Appendices

20 May, 2024

Appendix I: An overview of Japanese insurers

A closer look at Japanese life and P&C insurers

As mentioned previously, life and P&C insurers hold most Japanese insurance assets. According to 生命保険事業概況 | 生命保険協会 (seiho.or.jp) and 統計|日本損害保険協会 (sonpo.or.jp), among life insurers the dominance is primarily attributed to Nissay, Japan Post, Meiji-Yasuda, Dai-ichi, and Sumitomo. These traditional five companies account for 63% of the total life market. For P&C insurers, dominance among the major players is further accentuated, with Tokio Marine, Mitsui Sumitomo, Sompo Japan, and Aioi Nissay Dowa accounting for 87% of the total P&C market.

Table 1: The top life and P&C insurers in Japan

  Asset size
(in JPY trillions)
Percentage of
total market 
 Top 5 life insurers
1Nissay75.318%
2Japan Post67.216%
3Meiji-Yasuda 43.511%
4Dai-ichi37.59%
5Sumitomo 35.79%
 Total26863%
 Top 4 P&C insurers
1Tokio Marine8.830%
2Mitsui Sumitomo6.924%
3Sompo Japan6.623%
4Aioi Nissay Dowa3.211%
 Total2687%

*Data as of December 2021.
Source: US Financial Stability Board, World Gold Council

Life insurers’ asset management 

Life insurers offer a diverse array of products to cater to customer needs. These include whole life insurance, term insurance, endowment insurance, health insurance and annuity products for both individuals and corporates. 

For life insurers’ general accounts (GAs), premiums are pooled from different types of policies and invested collectively to meet the fixed guaranteed rates of return (crediting rates) to policyholders. The combination of long-term liabilities and guaranteed crediting rates means life insurers face a number of risks, namely market risk, credit risk and liquidity risk. To mitigate these, life insurers construct their investment portfolios through Asset-Liability Management (ALM).

Historically, whole life policies accounted for the majority of life insurers’ value of in-force as they were viewed as a reliable and stable form of coverage. This also aligns well with the Japanese society’s cultural values around financial security. However, since there was limited availability of long-term fixed-income assets at the time, Japanese insurers were often exposed to a large duration mismatch between assets and liabilities. After the collapse of the Japanese bubble economy in the early 1990s, long-term interest rates declined sharply. 

 

Chart 8: The Japanese 10-year bond yield plunged after 1990

The Japanese 10-year bond yield plunged after 1990

The Japanese 10-year bond yield plunged after 1990
*Data as of December 2023. Source: Bloomberg, World Gold Council

Sources: Bloomberg, World Gold Council; Disclaimer

*Data as of December 2023.

Due to the large duration mismatch, this sharp decline in interest rates led to a huge burden for Japanese insurers. They struggled to meet their obligations to policyholders, which led to the insolvency of seven small- and medium-sized life insurers between April 1997 and March 2001. These events acted as a catalyst for the remaining life insurers to place greater emphasis on Liability-Driven Investments (LDIs) in order to mitigate as much risk as possible.

For instance, they increased their allocation to longer-term fixed-income assets with extended maturities, mostly 30-year and 40-year Japanese government bonds (JGBs). This highlights the active efforts by insurers to reduce the duration mismatch.

Japan’s ultra-low interest rate environment, which has now persisted for many years, has led to an increasing need for life insurers to invest in alternative assets for yield-enhancement. Such needs have seen insurers turn to global private equity funds and hedge funds, as well as increase their allocations to foreign stocks and other alternative assets. 

 

Chart 9: Life insurers’ allocations to long-term assets have increased sharply over the past 20 years*

Top 4 life insurers’ aggregated asset maturities

2022

Life insurers’ allocations to long-term assets have increased sharply over the past 20 years*

Life insurers’ allocations to long-term assets have increased sharply over the past 20 years*
Top 4 life insurers’ aggregated asset maturities
*Data as of December 2022. Source: Insurers’ disclosures, World Gold Council
 

2003

Life insurers’ allocations to long-term assets have increased sharply over the past 20 years*

Life insurers’ allocations to long-term assets have increased sharply over the past 20 years*
Top 4 life insurers’ aggregated asset maturities
*Data as of December 2022. Source: Insurers’ disclosures, World Gold Council

Sources: Insurers’ disclosures, World Gold Council; Disclaimer

*Data as of December 2022.

P&C insurers’ asset management 

P&C insurers exhibit distinct liability characteristics compared with life insurers. P&C insurers, whose main products are auto, fire, marine and cargo, accident and other casualty insurance products, have much shorter liability durations as their coverage usually spans only one year before policy renewal. Consequently, P&C insurers consider their primary risk to be the potential liquidity risk stemming from unforeseen catastrophic accidents.

P&C insurers allocate a meaningful share of around 23% to domestic bonds, including JGBs, highlighting this need for liquidity. While P&C insurers, similar to life insurers, invest in safe JGBs and high investment grade corporate bonds from an ALM perspective, they have much greater room and risk appetite for yield-enhancing assets. This is due to the fact that most P&C policies are short term without any guaranteed crediting rates. Thus, interest-surplus losses arising from interest rate declines in the 1990s had a lower impact on them compared with life insurers. With this historical perspective in mind, Japanese P&C insurers are in a relatively better position to use their capital to invest in higher yield assets with lower liquidity, such as private equity funds.

Appendix II: The liquidity of gold

A closer look at the liquidity of gold 

As mentioned previously, the gold market is large, global, and highly liquid. Its daily trading volume is higher than many other major financial markets and assets (Chart 10). 

 

Chart 10: The trading volume of gold is higher than many other major financial markets and assets

Average daily trading volume, trillion yen*

The trading volume of gold is higher than many other major financial markets and assets

The trading volume of gold is higher than many other major financial markets and assets
Average daily trading volume, trillion yen*
*Based on estimated average daily trading volumes from 1 January to 31 December 2023. The liquidity of gold includes estimates of OTC transactions and published statistics on future exchanges and gold-backed exchange-traded products. See: Gold Trading Volume. Source: Bloomberg, Bank for International Settlements, FINRA TRACE, International Capital Market Association (ISMA), Nasdaq, World Gold Council

Sources: Bloomberg, Bank for International Settlements, FINRA TRACE, International Capital Market Association (ISMA), Nasdaq, World Gold Council; Disclaimer

*Based on estimated average daily trading volumes from 1 January to 31 December 2023. The liquidity of gold includes estimates of OTC transactions and published statistics on future exchanges and gold-backed exchange-traded products. See: Gold Trading Volume

The scale and depth of the market means that it can comfortably accommodate large, buy-and-hold institutional investors. One market expert suggests that gold should be tradable up to around 25% of the average daily trading volume. For more information, see: Gold Trading Volume | Gold Daily Volume | World Gold Council.

“Typically, it is recommended to remain under 20-25% of the average daily volume to minimise the market impact…

Quote from a gold ETF Manager on the liquidity of gold

Appendix III: The inflation hedging effect of gold

A closer look at the inflation hedging effect of gold 

As mentioned previously, many insurers have argued that assets such as REITs and equities already provide the inflation hedge required. Thus, they do not see the need to invest in gold. To test their argument, we compared the correlations between the CPI change and returns for each asset class as shown in Chart 11, where the two colours indicate periods with a CPI change of greater than 5% y/y (dark blue), and less than or equal to 5% y/y (light blue). 

Chart 11 demonstrates that gold and the US CPI have a historically weak linear relationship. In fact, when calculating R-squared linear regression of the whole data set only 19% of the variation in gold prices since 1971 – when the US Gold Standard ended – can be explained by a linear relationship to changes in CPI inflation. However, it can be inferred from the two plotted regression lines that the correlation between the two variables becomes significantly stronger during periods of extremely high inflation, i.e. when the change in the CPI is higher than 5% year-on-year. The positive correlation of the regression line for periods of high inflation, signified by the dark blue line, indicates that, on average, the higher the CPI level, the higher the nominal return of gold, as long as the inflation rate is greater than 5%. Moreover, in comparison with the data for the FTSE REITs and S&P 500, it is clear that the high performance of gold during periods of extremely high inflation is a unique advantage that is not matched by the other two asset types (Table 2).

Table 2: Summary of the key variables comparing the three asset classes

  Gold FTSE REITsS&P 500
High inflation
(ΔCPI>5%)
Value of R20.3830.0000.012
Gradient of regression  10.258-0.0476-0.736
Real return12.4%-6.5%-4.7%
Low inflation
(ΔCPI≤5%)
Value of R20.0030.0290.039
Gradient of regression  0.7512.6272.733
Real return4.0%4.8%8.2%

Source: Bloomberg, Oliver Wyman, World Gold Council

Chart 11: Gold is an effective hedge against inflation

The relationship of gold, REITs and the S&P 500 with CPI

 

 

Gold:

Chart 11: Gold is an effective hedge against inflation

Chart 11: Gold is an effective hedge against inflation
The relationship of gold, REITs and the S&P 500 with CPI
*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data. Source: Refinitiv, Oliver Wyman, World Gold Council

Sources: Refinitiv, World Gold Council; Disclaimer

*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data.

 

FTSE REITs:

Chart 11: Gold is an effective hedge against inflation

Chart 11: Gold is an effective hedge against inflation
FTSE REITs:
*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data. Source: Refinitiv, Oliver Wyman, World Gold Council

Sources: Refinitiv, World Gold Council; Disclaimer

*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data.  

 

S&P 500:

Chart 11: Gold is an effective hedge against inflation

Chart 11: Gold is an effective hedge against inflation
S&P 500:
*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data. Source: Refinitiv, Oliver Wyman, World Gold Council

Sources: Refinitiv, World Gold Council; Disclaimer

*For Gold and S&P 500 data is from Q1 1971 to Q1 2023, where changes in the gold price and the US CPI are computed from the Gold Bullion LBM $/t oz Delay and the All Urban US CPI, respectively. For FTSE REITs data from the FTSE Nareit All Equity REITs Index from Q1 1973 to Q1 2023 is used due to the lack of other available data.  

Important disclaimers and disclosures [+]Important disclaimers and disclosures [-]