Should insurers leverage

Solvency 2 discount rate

techniques when valuing
insurance liabilities under
IFRS17?
July 2020
Brochure / report title goes here |
 Section title goes here

1. Discounting under IFRS17

2. D
 ifferences between Solvency 2 and IFRS17 discounting approaches

3. Alignment between EIOPA’s techniques and IFRS17 requirements

4. Assessment of the liquidity characteristics of insurance contracts

5. Conclusion

02
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

Introduction and structure

of this paper
Discount rates are usually seen as a technically challenging topic for
Preamble
insurers, especially given the impact they could have when valuing the
This document
time value of money and guarantees of long-term life insurance contracts.
is based on first
market orientations With the upcoming application of IFRS17 to insurance contracts, the
reflecting discussions measurement of insurance liabilities will be a key factor in determining the
and benchmarks level of technical provisions and may influence the pattern of recognition
with insurance and of insurers’ IFRS17 profits.
reinsurance companies
across the world. It is
In Europe, while Solvency 2 sets very clear guidelines regarding
not a reflection of either
discounting, long debates and discussions have nonetheless taken place
a Deloitte opinion or
guidance on setting around the effect of long-term guaranteed measures (for example in the
discount rates under European Commission’s review of the Directives, as presented in the last
IFRS17. consultation paper issued by EIOPA1). For IFRS17, the Standard2 describes
general principles rather than rules for discounting, leaving various
possibilities for insurers to consider.

In this paper, we start by introducing the principle-based IFRS17

requirements regarding discounting. We then address the differences
between these requirements and techniques described under Solvency 2.

We also assess the possibility of using Solvency 2 prescribed techniques

for IFRS17 when defining the discount rate, especially when considering
the bottom-up approach and the volatility adjustment as a measurement
for liquidity premium for liabilities.
Finally we describe alternative approaches to estimate the liquidity
premium when applying the bottom-up approach for IFRS17, based on the
characteristics of the relevant insurance contracts.

1 2020 Review of Solvency 2 – Oct 2019 LINK

2 IFRS17 Insurance Contracts – May 2017 and Amendments to IFRS17 – June 2019
03
Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

1. Discounting under IFRS17

IFRS17 will replace IFRS4 for insurance • A bottom-up approach, where cash flows
contracts starting from January 1st 20233. are discounted using a yield curve that
exposes the holder to “no or negligible
Paragraph 36 of the Standard states credit risk” (paragraph B79), adjusted to
that an entity shall use a discount rate to reflect “the liquidity characteristics of the
estimate the present value of its future group of insurance contracts” to which
cash flows. The discount rate used should: the curve is applied;
(1) r eflect the characteristics of the cash • A top-down approach, where the discount
flows and the liquidity characteristics of rate is defined by adjusting the yield
the insurance contracts, curve that reflects the current market
(2) b e consistent with observable market rates of return, implicit in a fair value
prices, and measurement of a reference portfolio
(3) e xclude the effect of factors that of assets (paragraph B81), adjusted to
influence such observable market prices eliminate any factor that is not relevant to
but do not affect the future cash flows the insurance contracts.
of the insurance contracts.

Paragraph B82 further states that the yield Illustrative example of top-down and bottom-up approaches
curve must reflect certain characteristics
such as observable market prices wherever 3.5%
possible. Additionally, paragraph B83 Reference assets portfolio return = 3.2%
states that an entity should adjust the
yields observed in the market to match
3.0%
the characteristics of liability cash flows.
In particular, for cash flows that do not
vary based on the returns of the assets of Factors that
reference, those adjustments must: are not related
2.5% to insurance
• Include differences between the amount, contracts
the timing and uncertainty of cash flows (e.g. expected
and non
of the assets in the portfolio, and the
2.0% expected credit
uncertainty of the cash flows of the loss) = 1.8%*
insurance contracts, and
• Exclude market risk premiums for credit
risk, which are only relevant to the assets Bottom-up approach = 1.5%
1.5%
included in the reference portfolio.
Liquidity
premium Top-down approach = 1.4%*
More broadly from paragraphs B72 to B85, = 0.5%
the Standard provides some guidelines to 1.0%
define a methodology for discounting. Two Risk-free rate = 1%*
potential approaches are proposed:

0%

*Percentages as examples

3 As published on June 25th 2020

04
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

According to the Standard, both are traded. Therefore, discounting under

approaches could be used for all types IFRS17 is expected to be supported
of insurance contracts (i.e. life, health, by expert judgment and estimation
protection and P&C), either with cash techniques that will most probably differ
flows that vary based on the returns of from one insurer to another and from one
underlying items or not. The discount country to another. This does not allow for
rate could be determined by adjusting easy comparison between two disclosures
the liquid risk-free yield curve to reflect even if companies are asked to provide
the differences between the liquidity details in their communication.
characteristics of the financial instruments In Europe, companies will most likely rely
and the insurance contracts (paragraph on Solvency 2 techniques for IFRS17 needs
B80), or the entity may adjust the return of (see market trend box), but will need to
a reference portfolio (real or theoretical) adjust the characteristics of their own
by removing factors that are not relevant business in order to comply with IFRS17.
to insurance contracts. In the latter case,
the Standard does not require any further
Market Trend
adjustment of the top-down yield curve
Based on a Deloitte survey among 15 global insurers4, it appears that at the end of 2019
in respect of the liquidity characteristics
the market had just begun to discuss discount rate methodologies in detail, and that
of the insurance contracts; the liquidity
many grey areas were seen to remain in the interpretation of the Standard’s requirements
characteristics of the reference portfolio
and usage under IFRS17.
can be retained unadjusted (paragraph
However 8 of the 9 European insurers surveyed were keen to use known methodologies
B81).
from Solvency 2 or QIS 5 techniques to define the basics behind discounting under
IFRS17. Nevertheless, there is some doubt still as to the adjustments they will need to
It can be seen that IFRS17 provides clear
make on these known methodologies in order to adapt to IFRS17 expectations.
principles but no detailed technical
approach to be applied when defining the
Among the drivers that will help companies to decide between methods, the most listed
yield curve. Companies are then expected
were:
to define their own approach, as long as it
• the level of future profit;
complies with the principles of IFRS17. In
• the volatility to which their P&L is exposed;
doing so, the key guiding principles are:
• the simplicity of the disclosed communication;
• Consider the characteristics of the cash
• the impact in terms of process change.
flows for both approaches and the
liquidity characteristics of the insurance
As a consequence, in Europe many insurance companies are interested in leveraging
contracts when bottom-up approach is
discount rates methods defined under Solvency 2. Many of the IFRS17 requirements
applied (B78)
are met by using Solvency 2 methods, the methods are already known by investors and
• Maximize the usage of observable market supervisors, and the impact on processes could be reduced if insurers optimize the usage
data (B82 a and b) of current tools and methodology from Solvency 2.
• Use estimation techniques when data is
not available or not robust enough (B82c).

While many market instruments can

be used to derive risk-free discount
rates (mostly interest swap rates and
government bonds), robust, deep and
liquid market data is rarely available
especially for very long maturities.
Additionally, the assessment of liquidity
characteristics of insurance contracts
is not a simple exercise as there is no
liquid market where insurance liabilities

4 Benchmark performed by Deloitte in September 2019 among 15 international (re) insurers, of which 9 are based in
Europe, 3 in Americas and 3 in Asia, of which 10 are composite (re) insurers and 5 are pure insurers (either life or P&C)
05
Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

2. D
 ifferences between
Solvency 2 and IFRS17
discounting approaches

Under IFRS17, insurers are required to define a methodology that is appropriate to their own business, whereas
under Solvency 2, EIOPA publishes the risk-free yield curve to be used by currency, as well as the adjustments to
be performed on the risk-free rates.

The table below presents the main differences between the two standards:

Item Solvency II (for Euro economy) (5) IFRS17

General approach Risk-free curve provided by EIOPA to which a Discount rates determined either by adjusting a
volatility adjustment or a matching adjustment is portfolio’s total return (top-down) or by adding a
added. liquidity premium to a risk free rate (bottom-up).
Granularity per liability The volatility adjustment is set per currency and IFRS17 does not set any requirement in terms
characteristic is the same for all insurance and reinsurance of granularity; however, as the liquidity premium
obligations, unless a country-specific adjustment is reflects the characteristics of the insurance
applied. contracts, it is expected to have different discount
rate curves depending on the currency and liquidity
characteristics of underlying portfolios.
Frequency EIOPA publishes the risk-free discount rate as well Insurers need to estimate the discount rate under
as the volatility adjustment and all data needed on a IFRS17 for at least each closing period.
monthly basis.
When applying the bottom-up approach
Risk-free discount rate The risk-free yield curve is based on 6-month The initial risk-free yield curve can be determined
Euribor swap rates6 – corrected using an based on several financial instruments: swap rates,
adjustment defined by EIOPA. EONIA rates, government bond rates,... corrected
to reflect no or negligible credit risk exposure for
the holder. However, observability, liquidity and
robustness of the data used have to be justified.
Last Liquid Point The Last Liquid Point for the Euro zone is fixed by Not specifically defined, but should make reference
EIOPA to 20 years7. to the liquidity of financial instruments observed on
the market.
Ultimate Forward Rate EIOPA defines a methodology for calculating the The methodology for setting an Ultimate Forward
UFR based on historical observed rates as well as Rate is not defined. The insurance company is
expected future inflation. expected to maximize the use of observable data.

In addition, variations in the UFR from one year to

another are capped and floored.

06
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

Extrapolation technique EIOPA sets Smith-Wilson extrapolation method to The Standard leaves it to expert judgement to set
address the unavailability of data from the last liquid the extrapolation technique to estimate rates which
point to the UFR. cannot be determined from available market data.
Adjustment to the risk- For the volatility adjustment calculation, EIOPA The Standard expects the discount rate curve to
free yield curve defines a reference bonds portfolio that reflects reflect the liquidity characteristics of the insurance
the average investments of European insurers. An contracts.
adjustment is then applied to eliminate any credit
risk related to the reference portfolio. EIOPA then
sets an adjustment of 65% on the risk-corrected
spread.
In some restrictive conditions, EIOPA allows for the
use of a matching adjustment (MA) technique. The
use of matching adjustment requires approval from
local insurance supervisors.
When applying the Matching Adjustment technique,
insurer use their own investment portfolio.
When applying the top-down approach
Reference portfolio IFRS17 does not set any requirements regarding the
portfolio, either real or theoretical, to be used in the
top-down approach.
Real estate and mortgage Solvency 2 does not have a top-down approach There are no requirements regarding the types
to derive the overall yield curve. There are some of assets to be included in the reference portfolio
similarities to a top-down approach in the derivation when estimating the discount rate; real estate and
of the Volatility Adjustment or the Matching mortgage assets might be permissible.
Risk premium Adjustment - see above. There is no explicit restriction on the usage of a
consideration risk premium for liabilities with cash-flows that vary
based on the returns of the reference assets.
Adjusting liquidity When applying the top-down approach, an entity
premium calculated based shall adjust the market consistent yield curve
on assets to reflect liability to eliminate irrelevant factors to the insurance
characteristics contracts, but is not required to adjust the yield
curve for differences in liquidity characteristics of
the insurance contracts and the reference portfolio.

5 Technical documentation of the methodology to derive EIOPA’s risk-free interest rate term structures LINK
6 The working group on euro risk-free rate from European Central Bank recommended the use of euro short-term rate (€STR) as a risk free rate alternative to
replace the benchmarks used for variety of financial instruments and contracts in the Euro area LINK
7 The 20 years LLP as well as the VA methodology are challenged in EIOPA’s Consultation Paper on the Opinion on the 2020 review of Solvency II
07
Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

Many conceptual similarities exist between the principal based IFRS17 and the techniques prescribed by Solvency
2. Nevertheless, EIOPA’s technique are not completely8 aligned with IFRS17 and some adjustments will need to be
performed.

How is long-term discounting considered within the ICS Standards9

For the final year of field testing of its Insurance Capital Standard (ICS), the IAIS has shared its
approach on assessing long-term discount rates.
The conclusions are that:
• while the ICS follows a broadly similar approach to Solvency II, the resulting Euro and Sterling discount
curves could nonetheless differ, potentially introducing significant valuation differences between both
regimes;
• the need to move away from reliance on LIBOR and EURIBOR in the future could introduce new
differences in methodology for risk-free rates; and
• 2019 field testing may provide an illustration of how a “single adjustment mechanism”, replacing the VA
and MA, could work in Solvency II, as recently envisaged by the European Commission.

8 See next section

9 ICS 2019 Field Testing: shedding light on Solvency II long term discounting – July 2019 – Deloitte LINK

08
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

3. Alignment between
EIOPA’s techniques and
IFRS17 requirements
In one of its issued papers10, EIOPA For IFRS17 purposes, the 6-month swap
The UFR in EIOPA’s methodology
declared that the overarching principles rates as well as the relevant Overnight
is determined by summing the
of the volatility and matching adjustments Index Swap (OIS) rate that is used for
expected long-term nominal expected
techniques appear to be in line with the the CRA estimation are considered to be
inflation and expected real interest
IFRS17 guidance on calculating discount “observable market data”. In addition,
rates.
rates. However, EIOPA acknowledges that EIOPA uses interpolation techniques
The variation of a UFR from one year
a “slight” difference in requirements also for non-relevant market data for some
to another cannot exceed 15bps.
exists between the two Standards and that maturities based on expert judgment that
For the last 3 years, UFR has been
methods defined for Solvency 2 need to be can be reused in the context of IFRS17.
decreased by 15bps per year, which
adapted to align with IFRS17 requirements. However, under IFRS17, some assumptions
corresponds to the maximum
and techniques used by EIOPA are
authorized by the technique as
The VA technique set by EIOPA appears challengeable. For example, the choice
defined by EIOPA.
to match the description of a bottom-up of a LLP set to 20 years when market
approach11 defined by the IFRS17 Standard, data are available up to 30 years12 could
as first a risk-free rate curve is defined, be questioned under IFRS17. The same
then an adjustment, that reflects the observation concerns the methodology
misalignment between assets and liabilities and inputs used by EIOPA that lead to
in terms of liquidity and uncertainty, is a material and volatile UFR (UFR was at
added. 4.2% in 2017 for Euro currency, decreased
to 4.05% in 2018, then to 3.9% for 2019
calculations for the Eurozone).
How does the Solvency 2 risk free rate
comply with IFRS17?
The risk-free rate as set by EIOPA is
based on 6-month swap rates that are
observable on the market, until the last
liquid point (LLP) set to 20 years for Euro
currency. Starting from 20 years maturity,
the rate is extrapolated until the Ultimate
Forward Rate (UFR), using the Smith
Wilson extrapolation method. The rates
obtained are adjusted using a Credit Rate
Adjustment (CRA) applied as a parallel
downward shift to the risk-free curve
observed for all maturities until the LLP.

10 EIOPA’s analysis of IFRS17 Insurance Contracts LINK

11 S
 ome experts consider the VA/MA approaches as a mixed (top-down and bottom-up) approach as the liquidity premium added to the risk-free rate is deduced
from a reference asset portfolio
12 The level of bid/ask on 20 years European CMS is equivalent to the 30 years European CMS over the last 5 years (source Bloomberg)

09
Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

How does the volatility adjustment

The Solvency 2 volatility
comply with IFRS17? Another question can be raised regarding
adjustment is aimed at dampening
Similarly to the situation with the risk-free the usage of a risk premium on non-fixed
the “own funds’ artificial volatility” that
rate, the use of the EIOPA approach to income assets when determining the
is caused by the stressed fixed-income
set the volatility adjustment for IFRS17 adjustment related to insurance contracts
financial markets. It is calculated on
purposes is subject to discussion. Whilst with cash flows that are asset-dependent.
a generic European bonds portfolio
it seems clear that IFRS17 allows a VA in This option has been excluded from EIOPA’s
(government and corporate), ensuring
concept, it is less clear if the EIOPA VA could methodology where only fixed income
convergence in the calculation of
be used without adjustment. In particular, assets are used when assessing the VA or
Solvency 2 pillar 1 quantitative
the VA based on EIOPA’s methodology is the MA. Non-fixed income assets (equities
requirements
assessed using a generic assets portfolio and properties for the most part) are
that represents the average investment of held by insurers in order to capture a risk
European insurers: would this be permitted premium in a real-world environment; the
under IFRS17? risk premium considered is shared with
the policyholders for contracts that are
In the Deloitte paper “Volatility adjustment asset-dependent which may leave open
under the loop”13, the authors explain the possibility under IFRS17 to add an
that deriving the monetary impact of the additional element of risk premium to the
volatility adjustment directly from the liquidity premium.
asset and liability portfolios owned by the
insurer itself allows for better capturing Finally, while EIOPA uses a 65% fixed factor
the characteristics of insurance contracts, to adjust the asset liquidity premium
in terms of duration, credit exposure, calculated based on the reference portfolio
and liquidity. In this context, the usage of to derive the VA, IFRS17 clearly requires that
other techniques like a dynamic volatility the liquidity premium captures the liquidity
adjustment (DVA) allows the size of the VA characteristics of the insurance contracts
to change with the characteristics of the held by the insurance company, meaning
insurers’ own portfolios over time, and that more consideration of the appropriate
therefore to be accordingly impacted by the factor to move to the liability-based
changes in duration. This alternative could assessment is needed.
be considered as more aligned with the
IFRS17 principles outlined above. Hence, when defining the adjustment to be
applied to the risk-free yield curve under
Furthermore, the use of a Solvency 2-style IFRS17, insurance companies will need to
country-wide14 volatility adjustment for assess the liquidity characteristics of their
IFRS17 discounting is challengeable. insurance contracts portfolio by portfolio,
Country-wide15 asset portfolios might and potentially at a more granular level
not be seen as fulfilling the criteria of the defined by groupings of policies with similar
bottom-up IFRS17 approach considering liquidity profiles.
they do not reflect the liquidity of the
insurer’s liabilities, and under the top-down
approach the country-average portfolio of
assets underlying the Solvency 2 VA might
not be seen as an appropriate reference
portfolio.

13 Volatility adjustment under the loop – February 2018 – Deloitte LINK

14 T he country “VA specific” is being reviewed by EIOPA for its 2020 review of Solvency II LINK
15 In November 2018, the country-specific VA in Italy increased the VA to 54bps vs. 22bps without specific VA. Such difference (32bps) can have a material impact
on IFRS17 measures

10
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

4. A
 ssessment of the
liquidity characteristics
of insurance contracts
Usually, when applied to tradeable assets, The liquidity of insurance liabilities is here In this context, different factors for
the liquidity premium refers to the financial considered from a policyholder standpoint determining to which extent an insurance
excess demanded by investors when the rather than from the “tradability” liability is liquid or not have to be assessed.
asset held cannot be easily converted into perspective of the company, and focuses In its Solvency 2 consultation paper, EIOPA
cash for its fair market value. However, on the predictability of cash flows: if a proposes 2 sets of high-level guidelines in
there is no transparent and liquid market liability is highly illiquid, the corresponding order to measure the liquidity of a given set
for trading insurance liabilities, and this cash flows are more predictable and the of liabilities:
definition is difficult to apply. The MCEV insurance company is more disposed to 1. L
 iquidity buffer based on the terms and
(Market Consistent Embedded Value) hold the backing assets to maturity in order conditions of the insurance contract ;
principles use another definition of what to target a higher investment return. On the
2. Liquidity buffer based on duration.
liquidity could mean in relation to insurance other hand, if a liability is liquid, the cash
liabilities16 : “A liability is liquid if the liability flows are less predictable and the insurance
cash flows are not reasonably predictable”. company would not be disposed to hold to
maturity the assets backing this liability.

1. Liquidity buffer based on the terms and conditions of the insurance contract
Based on the assumption that a contract is totally liquid when its underlying cash flows are highly unpredictable, the first step is to consider
whether or not the insurance policy includes features that could be seen as highly unpredictable.

In its consultation paper, EIOPA has proposed the following grouping:

Group of illiquidity Features of the contracts Typical examples of contracts

High illiquidity • Without any surrender/cancellation option or • Annuities in payment phase
where the surrender value does not exceed the • Term life insurance (without savings component)
market value of the assets • Disability insurance
Medium illiquidity • Contracts with limited surrender risk: • State subsidized pension products
- including disincentives for surrender
- low risk charge for the risk of a permanent
increase in lapse rates…
• Contracts with low mortality risk and
catastrophe risk...
Low illiquidity • Contracts that do not fall into the first two • Unit linked contract
categories

Depending on the degree of liquidity, a different liquidity factor would be applied to the liquidity premium calculated
on the assets portfolio backing these liabilities.

16 Paragraph 138 of the Basis for Conclusions from CFO Forum – April 2016

11
Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

2. Liquidity buffer based on duration duration: the more sensitive a contract’s duration is
to stressed scenarios, the more liquid is the contract,
Another feature that could influence the liquidity of a
or on an assessment of the duration variation using
liability is its duration, or more precisely the change in
stochastic scenarios.
its duration when stressed events occur (mass lapse
due to a systemic event for example). The assessment
of this type of liquidity can be based on Macaulay

17 Liquidity Premium buckets EIOPA Report on the fifth Quantitative Impact Study (QIS5) for Solvency II – March 2011 LINK

12
 Should insurers leverage Solvency II discount rate techniques when valuing insurance liabilities under IFRS17?

5. Conclusion
Well before the IFRS17 Standard was published, Finally, whilst Solvency 2 discounting methodology
the European Commission has defined guidelines shares much in common with the discounting
and techniques related to the long-term estimation requirements of IFRS17, important differences remain.
of the time value of insurers’ liabilities, based on The more principles-based philosophy of IFRS17
either observable market data (historical or market- presents an opportunity to insurers to examine where
consistent) or on expert judgment. These estimation these differences allow them to consider their own
techniques present many advantages that incentivise liabilities and potentially assets to drive their discount
insurers in Europe and beyond to think about reusing rates. The comparative freedom of interpretation,
them in the context of IFRS17. It is clear that these have however, means that a range of outcomes from one
a common influence on the methods underlying the insurer to the next is to be expected, and time will
prescribed Solvency 2 discount curves as well as the be needed for insurers to settle on their final chosen
principles-based IFRS17 guidance. approach.

In relation to this, it will be of interest to insurance

companies to monitor the proposed 2020 EIOPA
Solvency 2 review of long-term guarantees, where
the techniques behind setting discount curves are
discussed.

13
Contacts

Imène Seghouani Simon Walpole Baptiste Brechot

Director Partner Partner
Tel.: +33158370558 Tél.: +41582797149 Tél.: +33155617912
Mail: iseghouani@deloitte.fr Mail: swalpole@deloitte.ch Mail: bbrechot@deloitte.fr

About Deloitte
Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited (“DTTL”), its
global network of member firms, and their related entities. DTTL (also referred
to as “Deloitte Global”) and each of its member firms are legally separate and
independent entities. DTTL does not provide services to clients. Please see www.
deloitte.com/about to learn more. In France, Deloitte SAS is the member firm of
Deloitte Touche Tohmatsu Limited, and professional services are rendered by its
subsidiaries and affiliates.

Deloitte
6, place de la Pyramide – 92908 Paris-La Défense Cedex

© July 2020 Deloitte SAS – Member of Deloitte Touche Tohmatsu Limited

All rights reserved – Studio Design Paris