ECB Balance Sheet: Composition, Scope and Limits

Having considered some of the inner workings of how commercial and investment banking works from the point of view of (1)Banking intermediation, Fractional Reserve Banking and Bank runs, (2) Asymmetries of Information, Overlending and the Output gap, and  (3) Leverage and Balance Sheet Management: Reserve Ratios and VaR, I now turn my attention back to the operation of monetary policy, in particular from the Euro-Zone perspective of the ECB. I wrote a brief post previously on the Monetary Policy tools available to the ECB. This post seeks to introduce a wider discussion of the use of these tools in the operation and transmission of monetary policy in the EMU.

In the lines below I will describe the evolution of the ECB’s balance sheet and what it tells us about the scope and limitations of monetary policy. The point I make is that the ECB has a virtually limitless ability to expand its balance sheet. Ultimately, in the absence of hyperinflation, this is due to the fact that the relative size of the Euro-Zone and its linkages to other world economies make it a fundamental piece of the puzzle, the malfunction of which would threaten the global economy.


From the last week of 1998, the balance sheet of the ECB has expanded by close to 288.17%. Adjusting this for inflation (28.43% between 1999 and 2012), it still corresponds to a 259.74% expansion in 1999 prices. What accounts for this?


On asset side more or less 75% of the balance sheet is comprised by 4 items that contribute disproportionately to this growth, which takes place, mostly from 2004 onwards, gains momentum from 2007 and explodes in 2009:

Item “5”, “Lending to euro area credit institutions” is, the largest contributor representing an average 32.07% of assets over the period under consideration, with sub items “5.1”, MROs, and “5.2”, LTROs, correspondingly accounting for an average 18.67% and 13.28% of total assets.

MROs peaked in the last week of 2007(2007W52) and continues to dominates the balance sheet until the last week of September 2008 (“2008W40”), two weeks after the bankruptcy of Lehman Brothers, at which point LTROs gains momentum and takes over as the largest contributor to the Assets of the ECB.

The LTRO fluctuated between a minimum of €22.230 Bn in the first week of 2009 (2009W1) and a maximum of €728.598 Bn Euro’s in the week of June 28 to July 4, 2009 (2009W26) These items basically show that the ECB’s normal monetary policy activities account for the majority of its assets, which is as it should.

The next highest asset category is item “2”, “Claims on non-euro area residents denominated in foreign currency”, which probably accounts for all loans made to foreign banks in foreign currencies. On average it accounts for 19.97% of the ECB’s assets, fluctuating, in absolute terms, between a peak of € 283.395 Bn in (2000W40) and a trough of €131.150 Bn on the week of Lehman Brother’s bankruptcy.

The third highest asset component is “1”, “Gold and gold receivables, World not allocated”, which corresponds to the value of gold bullion held by the ECB. This is an important asset for people like the Austrian School who take issue with the post-Bretton Woods, non-gold standard approach to monetary policy. Although the rest of the balance sheet will scare them, I hope it brings them some comfort to know that as the third largest asset category, it corresponds to an average 15.22% of the balance sheet, and experiences one of the lowest volatilities of the whole balance sheet.

The fourth item of interest is “9. Other assets, World not allocated (geographically), All currencies combined”, which accounts for 14.33% of the ECB’s assets. This asset peaked at €74.249 Bn in 2008W23 and fell as low as €38.7152 Bn in 1999W21.

The last asset category is item “7”, “Securities of euro area residents denominated in euro, Euro” which corresponds to the stock and credit markets interventions of the ECB and accounts, on average for 9.34% of its assets. This is a very interesting item, itself composed in two further sub-items, “7.1 – Securities held for monetary policy purposes, Euro” and the vaguely termed “7.2 – Other securities, Euro”. “7.1” tends to receive a lot of attention because it is the asset category which accounts for the special security buying programmes, the CBPP and the SMP, at the advent of the financial crisis. This asset class contributes very little on average to the balance sheet of the ECB because it only comes into being in the week between July 12 and 18, 2009 (2009W28). Since then this account has been continuously growing as the ECB accumulates Covered Bonds and government debt. As of January 27, 2011, the SMP and CBPP correspond to a cumulative €282.215 Bn.

Item “7.2” is less special but more intriguing to me because of the limited attention it has received despite its relatively larger size. It corresponds to 8.28% of the balance sheet, it peaked at €351.858 Bn in the middle of 2011 (2011W27) and fell to a trough of €20.914 Bn on the first week of 1999.

Comments on ECB Assets

Regarding the intervention of the ECB, there are three very interesting points that should be made.

The first such argument is about what items are better suited to account for the intervention of the ECB in sovereign and banking debt markets to mititgate the negative shock of the sovereign debt crisis. Although, in principle, the ECB uses item 7.1 to account for its purchases under the SMP and the CBPP, which are its declared interventions the ECB’s policy is more complicated than this. Indeed, in the annex describing the Accounting conventions and content of the consolidated weekly financial statement of the ECB, the following two statements regarding items 7.2 and 9. are made:

 “Item 7.1 Securities held for monetary policy purposes reflects securities issued in the euro area held for monetary policy purposes and ECB debt certificates purchased for fine tuning purposes. Sub-item asset 7.2  Other securities, shows holdings of securities and government debt other than those under asset item 7.1 and under asset item 9 Other assets and contains certain categories of marketable securities and equity instruments, which may potentially be used for monetary policy operations.”

“Other assets is a collective item that includes items in the course of settlement, coins of the euro area if an NCB is not the legal issuer, tangible and intangible fixed assets and other financial assets. Other financial assets comprises participating interests and investments in subsidiaries; equities held for strategic/policy reasons, securities, including equities, and other financial instruments and balances (e.g. fixed-term deposits and current accounts), held as an earmarked portfolio: reverse repo transactions with credit institutions in connection with the management of securities portfolios. This item also contains revaluation differences arising on off-balance-sheet instruments and accruals and prepaid expenditure.”

The interesting point however is that this has changed since the onset of the Eurozone sovereign debt crisis. Indeed, whereas the average observations above are representative of most periods, a clear change has occurred in the last year and a half which has caused the size of items 7 and 9 to explode. This is just one clue to the amount of debt monetisation that ECB has been forced to conduct in order to contain the shock to bank balance sheets created by the sovereign debt crisis.

Moreover, the marginal lending facility, which accounts for all overnight lending by the ECB, never corresponds to a high value, meaning that in general the ECB’s main role is not to facilitate the overnight market, but rather to control it. It’s peak value was in the last week of 1999, when it represented a little above 1.4% of the ECB’s balance sheet.

Finally, and for purposes that shall become clearer closer to the end of this post, it is important to consider the size of assets denominated in foreign currencies. The relevant accounts here are the number 2., 3., and 9. As of February 6, the sum of these accounts represents 25% of GDP, 20% less than it did at the inception of the EMU.  This can be understood as the result of a very progressive, but consistent sell-off of such assets, beginning in 2003.


On the Liabilities side, the largest there are also 4 categories which account for more or less 75% of the balance sheet:

The largest liability of the ECB is, on average item “1. Banknotes in circulation, World not allocated (geographically), Euro”, ie: bills and coins either in your pocket or stored in bank vaults. It represents 46.25% of the ECB’s liabilities, on average, and as much as 32.40% as of February 6, 2012.

It corresponds to the fraction of the monetary base M1 that excludes overnight deposits at the ESCB through the Deposit facility. I suppose we could call it M0. Since the birth of the Euro it’s stock has grown by 1.55%. As the figures below show, interestingly, inflation does not seem to be exclusively determined by M0 supply, which has on average grown 0.02 % slower.

Item “2. Liabilities to euro area credit institutions related to MPOs denominated in euro, Euro area (changing composition), Euro” Is the second largest liability of the ECB, representing on average 16.57%, and as much as 29.74% on February 6. It peaked at 32.49% two weeks before the end of 2011 and has remained at similarly high levels since then. It consists of several other items. Although ”2.1 – Current accounts (covering the minimum reserves system), Euro area (changing composition), Euro” accounts for overwhelming majority of this liability category, “2.2 – [Overnight] Deposit Facility” is the next largest sub-category. The disproportional size of 2.1 is not complicated to understand in light of the fact that every financial institution operating in the Euro-Zone must comply with reserve requirements. The rest of the items are very small. They are composed majoritarily of only used very infrequently by the ECB

The next Item is “11. Revaluation accounts, World not allocated (geographically), All currencies combined”, which represents 11.95% of the ECB’s balance sheet, on average. Because the meaning of this account is not intuitive, I had to access the “WFS User Guide”, which describes it as

“unrealised gains related to price movements, foreign exchange rate movements and market valuation differences related to interest rate risk derivatives”. So in a sense these are the liabilities that the ECB has incurred due to the fact that it operates in several international financial markets. It peaked at 17.8% during the week of October 16 to 22, 2000.”

The last of the average largest four liabilities of the ECB is item “10. Other liabilities, World not allocated (geographically), All currencies combined”, which represents 7.94% of the ECB’s liabilities. According to the “WFS User Guide”, it is

“a collective item including items in the course of settlement, the revaluation differences of off-balance-sheet instruments, accruals and income collected in advance. This position also includes provisions and profit for the year”.

It peaked at 11.19% during the end of January 1999.

Aside from these, there are 2 other categories, limited in size but important in scope, which are worth mentioning. The first, is the liability counterpart to the monetary programmes covered by the Asset item “7.1”, namely Liability item “2.3. Fixed-term deposits, Euro”.

Finally I would like to consider all those liabilities denominated in a foreign currencies. This group includes Items 7. to 12. On average these Items have fluctuated between representing 20-35 % of the ECB’s balance sheet and as of February 6, they represented 28.3%, which is somewhat midrange.


These insights provide a useful template for understanding the limits of ECB intervention, of which there are several. Some such as price stability are mandated by political institutions, others, such as those arising from the Moral hazard cum hyperflation fears created by seignoreage/debt monetisation are self-imposed. However, a third class is created by the market and by the ECB’s balance sheet. It is with this last one that the end of this post is concerned with. The next few lines are heavily influenced by this highly recommended CEPR policy insight article from Willem Buiter.

As is noted in that article, any central bank can theoretically expand its balance sheet infinitely and immediately. The only two constraints facing it, from a purely balance sheet point of view are the proportion of inflation-linked and foreign currency denominated liabilities it holds on its balance sheet. This is because in the expansion of its balance sheet, the central bank issues currency to purchase assets that are threatening the stability of financial markets and so need to be purged from them. Whether this is done through QE or through Debt Monetisation, this seignoreage creates a liquidity flood which will internally devalue the currency (inflation) as well as externally (depreciation).

Limits imposed by Inflation-linked Liabilities

If a central bank holds too many inflation-linked liabilities, the potential (hyper-)inflation created by the expansion of base money necessary to pay for the asset purchases of the central bank would automatically increase the value of liabilities and threaten the solvency of the central bank. As Buiter puts it

“if we start from low rates of inflation, the central bank can, if it suffers a capital loss, boost its solvency by increasing the growth rate of the nominal base money stock and the rate of inflation. For a while, such increases in the growth rate of base money and the rate of inflation will boost the real resources the central bank can extract from the economy through the issuance of base money. When inflation gets too high, however, the disincentive effects on desired holdings of real money balances associated with higher expected inflation will reduce the desired money-income ratio to such an extent, that higher inflation reduces the value of the real resources the central bank can appropriate by issuing base money.”

To gain a better understanding of this issue it would be beneficial to know the volume of such assets in the ECB’s liabilities as well as the volume of such assets in the markets. However, the first step, which for now should suffice, is to consider the existing rate of inflation. Clearly, the inflationary problem is still a long way away, as the Euro-Zone’s inflation, at 2.7% annually, is below that of the USA (3%), the UK (4.2%), China (4.1), India (6.5%) or Brazil (6.5%). Of course if the ECB begins expanding its balance sheet at an exponential rate, these facts should change quickly, at which stage the other two issues will become relevant.

Limits imposed by foreign denominated Liabilities

If the expansion of the central bank’s balance sheet is done through the purchase of assets denominated in a foreign currency, a similar problem arises, where by the foreign currency reserves at the disposal of the central bank may not be sufficient to fulfil this purchasing need. In order to deal with this, the central bank would begin to sell (assets denominated in) Euros in favour of the relevant currencies, thereby incrementally increasing the amount of Euro’s necessary to purchase those assets. This would not just increase the (foreign denominated) asset expansion cost to the ECB but also it would increase the value of (foreign denominated) liabilities in its balance sheet.

 As noted above the ECB holds more foreign denominated liabilities than it does assets. This means that if it needed to bailout banks or countries whose bad debt/assets were denominated in dollars, the ECB would have a lot of assets to sell-off, but it would be likely to experience exchange rate costs very fast. Nonetheless, the relevance of those assets and liabilities, although relevant, does not represent a majority of the assets of the ECB.

Despite the limited (albeit sizeable) volume of these exposures, there is one more reason why the ECB is likely not to suffer from external financing problems. This has to do with systemic relevance, bargaining power and the ability to project influence.

Unfortunately, it is impossible to obtain up-to-date data on this, as economic measures for the entire world are aggregated with a rather substancial lag. However, as of 2010, assets (M3) held by MFIs operating in Euro-Zone amounted to +/-  €9.5 Tn in nominal seasonally adjusted terms, or $US15.62 Tn in constant 2000 prices. This corresponds to about one third of the total amount of broad money circulating in the whole of the World economy (+/- $US 45 Tn in constant 2000 prices) or 36% of total World GDP in the same units.

The point I wish to make is that the Euro-zone is too important an economic zone to allow it to experience such problems. In the case where such an event would be triggered the ECB would, as it has, enter into forex swaps with other central banks in order to coordinate the response to such a crisis.


In light of the facts described above, I feel comfortable in stating that the ECB’s balance sheet is relatively healthy, in light of its operations and of the present crisis. As a result, I would argue in favour of an expansion of the assets held under category 7., through the SMP and the CBPP, as well as through item 7.1, which I suspect might actually being used for those same debt market stabilising purposes. The foreign exchanges as well as the inflationary risks are, at the moment, moderate enough to warrant such an expansion.

This entry was posted in ECB and tagged , , , , , , , , , , , , . Bookmark the permalink.

Leave a Comment

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.