Chapter II Is Asset Allocation a Myth or a Reality? An Empirical Study on Taiwan
2.7 Concluding Remarks
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2.7 Concluding Remarks
There are two opposing views as to whether or not international investment portfolios should enjoy enhanced performance. Taiwan life insurance industry ratio of foreign investment has already exceeded 30%. Therefore, it is worth exploring in depth whether or not the foreign investment strategy of the life insurance industry conveys positive benefits.
This paper aims to examine the relationship between foreign investment and asset performance in Taiwan life insurance industry. 25 Taiwan life insurance companies, between 2004 and 2008, are evaluating with a view towards the home bias puzzle and its financial impact. However, if the observed sample is divided into two categories according to ownership structure, i.e., domestically owned and foreign-owned companies, the results show that domestic companies are more aggressive on foreign investment than are foreign-owned firms, and enjoy more advantages than do foreign-owned firms. The major results and policy implications are as follows:
1. Asset allocation in foreign assets can improve a company's investment performance. The study has also found, in 2008, global financial turmoil had a massively negative impact on the foreign investments of Taiwan life insurance companies. Thus, when considering foreign investment, domestic companies should not only take into account the exchange rate risk, but also pay attention to credit risk management.
2. Asset size and rate of return on investments are negatively correlated, i.e., the larger a company’s assets, the lower its efficiency in the use of funds. As a result, life insurance companies should strive to improve their efficiency in asset management as their market share increases.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Chapter III
Striving for Home Advantages? An Empirical Study of Currency Hedging of Taiwan Life Insurers
3.1 Introduction
Two opposing views debate whether international investment portfolios should have currency hedging strategies. According to the proposition in Modigliani and Miller (1958), hedging strategies in perfect capital markets, i.e., those with no asymmetric information, no taxes, and no transactions (including bankruptcy) costs-- should not affect the value of the firm; the firm value is independent of hedging policy.
Conversely, Kaplanis and Schaefer (1991) proposed that hedging could reduce the volatility of a company's cash flow and currency risk, thus increasing the value of the company. The studies of Glen and Jorion (1993), Froot et al. (1993), Eun and Resnick (1994) and Odier and Solnik (1993) support foreign stock investment hedging. Several empirical papers, including Nance et al. (1983), Allayannis and Weston (2001), Haushalter (2000), Bartram et al. (2004), and Ben Khediri and Folus (2009a, b) report results supporting the value-enhancing hypothesis. But Karim’s (2010) study shows that using derivatives has no effect on firm valuation in French non-financial firms.
Hedging can stabilize a company’s value and maximize lifelong expected utility;
therefore, managers often use hedging strategies (Gunther and Siems, 1995; Colquitt and Hoyt, 1997).
Ten-year government bond yield rates fell from 5.25% to 1.52% from the year 2000 to 2004. To reduce the negative interest-rate spread problems, Taiwan life insurance firms increased their asset allocation in the U.S. bond and equity securities market12. The foreign investment allocation rose from 4.62% in 2000 to 27.21% in
12 The world’s major capital markets come from the regular issuance of the U.S. bond market and the development of the secondary market.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2004 (see Figure 3.1). To reduce the negative interest-rate spread problems, Taiwan life insurers changed their asset allocation from home-biased13 to internationally diversified starting in 2000. In 2008, foreign assets already exceeded 30.31% of the industry asset portfolio and reached 2.4 trillion NT$ (also see Figure 3.1). However, currency risk exposure has risen and NTD against USD exchange rate volatility was a major currency risk14. The investment asset evaluation rule changed from historical cost to fair value when the Statement of Financial Accounting Standard (SFAS) No. 34 was implemented in 2006. Currency hedging derivatives positions should mark to market, which affects the volatility of reported earnings. Currency risk management thus became an important issue for Taiwan life insurance industry. Our paper aims to examine the relationship between the investment performance of Taiwan life insurers and their currency hedging strategies.
Hedging can reduce the volatility of investment returns, but hedging cost will decrease investment performance, and hedging strategy effectiveness is influenced by interest rate fluctuations and exchange rate policies of the Central Bank. Taiwan is a small open economy, and the exchange rate has a large impact on economic and financial activities15. Due to the trade-off between hedging cost and hedging efficiency, an optimal hedging strategy is difficult to select (Chen and Liang, 2010).
Short-term nominal interest rates and currency fluctuations were positively correlated (Solnik, 1998). Using currency forward contracts for hedging involves higher costs due to Taiwan interest rates being lower than those in the U.S.A., and long-term exchange rate exposure was not effectively eliminated through financial instruments, which means that even if a portfolio had hedging strategies, it was still exposed to currency risk (Pringle and Connolly, 1993; Chow et al., 1997, and Li et al., 2009).
In the early 1990s, larger European and American pension funds left their international assets exposed to currency risk. Until the mid-1990s, international asset
13 Since the US fixed income market is the deepest and most liquid bond market in the world, the Taiwan life industry has heavily invested its foreign assets in America.
14 Since the US fixed income market is the deepest and most liquid bond market in the world, the Taiwan life industry has heavily invested its foreign assets in America.
15 Wang (2005) discovered that in Taiwan, short-term foreign currency fluctuation was significantly affected by changes in the Central Bank’s net foreign assets.
‧
pension funds prompted these funds to pursue optimal currency hedging strategies (Solnik, 1998). There was a negative relationship among currency fluctuation, foreign direct investment, and citizens with foreign bond investments (Wang, 2005). In 2008, the Taiwan life insurance industry’s foreign assets ratio already exceeded 30%.Foreign exchange risk has become one of the major risks that Taiwan insurers face.
NTD against USD appreciation trends and hedging strategy effectiveness are both influenced by interest rate fluctuations and the exchange rate policies of the Central Bank. In 1999 to 2003, the NTD showed a slight depreciation against the USD, so that foreign assets had unrealized gains for which there was no significant value in hedging. However, in 2004 and 2008, the NTD reversed course and appreciated (see Figure 3.2), causing huge losses on foreign exchange, and currency hedging gradually led to an import issue. The aim of this study was to answer the following question:
Does currency risk hedging enhance investment return for Taiwan life insurers and, if so, how?
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
0
Figure 3.1 Taiwan Three-Year Deposit Interest Rate and Foreign Investment Ratio, and U.S. Ten-Year Government Bond Yield. Source: Deposit rate is from TEJ; foreign investment ratio is from author's calculations. See also Chang et al., (2010).
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
2000 2001 2002 2003 2004 2005 2006 2007 2008
30 31 32 33 34 35 36
EXCHANGE
Figure 3.2 Exchange Rates of the New Taiwan Dollar against the U.S. Dollar. Source:
Central Bank of Taiwan. Exchange Rates of the NTD against the USD Interbank spot market closing rates. See also Chang et al., (2010).
Financial reporting and regulatory requirements have made insurers more sensitive to the risks inherent in their asset and liability portfolios. The most prominent changes have been the adoption of Risk-Based Capital (RBC) requirements, Financial Accounting Standard (FAS) 115, requiring mark-to-market accounting for fixed-income securities held in the “trading” or “available for sale” categories, and FAS 119, requiring disclosure of the purpose of derivative transactions (Cummins et al., 1997). Taiwan life insurance industry provides a natural experiment for investigating the relationship between hedging and investment performance for a number of reasons: (1) to reduce the negative interest-rate spread problems; Taiwan life insurers have gradually increased their asset allocation in the U.S. bond and equity securities market. The foreign investment allocation increased from 4.62% to 27.11% from 2000 to 2004. The international asset ratio already exceeded 30% in 2008. The life insurance industry is exposed to foreign currency risk, which affects financial report performance for all insurers; (2) Capital regulation changed from fixed amount to RBC requirements in 2003, and Statement of Financial Accounting
‧
Standard (SFAS) No. 34 was implemented in 200616. These changes have affected life insurers’ asset-liability management. Therefore, currency hedging instrument cost and mark to market evaluation results affect the effectiveness of financial report performance; (3) the difference between domestic-owned and foreign-owned firms in terms of whether or not stocks must be publicly issued17 affects the volatility tolerance of the currency risk on income statements18. Hence, the hedging strategy of the Taiwan insurers is examined to test whether these strategies have any impacts on their investment performance.
This chapter aims to examine the relationship between Taiwan Life Insurers’
investment performance and currency hedging strategy, and whether SFAS No. 34 has made insurers sensitive to currency hedging behaviors. We have found that the asset-liability management of foreign-owned companies differs from that of domestic-owned companies19. Furthermore, foreign-owned companies with stocks not publicly listed on the TAIEX did not need to disclose financial reports, so volatility tolerance of the currency risk on income statements is higher than that of
16 SFAS No. 34 like Financial Accounting Standard (FAS) 115, requires mark-to-market accounting for financial assets held in the “trading” or “available for sale” categories. Most of Taiwanese life insurance firms’ foreign assets are in “available for sale” and “hold to maturity” to match long-term life insurance obligations, and the evaluation results of assets in “available for sale” affects the volatility of ownership. However currency derivatives must mark to market, affect the gain or loss. This characteristic means that the result of currency hedging significantly affects income statement report earnings, and gives managers the incentive to adopt hedging strategies.
17 Insurance Act Article 136: An insurance enterprise organized as a company limited by shares shall issue its stock publicly unless another law provides otherwise or the competent authority has granted permission. But in the Financial Supervisory Commission’s response in 2006 to the American Chamber of Commerce’s Taiwan industry white paper, because foreign-owned company stockholders' rights are concentrated, the issue of minority shareholder rights and interests does not arise.
18 Companies with publicly issued stocks are required to disclose financial report performance periodically, whereas foreign-owned firms without publicly issued stocks do not have this requirement.
The volatility tolerance of the currency risk on income statements is higher for foreign-owned than for domestic-owned firms.
19 For example, more than 50% of the assets of PCA Life, Aegon, and Prudential of Taiwan are allocated in long-term government bonds.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
domestic-owned firms20. Domestic-owned firms’ managers are insensitive to the need for hedging currency risk in order to stabilize financial statement performance (Colquitt and Hoyt, 1997).
However, if we divide the sample into two groups based on ownership structure, the results show that domestic-owned companies with currency hedging strategies have more advantages than foreign-owned firms that left their international assets exposed to currency risk. Furthermore, if the sampled companies are divided into companies publicly listed on the TAIEX and others, the results show that hedging strategies have positive effects too. The results support the managerial discretion hypothesis in Stulz (1990), and show that SFAS No. 34 has significant effects on the currency hedging behaviors for the domestic-owned and the listed companies. The policy implications are as follows:
1. To reduce negative interest-rate spread problems, international diversification of asset allocation can enhance investment performance, but concentration of currency in USD makes currency risk management difficult.
2. Foreign asset allocation is mostly held long term, but currency hedging derivative transactions need to be marked to market monthly. Financial asset evaluation by SFAS No. 34 leads Taiwan life insurance firms to hedge currency risk just for financial reporting purposes, rather than to control real currency risk. We suggest that the Financial Supervisory Commission adopt a suitable life insurance accounting rule to reduce currency risk hedging result distortion financial report information.
The rest of this chapter is organized as follows. In Section 3.2 presents the related literature. In Section 3.3 we describe the sample, variables and methodology employed in this study. In Section 3.4 presents empirical result. Finally, we conclude and provide some policy implications.
20 Foreign-owned firms can apply for permission to not publicly issue their stock, whereas domestic-owned companies are required to publicly issue stock and to disclose financial reporting periodically. Profit and loss on exchange rates affect investment performance, so domestic-owned companies usually hedge currency risks.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
3.2 Literature Review
3.2.1 Hedging Motives
Four main hedging hypotheses have influenced the theories on motives for hedging in the insurance industry (Gunther and Siems, 1995): the economics of scale and information economics hypothesis; the bankruptcy cost hypothesis; the underinvestment hypothesis; and the managerial discretion hypothesis. Volatile fluctuation in the exchange rate will affect managers’ compensation from the company and their lifetime utilities will be changed (Solnik, 1974). Hedging can stabilize a company’s value and maximize lifelong expected utility; Firm managers seek to manage currency risk due to their desire to control cash flow volatility (Bodnar et al., 1998), so managers often use hedging strategies (Gunther and Siems, 1995). Colquitt and Hoyt (1997) verified the purpose of life insurance hedging which supports the managerial discretion hypothesis.
There is a positive relationship between hedging derivatives and the size of a company (see Warner, 1977; Nance et al., 1993; Gunther and Siems, 1995, and Cummins et al., 1997). However, various studies yield different views. Bankruptcy costs are more expensive to small companies than large ones, so small companies hedge more aggressively. Highly leveraged companies have a high probability of bankruptcy with underinvestment problems (see Myers, 1977; and Mayers and Smith, 1987). Therefore, there is a negative correlation between the size of the company and its hedging strategy (Altman, 1984), and companies with higher leverage ratios will often choose hedging derivatives (Colquitt and Hoyt, 1997). Purchasing insurance can reduce the risk of insurance liabilities, increase company value, and decrease underinvestment problems (Mayers and Smith, 1987; Nance et al., 1993). The insurance industry should adopt hedging strategies to ease the high debt amount produced by underinvestment.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
3.2.2 The Relationship between Hedging and Performance
Hedging reduces the volatility of cash flow and currency risk in a company (Glen and Jorion, 1993). If external financing costs are higher than internal financing, in the pursuit of the goal of maximizing shareholders’ value, it would be beneficial for companies to hedge (Froot et al., 1993). Other studies that support hedging include Stulz (1984), Smith and Stulz (1985), Stulz (1990), Breeden and Viswanathan (1990), Froot et al. (1994), and Campbell et al. (2003). By contrast, Modigliani and Miller (1958) claimed that having a perfect hedging strategy in the financial market is futile.
Pringle and Connolly (1993), Chow et al. (1997), Li et al. (2009), and others claimed that long-term foreign exchange exposure cannot usually make the outflow of foreign currency equal the inflows. Effective hedging through long-term exchange rate exposure financial instruments is not easy; the portfolio remains exposed to currency risk even after hedging.
Froot (1993) uses data showing that while over short horizons hedging reduces risk substantially, over long horizons, hedging often does not reduce risk at all. In fact, at long horizons, many fully hedged international investments actually have greater return variance than their un-hedged counterparts. Hedge ratios should primarily reflect real exchange rate exposure and be quite sensitive to the investment horizon (Froot, 1993). The results show that selective hedging strategies are superior under some conditions, and that transaction costs are very important in determining the various currency risk management strategies for both stocks and bonds invested at the one-month horizon (Beltratti et al. 2004).
3.2.3 Optimal Hedging Strategy
According to our literature review, three types of hedging strategies exist, i.e., fully currency hedged (100% hedged), fully currency un-hedged (hedge ratio 0%) and optimal hedge. When real investment positions are unrelated to the fluctuation of foreign currency return, full currency hedge is optimal (Solnik, 1974). However, this full currency hedging does not distinguish between strong or weak currencies, and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
investors may regret the lost opportunity to hold strong currencies (Michenaud and Solnik, 2008). Harris (2004) surveys the distribution of currency-hedge ratios for 563 institutional investors delegating the currency hedging decision to overlay managers.
On average worldwide, 39% of investors adopt a no-hedging policy, 34% adopt a 50%
hedging policy, 14% adopt a 100% hedging policy, and 13% adopt some other hedge ratio policies.
Froot et al. (1993) found that long-term exchange rates will approach purchasing power value, and thus recommended that pension funds not hedge.
Nevertheless, based on 150 years of observation, their study discusses limitations when applying real world investment practices. Ederington (1979) proposed the optimal hedge ratio. Glen and Jorion (1993) have derived the optimal currency hedging strategy for international equity and bond portfolios. Campbell et al. (2010) verified that selective hedging strategies can reduce the investment risks of equity investors. Michenaud and Solnik (2008) used regret theory and proposed that the appropriate hedging ratio to avoid the regret is 50%. The value of companies that hedged was higher than those that did not or that stopped doing so (Allayannis and Weston, 2001).
3.3 Research Design
3.3.1 Relationship between Currency Hedging and Investment Performance
Hedging can reduce the volatility of cash flow and currency risk in a company (Glen and Jorion, 1993). If external financing costs are higher than internal financing, in the pursuit of the shareholders’ value maximizing goal, it would be beneficial for the company’s value to hedge (Froot et al., 1993; Stulz, 1984; Smith and Stulz, 1985; Stulz, 1990; Breeden and Viswanathan, 1990; Froot et al., 1994; and Campbell et al., 2003).By contrast, Modigliani and Miller (1958) claimed that having a perfect hedging strategy in the financial market is futile. A hedging strategy is more advantageous than a non-hedging strategy (Campbell et al., 2010). The foreign assets of Taiwan life insurance industry account for approximately 30% of their funds, and 90% of assets
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
are allocated to USD assets. In addition, the currency derivatives position evaluation rule has changed from off-balance-sheet footnote to mark to market. These factors, together affect the volatility of report earnings, so we expect that adoption of currency hedging can enhance life insurance firms’ investment performance. We form the following hypothesis:
H3: Utilizing currency hedging can enhance life insurance firms’ investment
performance.
3.3.2 The Relationship among Currency Hedging, Manager Incentive and Economy of Scale
Hedging can stabilize a company’s value and maximize the managers’ lifetime expected utility. Therefore, managers often adopt hedging strategies (Gunther and Siems, 1995; Hoyt, 1997). After the implementation of SFAS No. 34, hedging cost and mark to market results are included in the exchange on gains and losses, which affect the monthly profit and loss. Currency risk significantly affects company investment performance, and changes the compensation plans of managers and their lifetime utility. Hedging can reduce the volatility of investment, so managers of the listed companies are more likely to adopt hedging strategies than unlisted companies would. This section will examine whether the high hedging21 in domestic-owned companies is more favorable than the foreign-owned firms’ non-hedged strategies, and whether this study’s results support the optimal currency hedging strategies of Solnik (1995) and Campbell et al. (2010).
H4: The currency hedging strategy of domestic-owned firms’ investment performance
is more favorable than non-hedging strategies of foreign-owned firms.
21 In an investor conference of Cathay Financial Holdings on March 21th, 2008, currency hedging of Cathay Life Insurance in 2007 was reported as NT$ 17.7 billion. The traditional hedging ratio is 40%; a currency basket is 50%. In 2008, Cathay Life Insurance increased its traditional hedging ratio.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Derivative hedging decisions have a positive correlation with the size of the company (Warner, 1977). When life insurance companies wish to implement and manage derivative transactions, educating the managers becomes the most critical issue (Hoyt, 1989). Transaction costs are relevant to the exposure of derivatives;
hedged positions must be higher than 5-10 million USD to play a cost-effective hedging role (Mian, 1996). Among the five listed life insurance companies, three of them are the subsidiaries of financial holding companies, with experienced investment professionals and mature risk management systems. The scale of asset management is larger than that of unlisted companies22. This study will examine whether the hedging strategies of listed companies are aligned with the information
hedged positions must be higher than 5-10 million USD to play a cost-effective hedging role (Mian, 1996). Among the five listed life insurance companies, three of them are the subsidiaries of financial holding companies, with experienced investment professionals and mature risk management systems. The scale of asset management is larger than that of unlisted companies22. This study will examine whether the hedging strategies of listed companies are aligned with the information