refinancing Operations (MRO), which is used to remunerate the deposits corresponding to the sterilised amounts, the incentive to participate in the sterilising operations is much reduced; second, while there is probably a stigma implied in borrowing from the ECB under its Very Long Term Refinancing Operations (VLTRO), especially in view of the AQR, there is, obviously, no stigma in non participating to sterilisation operations, so banks prefer, for similar financial cost, to keep the liquidity obtained through the SMP than to borrow through the refinancing operations, in particular through the VLTRO. So one can explain the apparently contradictory development of banks returning the money borrowed under the VLTRO and, at the same time, refusing to give back the SMP liquidity to the ECB.
This behaviour is of course tied to the end of year tensions, but raises a more general issue: as money market rates approach the level of the MRO, the incentive to participate to sterilising operations decreases and the possibility of underbidding at these operations increases. In addition, if there is a stigma in borrowing from VLTRO and other ECB refinancing operations, the phenomenon of banks returning funds borrowed under the LTRO but keeping the liquidity obtained against sales under the SMP could be recurrent. The homeostatic mechanism whereby banks borrow more liquidity from the ECB when it becomes more valuable would find another channel of functioning.
Beyond the end of year tensions, what is important for monetary policy is what will happen in the new year: will liquidity maintain its higher value, which translated in monetary policy terms means that there will be, contrary to the forward guidance of the ECB, a lasting increase of interest rates? And if this was the case what could the ECB do?
As I mentioned in a previous post  “Is the price stability target of the ECB at risk?” , the ECB has a number of tools at its disposal.
A first one would be to stop absorbing the liquidity created through the SMP: what started this month as a decision of the banks could be transformed into a policy decision. In this way more than 180 billion € of permanent liquidity would be added to the market. Of course, this could accelerate the reimbursement of the two VLTRO, but the reimbursement would take time and the impact of a doubling of excess liquidity would have a lasting effect on the EONIA. A provision of liquidity could also be engineered by a reduction, say to 0.5%, of the coefficient of required reserves, equivalent to around 50 billion €.
Alternatively, and possibly with a higher probability, the ECB could offer an other VLTRO. Draghi has explicitly referred to it [1] but has also clearly said that it can only be done if it would not lead to another round of carry trades. It was not so clear, when the 2 VLTROs were launched between the end of 2011 and the beginning of 2012, that there was so much displeasure at the ECB that banks, especially in Italy and Spain, bought a large amount of national government securities. Still, this is now the clear policy line: a new VLTRO should not lead to more carry trades.
The problem is how to achieve this? If you give liquidity to banks how can you control that they use it for lending to firms and not for lending to governments?
The wrong answer to this question would be a resurrection of direct controls on banks balance sheets, like the Corset used by the Bank of England in the seventies and the Massimale and the Vincolo di portafoglio used by Banca d´Italia in the seventies and eighties. These measures were used to try and restrict commercial bank lending that the two central banks were not capable of controlling by means of interest rate policies and, in the case of Italy [2], they were particularly aimed at favouring the funding of the budget deficits. In principle, however, the measures could be turned on heir heads and stimulate the funding of firms instead of governments.
However, even if sometimes I have the impression that direct measures seem to get a new lease of life under the nobler name of macroprudential tools, such measures would not be allowed under the ECB Statute, that prescribes the use of market instruments [3]  and would, in any case, not be a smart idea.
Still, one can find a trace of a subtler solution to avoid a new wave of carry trades, with two components.
First, one heard both Draghi and Praet [4] hinting at a penalisation of sovereign holdings in the Stress Test phase of the Comprehensive Balance Sheet Assessment that the ECB has started. The utterances were not very detailed and it is not clear whether the penalisation would affect both the “held to maturity” and the “trade” portfolios of securities or only the latter category, nor is it clear whether the stress scenario would be so harsh as to reproduce the peak of the sovereign crisis in the summer of last year. Still the message must have reached banks and reduced their willingness to borrow cheap from the ECB to enjoy the carry available purchasing peripheral bonds.
Second, if the new VLTRO was seen as a rollover of the maturing current VLTROs rather than an addition to them, one could argue that there is no increase of carry trades but just their prolongation. In addition, a VLTRO rolling over existing ones would also match the fact that, notwithstanding the large reimbursement effected by banks so far (around 425 billion €), there remains a large demand for medium term funding by the ECB, which would not be satisfied if the existing LTROs were just left to mature. As an additional incentive, to make its offer more interesting, the ECB could add some optionality to a new VLTRO, for example by capping the interest rate at the current low level, thus allowing a further possible rate cut to reduce its cost but protecting against an increase.
In conclusion, should the tensions on the money market prove not to depend just on the reinforced end of year effect and should the ECB want to buttress with facts its forward guidance, It would have two fairly effective and not too costly tools to use: stopping the absorption of liquidity injected through the SMP or providing a new VLTRO that would roll over existing ones.


*** Research assistance was provided by Mădălina Norocea

[1] “if we are to do an operation similar to the LTRO, we will want to make sure that this is being used for the economy. And we will want to make sure that this operation is not going to be used for subsidising capital formation by the banking system under these carry trade operations.”  Mario Draghi, December 2013. 

[2] F.Cotula and S. Rossi, Il controllo amministrativo dei flussi finanziari in Italia, in La Politica Monetaria in Italia. F. Cotula ed. Il Mulino, 1989.
[3] “The ESCB shall act in accordance with the principle of an open market economy with free competition”, Art 2, Statute of the ESCB and of the ECB.
[4] “Sovereign debt is going to be stressed like all other categories in banks’ balance sheets,” Mario Draghi, 2013.
“Appropriately treating banks’ holdings of sovereign debt according to the risk that they post to banks’ capital makes it unlikely that banks will use central bank liquidity to excessively increase their exposure to sovereign debt,” Peter Praet, 2013.