Evaluating total operational value and associated risks of financial holding companies in Taiwan

Yugoslav Journal of Operations Research Volume 20 (2010), Number 2, 276-292 DOI:10.2298/YJOR1002275C EVALUATING TOTAL OPERATIONAL VALUE AND ASSOCIATED RISKS OF FINANCIAL HOLDING COMPANIES IN TAIWAN Li-Hui CHEN Associate Professor of Departments of Business Administration, College of Management and Director of Accounting Office in Shu-Te University. Taiwan, R.O.C. Received: August 2004 / Accepted: May 2010 Abstract: This study comprises several different parts. The first part applies a normal benchmark valuation model established by Penman to assess the potential whole operational values of FHCs. The second part applies the concept of measuring financial risk as earnings variance to establish a financial risk measurement model. This model can be used to examine the degrees of financial risk before and after FHC’s establishment, and to distinguish different combinations of FHC based on risk diversion efficiency. The final part of this research constructs a new value-risk relation model that can be applied to cross-analysis for measuring total operation value of FHCs with different degrees of financial risk. Through completion of the above steps this study will demonstrate what combination of FHC offers the co-benefits of risk diversion and high whole operational value. Keywords: Financial holding companies, whole operational value, financial risk, value at risk, value-risk relation model. AMS Subject Classification: 91B30, 91B28. 1. INTRODUCTION An enterprise can use external growth strategies, including mergers, joint ventures, and the establishment of holding companies, to achieve objectives of expansion; increasing market share in the short-term, adjusting production scope to optimize economies of scale and operating multiple businesses to disperse operational risk during expansion [6]. The economic advantages of mergers are not obvious [31], L.,H., Chen / Evaluating Total Operational Value and Associated Risks 276 owing to risks involving site, production, personnel and executive decision-making [33]. In the U.S. financial markets, the deposit-loan ratio of commercial banks is restricted; only investment banks can underwrite securities; insurance companies can only supply insurance products and bank branch establishment is subject to geographical limitations [16]. These restrictions prevent financial institutions from offering timely and extensive services, restricting financial development [20]. Accordingly, financial institutions have established banking holding companies (BHC), which cannot disperse systematic risk [17]. To exploit economic synergies [4], financial institutions have devised the concept of integrating firms that operate in various financial industries as financial holding companies (FHC). The establishment of FHCs is a new trend in financial development [25]. Recently, the Japanese government lifted a ban on the establishment of holding companies. The U.S. government permitted financial institutions to found FHCs, and thus boosting financial system development. Financial firms can gain various advantages by transforming themselves into FHCs, including being able to supply customers with various products, enjoying economies of scale [14]; reducing overall enterprise risk and increasing total FHC operating value, and supporting risk management [7]. The Taiwanese government implemented the Law governing Financial Holding Companies in June, 2001, which allowed financial institutions to transform themselves into FHCs to increase their competitiveness. FHCs have the following characteristics. FHCs comprise a group of financial firms that combine into a single firm and offer distinct financial services. FHCs satisfy customers by offering one-stop shopping services, and thus resemble a financial supermarket [8]. FHCs have sufficient capital to support financial product research and organizational expansion, and integrate financial institutions to increase market share; promote sales of financial products, and boost earnings. FHCs also offer the advantages of a BHC [14, 25]. An insurance company that has been reorganized into an FHC can reduce its risk, and increase the operation value of its constituent parts [7]. Additionally, financial companies that restructure themselves into FHCs exploit preferential tax rules or offset profits against losses to reduce payable tax for the FHC to a value lower than the total sub payable by the constituent firms. Enterprises in an FHC provide mutual support with financing, and prepare consolidated financial statements to raise their capital adequacy ratios [11], thus increasing operation value [27]. Restated, financial institutions that reorganize themselves into FHCs have increased operational value [2], and consequently should enjoy increased stock value. However, the reality is that stock values generally reduce following FHC formation. Furthermore, the financial risk faced by an FHC increases with changes in operating procedures [28]. The persistence of improved business performance for FHCs is uncertain. This study thus examines operational instability and the increased financial risk of FHCs, and measures their overall operational value and financial risks. This work examines Taiwan based FHCs, examines overall enterprise operational value and financial risk in the year following establishment. Analysis is conducted to determine the structure of the relationships between the overall operating value and financial risk. This study has five main research objectives: first, to modify the business whole benchmarking value measurement model developed by Penman into a new FHC whole value measurement model adapted to evaluate the overall objective value of an FHC; second, to obtain the earnings, book value and market value of every FHC in Taiwan and then apply the whole value measurement model for FHCs estimate L.,H., Chen / Evaluating Total Operational Value and Associated Risks 277 the overall operating value for different combinations, and then compare it with that for the corresponding FHCs. The effect of FHC establishment on overall benchmarking value is also determined, to clarify whether FHC establishment creates financial synergies; third, to assess Z-scores for FHCs and their individual constituent enterprises to determine the efficiency of financial risk dispersal; fourth, to integrate the VaR (value at risk) and Z-score models into a single financial crisis measurement model, called a VaRZ model. The model is used to measure VaR for bankrupt FHCs; the final objective is to perform cross analysis to determine what combination of firms in an FHC minimizes financial risk while increasing overall operational value. The results of this study can help FHC managers not only to understand overall operating value but also perceive financial risk and thus adopt hedging strategies to mitigate extreme financial crises. Additionally, this study provides a good reference for investors in FHCs making risk- hedging decisions. 2. BIBLIOGRAPHY An FHC is generally formed when firms merge in order to reduce operational financial risk. Numerical methods are available for evaluating firm value. One such method is the Asset-based Valuation Model, which applies the Net Realization Value Method, Modified Book Value Method, Fair Market Value Method, Replacement Cost Method and Strategy Value Method [6]. These methods focus on evaluating assets, but ignore the influence of liabilities and net value on firm value, profitability and future cash flows, leading to the underestimation of firm value. The second method is the Cash Flow Discounted Model, which considers enterprise value to be the total net value of future expected cash flows. This method uses weighted-average cost of capital (WACC) as the discount rate [24]. The advantages of this method include its consideration of the time value of currency, the effect of accounting method on earnings, business and financial risk, perpetual operation value and demand for working capital. However, this method also suffers limitations associated with the difficulty of estimating prospective cash flows and the lack of objective methods for determining the discount period. The third method is the Adjusted Cash Flow Discounted Method, which is based on the Cash Flow Discounted Model, Adjusted Present Value Method and Substance Option Method. The first method simultaneously considers all present values of equity funds and the way in which the finance strategy influences the firm’s present value [17]. The latter method focuses on the values of the firm’s accounts [6]. However, this method suffers the same weaknesses as the Cash Flow Discounted Method. Stockholder equity has an objective market value. Virtually all suppliers and buyers in financial markets thus consider stock values in their investment decision- making [32]; Thomson [30] considered return on equity (ROE), return on assets (ROA) and ratio of capital to assets to measure FHC performance. As ROE or ROA increase, or the capital to assets ratio of an FHC decreases, operational performance increases, as does shareholder wealth. Shareholder benefits can be increased if FHCs effectively use derivatives for hedging. Hedging strategy quality thus influences the wealth of FHC owners [7]. Most studies on this area use a single numerical measurement of business value, for example market value, book value or earnings. These studies assume that these numerical values are unrelated [5], but some researchers see this assumption as false L.,H., Chen / Evaluating Total Operational Value and Associated Risks 278 ([10]; [21]). Penman [21] generated a whole business benchmark value measurement model that not only combined three methods for measuring firm value - discounted dividends, value of growth opportunities and multi-step development - and also modified the traditional appraisal method that simply selected the unit variable used to determine the entity value. Penman claimed that book value and earnings jointly determined the business value. Earnings are obtained from the Income Statement and book value is obtained from the Balance Sheet, each of which reflect firm overall working situation. The book value and earnings, reasonably weighted, correspond to the overall entity value. When business earnings increase or book value decreases, expected ROE increases, and future earnings can be more heavily weighted in the measurement of firm value. Several factors influence business, most notably financial risk [4]. High returns motivate managers to accept high risk [26]. A business that cannot disperse financial risk and avoid bankruptcy has a value of zero [1]. Risk measurement methods include stationary and dynamic measurement methods; the former characterize risk using statistical values such as probability, expected value, variation, bias and peak, while the latter represent expected risks and variations thereof using time series. Such measurement methods focus on business operations variation or uncertainty [22]. Value at risk (VaR) is already an important index for assessing market risk. VaR is the greatest loss that can be borne, within confidence limits, during a particular period in a normal market situation. VaR can estimate risk more immediately and explicitly than can traditional methods. Furthermore, VaR can simply express the maximum loss and associated probability. Methods commonly used to calculate VaR include the variable-covariable method, historical simulation method and Monte-Carlo simulation model [29]. Jackson et al. [13] claimed that the historical simulation method was the best among these because it accurately predicts fluctuations in the data and has an easily understandable numerical model. 3. NUMERICAL MODEL 3.1 Variables and assumptions The notation used for the numerical model in this paper is detailed below. T : the end of the period in this case a season considered in the models. τ :τ th season in the period considered by the models with τ =0,1,2,,T . k : cost of equity capital, plus 1,such that 1+= kρ . T tV : equity value discounted based on perpetual dividends during period t . τ+td ~ : expected reinvestment dividend. τ+tX ~ : expected net earnings less dividends. c tX τ+ ~ : expected earnings, including expected reinvestment dividends. cd tX τ+ ~ : expected earnings combined with expected and reinvested dividend during imP : stock market price of the i th FHC. *iB : per book value of the i th FHC in the whole benchmark value measurement *iX : earnings per share of the i th FHC in the whole benchmark value measurement τjW : actual weight of the j th firm during the model period. L.,H., Chen / Evaluating Total Operational Value and Associated Risks 279 ∗iwˆ : assessed weight of earnings after the establishment of FHC in the whole τφ : discount factor of earnings; that is )1( −ττ ρρ . *ˆ iP : estimated whole benchmark value of the i th FHC in the whole benchmark *ˆ ijp : estimated whole benchmark value of the j th subsidiaries in the i th FHC in ijC : contribution of each of the jth subsidiaries in the i th FHC to whole benchmark iG : accuracy of the whole benchmark value measurement model of the i th FHC. ije : total net value of assets of the j th subsidiaries during the last season before ∗ ipV : whole benchmark value of i th FHC before its establishment in the whole )(tPi : price of the i th firm at time t . )(tPiΔ : vector daily loss or gain in stock prices (change in prices). −HS )(P : q th probable simulated price on the first day in the future period ( )1(iP ). tVaR1 : Value at Risk during period t . 0PV : mixed price of each subsidiary within an FHC. )(qPV : stock price under the q th loss state of FHC. PVΔ : mixed loss or income value of each subsidiary of the FHC. ntVaR : VaR of a FHC with n subsidiaries at period t . nVaR : mean VaR during the change wicket of FHC before its establishment. 1VaR : average VaR after FHC establishment. * VaRP : stock price of FHC, accounting for VaR and *ˆ iP . lr : coefficient of each factor that explains the degree of risk, l =1,2,3,4. EBIT : earnings before tax and interest. TA : total assets. NWC : net working capital, namely current assets minus current liabilities. MVE : market value of equity, namely the product of the common stock market price BVE : book value of equity. TD : total liabilities. ARE : accumulated retained earnings. FON : outstanding quantity of common stock. The following assumptions are made to establish the numerical model and support effective research. 1. The price of equity is the discounted value of the dividend over an infinite horizon, and capital cost and current earnings determine future expected earnings. 2. The earnings and book value reflect all information regarding future earnings and price. 3. k denotes the total interest rate of the three-month Treasury bond and the historical risk premium of 6%. L.,H., Chen / Evaluating Total Operational Value and Associated Risks 280 4. The parameter used in the historical simulation method was %5=α ; the moving wicket was 250 days and the evaluation period was the following day. 3.2 Establishment of the whole benchmark measurement model The business benchmark measurement model developed by Penman [21] is applied initially to establish an FHC whole benchmark value measurement model. This model considers FHC book value and earnings. Based on the business benchmark measurement model, stockholder equity is the present value of future perpetual dividends. Therefore, stockholders equity has price: )~()1( 1 1 ∑−≡ = + − T c t T t T t XEV τ τ ρ (1) and ∑ ∑ ∑ −+≡ = = = + − ++ T T T t T ttt c tt dXEXE 1 1 1 ] ~ )1(~[)~( τ τ τ τ τττ ρ (2) Equation (1) specifies the equity value during period t as the sum of firm discounted expected earnings, based on the equity capital cost during period τ+t . Equation (2) demonstrates that part of the expected value results from the reinvestment of the current dividend, and the expected earnings are the combination of growth of current earnings and returns on reinvested dividends. As T tends to infinity, the equity value approaches the benchmark value, *P , for which the expected earnings equals the sum of expected net earnings and expected returns on the reinvested dividends. Using the book value and earnings to measure the business value is more reasonable than using a single variable, such as book value or earnings. Accordingly, Penman [21] changed Eq. (1) to: )()1( τττττττ φ jjjjjj dXWBWP −+−= (3) Equation (3) uses estimated weights to merge the earnings and book value into an inner value. Based on the hypothesis of the irrelevance of MM dividends (Ross et al, 2002), Eq. (1) is substituted into Eq. (3) to yield the expected whole benchmark value of one subsidiary company of an FHC : )(ˆ)ˆ1(ˆ ***** * iiiii XwBwP φ+−= (4) The weight in Eq. (4) enables the consolidation of earnings and book value to create a benchmark value measurement model for a firm. The weight τjW can be obtained from Eq. (3): )/()( ττττττ φ jjjjj BXBPW −−= (5) According to the theorem of Penman, the weight of FHC earnings can be assessed based on the median weight calculated during every sample period. The weight can be expressed as: ][ˆ * τji WMedianw = (6) L.,H., Chen / Evaluating Total Operational Value and Associated Risks 281 The model accuracy is measured by comparing the expected whole benchmark value from Eq. (4) with the actual market value, namely: %100)/ˆ1( * ×−−= imiimi PPPG (7) Accuracy exceeding 80% indicates acceptable evaluation efficiency [9]. If an FHC has n subsidiaries, then the contribution of the jth subsidiary to the overall operational value of the ith FHC is [24]: ∑= = n j ijijij eeC 1 / (8) Furthermore, the expected whole benchmark value of the FHC is: ∑ ×= = n j ijijip pCV 1 ** ˆ (9) Equation (9) describes the constructed FHC whole benchmark value measurement model, which is based on the concept of whole benchmark value developed by Penman. 3.3 Constructing a model for measuring financial risk 3.3.1 VaR model The historical simulation method for determining VaR considers historical series of stock value data and assumes that stock prices return completely during future evaluation periods. The interval during which the prices are altered is known as the "changing wicket", during which the daily return is the natural logarithm of the closing price for the current day divided by that closing price. The distribution of historical net incomes is also considered for simulating and predicting future returns. Initially, the distribution of stock-price returns for an individual firm is simulated. )(tPi denotes the price of the i th asset in period t , and the stock prices for the last N+1 days are simulated . After deducting the prices on the previous two days, the )(tPiΔ values of asset N are derived as a row of income data. This row represents the N types of income on the following day in the future. During a changing wicket 250 days [12], the stock price )0(iP plus historical income from )1(−Δ iP to )( NPi −Δ describe the N likely income situations on the next day: HS - )()0()( qPPqP iii Δ+= , q =1,2, .,N (10) Equation (10) describes the q th situation where the simulated price is the price on the following day in the future (Historical Simulation of S, HS-S). The distribution of stock returns is determined for every stock price, for various %αp to evaluate the loss of liquidation risk. If α ＝5%, enabling the VaR to be expressed as [12]: )]([%51 qPpVaR it Δ= (11) L.,H., Chen / Evaluating Total Operational Value and Associated Risks 282 Equation (11) can determine financial VaR of an FHC. When various assets and combinations of firms are considered, the historical income of each subsidiary must be simulated separately. The previous simulated income distribution of an FHC is then determined by multiplying the weighted sum of the historical simulated income of each subsidiary by the weights determined based on the net values of the assets of each subsidiary [3]. Consider n subsidiaries of an FHC, where 1C to nC denote the net weightings of the assets of each subsidiary. The price of the FHC, as a combination of its subsidiaries in the qth situation, is then: )0()0()0( 22110 nn PCPCPCPV +++= K (12) At time t=1(situation q ) : ))(()( 11 qPHSCqPV −= + ...))(( 22 +− qPHSC )),(( qPHSC nn −+ Nq ,...,2,1= (13) The stock income vector of FHC is: 0)()( PVqPVqPV −=Δ , Nq ,...,2,1= (14) PV denotes the evaluated stock price and ΔPV represents the income value of the FHC. If ΔPV values are sorted in ascending order, the income distribution of the portfolio on one day can be described. Given α ＝5%, VaR is as follows: )]([%5 qPVpVaRnt Δ= (15) To determine the whole risk of a subsidiary during a specific period, the following equation is used: TVaRVaR T t ntn / 1 ∑= = (16) while the whole VaR of an FHC is: TVaRVaR T t t / 1 11 ∑= = (17) Equations (16) and (17) can evaluate the VaR of any FHC. Finally, the stock price of the FHC can be determined using VaR: )1( 10 * VARPVPVaR −×= (18) Firm value varies inversely with risk [15], even when the risk and value are simultaneously high [23]. The concept of marginal risk probability developed by Luciano et al. [18] is applied in this study to derive the relationship between the business value and risk of an FHC [19]. Expanding Eq. (18) into a Value-risk Relation Model yields: α<≤= )ˆPr()( ** VaRiq PPVF (19) Equation (19) presents the relationship between the value and risk of establishing an FHC. L.,H., Chen / Evaluating Total Operational Value and Associated Risks 283 3.3.2 VaRZ model The VaR model of FHC presented above can measure only the market risk value, but does not indicate whether the FHC can operate perpetually. This study thus combines Eq. (18) with the Z-score model [1] due to establish a new VaRZ model. The model can be used to measure the marginal value of a financial risk crisis, and thus to assess whether an FHC experiencing a financial crisis is an insolvency risk given a certain VaR. The Z-score financial risk crisis model is: TA AREr TD MVEr TA NWCr TA EBITrZ 4321 +++= (20) Z-score<1.23 indicates that the firm has a high probability of insolvency; 1.232.9 indicates the absence of insolvency risk. The VaRZ model below was developed to examine whether an FHC facing financial risk is likely to face a bankruptcy crisis: TA AREr TD FONP r TA NWCr TA EBITrZ VaRVaR 4 * 321 +×++= (21) The VaRZ model can be used to evaluate the liquidation risk. 4. POSITIVE RESEARCH The models developed in Section 3 are used to demonstrate the hypotheses. Six hypotheses are posited. Hypothesis 1 is that the new FHC whole benchmark value measurement model can be used to objectively measure the whole operational value of an FHC; hypothesis 2 is that the establishment of an FHC generates financial synergy; hypothesis 3 is that the establishment of an FHC disperses financial risk; hypothesis 4 is that the new VaRZ model can measure the degree to which an FHC risks insolvency in a financial crisis, given a certain VaR, and hypothesis 5 is that FHCs tend to have high value and low financial risk. Financial data were obtained for 13 FHCs for the period 2002/Q1 to 2003/Q1 were obtained to test these five hypotheses. The FHCs considered all had an operating history exceeding) one year; included HuaNan FHC, FuBang FHC, GuoTai FHC, KaiFa FHC, YuShan FHC, FuHua FHC, JauFeng FHC, TaiShin FHC, ShinGuang FHC, GuoPiau FHC, JianHua FHC, JungShin FHC and JihSheng FHC. 4.1 Positive analysis This section described the content of analysis related to the hypotheses. Table 1: *ˆ iw of each FHC after it is established (1) Measurement of the whole benchmark value and accuracy Equations (5) and (6) evaluated the weight *ˆ iw of earnings of individual FHCs listed in Table 1, to measure the benchmark value of each FHC. Table 2 lists the whole benchmark values *iˆP of the FHCs, calculated from Eq. (4). Comparing all *iˆP values with L.,H., Chen / Evaluating Total Operational Value and Associated Risks 284 the market value imP gives accuracy of 87.38%. Accuracy exceeding 80% indicates that the model has good evaluation efficiency and can be accepted (Deng and Guo, 1996). Table 2: Accuracy degree examination for whole operational values of FHC FHC *iˆP imP %100)/ˆ1( * ×−−= imiimi PPPG HuaNan 20.39 23.61 86.36% KaiFa 18.72 13.89 65.23% GuoPiau 8.29 7.4 87.97% JungShin 27.91 28.97 96.34% FuHua 11.08 10.61 95.57% GuoTai 28.2 39.7 71.03% YuShan 11.81 16.73 70.59% JauFeng 15.91 17.45 91.17% JianHua 14.94 13.83 91.97% JihSheng 7.94 7.87 99.11% FuBang 32.72 27.86 82.56% ShinGuang 10.39 10.19 98.04% TaiShin 18.22 18.22 100.00% Mean of G -- -- 87.38% Unit：Dollar/ per stock) (2) Financial synergy Equation (6) yields the median of the weights, *wˆ =0.18, of each FHC subsidiary over a decade. This median can be considered to assess the whole benchmark value of an FHC before its establishment. The whole benchmark value of the ith FHC before its FHC *ˆ iw HuaNan -0.999 FuBang -1.129 GuoTai -2.717 KaiFa -0.443 YuShan -0.237 FuHua 0.156 JauFeng -0.047 TaiShin -0.364 ShinGuang -0.101 GuoPiau 0.351 JianHua -0.183 JungShin -1.234 JihSheng 0.440 L.,H., Chen / Evaluating Total Operational Value and Associated Risks 285 establishment is given by Eqs. (4) and (8), and is *ipV . The * ipV value of other FHCs are obtained similarly. After the FHC established, * iˆP and * ipV are compared to determine whether financial synergy exists (Table 3). Therefore, the hypothesis :oH 0ˆ ** ≤− ipi VP vs. :1H 0ˆ ** >− ipi VP is rejected because, when 05.0=α , the P-value was smaller, being 0.037. FHC establishment results in favorable increase in whole benchmark value. Table 3: Whole operational values of FHCs before and after their establishment FHC *ipV *iˆP * * *ˆ( ) / 100%i ip ipP V V− × HuaNan 19.13 20.39 6.59% KaiFa 15.82 18.72 18.33% GuoPiau 10.77 8.29 -23.03% JungShin 19.19 27.91 45.44% FuHua 8.93 11.08 24.08% GuoTai 16.59 28.2 69.98% YuShan 13.03 11.81 -9.36% JauFeng 15.29 15.91 4.05% JianHua 12.72 14.94 17.45% JihSheng 12.1 7.94 -34.38% FuBang 17.64 32.72 85.49% ShinGuang 12.69 10.39 -18.12% TaiShin 13.88 18.22 31.27% Mean -- -- 16.75%** (t-value) -- -- (1.707982) ** 5% of Significance level (Unit: Dollar/per stock) (3) Evaluating VaR and Z-score This work applied the Historical Simulation method to estimate VaR, with the estimation period being the following day in future, α of 5% and moving wicket of 250 days. Equations (16) and (17) were used to estimate the VaR, and to estimate the VaR of every FHC. Table 4 listed the detailed outputs. Subsequently, the significant statistical examination method verifies whether the establishment of FHC is associated with risk reduction. According to Table 4, as well as knowledge of the VaR of each FHC before and after establishment, the hypotheses :oH VaRafter - VaRbefore 0≤ v.s :1H VaRafter - VaRbefore >0 are examined and statistical significance tests are performed. The P-value was fond to be 0.0683, less than the level of significance of 10%, and thus hypothesis oH that represented the VaR after FHC established was exceeding before it established, was rejected. Equation (20) assesses the Z-score, considering the weights of the subsidiaries of each FHC, as a measure of the influence of FHCs establishment. Table 5 presents the L.,H., Chen / Evaluating Total Operational Value and Associated Risks 286 calculated difference between the Z-scores of established FHCs and those of the constituent firms prior to FHC establishment. Statistical testing is conducted to determine whether FHC establishment disperses risk. The hypotheses : 0o after beforeH Zscore Zscore− = vs. 1 : 0after beforeH Zscore Zscore− ≠ are tested. From the table, given a significance level of 05.0=α , the P-value was 0.0217 and thus oH was rejected, meaning that establishing FHCs increases financial risk. Table 4: Eestimated VaR of every FHC and their component companies FHC Before established VaR(1) After established VaR(2) After minus Before ((2)- (1)) HuaNan -0.039400 -0.065400 -0.026000 FuBang -0.039500 -0.033400 0.006100 GuoTai -0.033400 -0.036800 -0.003400 KaiFa -0.039100 -0.041200 -0.002100 YuShan -0.029600 -0.043100 -0.013500 FuHua -0.035300 -0.044900 -0.009600 JauFeng -0.041200 -0.049300 -0.008100 TaiShin -0.048800 -0.048900 -0.000100 ShinGuang -0.039100 -0.059100 -0.020000 GuoPiau -0.038700 -0.051300 -0.012600 JianHua -0.045300 -0.037800 0.007500 JungShin -0.038400 -0.033500 0.004900 JihSheng -0.047600 -0.045400 0.002200 Mean -0.039646 -0.045392 -0.00575** (t-value) -- -- (-2.003021) ** 5% of Significance level Table 5: Evaluated Z-scores of FHCs FHC scoreZ − [1] (before established) scoreZ − [2] (after established) [1]-[2] HuaNan 2.3413 1.8987 0.4426 FuBang 9.5931 4.1643 5.4288 GuoTai 6.6144 3.2385 3.3759 KaiFa 27.8838 10.2334 17.6504 YuShan 3.6436 2.5672 1.0764 FuHua 2.1880 2.0892 0.0988 JauFeng 13.5483 2.0448 11.5035 TaiShin 1.7742 1.4397 0.3345 ShinGuang 4.9317 1.8816 3.0501 GuoPiau 32.5322 6.0346 26.4976 JianHua 4.9206 2.2694 2.6512 JungShin 39.5642 2.6133 36.9509 JihSheng 3.4290 1.6869 1.7421 Mean 11.7665 3.2432 8.5233** (t-value) (2.6379) L.,H., Chen / Evaluating Total Operational Value and Associated Risks 287 ** 5% of Significance level (4) Measuring VaRZ This work applied historical simulation to estimate VaR, over an evaluation period of the following day in the future; α was 5% and the change wicket was 250 days. Equation (17) was used to estimate the VaR of each FHC, and Eq. (20) was used to yield the VaRZ numerical model (as shown in Eq. (21)) for measuring financial risk for specific VaR values. Table 6 lists the analytical results, and reveals that FuBang FHC, GuoTai FHC, KaiFa FHC and JungShin FHC did not hit a critical point of bankruptcy during the Asian financial crisis. Meanwhile, the other firms sampled came close to bankruptcy. TaiShin had the highest probability of insolvency, with a VaRZ value of close to 1.23. Table 6: VaRZ values of FHCs *** VaRZ <1.23 FHC that probably was exposed to risk of bankruptcy during financial crisis. ** 1.23< VaRZ <2.9 express uncertainty * VaRZ >2.9 express without bankruptcy crisis (5) Value-risk relationship and cross analysis The whole benchmark and VaR values of FHCs were determined, and Eq. (19), the Value-risk Measurement Model, was applied to determine the FHCs operational values given various risks, as listed in Table 7. Among the FHCs, after FHC establishment increased the value-risk, including FuBang, KaiFa, FuHua, TaiShin, FHC VaRZ NanHua 1.8655** FuBang 4.1082 GuoTai 3.1922 KaiFa 9.8339 YuShan 2.5297** FuHua 2.0482 ** JauFeng 2.0007 ** TaiShin 1.3998 ** ShinGuang 1.8603 ** GuoPiau 5.7622 JianHua 2.2339 ** JungShin 2.5792 ** JihSheng 1.6605 ** L.,H., Chen / Evaluating Total Operational Value and Associated Risks 288 ShinGuang, GuoPiau, JianHua and JihSheng. However, when the whole industry was investigated, there was no clear increase in value-risk following FHC establishment, which meant that FHC establishment could not enhance whole operational value of financial institutions and bring benefits of financial risk dispersion. A schematic curve was then plotted using the two dimensions of VaR, and the whole operational was expressed in Fig. 1. FHC combinations that disperse high risk and boost operational value were thus determined. Table 7: the VaR difference before and after FHCs establishment Company after before difference NanHua 20.39 22.07 -1.68 FuBang 32.72 26.93 5.79 GuoTai 28.20 38.24 -10.04 KaiFa 15.82 13.32 2.50 YuShan 11.81 16.01 -4.20 FuHua 11.08 10.13 0.95 JauFeng 15.91 16.59 -0.68 TaiShin 18.22 17.33 0.89 ShinGuang 10.39 9.59 0.80 GuoPiau 10.77 7.10 3.67 JianHua 14.94 13.31 1.63 JungShin 27.91 28.00 -0.09 JihSheng 7.94 7.51 0.43 Mean -0.001581 t-value -0.001471 ** 5% of Significance level -30 -20 -10 0 10 20 NanHua FuBang GuoTai KaiFa YuShan FuHau JauFeng TaiShin ShinGuang GuoPiau JianHua JungShin JihSheng Value vs. Risk Risk Value Fig. 1: Cross-analysis of values and risks L.,H., Chen / Evaluating Total Operational Value and Associated Risks 289 The values in Fig.1 represent the financial synergy associated with FHC establishment. The figure also shows risk dispersion. A value that exceeds the datum line (equal to zero) reveals financial synergy, and risk exceeding the datum line (equal to zero) is dispersed. Hence, Fig. 1 shows that FuBang FHC, JianHua FHC and JungShin FHC displayed high-risk dispersal and whole operational value. Table 7 also indicates that only FuBang FHC had high risk- dispersal efficiency and significantly increased whole operational value. In contrast, only YuShan FHC exhibited increased risk and reduced whole operational value. 4.2 Positive results The main study results are as follows. (1) This investigation developed a business whole benchmark evaluation model, which can be used to objectively measure the value of a FHC. Hypothesis 1 was supported. (2) The whole benchmark value of the business after FHC establishment was significantly increased compared to previously, implying that FHC establishment brings financial synergy, and hypothesis 2 thus was accepted. (3) FHC establishment significantly increases financial risk. The risks of the JungShin FHC, GuoPiau FHC, KaiFa FHC and JauFeng FHC were increased by more than 10 bases. FHC establishment increased firm financial risk, consistent with theory, and thus FHCs do not enjoy any advantages in terms of dispersed risk during the first year following their foundation. The related hypothesis was thus rejected. (4) The new VaRZ model was applied to objectively measure FHC insolvency risk. Besides FuBang FHC, GuoTai FHC, KaiFa FHC and JauFeng FHC, which have VaRZ values exceeding 2.9, the remaining FHCs faced possible bankruptcy crisis. The FHCs were ranked in order of descending probability of bankruptcy as follows: TaiShin FHC, JihSheng FHC, ShinGuang FHC, HuaNan FHC, JauFeng FHC, FuHua FHC, JianHua FHC, YuShan FHC and JungShin FHC. The VaRZ values can be compared to actual stock prices or earnings, enabling the VaRZ model to be used to measure FHC financial risk. Hypothesis 4 was accepted. (5) The Value-risk Relation Model was applied to perform cross analysis and determine that YSFHC faces high financial risk, with reduced operating value. Hypothesis 5 was firmly rejected. 5. CONCLUSION The establishment of FHCs has become popular in recent years. However, whether establishing FHCs creates value remains controversial. At the end of 2003/Q1, 13 FHCs were established and operated for over one year in Taiwan. This work focuses on designing a FHC whole benchmark value measurement model to assess FHC business value, and determine whether establishing an FHC generates financial synergy. A new financial crisis risk model is also designed to estimate FHC risk of bankruptcy at a particular VaR. The main contributions of research are the development of the FHC L.,H., Chen / Evaluating Total Operational Value and Associated Risks 290 Benchmark Value Measurement Model, the Value-risk Relation Model and the VaRZ Model, and the analysis demonstrated that FHC benefit from clear financial synergies. However, some FHCs do not benefit from dispersed financial risk, but instead suffer increased financial risk. The VaRZ Model can be used to evaluate the degree to which each FHC faces a financial risk associated crisis, and some FHCs have VaRZ values exceeding 2.9, while for others the values are below 1.23, indicating that none of the FHCs became insolvent. When the relationship between the values and risks of FHCs is considered, FHCs that combine banking, insurance and security firms (such as FuBang FHC) enjoy greater risk dispersal and high overall operational value. Notably, FHCs that comprise only a bank and a security firm (like YuShan FHC) face increased risk and reduced operating value. The results presented in this study can not only help FHC managers to understand the overall operational value of their enterprises, to help control business performance, but also to understand the risks they face. The findings also provide a useful reference for investors when evaluating an FHC or making decisions regarding the selection of FHC stocks as investment targets. REFERENCES [1] Altman, I.E., Corporate Financial Distress and Bankruptcy, Wiley, New York, 1993. [2] Anne, B., and David, G.H., “Intra-group, interstate strategic income management for tax, financial reporting, and regulatory purpose”, The Accounting Review, 76 (2001) 515-536. [3] Beder, T. S., “VaR: seductive but dangerous”, Financial Analysts Journal, 51 (1995) 12-24. [4] Boyd, J.H., “Bank holding company merger with non-bank financial firm: effect on the risk of failure”, Journal of Banking and Finance, 17 (1993) 43-63. [5] Burgstahler, D., and Dichev, I., “Earnings, adaption and equity value”, Accounting Review, 72 (1997) 187-215. [6] Copeland, T., Koller, T., and Murrin, J., Valuation: Measuring and Managing the Value of Companies, Wiley, New York, 1994. [7] Cummins, J.D., Phillips, R.D., and Smith, S.D., “Derivatives and corporate risk management: participation and volume decisions in the insurance industry”, Journal of Risk and Insurance, 68 (2001) 51-91. [8] Demsetz, R.S., and Strahan, P.E., “Historical patterns and recent changes in the relationship between bank holding company size and risk”, Economic Policy Review, 1 (1995) 13-26. [9] Deng, J.L., and Guo, H., The Principle and Application of Grey Forecast Theory, Chuan Wa, Taipei, 1996. [10] Feltham, G.A., and Ohlson, J.A., “Valuation and clean surplus accounting for operating and financial activities”, Contemporary Accounting Research, 11 (1995) 689-731. [11] Gaughan, P., Mergers, Acquisitions and Corporate Restructurings, Wiley, New York, 1999. [12] Hendricks, D., “Evaluation of value-at-risk model using historical data”, Economic Policy Review, Federal Reserve Bank of New York, 2 (1996) 39-69. [13] Jackson, P., Maude, D.J., and Perraudin, W., “Bank capital and value at risk”, The Journal of Derivatives, 4 (1997) 73-89. [14] Jerry, L.J., “Effective supervision and the evolving financial services industry”, Federal Reserve Bank of Cleveland. Economic Commentary, June (2001) 1-6. [15] Jianmin, J., and James, S.D., “A standard measure of risk and risk-value models”, Management Science, 42 (1996) 1691-1705. L.,H., Chen / Evaluating Total Operational Value and Associated Risks 291 [16] Kroszner, R.S., and Rajan, R.G., “Is the Glass-steagall Act justified? A study of the U.S. experience with universal banking before 1933”, The American Economic Review, 84 (1994) 810-832. [17] Laderman, E.S., The Potential Diversification and Failure Reduction Benefits of Bank Expansion into Non-Banking Activities, FRB of San Francisco Working Papers, 2000. [18] Luciano, E., Peccati, L., and Cifarelli, D. M., “VaR as a risk measure for multiperiod static inventory models”, International Journal of Production Economics, 81-82 (2003) 375-384. [19] Mark, L.B., David, M.L., and Matthew, G., Intermediate statistical methods and applications a computer package approach, Prentice-Hall, Englewood Cliffs, 1983. [20] Mester, L.J., “Repealing Glass-steagall: the past points the way to the future, business review”, Business Review-Federal Reserve Bank of Philadelphia, (1996) 3-18. [21] Penman, S.H., “Combining earnings and book value in equity valuation”, Contemporary Accounting Research, 15 (1998) 291-324. [22] Philippe, J., Value at risk-the new benchmark for managing financial risk, McGraw-Hill, Singapore, 2000. [23] Rakesh, K.S., and Martin, W., “Risk-value models”, European Journal of Operational Research, 70 (1993) 135-149. [24] Ross, A.S., Randolph, W.W., and Jeffrey, F.J., Corporate Finance, 6th, McGraw-Hill, New York, 2002. [25] Santomero, A.M., “The causes and effects of financial modernization”, Business Review- Federal Reserve Bank of Philadelphia, (2001) 1-4. [26] Stephen, A.R., Randolph, W.W., and Jeffrey, F.J., Corporate finance, McGraw-Hill, Singapore, 1999. [27] Steven, A.H., “Toiling in king Solomon’s mine: a study in business valuation for transfer tax purposes (part II)”, Valuation of Business Interests, Taxes/ November (1998) 13-29. [28] Sudip, D., Mai, I.D., and Kartik, R., “Executive compensation and corporate acquisition decisions”, The Journal of Finance, 6 (2002) 2299-2336. [29] Thomas J.L., and Neil, D.P., “Value at risk”, Financial Analysts Journal, 56 (2000) 47-67. [30] Thomson, J.B., “Unitary thrifts: a performance analysis”, Economic Review- Federal Reserve Bank of Cleveland, 37 (2002) 2-14. [31] Walsh, J.P., and Ellwood, J.W., “Mergers, acquisitions, and the pruning of managerial deadwood”, Strategic Management Journal, 12 (1991) 201-217. [32] West, H., “Hostile bidders, long-term performance, and restructuring methods: evidence from the UK”, American Business Review, 20 (2002) 71-81. [33] Westergren, B., “Risk management: a neglected aspect of M&A planning”, M&A Europe, Mar/Apr (1990) 47-52.