Is the liquidity management of the Eurosystem balkanising along national lines?

In a letter to the Editor in today’s Financial Times (28 January 2009), the Director-general of the Bank of Italy, and my esteemed former colleague at the EBRD, Dr Fabrizio Saccomanni, objects to some of the things I wrote in my post The Banca d’Italia and the re-nationalisation of cross-border banking on 24 January 2009, about the Bank of Italy’s new initiative in creating a collateralised inter-bank market or MIC (for Mercato Interbancario Collateralizzato).

I reproduce his letter here:

No protectionism in Bank of Italy’s initiative

Published: January 28 2009 02:00 | Last updated: January 28 2009 02:00

From Mr Fabrizio Saccomanni.

Sir, I am surprised to read in an article on the ft.com/maverecon website, that Professor Willem Buiter labels the Bank of Italy’s initiative to create a collateralised inter-bank market as “protectionist”. I believe this view is based upon inaccurate information and I would like to provide some clarifications.

First, no financial protectionism is embedded into the initiative. The scheme is open to any European Union credit institutions which satisfy a limited set of conditions equal for all participants. It is in fact in the very interest of the participating banks that the system be as open as possible.

Second, the new market could be used as a reference for setting up a similar scheme at a pan-European level, thus moving in the direction of a number of proposals including those of Professor Stephen Cecchetti, now economic adviser to the BIS, as long ago as the autumn of 2007.

Third, the Bank of Italy does not act as a market-maker of last resort. Market prices and trades will be driven only by market participants. Its role is that of a facilitator of trading, managing the custody of collateral, administrating and assessing the collateral, ensuring prompt settlement of transactions in the event of default by a participant, subsequently recovering the amounts from the collateral deposited.

Fourth, the criteria for evaluating the collateral and the mutualistic nature of the market substantially limit risks. Banks’ exposures will be correlated to their regulatory capital (up to a maximum of 30 per cent and in any case no more than €5bn) and cannot exceed the value of the collateral deposited with the Bank of Italy (with appropriate haircuts). Moreover, the scheme imposes a concentration limit of the individual exposure (20 per cent) and envisages a partial sharing of default risk. In the unlikely event in which the collateral provided were to be insufficient, the other market participants would jointly make up the difference within the limit of 10 per cent of their collateral.

Finally, in any case no fiscal back-up on behalf of the Italian Treasury is envisaged.

I hope that this clarification can allay your concerns on the new market.

Fabrizio Saccomanni,
Director-general,
Bank of Italy

It is possible that I misrepresented the true nature of the  Mercato Interbancario Collateralizzato (MIC), but nothing in Dr. Saccomanni’s letter convinces me that I did.

Is there equal and uniform access to the MIC scheme for all banks that are eligible counterparties of one of the NCBs in the Eurosystem?

I do not argue that the MIC scheme was protectionist in intent or design. It certainly is protectionist in effect or outcome, and risks balkanising the implementation of monetary and liquidity policy in the Eurosystem along national lines. Dr. Saccomanni states “The scheme is open to any EU credit institutions, which satisfy a limited set of conditions equal for all participants. It is in fact in the very interest of the participating banks that the system be as open as possible.”

The scheme was initiated by the Bank of Italy, together with e-MID SIM, the company that operates the e-MID electronic interbank market, and the Italian Banking Association.  The  advisory committee attached to MIC consists of advisory committee composed of the Bank of Italy, e-MID SIM, the Association of Bank Treasurers (ATIC) and the Italian Banking Association.  So far, so Italian.

The scheme mentions ‘participating banks’, but does not define this category.  I assume these are the same banks that participate as counterparties of the Banca d’Italia in its other monetary and liquidity operations.  The general eligibility criteria for eligible counterparties are defined by the ECB:  “An institution may access the Eurosystem’s standing facilities and open market operations based on standard tenders only through the national central bank of the Member State in which it is established. If an institution has establishments ….. in more than one Member State, each establishment has access to these operations through the national central bank of the Member State in which it is located, notwithstanding the fact that the bids of an institution may only be submitted by one establishment … in each Member State”.

The potential for balkanisation along national lines of the Eurosystem’s monetary and liquidity operations in contained in the statement that “An institution may access the Eurosystem’s standing facilities and open market operations … only through the national central bank of the Member State in which it is established”. This feature of the ECB’s and Eurosystem’s eligibility criteria for counterparties for national central banks (NCBs) means that unless the 16 Eurozone NCB’s apply exactly the same rules and procedures and offer exactly the same facilities, there is no uniformity of monetary and liquidity policy implementation across the different member states of the Eurozone.  The MIC is an Italian example of a monetary and liquidity arrangement offered by an NCB that is not available to banks that are not present (as parent banks, subsidiaries or branches) in the jurisdiction of that NCB.

If I am correct, a Dutch bank without a subsidiary or branch in Italy cannot access the MIC facilities.  Of course the Dutch central bank can start its own MIC.  But that does not mean the impact of the Banca d’Italia’s MIC is not discriminatory and protectionist. If country A subsidises its domestic industry, this will have a protectionist impact on country B, even if country B is free to subsidise its own industry also!

I would be proven to have erred in my post, if Dr. Saccomanni can confirm that any bank that is an eligible counterparty for one of the 15 non-Italian NCBs in the Eurosystem can access the MIC on the same terms as the Italian banks, even if the bank in question does not have a subsidiary or branch in Italy.

Is there a quasi-fiscal subsidy inherent in the MIC scheme?

Does the scheme present a quasi-fiscal subsidy to the participating Italian banks? I assumed that in the MIC scheme, the Banca d’Italia was taking on the joint credit and counterparty risk in these transactions (that is, the risk that both the borrowing bank and the issuer of the collateral would default and that the amount recoverable from the other participating banks in the MIC through the collective collateral feature, would not be enough to make up the shortfall).  However, Dr. Saccomanni’s letter suggests this is not the case.  “Its role is that of a facilitator of trading, managing the custody of collateral, administrating and assessing the collateral, ensuring prompt settlement of transactions in the event of default by a participant, subsequently recovering the amounts from the collateral deposited.”  Dr. Saccomanni also states that “In the unlikely event in which the collateral provided were to be insufficient, the other market participants would jointly make up the difference within the limit of 10 per cent of their collateral”.

Does this mean that, in the even more unlikely event that 10 percent of the collateral of each of the other market participants were still insufficient to cover the loss, the lending bank rather than the Banca d’Italia assumes the uncovered part of the loss? If this is indeed the case, I misrepresented that aspect of the scheme in my blog post.

Even if the Banca d’Italia does bear the residual counterparty risk, it may still not represent a correct use of terminology to describe a party that acts as a facilitator of trading, managing the custody of collateral, administrating and assessing the collateral, ensuring prompt settlement of transactions in the event of default by a participant, subsequently recovering the amounts from the collateral deposited  and takes the residual counterparty risk as a market maker of last resort, as I did in my post. In that case I would like a more accurate short description of that role.

If, as I assume, the Banca d’Italia would be responsible for any remaining losses, there would be an ex-ante quasi-fiscal subsidy involved in the arrangement if these potential losses were not appropriately reflected in the fees, charges and other payments the Banca d’Italia charges the banks that use the facility.  If these losses materialise, the ex-post loss incurred by the Banca d’Italia might, however, not be borne the Banca d’Italia alone.

For losses incurred in the normal market operations used for monetary and liquidity management by the Eurosystem, there is the rule that such losses are shared by all the NCB’s in the Eurosystem in proportion to their relative shares in the capital of the ECB.

For losses incurred because of an NCB acting as lender of last resort towards one or more specific banks in its jurisdiction/country, the national Treasury of that country would have to indemnify the NCB in question.  Indeed the NCB would only act as lender of last resort to assist specific institutions as an agent of the national Treasury, and not in its capacity as a member of the Eurosystem.  A prior commitment by the national Treasury to indemnify the NCB for any losses incurred during a lender-of-last-resort-operation would be required before the NCB could engage in such an action.

The MIC scheme seems to fall somewhere between these two simple cases.  There is a potentially large population of participating banks, and the purpose is to enhance market liquidity in the interbank market rather than to provide liquidity to individual floundering institutions.  This makes it look like the case of normal market and liquidity operations where the losses are shared by the Eurosystem as a whole.  On the other hand, the only participating banks are Italian banks (or Italian branches and subsidiaries of non-Italian banks). That makes it look like the case of lender of last resort operations aimed at supporting a specific set of domestic Italian banks, which would require the Italian Treasury to underwrite the losses.

Dr Saccomanni is rather ambiguous as to whether ultimately the Italian Treasury may have to be good for any losses incurred by the MIC scheme.  He writes “Finally, in any case no fiscal back-up on behalf of the Italian Treasury is envisaged.” , and I am sure he is correct in that.  However, few of use envisaged last August, that the UK  Treasury would own 70% of RBS and that the Dutch Treasury would own 100 percent of Dutch ABN and Dutch Fortis.  In this crisis, the unenvisaged happens all the time.  Perhaps Italy will be an exception.  I find that difficult to envisage.

I hope this clarifies the issues and any remaining areas of disagreement.

Maverecon: Willem Buiter

Willem Buiter's blog ran until December 2009. This blog is no longer active but it remains open as an archive.

Professor of European Political Economy, London School of Economics and Political Science; former chief economist of the EBRD, former external member of the MPC; adviser to international organisations, governments, central banks and private financial institutions.

Willem Buiter's website

Maverecon: a guide

Comment: To comment, please register with FT.com, which you can do for free here. Please also read our comments policy here.
Contact: You can write to Willem by using the email addresses shown on his website.
Time: UK time is shown on posts.
Follow: Links to the blog's Twitter and RSS feeds are at the top of the page. You can also read Maverecon on your mobile device, by going to www.ft.com/maverecon