IBOR reform in the euro area – progress in an ocean of uncertainty
To respond to the concerns about the reliability and robustness of the IBOR benchmarks, the Financial Stability Board (FSB) recommended the development of an alternative (nearly) risk-free reference rate (RFR) in its report in July 2014.
—UPDATE—Last week, on February 25, the European Commission announced that the EU institutions agreed on postponing the deadline for the benchmark rate reforms by 2 years. Providers of critical benchmarks, such as Euribor and EONIA interest rates, and third-country benchmarks are granted two extra years until 31 December 2021 to comply with the new Benchmark Regulation requirements.
We recently published an article focusing on expected changes and challenges related to this reform (Challenges associated with the replacement of IBOR benchmark interest rates). This new article summarizes the latest developments for benchmarks in the euro area.
The critical benchmarks in the euro area are the EONIA (overnight rate) and the EURIBOR (term structure). The final transition date for EONIA/EURIBOR was originally set for January 2020 – two years earlier than the transition date for other currency rates. Recently, the EC (European Commission) granted an extension of the transition period until January 2022.
In September 2017, the working group on euro risk-free rates was established by the European Central Bank (ECB), the Financial Services and Markets Authority (FSMA), the European Securities and Markets Authority (ESMA) and the European Commission (EC) to find alternatives for the euro risk-free rate. They recommended the introduction of a euro short-term rate (€str) which reflects the wholesale euro unsecured overnight borrowing costs of euro area banks. €str will serve as the RFR to replace EONIA. After two public consultations on the features of this benchmark, the ECB published the final methodology for €str.
Starting October 2019, €str will be reported daily; since June 2018 a pre-€str is already being published (see Figure 1).
Figure 1 Pre-€str and EONIA compared (spread approx. 9bps)
The main points of criticism on EONIA/EURIBOR rates was that they are based on a panel survey (with a declining number of participants) and therefore are not representative and prone to manipulation. €str shows many desirable characteristics:
- underpinned by a broad base and stable volume of transactions;
- good representation of the euro area;
- representation of nearly risk-free bank borrowing costs;
- rate shows high stability; and
- minimal opportunities for market manipulation.
How is €str calculated?
As opposed to EONIA, €str will be calculated using overnight unsecured fixed rate deposit transactions over €1 million. The data is therefore already available (and sufficient to create a benchmark). Based on the reported transactions the volume-weighted trimmed mean is calculated by
- ordering transactions from the lowest rate to the highest rate;
- aggregating the transactions occurring at each rate level;
- removing the top and bottom 25% in volume terms; and
- calculating the mean of the remaining 50% of the volume-weighted distribution of rates.
In specified cases of data insufficiency, a contingency procedure is defined1.
And what about the term structure? (EURIBOR reform)
€str is only calculated based on overnight rates, and therefore no term structure is easily available, which imposes the question on how to replace the different EURIBOR tenor rates. So far, no final decision has been made on the replacement of EURIBOR term structure. What we already know is that the current methodology is not compliant with the Benchmark Regulations (BMR). Given the current market conditions, a transition from the quote-based methodology to one fully based on transactions is not feasible according to a pre-live verification program conducted by the European Money Markets Institute (EMMI) in 2016/17.
EMMI suggests transitioning to a hybrid approach while the quote-based EURIBOR will be continued as long as necessary (contingent on the contributor’s willingness to provide input data). Approval by the authorities is still outstanding but considered very likely by a majority of the market participants.
The hybrid EURIBOR will be based on a waterfall methodology, prioritizing transaction data over expert judgement. Transaction-based data is used to the extent possible (level 1 and level 22 and enriched with expert judgement from panel banks (level 3) if necessary.
In many cases, sufficient data for submissions based on level 1 and 2 is not available, which will (again) make the benchmark rely on expert judgment and creates a conflict with the new BMR. Figure 2 shows to what extent the benchmark will still be based on the models and opinions of panel banks. Nevertheless, the regulator will likely approve the approach due to the lack of alternatives and the expected smooth transition from the current benchmark rates.
Figure 2 Reliance on different levels in waterfall approach for hybrid – EURIBOR (source: EMMI, second consultation on hybrid-EURIBOR)
Fallback rates for legacy derivatives
For derivative contracts with EURIBOR exposure3 that are outstanding after the end of the transition period, it is up to the contract parties to agree on a different rate. If this does not happen (in time), the EURIBOR will automatically be replaced by an official fall back rate.
As the following table shows, more than half of the derivatives with exposure to EURIBOR that were outstanding in 2017 will outlive January 2020. Since products with EURIBOR exposure are still issued we can expect that on a final transition date in January 2022 there will be a similarly high amount of outstanding derivatives with exposure to EURIBOR4.
Table 1: Exposures to EURIBOR (source: ECB)
The International Swaps and Derivatives Association (ISDA) conducted a consultation among 152 market participants to establish fallback rates based on the new RFRs (here: €str) for the term structures of the corresponding IBOR (here: EURIBOR) rates. To make those fallback rates comparable to the current (EUR)IBOR structure, adjustments are necessary to reflect the following:
- the difference of a term rate (here: adjusted €str) to an overnight rate (here: €str); and
- the spread incorporated in the EURIBOR (e.g. due to credit risk premium) as opposed to the RFR.
For both adjustments different options have been proposed and at the end of November 2018 the preliminary results of the survey were published5:
Regarding adjustment (1) the overwhelming majority of the contributors were in favor of using the compounded setting in arrears rate approach, which means that the daily rates of the RFR will be compounded over the relevant IBOR tenor.
For the spread adjustment (2) the favored approach was to use the historical mean/median spread of the relevant IBOR over the RFR during a significant, static lookback period (e.g. five or 10 years).
Note that the presented fallback rates differ significantly from the EURIBOR term structure as it is now and therefore impact on the market value of derivatives can be expected.
1 For details see ESTER methodology and policies
2 Level 1 contains solely transactions based on underlying interest at the defined tenor from the prior target dates while level 2 allows the use of transaction data across the money market spectrum and from recent target days.
3 Or exposure to other IBOR rates
4 At the time of publication of this article, no newer estimations were known to the authors.
5 see ISDA Publishes Preliminary Results of Benchmark Consultation