ECB Minimum Reserves - 10% or 10% less Government Bonds?

As expected, balance sheet policy has moved to the forefront of the monetary policy debate following the ECB’s – possibly last – interest rate hike in September and its decision to keep rates unchanged in October. This is consistent insofar as the ECB has pursued a bold unconventional monetary policy of balance sheet expansion in recent years. As a result, the stock of structural excess liquidity within the euro area has increased much beyond what is required for monetary policy in the medium or long term and under “normal conditions” (i.e. inflation around target, absence of external economic or financial market shocks). In this respect, after years of balance sheet expansion, a structural balance sheet reduction is now on the agenda, which has probably just begun hesitantly. That said the ECB will surely maintain a monetary policy regime of structural excess liquidity in the long term.

At first sight, the debate about a considerable increase of minimum reserve requirements in the euro area fits into the balance sheet adjustment discussion – even though other motivating factors seem to be at play too. Here we identify concerns about central bank profitability or “unjustified excess profits” in the commercial banking sector. From our understanding, however, a substantial increase of minimum reserve requirements – as currently prominently demanded by some actors, including the central bank governor of Austria and some academics1– would neither be a suitable instrument for structurally reducing excess liquidity nor distributing it more evenly among euro area banking sectors. We see the latter as key issues the ECB will have to deal with sooner or later, while preserving the homogeneity of monetary transmission within the euro area. The latter should be at the core of monetary policy in times of “high-for-longer” and possibly more substantive Quantitative Tightening (QT) – not so much short-term fine-tuning of the interest rate income in the euro area banking sector.

Figure 1: Euro area excess liquidity under various minimum reserve ratios

 

There is a history of central bank governors from Austria attracting special (market/media) attention with monetary policy statements. On the one hand, there have been cases where some of the ECB’s inside information has been spilled; on the other hand, Austrian central bank governors traditionally tend to belong to the hawkish monetary policy hawkish in Europe. And there has been some trench warfare between hawks and doves in the ECB Governing Council in recent years. As of late, the current central bank governor of Austria (Robert Holzmann) caused a stir with a statement that within the euro area a minimum reserve ratio of 5-10% could make sense for commercial banks, currently at 1%. However, this statement must be seen in a broader context of a monetary policy debate, actively joined by the German Bundesbank as well.

Central bankers’ analyses of minimum reserve ratios reflect the fact that the ECB has reached the end of the interest rate cycle and that further monetary policy impulses must be provided via balance sheet and/or liquidity policy. Conventionally, a (drastic) increase in minimum reserves has restrictive monetary policy effects. Above all, less lending, possibly some government bond selling and currently also a reduction in excess liquidity. As a reminder, as part of a broader package of measures, the ECB lowered the minimum reserve ratio from 2% to 1% at the end of 2011 to support bank lending and liquidity in the euro area money market. Away from the immediate reserve requirement issue, the ECB intends to address the issue of the medium-term size and composition of its balance sheet in the coming months anyway. Given the high level of excess liquidity, the possibility of a further increase in the reserve ratio (with zero interest rate payments by the ECB) has been discussed among market participants for some time. In addition, this would limit perceived “excess profits” in the European banking sector, a side-effect of the interest rate hikes at a breakneck pace as well as the outlook for key rates to remain high for longer (in restrictive territory).

Overall, euro area banks are currently depositing some 3,600 billion in excess liquidity at the ECB (deposit facility). That said, a return to the minimum reserve ratio of 2% (pre-2011 level) would reduce excess liquidity by “only” EUR 165 billion. A minimum reserve ratio at 10% would bring the excess liquidity position inside the euro area back to pre-COVID-19 crisis levels around 2,000 billion Euro. While an increase in the reserve ratio could be “sold” as part of a normalization of monetary policy (removal of the loose policy elements from the 2010s), it is hardly a pivotal factor for excess liquidity. Moreover, any increase in the minimum reserve ratio should be gradual to avoid jump effects and negative effects on lending and financial markets (risks of overtightening). This holds especially true, as excess liquidity is very unevenly distributed among euro area member countries. The implementation is, therefore, far from straightforward and runs into risks of triggering unintended side effects.

At first sight, every national euro banking sector could satisfy a minimum reserve ratio of 5-10% by reducing excess liquidity. Even Italian banks, which have the lowest (excess) liquidity position, hold EUR 197 billion within the ECB’s deposit facility, which is 10.6 times the minimum reserve requirement (Aug 2023). Yet, after accounting for the outstanding, but soon maturing, TLTROs this excess liquidity position shrinks to 1.2 times the minimum reserve requirement. Thus, if the ECB decides to go beyond a normalization of the minimum reserve ratio (increase well beyond 2%), the Italian banking sector would either need to substitute maturing TLTROs with alternative sources of liquidity to meet the minimum reserve requirements or a deleveraging in the sense of a balance sheet reduction would be in the cards. On the contrary, German and French banks hold excess liquidity positions of more than 20 times the minimum reserve requirements even after adjusting for maturing TLTROs. This very unequal distribution of excess liquidity in combination with a presumed forced approach of higher minimum reserve requirements, is likely to cause a certain degree of fragmentation on money and sovereign bond markets – something other monetary policy instruments are explicitly designed to combat (i.e. PEPP re-investment flexibility, TPI). Therefore, such a move could clearly undermine the homogeneity of monetary policy transmission within the euro area. Not to be forgotten, current very divergent lending trends between core Europe and the so-called (former) euro periphery (Italy, Spain) as well as sovereign bond spreads (mainly Italy) already point to certain fragmentation in the euro area or monetary policy transmission.

The ongoing controversial discussion about minimum reserves should not obscure the fact that the ECB will have to take key decisions regarding its balance sheet policy in the foreseeable future and possibly sooner-than-expected. In this sense, we do not believe that the minimum reserve is the appropriate instrument to structurally address the stock of excess liquidity in the euro area. The bulk of excess liquidity is represented by the ECB’s bond and especially government bond portfolios (currently close to 70% of the ECB’s balance sheet). An envisioned reduction in excess liquidity can no longer ignore this reality. In this context, it should be emphasized that the justification of simply prolonging (government bond) reinvestments within the PEPP until “at least the end of 2024” seems increasingly difficult to convey from a monetary policy perspective. We estimate that an earlier termination of reinvestments within the PEPP program (a move that would still represent a passive QT) would help to reduce excess liquidity by at least EUR 60 bn on a quarterly basis (i.e. a termination of PEPP reinvestments as of Jan 2024 would reduce excess liquidity by more than a minimum reserve requirements increase by 1% in 2024). Moreover, if the passive balance sheet rundown proves to be too gradual, a more active QT might be warranted. A reasonable solution could be possibly found in cooperation with the banking sector about how the ECB could dispose government bonds to the private sector (banks and here mainly in hold-to-maturity portfolios) without incurring too large losses. Moreover, such a step could be a precursor to a transition to a monetary policy regime as currently envisioned by the Bank of England (maintaining an increased central bank balance sheet, partly with government bonds, while at the same time providing collateralized lending/refinancing to banks to stabilize liquidity conditions). We see the BoE’s monetary policy setup more as a blueprint for the ECB rather than the Fed’s current monetary policy setup.

Figure 2: Towards a more balanced ECB asset mix: bond portfolio vs. refinancing operations

 

So far, the intellectual hypothesis has been that (substantially) higher minimum reserve requirements are intended to fulfill monetary policy targets, i.e. reducing excess liquidity. There is good reason to believe, however, that the main objective is to lower the interest paid by national central banks (to commercial banks). Thus, it’s possibly “just” profit / loss considerations rather than monetary policy contemplations, which are at the heart of the proposal or at least in the minds of some propagates. Also, the ECB’s July decision to lower the remuneration of minimum reserve requirements to 0% should be understood from the perspective of cost-efficiency. If the idea of a bumper increase in minimum reserves (in combination with zero interest rates) were not only aimed at fully absorbing “supposed” excess profits in the European banking sector, we believe there could be more targeted, market-friendly and balanced instruments. For instance, parts of the excess liquidity could be subject to somehow lower interest rates (e.g. by 150 basis points) than the depo rate (taking into account certain exemption limits to ensure money market interest rates are tightly linked to ECB key rates), so-called “reverse tiering”. In this context it is worth mentioning that banking sector profitability has to be seen from a through-the-cycle perspective, i.e. current profit readings are the basis for much needed future loss absorbing capabilities. Moreover, it shall not be forgotten that in the US monetary policy tools like minimum reserves are not in usage, while the Fed always remunerated US banks at market rates during all times of unconventional monetary policy and recent monetary tightening. Therefore, too much policy activism may put European banks once again in a less competitive position. In fact, very high minimum reserve requirements can be interpreted as extraordinary taxation of the banking sector, “levied” by the ECB. Against current liquidity positions, minimum reserve requirements of 5% could reduce the profitability of European banks by about 20-50%; orbitally high minimum reserve requirements of 10% would erode 36% to 100% of the net profits of recent years.

However, with regards to a “reverse tiring” the devil is in the detail, as the appropriate design is critical to keep a firm grip on money market interest rates. To limit interference with the conduct of monetary policy (money market interest rates) the share of excess liquidity which can be remunerated at 0% might be rather small and conditional on the liquidity preferences of commercial banks which are hard to quantify and do change over time. The total amount of interest which central banks can “save” without risking unintended effects for the conduct of monetary policy might be smaller than some central bankers currently think. Overall, designing a monetary policy friendly “reverse tiering” will be a non-trivial undertaking as it must be designed in a way that money market rates remain close to ECB key rates. Therefore, there are good reasons to think about alternatives.

What is not directly comprehensible is the argument that a significant hiking of minimum reserves in combination with zero interest rates would help to “build up reserves in the ECB balance sheet”. Of course, lower interest payments by the ECB to the private sector would allow for higher central bank profits. However, it is also the case that the ECB or the System of European Central Banks (ESCB) has made profits in recent years, some of which have been high, and has built up risk provisions. We think that, from a certain point onwards, the ECB and the public sector should also bear some costs of reversing the “special state” of monetary policy seen over the last decade. This is because, in our view, the private sector has so far borne the bulk of the costs of the ECB’s balance sheet/liquidity reduction (e.g. rapid reduction of LTROs, withdrawal of the ECB from the covered bond market). In contrast, only moderate exit steps are discernible in winding down (government) bond portfolios up to now.

Overall, we consider the discussion topic of minimum reserve ratios to be driven primarily by increasing balance sheet losses of individual central banks in the Eurosystem and consider an alienation of monetary policy instruments for profit-loss considerations to be hazardous, which ECB-President Lagarde views similarly according to a statement during the press conference after the October interest rate meeting. In this respect, a decision with regards to minimum reserves in the Governing Council may still take some time. An increase to 2%, would correspond to a normalization to pre-2011 levels. While this cannot be ruled out at one of the next ECB meetings in 2023 or 2024, an increase (well) above 2% will probably not take place before the ECB’s medium-term liquidity strategy has been revised.

Excessive utilization of minimum reserves would mean that we now want to limit negative effects of unconventional (crisis) monetary policy with instruments of “old school” conventional monetary policy. This does not seem to us to be very effective. To put it bluntly, the question arises: minimum reserve ratio at 10% or reduction of (government) bond portfolios by 10% more than intended? The latter step, a move towards more active QT, would rapidly reduce excess liquidity by at least EUR 500 billion. So far, QT has been handled in a very mechanistic way (let’s call it on autopilot), while we have seen several rounds of (market) surprise effects during Quantitative Easing (QE). We think the ECB could create a bit more symmetry between QT and QE if it sees the need to reduce excess liquidity faster than previously intended. Currently, market players surveyed by the ECB within the Survey of Monetary Analysts (SMA), heavily influenced by ECB announcements, expect the ECB to cautiously reduce its sovereign bond portfolio from just over EUR 5,000 billion currently to about 3,500 billion by the end of 2026. Over the next 3-4 years, this means a reduction of about EUR 1,500 bn in QT times, while in QE times a 2,500 to 3,000 billion increase in bond holdings had taken place in 2-3 years. Possibly QT has to be seen as a tool of more active monetary policy making for the euro area as a whole, while instruments to deal with country-specific market spillovers are in place (TPI). Moreover, any additionally created liquidity within the TPI could be sterilized by open market operations (i.e. fixed-term deposits), like ECB did during the Securities and Markets Program (SMP) in the early 2010s. Such sterilization should indeed be very easy given the completely different interest rate landscape today as compared to the early 2010s. Moreover, such a sterilization would keep holdings of central bank liquidity by commercial banks, possibly a valid monetary policy objective, unchanged at the aggregated euro area level.

Overall, we think that the ECB and national central banks of the ESCB should face the (subsequent) costs of ultra-expansive monetary policy and shall not easily pass them on ex-post (to the private sector). Only in this way we may expect a more measured use of unconventional monetary stimulus – if needed – in the future (especially bond purchases). Moreover, in our view, central banks are not profit-oriented asset managers, while lower central bank profits (i.e. lower transfers to state budgets) may possibly foster a much-needed enlightened discussion about the consequences of (excessive) usage of unconventional monetary policy tools at the national level and possibly within the EU political domain.