The value relevance of contingency reserves and catastrophe reserves included in insurance liabilities: Evidence from Japanese insurance companies

The value relevance of contingency reserves and catastrophe reserves included in insurance liabilities:

Evidence from Japanese insurance companies

Fuyu Kawasaki

Abstract

This paper investigates the value relevance of contingency reserves and catastrophe reserves. Japanese insurance companies report contingency reserves and catastrophe reserves as insurance liabilities (policy reserves) in the liabilities section of the balance sheet. Contingency reserves and catastrophe reserves are margins whose essential feature is like equity (net assets). So, I posit that contingency reserves and catastrophe reserves have positive impacts on stock prices.

Because of the small sample, I conduct correlation analyses by non-parametric tests. The results for all years show that contingency reserves and catastrophe reserves are positively and significantly associated with stock prices, consistent with my hypothesis. The results for each year are positively and significantly associated with stock prices, except for two years.

To conclude, I predict that contingency reserves and catastrophe reserves have

positive impacts on stock prices and conduct non-parametric tests. Most of the results

show that contingency reserves and catastrophe reserves are positively and

significantly correlated with stock prices, consistent with my hypothesis. The results

give an implication on the distinction between liabilities and equity to accounting

standard setters.

1. Introduction

The most important financial statement item for insurance companies is insurance liabilities which reflect unearned revenues and are presented in the liabilities section of the balance sheet. Insurance liabilities are basically measured as premiums from policyholders and consist of two elements: Expected payments to policyholders and margins which are expected profits to shareholders. Contingency reserves and catastrophe reserves are margins included in insurance liabilities.

This research investigates accounting for insurance liabilities, especially focusing on contingency reserves and catastrophe reserves. Although contingency reserves and catastrophe reserves are reported as insurance liabilities in the liabilities section of the balance sheet, these reserves are margins whose essential feature is like equity (net assets). So, I predict that contingency reserves and catastrophe reserves positively impact stock prices. This research is based on Japanese insurance companies listed on the Tokyo stock exchange.

This research is unique because Japanese insurance regulation requires Japanese insurance companies to report contingency reserves and catastrophe reserves as insurance liabilities (policy reserves), while International Financial Reporting Standard 4, Insurance Contracts published by International Accounting Standards Board (IASB) does not permit insurers to recognise catastrophe provisions as liabilities.

In section 2, I explain theories and prior research related to insurance accounting.

Then, I present my hypothesis. Section 3 describes the research methods. Section 4 represents the results of the analyses. The paper concludes in section 5.

2. Theories, prior research and hypothesis

In this section, first I describe two theories and prior empirical research related to

insurance accounting. Then, I present my hypothesis.

2.1 Theories

Prior studies on insurance accounting are based on two theories: Efficient market theory (hypothesis) and agency theory.

Fama (1970) develops efficient market theory (hypothesis), describing three efficient market forms, that is, weak, semi-strong, and strong forms. First, in markets under the weak form, prices reflect historical prices. Second, in markets under the semi-strong form, prices reflect publicly available information. Third, in markets under the strong form, prices reflect any information including private information.

Prior accounting studies are based on the semi-strong market form. Publicly available accounting information is one of main influences on stock market prices. Therefore, prior studies investigate the relationship between accounting information and stock prices (e.g. Khurana and Kim, 2003; Horton, 2007; Oliveira, Rodrigues and Craig, 2010; Ota, 2010). These studies are called value relevance studies. Kothari (2001) reviews capital markets research in accounting, including value relevance studies.

Another theory in prior studies is Jensen and Meckling’s (1976) agency theory.

Agency theory regards a firm as a nexus for contracting relationships between principals and agents. In firms, agents usually mean managers. Under agency theory, managers are assumed to be self-interested and to behave opportunistically. Principals such as shareholders are not sure whether managers work hard or not. So, indicators which show that managers execute the contract well or not are needed. Accounting information provides such indicators, specially earnings, which are a main indicator.

Consequently, managers might try to manage earnings, motivated by their own personal interests. Thus, prior research conducts earnings management studies (c.f.

Watts and Zimmerman, 1990; Healy and Wahlen, 1999).

2.2 Prior research

There are three features in the prior literature on insurance accounting.

First, studies on life insurers concentrate on the value relevance of accounting

information. Many life insurers disclose Embedded Value (EV). EV computes the

current value of insurance contracts to supplement the book value of insurance contracts. Prior studies find that EV has value relevance (e.g. Horton, 2007; Wu and Hsu, 2011; El-Gazzar, Jacob and McGregor, 2015; Tsukahara, Nakamura and Nishiyama, 2018).

Second, studies on general insurers tend to focus on earnings management using loss reserves. Loss reserves show the estimated unpaid claims already incurred at the balance sheet date. If loss reserves increase/decrease, earnings decrease/increase. Prior studies find that general insurers try to manage earnings by using loss reserves (e.g.

Petroni, 1992; Grace and Leverty, 2012).

The third feature of prior empirical research is that the prior studies do not deal with both types of insurer.

Prior studies on insurance accounting tend to separate general insurers from life insurers. However, I believe that some topics are common to both types of insurer. Both insurers receive premiums from policyholders in advance.

In addition, reporting unearned revenues as insurance liabilities is common to both types of insurer.

Bauman (2000, 2005) examines the relationship between stock prices and unearned revenues of firms in the US publishing industry and finds that unearned revenues have a positive impact on stock prices. Bauman (2000, 16; 2005, 58) concludes that stock prices behave as if unearned revenue liabilities represent an economic asset. He conjectures that his study would be applicable to other industries which report unearned revenues. His study is likely to be applicable to the insurance industry, where insurance companies report insurance liabilities and insurance liabilities are regarded as unearned revenues.

Thus, I investigate accounting for insurance liabilities which insurance companies

report. Especially, this research focuses on contingency reserves and catastrophe

reserves reported as insurance liabilities by Japanese insurance companies listed on the

Tokyo stock exchange.

2.3 Hypothesis

As I mentioned in section 1, insurance liabilities consist of expected payments to policyholders and margins which are expected profits to shareholders. Margins include contingency reserves and catastrophe reserves. Although contingency reserves and catastrophe reserves are presented as insurance liabilities in the liabilities section of the balance sheet, contingency reserves and catastrophe reserves are margins whose feature is like equity (net assets). So, I posit that contingency reserves and catastrophe reserves have positive impacts on stock prices. I present a hypothesis as below:

H1: Contingency reserves and catastrophe reserves included in insurance liabilities are positively related to stock prices.

IFRS 4 prohibits insurers from recognising catastrophe provisions as a liability because catastrophe provisions are not present obligations and are regarded as equity rather than a liability (IASB, 2004, paras. 14a, BC89a, b). On the other hand, Japanese insurance regulation requires Japanese insurance companies to report contingency reserves and catastrophe reserves as insurance liabilities (policy reserves) in the liabilities section of the balance sheet (MOF, 1996, articles 69-1-3, 70-1-2, 70-1-2-2).

This research uses accounting information provided by Japanese insurance companies and identifies how stock prices react to contingency reserves and catastrophe reserves, that is, whether investors regard contingency reserves and catastrophe reserves as liabilities or equity (net assets).

3. Research methods

3.1 Non-parametric tests

Statistical tests are categorised into parametric statistics and non-parametric

statistics (Smith, 2015, 74). As presented in section 3.2, the research has a small

sample, so I use non-parametric statistics which do not assume the normality of the

sample. In order to examine the relationship between stock prices and contingency

reserves and catastrophe reserves, I conduct correlation tests, including the Spearman and Kendall tau tests, similar to Horton (2007, 189).

3.2 Sample

I focus on Japanese listed insurance companies on the Tokyo stock exchange, including both general insurers and life insurers. There are 11 Japanese companies which sell their own insurance products, listed on the market.

Table 1 describes the sample for H1. The period is from 2012 to 2019 and 70 observations from 11 insurance companies are included. For all companies except Japan post, Ipet, and SBI, I collect their financial data from the fiscal year 2012 when they started to disclose contingency reserves and catastrophe reserves on a consolidated basis for calculating their consolidated solvency margin ratio. For Japan Post, Ipet, and SBI, I collect their financial data beginning when they were listed. Lifenet and Ipet do not have their subsidiaries, so I use their financial data on a non-consolidated basis.

Table 1 Sample selection

No Company Observation Year Listing date Data Source

1 Lifenet 8 2012-2019 15/03/2012 Annual securities reports 2 Japan post 4 2016-2019 04/11/2015 Annual securities reports 3 Ipet 1 2019 25/04/2018 Annual securities report 4 SBI

insurance 1 2019 27/09/2018 Annual report 5 Sompo 8 2012-2019 01/04/2010 https://www.sompo-

hd.com/ir/data/solvency/

6 Anicom 8 2012-2019 03/03/2010 Annual securities reports 7 MS&AD 8 2012-2019 01/04/2008 Annual reports (2012-2014)

Integrated reports (2015-2019) 8 Sony

financial 8 2012-2019 11/10/2007 Annual reports 9 Dai-ichi 8 2012-2019 01/04/2010 Annual securities reports 10 Tokyo

marine 8 2012-2019 01/04/2002 Annual reports (2012-2014) Integrated reports (2015-2019) 11 T&D 8 2012-2019 01/04/2004 Annual securities reports Total 70

All fiscal years are April to March. Commonly, Japanese companies select their fiscal years

which begin in April and end in March.

3.3 Dependent variable

The dependent variable is stock price (SP). I collect stock prices of the Japanese insurance companies at the last business day of each fiscal year-end from Bloomberg.

3.4 Independent variable

The independent variable is contingency reserve and catastrophe reserve per share (CCR). Contingency reserve and catastrophe reserve per share (CCR) are calculated by dividing the sum of contingency reserves and catastrophe reserves by shares outstanding at the last business day of each fiscal year-end. Data sources of contingency reserves and catastrophe reserves are shown in Table 1, section 3.2.

Shares outstanding at each fiscal year-end are collected from Bloomberg.

4. Results

This section represents the results of the analyses, using R.

4.1 Descriptive statistics

Table 2 shows descriptive statistics of the variables for H1. Stock price (SP) ranges from ¥368 to ¥5,362 with a median value of ¥1,939. Contingency reserve and catastrophe reserve per share (CCR) range from ¥17 to ¥3,958.

Contingency reserve and catastrophe reserve per share (CCR) have a median value of ¥523. In addition, I calculate that insurance liability per share has a median value of ¥16,800, indicating that contingency reserves and catastrophe reserves are about 3% of insurance liabilities.

Table 2 Descriptive statistics

(yen)

Variable N Mean Median SD Min Max

SP 70 2,172 1,939 1,218 368 5,362

CCR 70 823 523 915 17 3,958

SP is stock price, CCR is contingency reserve and catastrophe reserve per share.

4.2 Non-parametric tests

Panel A and Panel B of Table 3 show Spearman correlation coefficients and Kendall correlation coefficients for H1 respectively. The coefficients of 0.726 and 0.540 for all years 2012-2019 under the Spearman and Kendall correlation tests respectively are positive and significant at the 1% level. The results are consistent with H1 and indicate that contingency reserves and catastrophe reserves are positively correlated with stock prices.

Although the significance levels range from 1% to 10%, the coefficients for each year under the Spearman and Kendall correlation tests are positive and significant, except for the years 2012 and 2018. For the year 2012, the coefficient of 0.666 under the Spearman correlation test is positive and significant at the 10% level, but the coefficient of 0.5 under the Kendall correlation test is positive but not significant. For the year 2018, the coefficients of 0.466 and 0.388 under the Spearman and Kendall correlation tests respectively are positive but not significant.

Table 3 Spearman/Kendall correlations Panel A Spearman correlations: Coefficients between SP and CCR

2012 (N=8)

SP

2013(N=8) SP

2014(N=8) SP

2015(N=8) SP

2016(N=9) SP

CCR 0.666* 0.857** 0.904*** 0.761** 0.633*

2017(N=9) SP

2018(N=9) SP

2019(N=11) SP

Total (N=70) SP

CCR 0.783** 0.466 0.636** 0.726 ***

Panel B Kendall correlations: Coefficients between SP and CCR 2012(N=8)

SP

2013(N=8) SP

2014(N=8) SP

2015(N=8) SP

2016(N=9) SP

CCR 0.5 0.714** 0.785*** 0.571* 0.5*

2017(N=9) SP

2018(N=9) SP

2019(N=11) SP

Total (N=70) SP

CCR 0.611** 0.388 0.454* 0.540***

About the variables, see Table 2. *, ** and *** indicate p < 0.1, p < 0.05 and p < 0.01 respectively.

5. Conclusions

In this paper, I investigate the value relevance of contingency reserves and catastrophe reserves. Japanese insurance companies report contingency reserves and catastrophe reserves as insurance liabilities in the liabilities section of the balance sheet. Contingency reserves and catastrophe reserves are margins whose essential feature is like equity (net assets). So, I posit that contingency reserves and catastrophe reserves have positive impacts on stock prices.

I use non-parametric tests, that is, the Spearman and Kendall correlation tests because of the small sample. The results for all years 2012-2019 show that contingency reserves and catastrophe reserves are positively and significantly associated with stock prices, consistent with my hypothesis. The results for each year are positively and significantly associated with stock prices, except for the years 2012 and 2018.

To conclude, I predict that contingency reserves and catastrophe reserves positively impact stock prices and conduct non-parametric tests. Most of the results show that contingency reserves and catastrophe reserves are positively and significantly correlated with stock prices, consistent with my hypothesis, indicating that investors regard contingency reserves and catastrophe reserves as equity rather than liabilities. The results give an implication on the distinction between liabilities and equity to accounting standard setters.

There are some limitations to this research. First, this paper conducts non-parametric tests because of the small sample. In the future, the sample will be large, then parametric tests can be considered. I will adopt price models and return models for regression analyses (c.f. Kothari and Zimmerman, 1995; Ohlson, 1995;

Horton, 2007; Ota, 2010). Second, this paper focuses on contingency reserves and

catastrophe reserves which are margins included in insurance liabilities, but there are

other components of margins, such as excess of premium reserves. So, as a next step, I

will consider other components of margins. These are for future research.

Acknowledgements

This research was financially supported by Takachiho University. The author would like to thank Professor Niamh Brennan, Dr. John McCallig, Dr. Sean Power, Professor Eamonn Walsh, Dr. Makoto Tsukahara, and Associate Professor Takefumi Ueno for their helpful advice. The author is responsible for any errors.

Note

1 I am writing a literature review paper on empirical insurance accounting (Kawasaki, 2020).

In the paper, I describe three reasons why prior studies do not deal with both types of insurer; Feasibility of researching, different features between general insurance and life insurance, and path dependence.

References

Bauman, M. P. 2000. The unearned revenue liability and firm value: Evidence from the publishing industry, Available at SSRN:

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=238628: 1-31.

Bauman, M. P. 2005. A market-based examination of revenue and liability recognition: Evidence from the publishing industry, Review of Accounting and Finance 4(3): 52-63.

El-Gazzar, S. M., R. A. Jacob and S. P. McGregor. 2015. The relative and incremental valuation effects of embedded value disclosure by life insurers: Evidence from cross-listed firms in the U.S. Accounting Horizons 29(2): 327-339.

Fama, E. F. 1970. Efficient capital markets: A review of theory and empirical work. Journal of Finance 25(2): 383-417.

Grace, M. F. and J. T. Leverty. 2012. Property-liability insurer reserve error: Motive, manipulation, or mistake. Journal of Risk and Insurance 79(2): 351-380.

Healy, P. M. and J. M. Wahlen. 1999. A review of the earnings management literature and its implications for standard setting. Accounting Horizon 13(4): 365-383.

Horton, J. 2007. The value relevance of ‘realistic reporting’: Evidence from UK insurers.

Accounting and Business Research 37(3): 175-197.

International Accounting Standards Board (IASB). 2004. Insurance Contracts. International Financial Reporting Standard 4. London: IASB.

Jensen, M. C. and W. H. Meckling. 1976. Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics 3(4): 305-360.

Kawasaki, F. 2020. Empirical insurance accounting: A literature review. Working Paper: 1-26.

Khurana, I. K. and M.-S. Kim. 2003. Relative value relevance of historical cost vs. fair value:

Evidence from bank holding companies. Journal of Accounting and Public Policy 22(1): 19-42.

Kothari, S. P. 2001. Capital markets research in accounting. Journal of Accounting and Economics 31(1-3): 105-231.

Kothari, S. P. and J. L. Zimmerman. 1995. Price and return models. Journal of Accounting and Economics 20(2): 155-192.

Ministry of Finance (MOF). 1996. Order No.5 of 1996. Regulation for Enforcement of the Insurance Business Act. Tokyo: MOF.

Ohlson, J. A. 1995. Earnings, book values, and dividends in equity valuation. Contemporary Accounting Research 11(2): 661-687.

Oliveira, L., L. L. Rodrigues and R. Craig. 2010. Intangible assets and value relevance: Evidence from the Portuguese stock exchange. British Accounting Review 42(4): 241-252.

Ota, K. 2010. The value relevance of management forecasts and their impact on analysts’

forecasts: Empirical evidence from Japan. Abacus 46(1): 28-59.

Petroni, K. R. 1992. Optimistic reporting in the property-casualty insurance industry. Journal of Accounting and Economics 15(4): 485-508.

Smith, M. 2015. Research Methods in Accounting, Third edition. London: SAGE Publications Ltd.

Tsukahara, M., R. Nakamura, and K. Nishiyama. 2018. Value relevance of embedded value disclosure: Evidence from Japan (in Japanese). Accounting 194(5): 75-89.

Watts, R. L. and J. L. Zimmerman. 1990. Positive accounting theory: A ten year perspective.

Accounting Review 65(1): 131-156.

Wu, R. C.-F. and A. W.-H. Hsu. 2011. Value relevance of embedded value and IFRS 4 Insurance

Contracts. The Geneva Papers on Risk and Insurance-Issues and Practice 36(2): 283-303.