MILLIMAN RESEARCH REPORT
Life insurance capital regimes
in Asia
Comparative analysis and
implications of change
3
rd
edition
Summary report
July 2021
MILLIMAN RESEARCH REPORT
Table of Contents
INTRODUCTION ..................................................................................................................................................... 1
EXECUTIVE SUMMARY ........................................................................................................................................ 2
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 1 July 2021
Introduction
Capital regulations for life insurance companies in Asia are complex and varied. They are also subject to change,
with such changes often affecting how insurers manage their business. In many markets in the region, regulators
are introducing new risk-based capital (RBC) regimes or “upgrading” existing RBC frameworks, with increasing
consideration being given to consistency with the new International Financial Reporting Standard 17 (IFRS 17),
International Capital Standards (ICS) and other capital regimes worldwide.
In view of the pace of change and the increasing focus on regulatory capital across the region, we felt it was
timely to produce an update to the second edition of the report we published in 2020. This “3
rd
edition” report
covers the capital regimes in 13 markets in Asia plus “ICS Version 2.0 for the monitoring period.” The report also
makes reference to Solvency II, Bermuda Solvency Capital Requirements (BSCR), Canada’s Life Insurance
Capital Adequacy Test (LICAT) and the United States’ RBC regime (US RBC).
Our report aims to:
i) Compare and contrast life insurance RBC regimes across selected Asian markets
ii) Highlight some of the potential implications for life insurers arising from the future development of
capital regulations
iii) Contribute to the wider discussion on the potential impact of changes in regulation on the life
insurance industry in Asia
The report seeks to provide a comparison of key quantitative and qualitative aspects of life insurance capital
regimes in Asia and to show analysis of key capital results (e.g., capital ratio, risk charges, factors affecting
capital) based on information publicly available and from other market sources. It does not attempt to provide all
of the applicable details behind the capital regulations governing life insurance companies in the various markets
analysed. It is important to recognise that the regulatory environment in Asia is changing fast and, consequently,
the information contained in this report is time-sensitive. The various capital regimes covered in this report are
based on the applicable regulatory environment as at 31 May 2021. Some of these regulations may have
changed since this date. In addition, some markets have seen temporary changes in 2020 to capital regimes due
to COVID-19, and further changes may be expected in the future. All changes may not be fully captured for all
markets in this report.
We have produced an executive summary of the full report, which we are sharing here.
Please contact one of the Milliman consultants listed at the end of the report to request a copy of the full report or
to discuss the RBC frameworks in any of the markets in more detail.
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 2 July 2021
Executive summary
Comparison of technical specifications of capital regimes
Overview
Most insurance markets in Asia follow some form of RBC regime, although some of them, including Hong Kong,
India and Brunei, are still currently using an EU Solvency I type of approach. In some markets, insurance
regulators are reviewing the existing capital regulations, with Hong Kong, Taiwan, South Korea and Brunei in the
process of developing a new RBC regime, and China and Malaysia looking to “upgrade” their existing RBC
requirements. We have included in this report the latest details of China C-ROSS Phase II, an upgrade from the
existing China C-ROSS (Phase I). The latest information is based on the draft technical specification released by
the China Banking and Insurance Regulatory Commission (CBIRC) in December 2020, and is subject to change.
In Taiwan, the industry is undergoing quantitative impact studies (QIS) for the upcoming new RBC regime,
Taiwan ICS or T-ICS, the details of which are shown in this report. Table 1.1 provides an overview of the current
status of capital regimes for the markets covered in this report.
TABLE 1.1: STATUS OF THE CAPITAL REGIMES
MARKET
INSURANCE
REGULATORY/
GOVERNING BODY
EXISTING
CAPITAL
REGIME /
APPROACH
DEVELOPMENTS
BRUNEI RBCS
Brunei Darussalam
Central Bank (BDCB)
EU Solvency I
Not risk-based
RBC framework is to be incorporated in the near future. The second
parallel run will be conducted for financial year-end 31 December 2020
in Q2 2021 for submission by 30 July 2021.
CHINA C-ROSS
China Banking and
Insurance Regulatory
Commission (CBIRC)
C-ROSS
Risk-based
The CBIRC is currently reviewing C-ROSS formulae and parameters,
and field-testing is currently ongoing. The exact timing of C-ROSS
Phase II remains uncertain, but the final quantitative requirements are
expected to be released later in 2021.
HONG KONG
RBC (RBC 2020)
Hong Kong Insurance
Authority (IA)
EU Solvency I
Not risk-based
Hong Kong is introducing a RBC framework, targeted for tabling to the
Legislative Council in 2022 and to be effective in 2024 (depending on
time spent on legislative process), although there have been
discussions around possible early adoption for some insurers. There
have been three rounds of industry quantitative impact studies (QIS) to
date plus more voluntary studies on different refined approaches. The
latest technical specification (named “RBC 2020”) was released for
companies to perform stress and scenario testing as part of the ORSA
requirements, and further refinements are possible before the
framework is put forward to the Legislative Council.
JAPAN
(REGULATORY)
Financial Services
Agency (FSA)
Risk-based
(US risk-based)
The FSA is contemplating the introduction of an economic value-based
solvency regime. A recent field test was based on the ICS field test,
although the FSA reminded the industry that this should not be
interpreted as a final direction. The exact timing of the introduction of
this new regime remains uncertain.
INDIA
SOLVENCY I
Insurance Regulatory and
Development Authority of
India (IRDAI)
EU Solvency I
Not risk-based
The IRDAI is contemplating the introduction of a RBC regime.
However, the exact framework to be adopted has yet to be defined,
and the timing of implementation remains uncertain.
INDONESIA
RBC
Otoritas Jasa Keuangan
(OJK)
Risk-based
We understand that no material future developments to the current
RBC framework are expected in the near term.
MALAYSIA RBC
Bank Negara Malaysia
(BNM)
Risk-based
BNM has initiated a review of its current RBC framework, conducted in
phases since 2018. The first phase will focus on reviewing the
prudential limits on assets and counterparty exposures, followed by a
review of the standards for the valuation of liabilities and capital
adequacy components. In December 2019, BNM issued an updated
exposure draft of the life insurance liabilities valuation guideline. An
exposure draft for updated RBC may be released later following the
release of the valuation guideline. The exact timing of updated rules
remains uncertain.
In addition, in March 2020, BNM revised the stress parameters for the
computation of interest rate capital charge to reflect prevailing market
conditions.
PHILIPPINES
RBC 2
Insurance Commission
(IC)
Risk-based
We understand that no material future developments to the current
RBC framework are expected in the near term.
SINGAPORE
RBC 2
Monetary Authority of
Singapore (MAS)
Risk-based
MAS is considering the allowance for countercyclical buffers within the
existing RBC2 framework.
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 3 July 2021
MARKET
INSURANCE
REGULATORY/
GOVERNING BODY
EXISTING
CAPITAL
REGIME /
APPROACH
DEVELOPMENTS
SOUTH KOREA
RBC
Financial Supervisory
Service (FSS)
Risk-based
(US risk-based)
The FSS has announced its plan to adopt K-ICS, a principle-based
capital framework, which is similar to ICS. The target effective date is
expected to be the same as the effective date of IFRS 17.
There have been three rounds of quantitative impact studies (QIS) to
date, and further refinements are possible before the framework is put
forward to the legislative council.
SRI LANKA
RBC
Insurance Regulatory
Commission of Sri Lanka
(IRCSL)
Risk-based
There may be some tightening of the capital requirements in the near
future, potentially leading to higher capital charges.
TAIWAN
CURRENT RBC
Financial Supervisory
Commission (FSC)
Risk-based
(US risk-based)
The current RBC approach is based on prescribed risk factors
multiplied by risk exposures. Going forward, Taiwan is set to move to
an ICS-based regime, with the industry currently undergoing
quantitative impact studies (QIS). Taiwan ICS (T-ICS) is scheduled to
come into effect on 1 January 2026.
THAILAND RBC
2 (95
TH
PERCENTILE)
Office of Insurance
Commission (OIC)
Risk-based
The current Thailand RBC 2 framework is based on a 95
th
percentile
confidence level. It is understood that the OIC may plan to introduce a
99.5
th
percentile confidence level framework two years after IFRS 17
applies in Thailand.
A move towards an economic balance sheet framework across the region, but key differences exist
The assessment of required and available capital using an economic balance sheet approach has underpinned
most of the recent changes in Asian capital regulations. A fundamental premise of the economic balance sheet
framework is the endorsement of the concept that assets and liabilities should be valued on a consistent
economic basis, leading to a reduction or elimination, where possible, of accounting mismatches. This economic
balance sheet approach is also consistent with the approach used under Solvency II, ICS and IFRS 17 principles.
In particular, for solvency purposes, an increasing number of Asian capital regimes require companies to:
- Assess their assets on a market value basis (e.g., Hong Kong's proposed RBC framework, Indonesia,
Singapore, Thailand, Malaysia), although some are still measuring their assets using different
accounting bases (e.g., for China’s C-ROSS, Japan’s regulatory capital)
- Value their liabilities using a gross premium valuation (GPV) approach allowing for an additional risk
margin and, potentially, a time value of options and guarantees (TVOG), using a fair value approach
based on “relatively market consistent” discount factors
Although there is a trend towards the use of an economic balance sheet framework, markets are moving at
different paces, and many regulators in Asia seem to have taken a more practical approach that reflects market
specifics, while ensuring a reasonable degree of conservatism (e.g., the flooring of reserves in some markets).
This leads to inconsistencies between RBC regimes across the region. Table 1.2 gives an overview of some of
these differences when assessing liabilities.
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 4 July 2021
TABLE 1.2: APPROACH OF EVALUATING DETERMINISTIC INSURANCE LIABILITIES
CAPITAL REGIME
GENERAL
RISK MARGIN
TVOG
APPROACH
LIABILITY FLOOR
ALLOWED?
APPROACH
ALLOWED?
APPROACH
BRUNEI RBCS
GPV
Reserves floored to zero
at policy level
√
PAD
X
None
CHINA C-ROSS (PHASE
I AND II)
GPV
CSV less capital
requirement
√
PHASE I: PAD
PHSAE II:
Percentile
method
√
Deterministic
only
(b)
HONG KONG RBC (RBC
2020)
GPV
None
√
PAD
√
Stochastic /
Deterministic
JAPAN (REGULATORY)
NPV
Reserves floored to zero
at policy level
X
Considered
implicitly
√
Stochastic /
Deterministic
JAPAN - ICS VERSION
2.0 FOR THE
MONITORING PERIOD
GPV
None
√
PAD
/Percentile
method
√
Stochastic /
Deterministic
INDIA SOLVENCY I
GPV
CSV (if there is a
surrender value) or
reserves floored to zero at
policy level
√
PAD
√
Not explicitly
specified
INDONESIA RBC
GPV
Reserves floored to zero
at policy level
√
PAD
X
N/A
MALAYSIA RBC
GPV
Reserves floored to zero
at fund level
√
PAD
√
Stochastic /
Deterministic
PHILIPPINES RBC 2
GPV
None
√
PAD
X
N/A
SINGAPORE RBC 2
GPV
Reserves floored to zero
at policy level
(a)
√
PAD
X
N/A
SOUTH KOREA RBC
NPV
Reserves floored to zero
at policy level
X
Considered
implicitly
√
Stochastic
SRI LANKA RBC
GPV
No floor for the liability.
However, the sum of
reserves and required
capital should not be less
than the total surrender
value of policies.
√
PAD
√
Stochastic /
Deterministic
TAIWAN CURRENT RBC
NPV
Reserves floored to zero
at product level
X
Considered
implicitly
X
N/A
TAIWAN ICS
GPV
None
√
Percentile
method
√
Stochastic /
Deterministic
THAILAND RBC 2 (95
TH
PERCENTILE)
GPV
Reserves floored to zero
at product group level
√
PAD
X
N/A
SOLVENCY II
GPV
None
√
CoC
√
Stochastic
BERMUDA BSCR
GPV
None
√
CoC
√
Stochastic
CANADA LICAT
GPV
Cap on credit taken for
negative reserves and if
CSV greater than
reserves
√
PAD
X
N/A
US RBC
NPV
Reserves floored to zero
at policy level
X
Considered
implicitly
X
N/A
Notes: GPV = Gross Premium Valuation, NPV = Net Premium Valuation, CSV = Cash Surrender Value, PAD = Provision for Adverse Deviation, CoC = Cost
of Capital
(a) Singapore RBC 2 regime continues to floor policy reserves to zero but recognises negative reserves as an increase to financial resources
N/A: not appropriate
(b) Although C-ROSS Phase II uses deterministic factor approach to TVOG calculation, the factors only depend on the guaranteed interest rate while both
remaining liability duration and guaranteed interest rate are considered in C-ROSS Phase I.
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 5 July 2021
TVOG is a good example of such discrepancies. Universal life products offering guarantees are prevalent in
many markets in Asia including China, Hong Kong and Singapore, but TVOG is only included in the newly
proposed Hong Kong RBC (RBC 2020) and China C-ROSS regimes (Phase I and Phase II). Moreover, under
C-ROSS I and II, TVOG is assessed using a prescribed deterministic formula that applies to the whole industry,
whereas the Hong Kong regulator is encouraging companies to assess TVOG using stochastic asset liability
management (ALM) models to better reflect their own cost of financial options and guarantees. The same
discrepancies in TVOG methodology apply to participating business, which is material in many markets in Asia
(e.g. Hong Kong, Singapore, Malaysia, China, India and Sri Lanka).
The risk margin is another example of discrepancies across RBC regimes in Asia. While provisions for adverse
deviation (PADs) are adopted in most of the capital regimes in the region, the approach to derive the PADs, (and in
particular the underlying risk charges used to calculate the PADs) differs from one market to another. In addition, the
PAD approach (which is determined by recalculating liabilities by including an additional margin on top of the best
estimate assumptions) is not consistent with the cost of capital (CoC) approach used for Solvency II and Bermuda
BSCR. It may also not be in line with the approach adopted by some Asian life insurance companies under IFRS 17
(although some companies may also decide to use a PAD approach) or for economic capital purposes. It is worth
noting that the method of determining risk margin based on a ratio with regard to insurance risk capital will be
considered in the upcoming regimes such as C-ROSS Phase II and Taiwan ICS.
Discount rate: Market consistency and smoothing
Under RBC regimes, the yield curves used to assess the best estimate of liabilities (BEL) are typically defined
using a “bottom-up” approach, whereby the discount rate reflects a market consistent risk-free rate plus an
adjustment for illiquidity and smoothing prescribed by regulators. However, the valuation of liabilities requires the
use of a yield curve that extends to very long durations, reflecting both market conditions and long-term economic
views. This poses a challenge in Asia where available market data is often covering a much shorter duration than
the projected cash flows. The reference yield curve is typically extrapolated from the last liquid market point (LLP)
to some long-term equilibrium rate (ultimate forward rate or UFR). Table 1.3 compares the parameters used by
the various regimes.
TABLE 1.3: DETERMINATION OF THE DISCOUNT CURVE
CAPITAL REGIME
BASIC YIELD
ILLIQUIDITY
PREMIUM
/SMOOTHING
LLP
UFR
INTERPOLATION/
EXTRAPOLATION
BRUNEI RBCS
Government bond
yield curve
(Singapore is used as
a proxy)
N/A
20 years
3.8%
Smith-Wilson
CHINA C-ROSS
(PHASE I AND II)
Government bond
yield
30 / 45 / 70 bps
depending on product
and issue date
20 years
4.5%
Quadratic
HONG KONG RBC
(RBC 2020)
Government bond
yield for USD, swap
for HKD
Matching adjustment
with additional Long-
term Adjustment
(LTA) to equity and
property under
segregated
participating /
universal life
portfolios.
HKD: 15 years
USD: 30 years
HKD: 3.8%
USD: 3.8%
Smith-Wilson
method
JAPAN
(REGULATORY)
Stipulated interest rate for policies issued after March 1996 with some exceptions. Otherwise, the (guaranteed)
interest rates filed with FSA upon product launch.
JAPAN - ICS
VERSION 2.0 FOR
THE MONITORING
PERIOD
Swap rate or
government bond
yield
Prescribed illiquidity
premium (three-
bucket approach)
JPY: 30 years
USD: 30 years
JPY: 3.8%
USD: 3.8%
Smith-Wilson
method
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 6 July 2021
CAPITAL REGIME
BASIC YIELD
ILLIQUIDITY
PREMIUM
/SMOOTHING
LLP
UFR
INTERPOLATION/
EXTRAPOLATION
INDIA SOLVENCY I
Best estimate
investment return (net
of PAD)
N/A, although risk-
adjusted corporate
bond spreads may be
included in the best
estimate investment
return
N/A
N/A
N/A
INDONESIA RBC
Government bond
yield
Averaging of
government bond
yield plus a
discretionary
adjustment of up to
50bps
N/A
N/A
N/A
MALAYSIA RBC
Government bond
yield
N/A, yet volatility
adjustment and
matching adjustment
are introduced in the
latest drat exposure
for liability valuation,
which may be a
change of direction
15 years
Same level as at
LLP
Based on forward
rate
PHILIPPINES RBC 2
Bloomberg PHP
BVAL reference rate
for PHP
Bloomberg
international yield
curve for USD
N/A
N/A
N/A
N/A
SINGAPORE RBC 2
Government bond
yield
Allowance for
illiquidity premium or
matching adjustment
SGD : 20 years
USD: 30 years
SGD : 3.8%
USD: 3.8%
Smith-Wilson
method
SOUTH KOREA RBC
Assumed
(guaranteed) interest
rates filed with FSS at
a product launch.
N/A
N/A
N/A
N/A
SRI LANKA RBC
Government bond
yield curve as
specified by IRCSL
N/A
10 years
Same as the spot
rate at the LLP
N/A
TAIWAN CURRENT
RBC
US government bond
yield
N/A
N/A
N/A
N/A
TAIWAN ICS
Swap rate or
government bond
yield
Prescribed illiquidity
premium (three-
bucket approach)
TWD: 10 years
USD: 30 years
TWD: 4.4%
USD: 3.8%
Smith-Wilson
method
THAILAND RBC 2
(95TH PERCENTILE)
Government bond
yield
Averaging of
government bond
yield
50 years
Same level as at
LLP
N/A
SOLVENCY II
Swap rate or
government bond
yield
Volatility adjustment
or matching
adjustment
Euro: 20 years
USD: 50 years
Euro 3.75% (Dec
2020)
USD: 3.75% (Dec
2020)
Smith-Wilson
method
Given the long-term nature of many life insurance contracts, life insurers typically require long-term assets to
match their liabilities. Where those liabilities are “illiquid,” such that they have relatively predictable cash flow
profiles, insurers can invest in such a manner that recognises that a forced sale of assets, in most cases, would
not be required. The insurers can then potentially benefit from the risk premium that can be available to long-term
investors, typically called an illiquidity premium. Furthermore, insurers are typically not exposed to short-term
fluctuations in the price of assets, albeit the insurer is exposed to changes in the fundamental value of the cash
flows on the assets, for example an increased probability of defaults. Illiquidity premium adjustments and
smoothing adjustments (e.g., volatility adjustment, UFR, averaging of spot yield curve) are, therefore, applied in
the discount rate to reduce the short-term economic balance sheet volatility, stabilise the net asset value (i.e.,
difference between assets and liabilities) and better reflect the long-term nature of insurance businesses, in
particular the illiquid nature of liabilities. RBC capital adequacy ratios (CAR) and the different blocks of the
economic balance sheet are usually sensitive to the discount rate, which is often a key component in different
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 7 July 2021
phases of quantitative impact studies and testing from regulators. Some RBC regimes are currently reviewing the
approach to determine the discount rate in order to more appropriately reflect the asset and liability management
position of insurance companies and dampen the impact of the prevailing low (and potentially volatile) interest
rate environment. With IFRS 17, this topic has also become increasingly important as insurance companies need
to reflect the characteristics of the liability cash flows when setting the IFRS 17 discount rate, and in particular the
level of liquidity.
Capital requirement modules and submodules are broadly consistent across RBC regimes in Asia,
but underlying parameters differ
The exhaustive list of risks considered in determining capital requirements varies across different capital regimes.
However, key material risks considered are typically similar, and include insurance risk, market risk, counterparty
default risk and operational risk.
- Life insurance risks include mortality or longevity risk, morbidity risk, lapse risk (long-term and mass
lapse) and expense risk. Mortality catastrophe risk is also sometimes explicitly considered.
- Market risks typically consist of equity risk, interest rate risk or ALM risk, credit spread risk, property risk
and foreign exchange risk. (Note that equity volatility and interest rate volatility risk are typically not
considered within RBC regimes in Asia.)
- Operational risk is normally quantified by applying risk factors to risk drivers, with premiums being one of
the most common risk drivers.
As there are natural hedges between different risks, correlation matrices are usually considered to reflect
diversification benefits across various risk modules and sub-modules. In particular, most of the RBC regimes in
Asia (and in particular all of the RBC regimes revised recently) consider diversification benefits when aggregating
the sub-modules under the insurance and market risk modules. Some RBC regimes consider diversification
between all risk components other than operational risk, while some others only consider diversification between
asset risk and insurance risk.
There is generally a trend towards making risk charge parameters and stress factors more consistent from one
regime to another, to the extent possible. However, material discrepancies remain, as illustrated by the
comparison of interest rate stress factors for selected markets in Asia in Table 1.4.
TABLE 1.4: KEY PARAMETERS COMPARISON FOR INTEREST RATE FOR SELECTED TERM TO MATURITY, SHOCK DOWN
CAPITAL REGIME
INTEREST RATE/ALM, STRESS-BASED
APPLIES TO INTEREST RATE OR OTHERWISE AS STATED
TERM TO MATURITY
(YEAR)
1
3
5
7
10
15
20
BRUNEI RBCS
-60%
-55%
-55%
-50%
-40%
-30%
-20%
CHINA C-ROSS (PHASE I)
(a)
-73%
-68%
-58%
-50%
-37%
-28%
-24%
HONG KONG RBC (RBC 2020)
-75%
-64%
-61%
-57%
-53%
-49%
-43%
MALAYSIA RBC
(b)
-15%
-15%
-15%
-15%
-15%
-15%
-15%
PHILIPPINES RBC 2
-100%
-59%
-54%
-54%
-54%
-51%
-51%
SINGAPORE RBC 2
-70%
-65%
-60%
-50%
-40%
-30%
-25%
SRI LANKA RBC
-75%
-56%
-46%
-39%
-31%
-27%
-29%
THAILAND RBC 2 (95TH
PERCENTILE)
-40%
-38%
-36%
-34%
-31%
-26%
-21%
SOLVENCY II
-75%
-56%
-46%
-39%
-31%
-27%
-29%
Notes:
(a) China has different shocks for assets and liabilities. The asset shocks are shown in the table. The liability shocks are generally lower.
(b) For Malaysia, the stress is formula-based and depends on the MGS yield. The stress shown above for comparison purposes is applicable
as at end of 2020.
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 8 July 2021
Comparative analysis of key capital results across Asia and impact of new RBC regimes on
life insurance companies
Comparative analysis of capital adequacy ratios across Asia
Figure 1.5 shows the industry average capital adequacy ratios for each market covered in this report, except for
China, Brunei and the Philippines, where there are data limitations. Most of the markets have an average
regulatory solvency ratio within the range of 180% to 400%, except for Japan and Indonesia, which have
relatively higher average solvency ratios above 450%.
FIGURE 1.5: TYPICAL INDUSTRY SOLVENCY RATIO LEVEL
Source: Estimates based on public information and Milliman internal data. Specific companies may have different solvency ratios to the typical industry
solvency ratios in Figure 1.5.
Note 1: The solvency ratios shown above are as at 31 December 2020, based on prevailing RBC regimes in each market except: a) the Japan regulatory
solvency ratio and India Solvency I ratio are as at 31 March 2020 to reflect their financial year-ends; b) Japan 2019 FSA field test results are as at 31 March
2019; c) Sri Lanka and Singapore results are as at 31 December 2019; and d) Hong Kong RBC QIS 3 results are as at 31 December 2018.
Note 2: The latest industry-wide solvency assessment was carried out by the IA via QIS 3 in 2019, and the resulting average industry solvency ratios as at
the end of 2018 were in the range of 100% to 200% based on information gathered from the industry. Similarly, Japan’s FSA carried out an economic
balance sheet RBC field test in 2019, and the resulting average solvency ratios were in the range of 150% to 200%. However, both quantitative impact
studies were conducted using parameters and approaches that are subject to review and further industry consultation. The typical industry solvency ratios
under the final implemented RBC requirements are likely to differ (potentially significantly) from those shown.
Note 3: For Singapore, the ratios shown are based on the RBC 1 regime, as statistics under the new RBC 2 regime are not publicly available
In general, industry-level solvency ratios in Asia have been relatively stable over the past few years, with small
changes driven primarily by changes in the interest rate environment (with government bond yields typically used to
determine the discount rate, as discussed above). Since early 2020, the outbreak of the COVID-19 pandemic has hit
the global economy, with many Asian governments cutting interest rates in order to stimulate economic activity,
with government bond yields falling. The downward pressure on fixed income yields has affected both assets and
liabilities of life insurance companies, leading to a decrease of solvency ratios under an economic balance sheet
framework in most markets across Asia, especially in the first half of 2020.
It is worth noticing that regulators introduced relief measures in several markets in 2020 to help offset some of the
negative impacts of the COVID-19 pandemic. The MAS introduced a transitional measure in Singapore that came
into effect from 31 March 2020 and is planned to be gradually phased out by the end of 2021. The OJK in Indonesia
introduced a new regulation entitled “Countercyclical Policies against the Impact of the Coronavirus Disease 2019
RBC QIS3
2019 FSA
field test
Solvency I
RBC
Takaful
RBC
RBC RBC
RBC
RBC
Regulatory
Regulatory
RBC
0%
50%
100%
150%
200%
250%
300%
350%
400%
450%
500%
Hong Kong Japan India Indonesia Malaysia Singapore South
Korea
Taiwan Thailand Sri Lanka
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 9 July 2021
for Non-banking Financial Services.” effective from April 2020, which has been extended until April 2022. The IC in
the Philippines provided temporary relief to insurers by lowering the minimum CAR while the IA in Hong Kong
relaxed some aspects of the methodology to determine the valuation interest rate in 2020.
As shown in Figure 1.6, for markets with RBC regimes, the total capital requirement tends to be mainly driven by
market risks (i.e., interest rate, equity and credit spread), although lapse risk and morbidity risks are also key
contributors. In some markets such as Japan, currency risk can also be material.
FIGURE 1.6: RISK CHARGE BREAKDOWN – INSURANCE RISK VERSUS OTHER RISKS
Source: Estimates based on public information and Milliman internal data.
Note 1: The figures above are as at 31 December 2019 based on prevailing RBC regimes of each country except: a) Japan 2019 FSA field test result is as at
31 March 2019, and b) The IA carried out QIS 3 for the developing RBC regime in 2019. Since then, there has been no further industry wide assessment for
Hong Kong.
Note 2: For Singapore, the above breakdown is based on Singapore RBC 1 parameters.
The industry-level CARs and the breakdown of risk charges can be explained largely by the nature of assets, the
nature of liabilities and the matching (or lack of matching) of assets and liabilities.
More than half of the life insurance assets across these markets are invested in bonds, with insurers in some
markets investing a high proportion in government bonds (e.g., Thailand), while others are investing higher
proportions in corporate bonds (e.g., Hong Kong) and alternative credit (although this remains small). The
proportion of equities varies by jurisdiction, with markets having a material proportion of participating business
(e.g., Singapore, Malaysia, Hong Kong) typically investing more in equities with less in liquid asset classes (e.g.,
private equity, debt/equity/property funds).
Liabilities also differ significantly from one market to another due to product mix differences. The proportion of
unit-linked business is significant in some markets (e.g., Indonesia, India and Malaysia), while universal life
business has been popular in Hong Kong, Singapore and South Korea. Non-participating traditional business
(e.g., endowments, whole life, credit life, term life) remains a material product category for all the markets studied.
Participating business (e.g., endowments, whole life) is also a popular line of business for some markets across
the region, including Japan, Hong Kong, Singapore, India and Sri Lanka. Unit-linked business and insurance
products with lower investment guarantees and more protection benefits typically look more attractive under an
economic balance sheet framework, whereas savings products with higher investment guarantees (implicit or
explicit) generally look less attractive (the degree of attractiveness being typically measured in terms of new
business margin). As a part of the liability in the economic balance sheet framework, TVOG measures the in-the-
moneyness of the investment guarantees embedded in the products. Table 1.7 provides a high-level overview of
the materiality of TVOG for selected markets.
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Hong Kong RBC QIS 3
Japan 2019 Field Test
Indonesia
Malaysia
Singapore RBC
South Korea
Taiwan
Thailand RBC
Sri Lanka
Insurance risk Other risks
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 10 July 2021
TABLE 1.7: OBSERVATIONS ON TVOG IN SELECTED MARKETS
MARKET
CAPITAL
REGIME
TVOG
CONSIDERED?
MATERIALITY OF TVOG
HONG KONG
Solvency I
(moving to RBC)
√ (under RBC
QIS)
TVOG could be relatively material for participating and universal life products,
two of the main product categories sold in Hong Kong.
INDIA
Solvency I
√
Generally not material as:
The level of guarantees for participating products are typically low and
interest rates are still relatively high. Hence, participating product
guarantees are typically out-of-the-money.
Capital guarantees are not widespread for unit-linked business.
However, for non-linked group funds management business, guarantee costs
may be significant depending on the level of asset/liability duration mismatch.
INDONESIA
RBC
X
Generally not material for multinationals as a high proportion of products sold
by these players are unit-linked without investment guarantees. The
traditional savings products sold by domestic players may have a significant
TVOG.
MALAYSIA
RBC
√
Generally not material as:
TVOG for participating products are currently out-of-the-money.
Other products typically do not have material TVOG.
SINGAPORE
RBC
X
TVOG is not assessed as part of the RBC framework, hence no formal
quantification of TVOG is publicly available.
While TVOG is not expected to be material for most products (as investment
guarantees are generally low and out-of-the-money), it is expected to be
material for some products such as universal life, single premium participating
products and recent tranches of new participating products where
investments guarantees can be high.
TAIWAN
RBC
X (might be
considered
under T-ICS)
TVOG is not assessed as part of the current RBC framework, hence no
formal quantification of TVOG is publicly available.
When moving to T-ICS, TVOG is expected to be material given the nature of
products sold in the market. However, as the industry is currently undergoing
QIS, the exact impact is not known at present.
THAILAND
RBC
X
Generally not material as:
Most products are non-participating in nature.
The participating component is typically not material and does not lead to a
material TVOG.
Unit-linked (without investment guarantee) are also becoming more
material for some companies.
Source: Estimates based on public information and Milliman market intelligence.
The comments regarding the materiality of TVOG in the table above are general comments related to the relevant market in question, based on
our observations. The situation for individual companies within the market may vary.
Potential impact of changes in capital regimes for life insurance business in Asia
A move to a more “economic” RBC regime tends to incentivise life insurers to optimise and potentially de-risk
their balance sheets by shifting more risks to policyholders and third-party asset managers, reducing the level
and cost of guarantees, tailoring existing insurance product features to be more RBC friendly, improving ALM,
and optimising investment and hedging strategies. In particular, the management of RBC balance sheet volatility
becomes increasingly important as a result of:
(i) The typical fair value approach used to value assets and liabilities
(ii) The current more volatile and unpredictable economic environment
These new capital regimes necessitate insurers to use more sophisticated and value-risk based techniques to set
and validate strategic decisions and manage their business.
Strategic planning and risk management. In line with shareholder expectations, many insurers currently
conduct their strategic planning with a key focus on traditional top-line revenue and bottom-line profitability
growth metrics, e.g., annualised premium equivalent (APE) growth, (traditional) embedded value (EV)
growth, value of one year’s new business (VONB) margin or growth using deterministic investment return
assumptions. Under the new RBC regimes (and IFRS 17), these measures would need to be accompanied
MILLIMAN RESEARCH REPORT
Life insurance capital regimes in Asia – 3
rd
edition 11 July 2021
by additional risk-based metrics that clearly identify the trade-off between shareholder value (e.g., measured
in terms of EV or VONB) and risk (e.g., measured in terms of RBC requirements and return on capital).
Strategic planning will not only be a matter of finding the appropriate business strategy to grow revenue and
profitability, but also a matter of optimising the allocation of capital and controlling and reducing risk, via
potentially the definition of a “return on capital” type of metric. For new business in particular, life insurers will
need to find the right balance between maximising top line (by selling products with attractive returns to
customers but with potentially expensive financial options and guarantees) and optimising capital (by selling
products that are more capital-efficient but which may not be so attractive to customers). Ultimately, more
emphasis is likely to be placed on recognising diversification benefits (both product and risk) for a given line
of business.
Capital management, strategic asset allocation and hedging strategy. Changes in capital regulations
will likely prompt insurers to revisit their existing capital management, strategic asset allocation and hedging
programs. In particular,
− Optimising capital requirement and return on capital will become an increasingly key priority.
Management actions will need to be tailored to better reflect management decisions under stress
scenarios that affect the risks faced by the company, and ultimately to make allowance for this within
the assessment of RBC capital.
− Strategic asset allocations will need to be revised, with potentially less focus on levels of asset returns
and more emphasis on risk-based metrics. More dynamic hedging programs may become
increasingly relevant, targeting a certain level of volatility whilst keeping a material exposure to
achieving upside.
− The financing strategy of insurance companies may also be impacted as a result of the introduction of
new definitions of eligible capital, typically grouped into tiers.
MILLIMAN RESEARCH REPORT
Milliman is among the world’s largest providers of actuarial and
related products and services. The firm has consulting practices in
life insurance and financial services, property & casualty insurance,
healthcare, and employee benefits. Founded in 1947, Milliman is an
independent firm with offices in major cities around the globe.
milliman.com
CONTACT
Clement Bonnet
Hong Kong & South East Asia
clement.bonnet@milliman.com
Brian Colgan
Indonesia
brian.colgan@milliman.com
Michael Daly
Hong Kong & South East Asia
michael.daly@milliman.com
Sharon Huang
China
sharon.huang@milliman.com
Farzana Ismail
Malaysia & Brunei
farzana.ismail@milliman.com
Philip Jackson
India & Sri Lanka
philip.jackson@milliman.com
Sung Hoon Kim
South Korea
sung.hoon.ki[email protected]m
David Kong
Singapore & South East Asia
david.kong@milliman.com
Wen Yee Lee
Singapore & South East Asia
wenyee.lee@milliman.com
Atsushi Okawa
Japan
atsushi.okaw[email protected]m
Wing Wong
Taiwan & China
wing.wo[email protected]
© 2021 Milliman, Inc. All Rights Reserved. The materials in this document represent the opinion of the authors and are not representative of the views of Milliman,
Inc. Milliman does not certify the information, nor does it guarantee the accuracy and completeness of such information. Use of such information is voluntary and
should not be relied upon unless an independent review of its accuracy and completeness has been performed. Materials may not be reproduced without the
express consent of Milliman.
