期末報告
負債導向基金之動態資產配置：下檔風險限制與違約選擇權
計 畫 類 別 ： 個別型計畫 計 畫 編 號 ： MOST 103-2410-H-004-092-執 行 期 間 ： 103年08月01日至104年08月31日 執 行 單 位 ： 國立政治大學風險管理與保險學系 計 畫 主 持 人 ： 張士傑 報 告 附 件 ： 出席國際會議研究心得報告及發表論文 處 理 方 式 ： 1.公開資訊：本計畫涉及專利或其他智慧財產權，2年後可公開查詢 2.「本研究」是否已有嚴重損及公共利益之發現：否 3.「本報告」是否建議提供政府單位施政參考：否中 華 民 國 104 年 10 月 28 日
之破產賠付金額表示，有別於以往文獻利用財務指標作為違約考量 因素，本研究考量加入風險導向資本監理條件，分析安定基金所承 受風險與違約風險間之關係。
加入Yang et al. (2012)與Hwang et al. (2015)假設，考量資本市 場系統性風險下機構投資人資產配置與違約價值關聯性，本研究建 立資本市場之資產收益模型，分析下檔風險與違約價值關連性。嘗 試依台灣人壽保險市場之公司資產負債資訊建立模型，依隱含選擇 權模型分析投資風險與違約成本於基金資產配置效果與投資之影響 。 中 文 關 鍵 詞 ： 公平保費、跳躍過程、隨機波動、監理寬容
英 文 摘 要 ： In this paper the risk-weighted sliding scale of policy reserves as a basis for the contribution to the Taiwan Insurance Guaranty Fund (TIGF) is evaluated. Through Monte Carlo simulations, a detailed cash flow of an insurer’s asset allocation can depict the risk preference of the life insurer. We consider the jump diffusion process and
stochastic volatility in our stock model to reflect the increasing volatility that a life insurer encounters in the capital market.
We also introduce regulatory forbearance from the regulator in Yang et al. (2012) and Hwang et al. (2015) as an
external factor and its effects on the life insurance
industry. We find that as the supervisor extends the period of regulatory forbearance, the contribution premium towards the TIGF increases. Whereas the supervisor raises the
regulatory criteria, the contribution premium rises as the regulatory criteria reaches a certain level; as the life insurer increases its leverage ratio, its contribution premium also increases.
英 文 關 鍵 詞 ： fair premium, jump diffusion, stochastic volatility, regulatory forbearance.
科技部補助專題研究計畫成果報告
（期末報告）
負債導向基金之動態資產配置:下檔風險限制與違約選擇權
計畫類別：個別型計畫
計畫編號：MOST 103-2410-H-004-092
執行期間： 2014/08/01 ~ 2015/8/31
執行機構及系所：國立政治大學風險管理與保險學系
計畫主持人：張士傑
計畫參與人員：宣葳、林佳儀、鐘昀珊
本計畫除繳交成果報告外，另含下列出國報告，共 __1_ 份：
出席國際學術會議心得報告
期末報告處理方式：
1. 公開方式：
非列管計畫亦不具下列情形，立即公開查詢
2.「本研究」是否已有嚴重損及公共利益之發現：否
3.「本報告」是否建議提供政府單位施政參考 否
中 華 民 國 104 年 10 月 31 日
摘要 本研究探討人壽保險安定基金的風險保 費，考量人壽保險公司之資產負債組 合，如何影響公司之違約風險，違約風 險則以基金所承受之破產賠付金額表 示，有別於以往文獻利用財務指標作為 違約考量因素，本研究考量加入風險導 向資本監理條件，分析安定基金所承受 風險與違約風險間之關係。
加入 Yang et al. (2012)與 Hwang et al. (2015)假設，考量資本市場系統性風險下 機構投資人資產配置與違約價值關聯 性，本研究建立資本市場之資產收益模 型，分析下檔風險與違約價值關連性。 嘗試依台灣人壽保險市場之公司資產負 債資訊建立模型，依隱含選擇權模型分 析投資風險與違約成本於基金資產配置 效果與投資之影響。 關鍵字：公平保費、跳躍過程、隨機波 動、監理寬容 Abstract
In this paper the risk-weighted sliding scale of policy reserves as a basis for the contribution to the Taiwan Insurance Guaranty Fund (TIGF) is evaluated. Through Monte Carlo simulations, a detailed cash flow of an insurer’s asset allocation can depict the risk preference of the life insurer. We consider the jump diffusion process and stochastic volatility in our stock model to reflect the
increasing volatility that a life insurer encounters in the capital market.
We also introduce regulatory forbearance from the regulator in Yang et al. (2012) and Hwang et al. (2015) as an external factor and its effects on the life insurance industry. We find that as the supervisor extends the period of regulatory forbearance, the contribution premium towards the TIGF increases. Whereas the supervisor raises the regulatory criteria, the contribution premium rises as the regulatory criteria reaches a certain level; as the life insurer increases its leverage ratio, its contribution premium also increases.
Keywords: fair premium, jump diffusion,
stochastic volatility, regulatory
forbearance.
一 、 計 畫 緣 由
The purpose of the insurance guaranty fund is to provide capital injection or advancement upon an insurer becoming insolvent, in order to ensure market and sector stability such that it would not result in systematic risk cascading through the industry affecting business continuity of other insurers. The fund not only provides protection to policyholders but also plays an important role in the exit mechanism of the insurance industry. Separate life and non-life insurance guaranty funds are set up under the insurance guaranty umbrella fund. Life insurers in Taiwan have significant offshore holdings in risky assets. The global financial crisis in 2008 has caused significant loss in value to various types of financial products held by the insurers. As at 2008, data from the Taiwan Insurance Institute (TII) has shown life insurers had lost NT$ 144.6 billion in securities. This not only eroded the life insurer’s balance
sheet, but also caused 11 life insurers’ capital to Risk-Based Capital (RBC) ratio to fall below the statutory capital requirement of 200%, and thus the need to increase their capital for the continuation of business. Subsequent to the data published by the TII, the Financial Supervisory Commission (FSC) had announced regulatory forbearance as a temporary relief plan in light of the difficulties the life insurers are facing with respect to raising capital and meeting the statutory capital requirement under global economic recession.
The trend in net liability of life insurers from 2002 to 2012 suggests that there is an upward trend in written premiums, meaning there are also corresponding liabilities. However, the growth in owner’s equities is less than its total liabilities, resulting in the financial leverage ratio (total liabilities/owner’s equity) to be as high as over 25. To illustrate the significance of these numbers, the average leverage ratio between 2002 and 2007 was 20, where it more than doubled in 2008 and was
as high as 30 at the end of 20111
. This suggests that cash inflows are significantly higher than its capital holdings, management is also at risk of moral hazard where they may lean towards maximizing shareholder value by employing higher risk strategies at the expense of the interest of the insured.
The current rate of contribution towards the insurance guaranty fund is based on the provisions as set out by the Ministry of Finance. As of 1 July 2014, the contribution rate towards the non-life insurance and life insurance guaranty fund is based on a
percentage of the total premium received2. This
percentage is determined according to two main criteria, namely the Capital Adequacy
1, 3 Source: Indicators of Insurance publication, Taiwan Insurance Institute. www.tii.org.tw, 2014. 2 Source: Taiwan Insurance Guaranty Fund.
www.tigf.org.tw, 2014.
Ratio (CAR) and its Management Performance Indicators (MPI) ratings; the CAR is divided into five tiers and the MPI ratings are further subdivided into four categories and eight indicators. These four categories are, risk management, financial structure, type of operations, and legal compliance.
This revised contribution rate has already doubled the average contribution rate from 0.1% to 0.2%. Although it is a step towards minimizing the gap in the TIGF, the fund itself has a balance of NT$ 19 billion as at the end
of 20123. However, after the takeover of
insolvent life insurers by the FSC in 2013, the fund had paid out over NT$16 billion from the Life Insurance Guaranty Fund. Thus whether the current contribution regime towards the TIGF passes the reasonability check and that regulatory forbearance measures are fair and proportionate is the focus of discussion.
The fund adopts a pre-assessment approach, where the basis of the contribution towards the TIGF is based on total annual premium received. The purpose of this study is to determine the fair premium of such contribution. Our opinion is that it seems more reasonable to base this premium on the liability reserves rather than actual premiums received (written), due to the liability reserving is based on those policies that are currently in-force, which provides a more holistic view of the business and not merely on recurring premiums or new businesses written. By utilizing liability reserves as a basis of calculation, it is closer in form in the pricing of risk-based premium and a true representation of the size of a life insurer’s liabilities if it does become insolvent. Therefore, this study is modeled after the provisions of the German insurance stabilization fund and that the contribution towards the fund is based on liability reserves and not total premiums
received.
Lastly, we look at the how one would determine the fair premium under regulatory forbearance, where it takes the risk appetite of the individual life insurer into account such that insurers who take on higher risks will be penalized. The purpose of this study is to propose a methodology to determine a fair premium contribution towards the TIGF where the fund’s existence is not only to provide and protect the interest of the policyholders but also to provide stability in the insurance industry.
二. 計劃目的
In this paper we extend the single-period pricing model of Cummins (1988) to a multi-period pricing model and with reference to Duan and Yu (2005), we also consider factors such as interest rate uncertainty and the regulatory constraint of risk-based capital (RBC) requirements. Furthermore, the jump in asset prices is also incorporated into our pricing model.
Both Cummins (1988), Duan and Yu (2005) used jump diffusion process in their liability model to illustrate the impact of catastrophe risk, however, this jump process was not mirrored to its asset model where there would also be significant movements in the capital
market. Unfortunately, these modelling
assumptions are closely related to the property and liability insurance sector. We rely on the insights from Kou (2002), Heston (1993) and Chang (1999) to express the impact to asset values and the sustainability of life insurers under extreme market movements, together with Duan and Yu (2005) we establish a more complete asset and liability pricing model better suited to life insurers.
With reference to Merton (1977) and his subsequent publications, we assume the
insurance guaranty fund in this paper has the characteristics of a put option and the use of Parisian options in pension liability as put forward by Broeders and Chen (2010). We also look at the following literary studies from Merton (1978) and Pennachhi (1987a) on the cost and pricing of deposit insurance, Duan and Yu (2005), Broeders and Chen (2010), Yang et al. (2012) and Hwang et al. (2015) for the premium pricing of insurance guaranty fund in the presence of regulatory provisions.
Past studies of Cummins (1988) and Duan and Yu (2005) focuses on the allocation of asset and liability of property and liability insurance, and its pricing model is constructed under the deposit insurance framework suited for property and liability insurance. Thus, in this study we attempt to look at these structures under the life insurance framework where we incorporate the jump diffusion process and stochastic volatility into our asset model to express the effects of capital market movements to a life insurer’s asset base. In the case when a life insurer becomes insolvent and requires financial assistance from the insurance guaranty fund, the cleanup process becomes much more complex than the deposit insurance mechanism of property and liability insurance. Furthermore, this paper also utilizes the Monte Carlo methodology to analyze the effect of regulatory forbearance measures employed by the regulators and a life insurer ’ s risk strategy have on the premium pricing of the insurance guaranty fund.
Policies underwritten by life insurers are often of a long-term nature, and it is the life insurer ’ s obligation to manage its assets adequately to ensure its ability to meet future policyholder claims. It is not uncommon for life insurers to hold certain positions in fixed-income type assets and other risky assets on their balance sheet. We thus use the following three asset classes, bonds, stocks, cash, and net cash flows of the life insurer in constructing a
generic asset portfolio for the purpose of this study.
The asset model
We have adopted the Cox, Ingersoll, and Ross (abbrev. CIR) model (1985) as our interest rate framework. The interest rate behavior implied by the CIR model has the following properties: i) given the square root process, negative interest rates are precluded. ii) If the interest rate reaches zero, it can subsequently become positive. iii) The absolute variance of the interest rate increases when the interest rate itself increases. iv) There is a steady state of distribution for the interest rate.
The interest rate dynamics of the CIR process reads as follows:
,
(1)
where is the long-run mean of the interest rate; the interest rate volatility; is a positive constant measuring the mean reverting intensity,
while denotes a Wiener process. Under
risk neutral probability measure, we can convert equation (1) and express it as follows:
,
(2)
, ,
.
For is a constant and denotes the market price of the interest rate risk as stated under
the CIR model assumption. is the Wiener
process as defined under Q-measure.
The bond price at time t with
maturity T in the CIR model is
, is a constant and refers to the market price of risk where
.
The differential form is,
, (3)
We re-write equation (3) under risk neutral probability measures:
,
(4)
.
Stochastic processes has been extensively used in the modeling of financial products. Many assume the stock price follows a diffusion process, meaning that the change in the stock price is continuous and non-discrete, thus there will not exist a jump or discontinuity in the process. However, the above hypothesis had been subject to many a challenge post the 2008 financial crisis and the 2011 sovereign debt crisis, as these events had caused significant price movements in the various markets.
We have considered the following two cases to depict the effect of asset movements to a life insurer ’ s balance sheet position, by introducing the jump term to the dynamic stochastic process of the asset to illustrate systemic risk; and that the rate λ in the Poisson process is an independent stochastic function, and not a constant. Meaning outside news, whether good or bad arrive according to a Poisson process, and that the asset price changes in response according to the jump size distribution.
Due to the leptokurtic, skewness and fat tail characteristics of asset returns, one cannot capture these phenomena under the standard normal distribution assumptions. Asset prices
are often influenced by external factors such as
financial and political developments,
government policies, resulting in instantaneous jump and discontinuity in asset prices. Normal stock prices follow a geometric Brownian motion which is unable to explain this instantaneous jump and discontinuity, Merton (1976) had put forward a theory of where underlying stock returns are discontinuous; thus by introducing an independent Poisson process into the geometric Brownian motion, it produces a jump-diffusion process to describe the jump and discontinuity in the returns of the underlying stock.
However, normal distribution and Brownian motion does not address the leptokurtic and asymmetric features of asset returns. Kou (2002) proposed that the asset price is assumed to follow a Brownian motion plus a compound Poisson process with jump sizes double exponentially distributed to incorporate both these features in the modelling of the asset price. Thus, the dynamics of the asset price can be described as follows:
, (5)
where denotes the risk premium on the
underlying asset; is a Poisson process with
rate and is assumed to be independent with other variables; is a series of independent identically distributed (i.i.d) non-negative random variables, representing the percentage change in asset price. It also has an asymmetric double exponential distribution with density function
,
, .
where p and q denote the probabilities of
upward and downward jumps; is the
upward jump percentage and ; the
downward jump percentage and positive
. and are independent
Weiner processes with the correlation coefficient and can be expressed as follows:
Under the risk neutral probability measure, (5) can be written as:
, (6)
,
.
With stochastic volatility process
Systematic risk is measured as the volatility in the capital market. This type of risk is both unpredictable and is not possible to avoid. It cannot be mitigated through diversification. Systematic risk is vulnerability to events that affect aggregate outcomes such as market returns. It arises from market structure that produce uncertainty faced by agents in the market. Such uncertainties might arise from international economic factors.
We assume the risky asset’s variance to be stochastic in this section instead of a deterministic variance as assumed in the previous section. Here we adopt the Heston (1993) model to describe the asset pricing dynamics:
,
, and are
independent,
where denotes the risk premium on the
underlying asset; is the variance of the
underlying asset price, which is a random variable; is the mean reversion speed; is the long-run mean of the asset price variation; describes the variation of . Under risk-neutral probability measure, (7) can be rewritten as:
, (8)
From the above, the asset allocation in
bonds, stocks and cash, with weightings ,
, respectively can be
written as:
,
(9)
The term is the expected net cash
flow to assets ratio at time t, while is
the volatility of NCIF at time t.
The liability model
Due to the long-term nature of life insurance policies, interest rate movements plays an important role in the management of a life insurer’s liabilities. With reference to Duan and Yu (2005), is the present value of all
future claims, thus can be specified as
follows:
, (10)
The term is a constant measuring the net
increase in the underwriting activities; is the
risk premium of the liabilities; is the
elasticity of liabilities to interest rate, meaning the ratio of the percentage change in liabilities to the percentage change in interest rate.
Fair premium of the insurance
guaranty fund
We can see the insurance guaranty fund as a put option on a life insurer’s assets. We set the periodic audit times as t = 0, 1, … , where t is denoted in years, to be in line with the annual financial reports; and the put option maturity can be expressed as,
.
Under current legislation, regulatory capital to risk-weighted asset ratio (RBC) must exceed a certain ratio and there is a regulatory forbearance period before intervention by regulators, where the life insurer must raise sufficient capital before becoming insolvent. We define the following:
The term is defined as the time closest to
time where the asset to liability ratio reaches
the barrier that warrants
regulatory intervention; and that is the first
occurrence of when asset value breaches
exceeding the regulatory forbearance period . Lastly, we define as the guaranty period used to determine the premium contribution towards the insurance guaranty fund, such that , as illustrated in the following example.
The cost of the insurance guaranty fund can be expressed in terms of a put option
payoff:
In our study, we argue that using the technical provisions as a basis of the annual contribution towards the insurance guaranty fund is sounder than the current practice of total premium received per financial year.
denote the fair premium rate per
period over an intended n-period (auditing
period) of guaranty coverage. The fair premium rate can be viewed as a risk-adjusted quantity
that equates the present value of -period
insurance levies with the present value of the total guaranty coverage. Specifically, the fair premium rate is the solution to the following system:
, (13)
where the right-hand side of Eq. (13) is the present value of the liability facing the guaranty fund. The left-hand side is the present value of insurance premium payments. This expression is consistent with the fact that the insurance premium payment is stopped when the insurer is found, upon audit, to be insolvent, and the fact that the premium payments are made at the beginning of every period when coverage is in effect. In other words, the premium payments are made on a pre-assessment basis.
Simulation
From Eq. (13), the fair premium rate can be expressed as,
.
(14)
From Eq. (14), the denominator is the cost
to the insurance guaranty fund should an insolvency event occurs and the numerator is the total contribution into the insurance guaranty fund based on technical provisions. Both the denominator and the numerator’s expected value can be computed by Monte Carlo simulations. The above pricing model is achieved by solving the stochastic differential equations (2), (4), (6), (8) and (9).
We set the auditing period between time
and to be a year, and further subdivide
this one-year period to 250 business days. The simulations are thus based on daily observations. Equations (2) and (4) can be written as follows:
, (15)
, (16)
Eq. (16) can be described as the 10-year rolling bond, and the stock price can then be expressed as,
,
(17)
. (18)
Finally, we can express the asset and liability processes as follows,
,
(19)
, (20)
Thus, through the above equations, the fair
premium rate can be computed using 50
000 sample paths.
三. 計劃成果自評
In this study we assumed an insurance guaranty fund coverage period to be ten years, and as the regulatory vigilance factor. We take a further look at the following scenarios
where i) the interest rate elasticity is set at
-3 and 0, ii) the correlation effect of initial funding level (A/L) and the regulatory forbearance period has on the fair premium
rate . Lastly, the correlation effect of the
initial funding level and the regulatory vigilance level has on the fair premium rate.
The purpose of this study is using the fair premium rate method as a basis for contributions towards the insurance guaranty fund, to determine its effects on the regulatory forbearance period, and further analyses the effects of a life insurer’s initial funding level and the level of regulatory vigilance has on the determination of the fair premium rate.
Our findings can be summarized as follows:
(1) Fair premium rate increases as one extends the regulatory forbearance period. (2) The fair
premium rate seems to at first decline then increase when the level of regulatory vigilance increases. (3)The fair premium rate declines at an exponential rate as the funding level increases.
Matrix of fair premium rate in
basis points
四、參考文獻
Broeders, D.W.G.A., and A. Chen (2010), Pension Regulation and the Market Value of Pension Liabilities - A Contingent Claims Analysis Using Parisian options, Journal of Banking and Finance 34(6), 1201-1214.
Chang, S .C. (1999), Option Pension Funding
through Dynamic Simulations: the Case of
Taiwan Public Employees Retirement System, Insurance: Mathematics and Economics 24, 187-199.
Cox, J, Ingersoll, J., Ross, S. (1985), A Theory of The Term Structure of Interest Rates, Econometrica 53, 385-407.
Cummins, J.D. (1988), Risk-Based Premiums for Insurance Guaranty Funds. Journal of Finance 43, 593-607.
Duan, J.C. and Yeh, C.Y. (2012), Price and Volatility Dynamics Implied by the VIX Term Structure. Working Paper, National University of Singapore.
Duan, J.C. and Yu, M.T. (2005), Fair Insurance Guaranty Premia in the Presence of Risk-Based Capital Regulations, Stochastic Interest Rate and Catastrophe Risk. Journal of Banking and Finance 29, 2435–2454.
Regulatory forbearance period ( ) 3 months 6 months 9 months
Interest rate elasticity Initial funding level
1.59 20.76 1.59 20.84 1.58 20.77
1.55 20.43 1.56 20.53 1.56 20.57
1.51 20.24 1.52 20.43 1.54 20.51
1.49 19.98 1.52 20.30 1.54 20.37
1.49 19.79 1.51 20.21 1.53 20.34
Initial funding level
0.85 17.92 0.85 17.97 0.85 17.92
0.84 17.70 0.85 17.79 0.84 17.77
0.82 17.41 0.84 17.65 0.85 17.65
0.80 17.16 0.83 17.41 0.84 17.50
0.80 17.12 0.82 17.40 0.84 17.54
Initial funding level
0.46 15.49 0.46 15.52 0.46 15.44
0.46 15.33 0.46 15.39 0.46 15.29
0.45 15.13 0.45 15.25 0.45 15.21
0.45 14.96 0.45 15.13 0.45 15.12
Heston, S. (1993), A closed-form solutions for options with stochastic volatility, Review of Financial Studies, 6, 327–343.
Heath, D., Jarrow, R., and Morton, A. (1992), Bond Pricing and the Term of Interest Rates: A New Methodology for Contingent Claims Valuation. Journal of the Econometric Society Vol. 60, No.1, Jan, 77-105.
Hwang, Y.W., Chang, S.C., and Wu, Y.C. (2015), Forbearance, Ex Ante Life Insurance
Guaranty Schemes, and Interest Rate
Uncertainty, North American Actuarial Journal. 19(2), 94–115.
Kou, S. G. (2002), A Jump – Diffusion Model for Option Pricing. Management Science 48 (8), August, 1086–1101.
Merton, R.C. (1976), Option Prices When Underlying Stock Returns Are Discontinuous. Journal of Financial Economics 3, 125-44. Merton, R.C. (1977). An Analytic Derivation of the Cost of Deposit Insurance and Loan Guarantee. Journal of Banking and Finance 1, 3-11.
Ramezani, C.A., and Zeng, Y. (2006), Maximum Likelihood Estimation of the Double Exponential Jump-Diffusion Process. Annals of Finance, 3, 487-507.
Yang, S.Y., Hwang, Y.W., and Chang, S.C. (2012), Bankruptcy Cost of the Life Insurance Industry under Regulatory Forbearance: An Embedded Option Approach, North American Actuarial Journal, 16(4) 513-523.
科技部補助專題研究計畫出席國際學術會議心
得報告
日期： 104 年 8 月 14 日一. 參加會議經過 世界風險管理與保險經濟會議(WRIEC 2015)與於第十九屆亞太風險與保險學會
年會2015 年 8 月 2 日至 6 日於德國慕尼黑大學(Ludwig Maximilian University of
Munich, also referred to as LMU or the University of Munich, in German)舉
行。此次會議共有超過300 多篇論文發表，與會學者專家分別來自歐洲、亞洲與
美洲各國，以4 天議程，採論文口頭報告的方式，分 9 個平行的議場進行。
此外大會特別於第一天開幕式由德國慕尼黑再保險公司董事長 Nikolaus von
Bomhard 致辭，報告“Current Challenges for the Insurance Industry”，依序的圓
桌座談討論『大數據與保險業發展』，由美國聖約翰大學校長Brandon Sweitzer 計畫編 號 MOST 103－2410 －Ｈ －004－092 － 計畫名 稱 負債導向基金之動態資產配置：下檔風險限制與違約選擇權 出國人 員姓名 張士傑 服務機 構及職 稱 國立政治大學風險管理與 保險學系教授 會議時 間 2015年 8 月 2 日至 2015 年 8 月 6 日 會議地 點 於德國慕尼黑大學 會議名 稱 (中文)世界風險與保險經濟會議
(英文)World Risk and Insurance Economics Congress Munich 2015 發表題
目
(中文) 系統風險下保險安定基金的風險保費
(英文) Fair Insurance Guarantee Premium in the Presence of
主持，參與與談人包含 Tammi Dulberger, Chief Pricing Actuary, Ironshore Hemant, Chief Executive Officer, RMS,
Analytics, XL Catlin。
當日下午的專題演講為 Nikhil Srinivasan, Chief Investment Officer, Generali 的主題為“Managing Insurer Assets through the Euro-Crisis”。
我的論文報告排在議程第2 天第 2 場的系統風險議題，由德國 Goethe University
Frankfurt Helmut Gründl 教授，共同報告人為政治大學風險管理與保險學系宣 葳博士研究生，研究內容有延續先前保險財務風險管理，負債導向基金(如人壽 保險公司或退休基金等機構法人)之策略性資產配置及破產成本（見 Chang and Li 2007, Chang and Hwang 2009, 2010, Chang and Yang 2010, Chang, Tsai and Hwang, 2011, Yang, Hwang 與 Chang 2012, Hwang, Chang and Wu 2015 等），以 隱含選擇權評價方法檢視基金之破產風險，其中違約風險則以基金所承受之可能 破產賠付金額表示，相關研究(見 Cummins 1988, Briys & Varenne 1994, Grosen & Jogensen 2002 等)主要以建立選擇權模型評估基金資產負債表之違約風險，為 實際反映市場之資產負債資訊，本研究考量加入流動性溢酬下之負債公平價值， 分析負債導向基金之資產配置，績效歸因與違約風險間之關聯性。
本研究延伸Hwang, Chang and Wu (2015)1之架構，並假設負債導向基金之資產
價值低於給定比例負債價值時，即進行接管監理程序。研究將建立資本市場之資 產指標收益模型，探討資產配置如何影響基金經理人之績效評估，分析投資風險
1Capital Forbearance, Ex Ante Life Insurance Guaranty Schemes, and Interest Rate
Uncertainty, North American Actuarial Journal, 19(2), 94–115, 2015. (Ya-‐Wen Hwang and Yang-‐Che Wu)
與違約成本關連性。於數值實證分析部分，將嘗試依台灣人壽保險市場之公司資 產負債資訊建立模型，依基金績效歸因與隱含選擇權模型分析投資風險與違約成 本於基金資產配置效果與投資能力之影響。資本市場假設將納入資產隨機波動模 型描述風險資產之市場風險，反映保險公司所持有資產之實際交易風險，用以表 達策略性資產配置下之績效歸因分析與下檔風險測度，並計算風險偏好對於基金 違約風險之影響程度，而為實際反映特許金融事業之資本監理因素，也將道德風 險與監理寬容等因素，於數值計算時納入模型中考量。與會學者除了於會場中發 問外，並私下進行廣泛的意見交談。 本文擔任的論文討論為第三天上午保險業併購場次，論文為 Mergers and
Acquisitions in the U.S. Insurance Industry，作者為韓國國立大學 Bum J. Kim and Sojung C. Park，有機會將對研究論文的未來看法有充分的溝通與討論。
二. 會議內容及心得
此次會議主要由歐洲風險與保險經濟學會(the European Group of Risk and Insurance Economists, EGRIE)與德國慕尼大學所負責主辦，世界三大主要保險 學會(美國、歐洲與泛太平洋風險管理學會)一起舉辦的年會，歐洲風險與保 險經濟學會為區域性的國際組織，此次會議參加者除學會主席、副主席、秘書等 重要學會工作人員，分別代表不同的與會國家，初步將以促進區域了解與推動風 險管理與保險教育與研究為主要宗旨。本次會議分別就子議題進行討論。 本人參與研討會整天的學會會議，積極參與國際事務與學術研究，與會期間參與 保險市場經濟結構，保險計價理論，危險理論，社會保險與退休金計畫，汽車保 險，人壽保險計價及座談討論等項目進行聆聽，收穫相當豐富。
三. 考察參觀活動 本次大會提供參觀德國慕尼大學的機會，優良的教育環境及校園建設與規劃，令 來賓稱讚不已。 四. 建議 風險管理與保險的研究發展在國際間已快速成長，相較於歐美各國於金融保險的 發展，台灣於風險管理與保險領域的研究已有顯著表現，下一年度 2016 年預 計將回歸至 APRIA 的年會，由中國大陸西南財經大學所舉辦。 五. 攜回資料名稱及內容 攜回大會光碟片包含所有論文及相關參考資料。
日期:2015/10/28
科技部補助計畫
計畫名稱: 負債導向基金之動態資產配置：下檔風險限制與違約選擇權 計畫主持人: 張士傑 計畫編號: 103-2410-H-004-092- 學門領域: 財務無研發成果推廣資料
計畫主持人：張士傑 計畫編號： 103-2410-H-004-092-計畫名稱：負債導向基金之動態資產配置：下檔風險限制與違約選擇權 成果項目 量化 單位 備註（質化說明 ：如數個計畫共 同成果、成果列 為該期刊之封面 故事...等） 實際已達成 數（被接受 或已發表） 預期總達成 數（含實際 已達成數） 本計畫實 際貢獻百 分比 國內 論文著作 期刊論文 0 0 100% 篇 研究報告/技術報告 0 0 100% 研討會論文 1 1 100% 專書 0 0 100% 章/本 專利 申請中件數 0 0 100% 件 已獲得件數 0 0 100% 技術移轉 件數 0 0 100% 件 權利金 0 0 100% 千元 參與計畫人力 （本國籍） 碩士生 0 0 100% 人次 博士生 0 0 100% 博士後研究員 0 0 100% 專任助理 0 0 100% 國外 論文著作 期刊論文 0 0 100% 篇 研究報告/技術報告 0 0 100% 研討會論文 1 1 100% 專書 0 0 100% 章/本 專利 申請中件數 0 0 100% 件 已獲得件數 0 0 100% 技術移轉 件數 0 0 100% 件 權利金 0 0 100% 千元 參與計畫人力 （外國籍） 碩士生 0 0 100% 人次 博士生 0 0 100% 博士後研究員 0 0 100% 專任助理 0 0 100% 其他成果 （無法以量化表達之 成果如辦理學術活動 、獲得獎項、重要國 際合作、研究成果國 際影響力及其他協助 產業技術發展之具體 效益事項等，請以文 字敘述填列。） 正在撰寫修改中
科 教 處 計 畫 加 填 項 目 測驗工具(含質性與量性) 0 課程/模組 0 電腦及網路系統或工具 0 教材 0 舉辦之活動/競賽 0 研討會/工作坊 0 電子報、網站 0 計畫成果推廣之參與（閱聽）人數 0
