# Are Philippine fixed income fund managers generating alpha for their clients?

INTRODUCTIONMutual funds, also known as investment companies, are pooled investment vehicles managed by an asset management firm that seeks to create value for shareholders through tactical allocation of investor funds in various asset classes or in securities within a particular asset category. Typically, mutual funds are classified based on investment style, and can be broadly identified as either a fixed income fund, an equity fund, or a balanced fund (combination of fixed income and equity instruments), depending on the specific assets the fund is allowed to invest in (as stated in the fund's prospectus). Investors consider mutual funds to be a viable alternative to direct investments in equity and debt securities due to a number of benefits that can be derived from pooled fund-investing, which include, but are not limited to, easy liquidity, economies-of-scale, access to otherwise inaccessible markets, diversification, and management of funds by a skilled and knowledgeable investments practitioner.

For an asset management firm to act as fund manager and administrator of an investment company, it must be awarded a management and distribution mandate by fund participants (owners). Mandated asset managers are expected to perform the following basic duties in exchange for a periodic management fee: distribute shares in the fund to potential investors, provide ready liquidity to clients who wish to divest their existing mutual fund holdings, act as investment advisor and administrator to the investment company, and manage the fund's securities portfolio for the collective benefit of all fund shareholders. However, asset managers are scrutinized by the investing community on the basis of their ability to generate returns not just relative to competitors but vis-a-vis a chosen benchmark portfolio as well. Since ownership in a mutual fund is dispersed given its pooled nature, the interests of the majority shareholders of the fund, just like in any registered corporation, are represented by a board of directors. And one of the key responsibilities of the board of directors is to prudently select a fund manager and periodically evaluate its performance in all of the five aforementioned aspects, with returns relative to peers and benchmark being one of the more critical aspects in determining investment success for the client. Asset managers that exhibit more than just modest performance are likely to have their mandate as fund manager renewed during each periodic review by the board of directors. This process of performance measurement and evaluation ensures that fund managers remain competitive in the pooled fund investments industry.

Mutual Fund Industry Growth in the Philippines

To put the competitive landscape of mutual fund investments in proper perspective, assets under management (AUM) in the Philippine mutual fund industry reached a historic high of PHP 95.7-billion as of the end of 2010. This is a dramatic increase from the PHP 1.35-billion AUM of managers of local mutual funds back in 1997 (Valderrama & Bautista, 2003). Of the PHP 95.7-billion, 19.8% represents investments in funds whose main investment style is long-term growth through equity investments, 58.3% is in the form of funds invested purely in short-term to long-term fixed income instruments, and the remaining 21.9% is allocated in balanced portfolios of both stocks and bonds (balanced fund category). In terms of number of players in the industry, the number of mutual funds in the Philippines has grown from 25 in the year 2003 to 45 in 2010, evidence to the intensifying competitive structure of the local pooled investments industry. The sudden surge of unit investment trust funds (UITFs) beginning 2005 due to the phase-out of common trust funds (CTFs) by the Bangko Sentral ng Pilipinas (BSP) likewise facilitated the rise in competition between fund managers, not to mention the improvement in domestic macroeconomic and financial market conditions since 2005 which increased demand for pooled investments. As of the end of December 2010, there were 79 UITFs operated by 16 trust departments of banks. This compares to the 10 asset management firms that provide advisory and management services to mutual funds.

Philippine Fixed Income Mutual Funds

It is notable that fixed income funds are the preferred investment avenue among Philippine mutual fund shareholders. And this is not at all unexpected given the relatively conservative profile of most Filipino investors. However, it is also noteworthy that the fraction of the mutual fund industry's AUM accounted for by fixed income funds (58.3% in December 2010) has fallen considerably from the 92.2% recorded in 2002 (Valderrama & Bautista, 2003). The decline in the share of fixed income funds in the net assets of the mutual fund sector can be attributed to the marked improvement in the performance of the Philippine stock market in recent years and the migration from accrual to mark-to-market accounting for fixed income funds in 2005, which created more volatility in the local fixed income industry. Notwithstanding the drop in AUM share of fixed income funds among the investment companies, the number of fixed income mutual funds continued to grow from the 4 bond funds recorded in the year 2000 to the 26 different types of fixed income funds in 2010. Of this 26, 10 are in the form of Peso-denominated bond funds, 9 are in Dollar-denominated bond funds, 2 are Euro-denominated bond funds, while 5 are Peso-denominated money market funds. The recent launch of global bond funds by some of the bigger players in the industry suggests that there is demand potential for fixed income funds in the Philippines.

Objectives of the Study

In view of the stiff competition and growth prospects of mutual funds in the pooled investments plane, it is clear that an objective and relative measure of fund manager ability is imperative to give clients a means of assessing whether or not the fund managers these clients have mandated to help them reach their investment goals are indeed performing up to par with expectations. The goal of this research undertaking is to quantitatively appraise the risk-adjusted returns performance of managers of mutual funds in the Philippines, specifically those that have been contracted by their clients to manage a fixed income portfolio on their behalf.

The study will look at the more commonly-watched parameters of mutual fund performance used by industry practitioners to be able to ascertain how effective Philippine fixed income fund managers are in utilizing available information and in taking advantage of market opportunities to further the returns performance of their managed portfolios. These composite portfolio performance measures are the Sharpe ratio, the Treynor ratio, the Jensen's alpha measure, and the information ratio. These metrics shall be applied to the historical returns of Philippine mutual funds over a 1-, 3-, and 5-year computation horizon to determine whether fund managers of fixed income funds are consistently providing returns beyond what is expected of them on a risk-adjusted basis and to establish which among these asset management firms consistently perform better than their peers and benchmark.

DESIRABLE ATTRIBUTES OF PORTFOLIO MANAGERS

There are two primary characteristics that investors seek from their portfolio managers. First is a fund manager's ability to generate above-average returns for a given risk profile. Superior returns can be attained if a fund manager has the ability to properly time the market and has the skill of identifying and trading incorrectly priced securities in the market. The former can be seen in a fund manager of a fixed income portfolio who increases the duration of his bond holdings prior to an anticipated drop in interest rates and decreases duration when he expects yields to rise. Such a market-timing strategy, if implemented properly, would make his fixed income portfolio valuations more sensitive to falling interest rates (have more bond price appreciation) and more defensive to rising interest rates (less bond price depreciation), hence, providing above-average returns for clients. The latter strategy of exploiting asset mispricing would likewise generate above-average returns, most especially if the portfolio manager is able to select undervalued securities for the portfolio (Reilly & Brown, 2009).

The second desired attribute is a fund manager's ability to rid a securities portfolio of all unnecessary unsystematic risks by virtue of diversification. This implies that investors in general would prefer to have their funds invested in a completely diversified portfolio of securities because they do not expect to be compensated for unsystematic risks. And the level of diversification of a portfolio relative to a benchmark or market index is measured by the correlation of its returns to the said market index. The relevance of these two portfolio manager attributes is that some portfolio performance assessment tools only consider one of the two characteristics while others indirectly consider both but fail to differentiate them (Reilly & Brown, 2009).

EARLY PORTFOLIO PERFORMANCE MEASUREMENT TECHNIQUES

Prior to the extensive work on portfolio theory in the 1960s, most investors evaluated investments solely on the basis of returns without explicitly accounting for risk. Although Modern Portfolio Theory introduced the concept of risk as the volatility of investment returns, analysts back in the 1960s would consider returns and risk as two separate and distinct realities and would make comparisons of portfolios within a particular risk class (those with similar returns variability). Portfolio analysts would group pooled funds together, not on the basis of investment style, but on the basis of variability of returns. This meant that funds that had similar returns variability would be categorized within the same peer group and pitted against one another to determine which fund generated the highest returns within the sample notwithstanding the fact that these funds may be invested in different asset classes. For example, a portfolio of high-yield corporate bonds could actually be grouped with a portfolio of equities with a totally different investment style and categorized within the same peer group just because they exhibit similar returns variability. An obvious problem to peer group comparisons such as these that only look at returns is that it disregards varying investment styles of fund managers and makes it difficult for investors to determine whether fund managers have met their clients' investment objectives after having satisfied certain investment constraints (Reilly & Brown, 2009).

COMPOSITE MEASURES OF PORTFOLIO PERFORMANCE

Some of the shortcomings of early portfolio performance assessment tools were the fact that they were very returns-centric and failed to properly account for risk. These shortfalls were a result of the lack of literature in portfolio theory during the time that combined both return and risk in performance appraisal. After the development of the Capital Asset Pricing Model (CAPM) by Sharpe, Treynor, and Mossin, in the 1960s, investors and financial economists alike recognized the need to address the issue of adjusting for risk in investment performance evaluation, which led to the development of a number of extensions to the CAPM. Three measures were proposed in the finance literature to account for risk when assessing portfolio returns--these are the Treynor, Sharpe, and Jensen portfolio performance measures, which are most commonly referred to as the composite measures of portfolio performance. Each method attempts to quantify a mutual fund's performance relative to a risk-adjusted return appropriate for a portfolio of the same risk. A portfolio's returns performance is deemed successful when it yields a return better than the benchmark equilibrium return (Chua, et al., 2008).

Treynor Portfolio Performance Measure

Jack Treynor in 1965 developed the first composite portfolio performance measure that included risk. The two components of risk he considered are the risk from general fluctuations in market returns and the risk from idiosyncratic fluctuations of securities in a portfolio. He proposed that all portfolios have a unique characteristic line that defines its degree of sensitivity to the returns on a market (benchmark) portfolio. The slope of the characteristic line is also known as beta and is a direct measure of a portfolio's relative risk attributes. Portfolios with beta coefficients equal to 1.0 have a risk that exactly mimics that of a market portfolio assumed to be fully diversified (with no unique risk).

He developed the Treynor measure because of his desire to introduce a performance parameter that could be utilized by all investors irrespective of their risk profiles. Building on the assumptions of Capital Market Theory, a risk-free asset is bundled with different portfolios to form a portfolio possibility line. Risk-averse investors would always prefer the portfolio possibility line with the highest slope since this would maximize an investor's return in excess of the risk-free rate for every unit of market risk. The Treynor ratio, denoted as Tt, is also known as the reward-to-volatility ratio and is the slope of the portfolio possibility line that rational investors seek to maximize. This is given in Equation (1).

[T.sub.i] = [[bar.R].sub.i] - [bar.RFR]/[[beta].sub.i] (1)

In this expression, [R.sub.i] is the average rate of return for portfolio i, RFR is the risk-free rate of return, and [R.sub.i] is the slope of the characteristic line during a specific time period. The numerator of this ratio is a risk premium that captures the extra return an investor earns from the portfolio during a specified time period over and above the return on a simple buy-and-hold strategy on a risk-free asset with a similar holding period. The denominator of this ratio, meanwhile, is a proxy for the portfolio's risk as captured by its beta coefficient, which measures only systematic risk. Treynor's rationale for using beta (and not total risk) as the relevant proxy for uncertainty is his assumption that investors do not require compensation for bearing unique risks that can be eliminated by investing in a completely diversified portfolio (which is the benchmark that investors and fund managers attempt to beat in terms of risk-adjusted returns). Putting returns and risk together, the Treynor ratio quantifies a portfolio's risk premium return per unit of risk (Reilly & Brown, 2009). A high and positive Treynor ratio shows superior risk-adjusted performance for a fund, whereas a low and negative Treynor ratio may be indicative of unfavorable performance (Chua, et al., 2008).

Plotting the Treynor ratios of a series of portfolios (including the market portfolio) on a graph allows investors to ascertain whether each portfolio is located above or below the Security Market Line (SML), which is a line that links the risk-free asset with the market portfolio and has a beta of 1.0. Such an exercise would facilitate investment analysis and aid investors in visually determining a portfolio's outperformance relative to peers and the market on a risk-adjusted basis. It must be noted however that the Treynor portfolio performance measure is a ranking criterion for mutual funds and provides relative, but not absolute, rankings of portfolios. Portfolios that plot higher than the SML, and hence, have a higher reward-to-volatility ratio relative to the market portfolio, have generated alpha for their clients, or risk-adjusted returns greater than that of the benchmark (Reilly & Brown, 2009).

One problem with the use of the Treynor measure is its use of beta as its measure of risk. Estimating a portfolio's beta coefficient using regression may introduce estimation bias because of the assumption that beta is constant over a specific period of investigation. An alternative method of arriving at the beta coefficient, which is a portfolio's covariance of returns with the market divided by the variance of the market's returns during a particular estimation period, shall be used for the purposes of this study.

Sharpe Portfolio Performance Measure

Just a year after Treynor introduced the first composite measure of portfolio performance, William Sharpe in 1966 proposed another method of evaluating the performance of mutual funds as an offshoot to his previous work on the CAPM and Capital Market Theory. The Sharpe ratio is similar to the Treynor measure in that it also quantifies reward-to-variability. But instead of considering only the risk of a portfolio relative to that of a fully diversified basket of assets, Sharpe sought to measure risk-adjusted performance by looking at a portfolio's total risk, which considers both market risks and risks specific or unique to the underlying assets in a portfolio. The proxy for risk he suggested was a measure of dispersion of a portfolio's returns series from its mean value, also known as a portfolio's standard deviation of returns. Similar to Treynor's measure of excess returns, Sharpe utilized the same definition, which is the incremental return on a portfolio during a specified time horizon over and above the return on a passive investment in a risk-free asset over a similar time frame. The Sharpe ratio, denoted by Si, is therefore a portfolio's risk premium return earned per unit of total risk (Reilly & Brown, 2009). This is given in Equation (2).

[S.sub.i] = [[bar.R].sub.i] - [bar.RFR]/[[sigma].sub.i] (2)

In addition to the earlier notation, [[sigma].sub.i] is the standard deviation of the return of portfolio i during the time period being analyzed.

The evaluation criteria for the Sharpe measure does not deviate much from Treynor's methodology. A higher Sharpe ratio is desired for a portfolio since this would suggest that a fund manager was able to generate incrementally higher excess returns for every unit of total risk. Sharpe's use of standard deviation of returns as the relevant risk measure makes Capital Market Theory the theoretical basis for comparing portfolio performance (compared to the CAPM for Treynor's measure). Combining a risk-free asset with different risky investment portfolios (whose total risk is defined by its standard deviation of returns) results in portfolio combinations that plot above, below, or along the Capital Market Line (CML). Mutual funds that plot above the CML are considered superior to the market portfolio in the sense that they provide shareholders with an excess return higher than the benchmark on a risk-adjusted basis (Reilly & Brown, 2009). These pooled funds are considered alpha-generating due to their benchmark outperformance. Portfolios with a higher reward-to-variability ratio relative to peers likewise exhibit returns outperformance vis-a-vis competitors after accounting for total risk.

It must be noted, however, that the Sharpe measure, just like the Treynor ratio, is a relative, and not absolute, measure of portfolio success. The use of Sharpe ratios can likewise facilitate a ranking of investment funds. The widespread use of the Sharpe ratio in the investments industry comes from its simplicity and straightforwardness. Its use of total risk makes the risk-adjustment process less dependent on stringent assumptions (such as the implicit presumption of full diversification for the Treynor ratio).

Jensen Portfolio Performance Measure

In 1968, Jensen proposed an extension to the CAPM that would address the inability of the Sharpe and Treynor portfolio metrics to quantify the investment ability of a portfolio manager on an absolute basis. He postulated that a fund manager's unique skill in forecasting market turns and selecting undervalued securities for his portfolio can be measured empirically by analyzing a portfolio manager's alpha coefficient. The alpha coefficient, symbolically represented by ([[alpha].sub.j]) is the portion of a portfolio's excess return (over the risk-free rate) that is beyond the market risk premium required or expected from an investment based on an equilibrium asset pricing model such as the CAPM or a multi-factor model of risk and return such as the Arbitrage Pricing Theory (APT). A portfolio manager is deemed to have outperformed his benchmark if the Jensen's alpha is positive and statistically significant (Reilly & Brown, 2009).

To arrive at Jensen's alpha coefficient, the basic CAPM expression, assuming it is empirically valid, is extended by deducting both sides of the equation by the risk-free rate of return. This would result in the following regression model given in Equation (3):

[R.sub.jt] - [RFR.sub.t] = [[beta].sub.f] [R.sub.mt] - [RFR.sub.t]] + [e.sub.jt], (3)

where [R.sub.jt] is the realized return on portfolio j over a given period, [RFR.sub.t] is the one-period risk-free interest rate, [R.sub.j] is the systematic risk for portfolio j, and [R.sub.mt] is the return on the market portfolio over the same holding period as portfolio j.

If the market were in equilibrium, an intercept (alpha) would no longer be expected. But if fund managers are anticipated to provide investors with superior returns due to market timing and security selection skills, a non-zero intercept (alpha) must be introduced in the regression model that would capture this unique skill of portfolio managers. A statistically significant positive (negative) alpha is reflective of a fund manager's superior (inferior) abilities. The regression model that incorporates the intercept (alpha coefficient) is given in Equation (4):

[R.sub.jt] - [RFR.sub.i] = [[alpha].sub.j] + [[beta].sub.f][[R.sub.mt] - [RFR.sub.t]] + [e.sub.jt] (4)

Rearranging Equation (4) results in the Jensen's alpha equation that measures a fund manager's absolute degree of outperformance (underperformance) relative to a benchmark's expected return required by a pricing model, as given in Equation (5).

[[alpha].sub.j] ={[R.sub.jt] - [RFR.sub.t]) - [[beta].sub.f][[R.sub.mt] - [RFR.sub.t]] (5)

Although Jensen's portfolio performance measure addresses the need for some analysts to quantify absolute performance of mutual funds, it is not without its share of shortcomings. It must be noted that the use of regression to estimate the value of Jensen's alpha may make it subject to statistical biases. Moreover, Jensen's alpha is sensitive to the choice of benchmark (CAPM or APT) and estimation method for calculating beta (Chua, et al., 2008). For this study, the proponent has elected to calculate beta for the Jensen's alpha measure using the same estimation method implemented in obtaining the systematic risk component of the Treynor ratio. For the choice of benchmark in Jensen's alpha, the single-factor CAPM shall be utilized to arrive at a portfolio's market risk premium.

Information Ratio Performance Measure

A fourth composite measure of portfolio performance that is widely used in the asset management industry is the information ratio, which is a risk-adjusted performance statistic that takes the return on a managed portfolio in excess of the returns on a benchmark portfolio and divides this by the standard deviation of this excess return. The numerator of the formula represents a fund manager's ability to use his professional asset management skills and information or knowledge of the markets to generate incremental returns over and above the benchmark clients use to assess his performance. The denominator, meanwhile, measures the amount of unsystematic risk the investors are exposed to because of the fund manager's active portfolio management efforts in pursuit of those excess returns (Reilly & Brown, 2009). This standard deviation of excess returns due to active management is also known as the tracking error. In notation form, the information ratio, denoted by [IR.sub.j], is expressed in Equation (6):

[IR.sub.j] = [[bar.R].sub.j] - [[bar.R].sub.b]/[[sigma].sub.R] = [[bar.ER].sub.j]/[[sigma].sub.ER] (6)

where [R.sub.b] is the average return for the benchmark portfolio during the period and [[sigma].sub.ER] is the standard deviation of the excess return during the period.

The information ratio is an example of a benefit-to-cost ratio because the tracking error is the cost of active portfolio management with the excess return of the portfolio relative to the market being the reward that is due to investors resulting from the portfolio manager's decision to deviate from a passive market benchmark portfolio-tracking strategy. Just as in both Treynor and Sharpe's measures, a higher information ratio is desired because it shows that a fund manager has maximized his active portfolio management skills to create value for his clients on a risk-adjusted basis.

The information ratio has its similarities to the three previously discussed portfolio performance metrics and can be considered a combination of the three. It is similar to the Sharpe and Treynor measures in that it is also a risk-adjusted measure of return that highlights a portfolio's relative, and not absolute, performance. The information ratio has a more noticeable similarity to the Sharpe ratio because of its use of standard deviation of returns as its proxy for uncertainty. Comparing the information ratio to the Jensen's alpha measure, it somewhat resembles Jensen's metric in the sense that it attempts to look at returns outperformance relative to a chosen benchmark.

RESEARCH METHODOLOGY

This study focuses on the application of the Treynor, Sharpe, Jensen, and information ratio portfolio performance measures on selected Philippine mutual funds. Due to the popularity and clear growth prospects of fixed income funds in the Philippine investments setting, the proponent has elected to evaluate the 1-, 3-, and 5-year risk-adjusted returns performance of all Peso-denominated medium- to long-term fixed income investment companies in the Philippines utilizing the four previously discussed composite measures of portfolio performance. Money market investment companies, although still part of the spectrum of fixed income funds, will be excluded from the analysis due to their insignificant size in the local mutual funds industry (money market mutual funds account for less than a percent of industry AUM). As of December 31, 2010, the study's evaluation date, there are 10 Peso-denominated bond funds that will form part of the research study's sample.

The following table enumerates the 10 different bond funds that will be appraised using risk-adjusted returns measures, including their respective asset management firms, the net asset values (NAV) of each fund as of 2010 yearend, and the proportion of each fund to the NAV of the entire industry of Peso-denominated bond (mutual) funds. To ensure anonymity and to remain impartiality in the analysis, the proponent has disguised the names of the fixed income funds and their asset managers and has assigned them fictitious names.

Returns performance will be analyzed using a 1-, 3-, and 5-year time horizon since this is the convention used by most asset managers in presenting historical fund performance. 1-year rolling returns shall be computed by taking the year-over-year percentage change in the Net Asset Value Per Share (NAVPS) of a mutual fund or portfolio during the last day of each month beginning January 2010 until December 31, 2010, the study's evaluation date, for a total of 12 monthly rolling returns. 3- and 5-year rolling returns, meanwhile, will be calculated in a similar manner by taking the percentage change in a mutual fund's NAVPS over a 3- and 5-year period respectively. Both the 3- and 5-year rolling returns will have 12 observations each covering the same monthly period of January to December 2010. To ensure that all returns figures are consistently presented in per annum terms, the 3- and 5-year rolling returns will be annualized by using a compound annual growth rate formula. Since all composite returns measures discussed in the previous section require the use of an average of realized portfolio returns, the arithmetic average of the 1-, 3-, and 5-year rolling mutual fund returns over the 12-month period from January to December 2010 will be calculated, and these arithmetic averages shall proxy for average portfolio returns. All official NAVPS data shall be obtained from the website of the local mutual funds association.

Because three of the composite portfolio measures that will be studied in this research paper analyze returns performance by considering average mutual fund and benchmark returns in excess of the returns on a risk-free asset, the yield on Peso-denominated Philippine government securities shall substitute for the risk-free rate. To maintain comparability and consistency in the analysis, the tenor of the risk-free asset will be commensurate with the computation horizon or holding period of the portfolio in scrutiny. For example, if the rolling returns of a mutual fund are being investigated over a 1-year evaluation horizon, then the risk-free rate that will be applied to calculate the portfolio's excess returns will be the fixing yield on a Philippine T-Bill exactly a year prior to the returns computation date. A positive excess return would then imply that investing in the mutual fund over the past year yielded a client a realized return greater than that of a passive investment in a 1-year T-Bill. For the 3- and 5-year portfolio returns evaluation, the fixing yield-to-maturity on a 3- and 5-year Philippine Fixed Rate Treasury Note (FXTN) exactly 3- and 5-years prior to the returns computation date respectively, will be utilized as the proxy for the risk-free rate. And similar to the process used to calculate average portfolio returns, the arithmetic mean of the 1-, 3-, and 5-year fixing yields over the 12 monthly periods from January to December 2010 shall represent the average risk-free rate of return. For all T-Bill and FXTN yields used, the rates shall be obtained from the website of the Philippine Dealing and Exchange Corporation (PDEx), the operator of the local fixed income exchange.

Three risk proxies have been introduced in this research undertaking, each having its own specific function in the risk-assessment process. For the Treynor ratio, which looks solely at a mutual fund's systematic risk component, quantification of uncertainty is achieved by calculating a fund's beta coefficient, which is simply the covariance of the return of the portfolio relative to the market divided by the covariance of the market's returns with itself. The most widely-used and accepted benchmark for the market portfolio by Philippine fixed income fund managers is the HSBC Philippines Local Currency Government Bond Total Return Index. This index is normally used as the benchmark for Peso-denominated fixed income funds primarily invested in medium- to long-duration government securities. The proponent has chosen not to compute beta using regression analysis to overcome potential problems of statistical bias and model misspecification during estimation. A mutual fund's beta aptly describes the linear co-movement of its returns to that of the HSBC Philippines Local Currency Bond Total Return Index and represents the sensitivity of the mutual fund's returns to changes in the said market index. Operationally, the beta of a mutual fund's rolling returns with respect to the benchmark index shall be computed by utilizing the rolling returns data of both the mutual fund and the HSBC benchmark over the most recent 12 months immediately prior to the study's evaluation date. Historical data on the HSBC Philippines Local Currency Bond Total Return Index shall be sourced from Bloomberg, LP.

Similar to the Treynor measure, the Jensen's alpha performance metric utilizes the beta coefficient in accounting for risk in the assessment of returns outperformance. Specifically, the beta is a necessary input in the determination of the required return implied by an asset pricing model that fund managers attempt to hurdle to generate alpha. The same beta coefficients computed for the Treynor measure will be used to quantify alpha based on Jensen's methodology. Although Jensen prescribes the use of regression when estimating a fund manager's alpha, for the purposes of this study, the proponent shall calculate alpha simply by taking the difference between the excess return of a mutual fund over and above the required risk premium return implied by the CAPM.

Both the Sharpe and information ratio performance assessment methodologies use a different approach to quantifying returns uncertainty when compared to Treynor's and Jensen's metrics. Instead of focusing on just the non-diversifiable portion of a portfolio's returns, the Sharpe and information ratios look at overall risk and unsystematic risk respectively in accounting for variability in mutual fund returns. Particularly, Sharpe operationalizes risk quantification in a mutual fund portfolio by computing for the standard deviation of its historical returns. As such, the total risk of the mutual funds to be investigated in this research paper shall be computed by taking the standard deviation of their annualized 1-, 3-, and 5-year monthly rolling returns over the 12 months immediately preceding the study's evaluation date. The information ratio, in contrast, measures uncertainty by zeroing-in on a mutual fund's active risk, calculated as the standard deviation of a mutual fund's active rolling returns or rolling returns in excess of that of the HSBC Philippines Local Currency Bond Total Return Index. The standard deviation of the active returns will be applied to the most recent 12 rolling returns observations as of the study's evaluation date for all three returns computation periods.

DATA PRESENTATION AND ANALYSIS

This section summarizes the results of the performance appraisal carried out on the 10 Philippine fixed income investment companies that were included in the sample. The presentation of the data shall be done by first listing in tabular form the risk-adjusted returns measures of all funds for the 1-, 3-, and 5-year returns time frames under each of the four composite portfolio performance criteria. A ranking based on these portfolio metrics shall likewise be shown to ascertain which of the local fund managers are industry leaders based on risk-adjusted returns and which are consistent performers over the short-, medium, and long-run. The proponent will then briefly discuss the results of the analysis and the rankings from a broad or industry standpoint and will proceed to highlight relevant facts or events in an attempt to rationalize the better (worse) than expected performance of specific funds. Benchmark figures shall also be shown to make the analysis richer.

Table 2 presents the historical returns of 10 Philippine bond funds after adjusting for their respective systematic risk attributes. The findings suggest that utilizing the Treynor ratio for ranking mutual funds would result in Coconut Fixed Income Fund consistently outperforming competition utilizing 3- and 5-year risk-adjusted returns as bases. Philippine Stock Peso Bond Fund was likewise one of the top performers using the Treynor criterion having been the best bond fund over the 1-year horizon and the 2nd top fund in the 3-year category. Education Fixed Income Fund similarly provided its investors with competitive risk-adjusted rates of return over a 1-year and 3-year returns evaluation time frame, being 2nd and 3rd respectively in terms of ranking.

The perennial underperformers in the industry are those bond funds whose rankings based on risk-adjusted returns are persistently in the 3rd and 4th quartile of performance and at the same time fail to beat the HSBC benchmark over the three returns evaluation periods. These underperformers are Church Mutual Fund and Pacific Ocean Fixed Income Fund. To some extent, Prudence Fixed Income Fund can be considered a laggard as well due to its 1- and 3-year returns both being in the 4th quartile of performance, but then it managed to get into the 2nd quartile for the 5-year returns.

It is also quite apparent from the Treynor ratios computed that with the exception of Coconut Fixed Income Fund, all other bond funds had negative Treynor ratios for the 3- and 5-year periods. Breaking down the Treynor ratios of the funds into their component parts would show that all these funds with negative Treynor metrics had 3- and 5-year average rolling returns that failed to beat on average a passive investment in 3- and 5-year FXTNs respectively. Hence, the underperformance was driven mostly by negative excess returns. The difficulty of most fund managers in exceeding the yield on 3- and 5-year FXTNs comes from the fact that interest rates 3- and 5-years ago, which form the basis for computing the average risk-free rate for 3- and 5year excess returns, were relatively higher during the time. The underperformance can also be attributed to the effect of a high base in the calculation of portfolio returns. The 3-year rolling returns formula utilizes the months of January to December 2007 as the base months in calculation while the 5-year rolling returns has the months of January to December 2005 as the computational base months. 2005 and 2007 were both relatively good years in the domestic fixed income market, which made NAVPS values of many fixed income funds already high to begin with during these years. 2005 was the final year local fixed income bonds funds were allowed to use accrual accounting as the valuation method for government securities. Because secondary market interest rates in 2005 were trending upward initially due to perceived fiscal risks, government securities were being issued at high rates during the auctions. Since bonds were not being marked-to-market then, fixed income funds were simply accruing the high coupons on the securities they held and were not booking any unrealized losses. Come the end of 2005, interest rates had fallen, and when investments funds were required by the new accounting standards to restate their accrual portfolios in favor of mark-to-market valuation, bond funds immediately realized gains during the migration. This led to 5 out of 9 bond funds having a full-year return that was in double-digit levels. 2007, on the other hand, was the year immediately preceding the global financial crisis. Asset valuations for both the equity and fixed income markets in the Philippines were elevated and brought about high NAVPS values for equity and bond funds alike.

One of the more interesting findings from the study is the fact that only Coconut Fixed Income Fund and Philippine Stock Peso Bond Fund have historically outperformed the HSBC benchmark using the Treynor ratio as the performance metric. For 1-year returns, only Philippine Stock Peso Bond Fund managed to provide systematic risk-adjusted returns higher than the HSBC benchmark. Whereas for the 3- and 5-year returns, only Coconut Fixed Income Fund's characteristic line had a slope that is steeper (hence, a higher risk-adjusted return) than the benchmark. This finding implies that only a minority of Philippine asset managers of fixed income investment companies has been beating their benchmark using market risk as the basis for risk-adjusting their returns.

One must exercise caution, however, when coming to the hasty conclusion that based on the Treynor ratio, Coconut Fixed Income Fund has successfully outperformed the HSBC benchmark. Its returns betas for the 1-, 3-, and 5-year returns are -0.12, 0.48, and -0.10 respectively. These negative and low betas imply that the fund's portfolio manager may have opted to go into asset exposures that are not completely captured by the HSBC index as a performance benchmark. After a thorough investigation of Coconut Fixed Income Fund's asset allocation as of the end of December 2010, it was discovered that this fund is heavily invested in loans (approximately 91% of the fund's NAV), which is not subject to daily mark-to-market revaluation, but rather, generates returns by way of accrual income. This significant allocation to loans may explain the low betas. One important takeaway from this is that the HSBC index may not necessarily be an appropriate performance benchmark for Coconut Fixed Income Fund given its bias towards loan exposures (the index construction/rebalancing criteria of the HSBC Philippines Local Currency Bond Total Return Index do not have a provision for loans). This digression in asset allocation from the index may create comparability issues when evaluating mutual fund performance vis-a-vis the benchmark. Although the risk-adjustment process of the Treynor ratio allows for funds with different styles and investment objectives to be pitted against one another, one must first dig deeper into the portfolio composition of the fund to ascertain its comparability with the benchmark before generalizing that it has indeed outperformed its benchmark after controlling for risk.

Another problem when using the Treynor ratio for appraising mutual funds against a benchmark is that negative betas may lead to incorrect conclusions about risk-adjusted performance. Case-in-point is the 1-year Treynor ratio of Coconut Fixed Income Fund of -0.53, which makes it the worst-performing mutual fund in the sample for this specific time horizon (underperforming the HSBC benchmark as well). This is surprising given that its 3- and 5-year Treynor ratios are both the highest among local bond funds and are better than the benchmark for the same time periods. In fact, looking at the returns alone of Coconut Fixed Income Fund (without any risk-adjustment) for the 1-year time horizon shows that it the best performing fund returns-wise. Indeed, it is the negative beta (-0.12) that deflates the risk-adjusted returns of the fund and complicates the analysis. Reilly & Brown (2009) suggest that when evaluating the performance of a mutual fund with a negative beta, it may be a better approach to compare the realized portfolio return of the fund with its required return based on the CAPM. Implementing this technique would yield a CAPM required return of 2.59%. And since the realized 1-year average rolling return of Coconut Fixed Income Fund was 11.38%, the fund manager has done a superior job.

The Sharpe ratios for the 10 Philippine bond funds in scrutiny were computed and ranked accordingly in Table 3. The process in which the Sharpe metrics were derived for each fund mimics that of the Treynor procedure in that excess returns over a risk-free asset's returns represents the average realized returns of a mutual fund which is subject to an adjustment for uncertainty. For the Sharpe ratio, the recommended proxy for uncertainty is total risk or the standard deviation of the portfolio's historical returns. The results of the Sharpe portfolio performance assessment is consistent with the findings of Treynor in that the fund manager of Coconut Fixed Income Fund has shown stability in terms of providing its clients with the highest risk-adjusted returns (having ranked 1st for the 1- and 3-year returns and 2nd in the 5-year returns category). Philippine Stock Peso Bond Fund was similarly ranked as a consistent top performer using the Sharpe measure, having placed 2nd in both the 1- and 3-year returns classifications.

Observing the results from a broader perspective, one would notice that almost all funds achieved positive Sharpe ratios in the 1-year returns horizon (with the exception of Church Mutual Fund and Prudence Fixed Income Fund). In the year 2010, domestic interest rates in the Philippines dropped to all-time lows due to the market's excess liquidity coupled with the central bank's achievement in keeping inflation well contained. Emerging markets in Asia were likewise a preferred destination of foreign funds flow due to the "Asian growth story" and this contributed to the very upbeat sentiment towards the local stock and bond markets. Only 3 bond funds were able to successfully beat the risk-adjusted returns of the HSBC index, and these were Coconut Fixed Income Fund, Philippine Stock Peso Bond Fund, and Pacific Ocean Fixed Income Fund with Sharpe ratios of 5.25, 2.12, and 2.01 respectively, higher than the total risk-adjusted return of the HSBC benchmark of 1.59. The inability of 2 investment companies to deliver positive risk-adjusted returns is due to their average realized 1-year returns failing to outperform the returns on a simple buy-and-hold strategy on a 1-year T-Bill on average.

For the 3- and 5-year returns computations periods, the results of the Sharpe performance review in general do not differ much from the findings under the Treynor metric in that a majority of the funds in the sample had negative excess returns after accounting for both systematic and unsystematic risk. An analysis of the reasons for the underperformance in the excess returns was tackled in the previous discussion on the Treynor ratio. Only Coconut Fixed Income Fund succeeded in realizing a positive risk-adjusted average 3-year return of 2.44, which was even higher than the 0.82 Sharpe ratio of the HSBC benchmark. After analyzing the 5-year risk-adjusted returns of all funds in the sample, one will observe that only a passive investment strategy of mirroring the returns of the benchmark would have been the only way for one to generate a positive risk-adjusted return.

A consistent underperformer using the Sharpe measure is Church Mutual Fund whose 1-, 3-, and 5-year risk-adjusted returns based on Sharpe's methodology rank it on the 3rd and 4th performance quartile throughout the analysis. This is similar to the findings under the Treynor performance analysis criterion, which puts Church Mutual Fund in the list of underachieving funds. This indicates that the fund manager of Church Mutual Fund was unsuccessful in utilizing his skill to improve returns performance relative to competition and even compared to average hold-to-maturity investments in safe government securities.

Another interesting observation is the long-term total risk-adjusted return of Prudence Fixed Income Fund, which makes it the best performing fund in the 5-year category using the Sharpe ratio. One will notice that this particular fund is ranked last for the 1- and 3-year returns horizons. The explanation for the fund's smallest negative 5-year Sharpe ratio among the 10 bond funds in the sample is the fact that its returns series is the most volatile in the industry. Its total risk or the standard deviation of its 5-year rolling returns is 0.71% compared to the industry average's 0.30%. Its higher total risk measure relative to competition deflates its negative excess returns the least. The fund's average 5-year excess returns of -3.47% is also 3rd best in the industry and was a factor in its superior risk-adjusted performance.

Among the 4 portfolio performance metrics applied in this study, it is only the Jensen's alpha measure that proposes a technique to quantify a mutual fund portfolio's absolute degree of returns outperformance. So far, the Treynor and Sharpe ratios that were computed has allowed the proponent to rank the 10 bond funds and determine their relative performance against one another and vis-a-vis the HSBC benchmark. The alpha coefficient that is obtained per fund using the methodology prescribed by Jensen captures the success of a fund manager in realizing an absolute excess return to fund shareholders that is beyond what is required by the CAPM given the fund's sensitivity to the returns on the HSBC benchmark (see Equation 5).

Table 4 exhibits the Jensen's alpha metrics computed for each of the 10 bond funds in the sample. The alpha coefficients have been rounded off to 4 decimal places to facilitate an easier analysis of the numbers (rounding off to 2 decimal places may make some funds ranked equally). Because the proxy for market returns to be used to compute the required return for each mutual fund is the return on the HSBC index, the alpha value for the HSBC benchmark must obviously be zero (since its beta is equal to 1.0).

By examining the rankings of the bond funds over the 3 returns computation horizons, it is evident again that Coconut Fixed Income Fund has consistently outdone its competitors in the industry, having been ranked 1st in the 1-, 3-, and 5-year returns classifications. Philippine Stock Peso Bond Fund is also one of the top performers, ranking 3rd and 2nd in the 1- and 3-year returns time periods. An underperformer using Jensen's approach is the Sunshine GS Fund, obtaining 4th quartile returns over the short- and medium-term. Prudence Fixed Income Fund similarly showed dismal Jensen's alpha measures over the short- and intermediate-term, giving it the worst ranking among all bond funds in the 1- and 3-year categories. The reason for the low ranking is its excess returns being the lowest in the industry for both 1- and 3-year returns. Notwithstanding its disappointing showing in these categories, its fund manager was somewhat able to redeem itself by realizing a Jensen's alpha that is 3rd best in the industry over a 5-year computation horizon.

Although the returns rankings based on Jensen's measure is a helpful tool in analyzing relative fund performance, the Jensen's alpha is a more meaningful metric when viewed from an absolute returns standpoint. As a matter of fact, there are a number of noteworthy findings from the Jensen's alpha computations in Table 4. First, over the year 2010, only the asset management firms of Coconut Fixed Income Fund, Pacific Ocean Fixed Income Fund, and Philippine Stock Peso Bond Fund managed to realize positive alpha for their clients amounting to 7.14%, 6.46%, and 0.98% respectively. These positive alpha measures suggest that only these 3 funds have outperformed the HSBC benchmark over the past year after considering the market return premium investors must theoretically demand compensation for. As mentioned previously, 2010 was a very good year for the domestic fixed income market as yields on government securities fell to historic lows. The resulting positive alphas of these 3 funds imply that very few Philippine fund managers were able to successfully time the market and increase duration prior to the market upturn and select underpriced bonds for their portfolios based on relative value in a way that would create abnormal returns for their funds. Second, only Coconut Fixed Income Fund has been able to provide positive realized alpha for its clients in the medium-term, as seen in its positive alpha of 1.7% over 3-years. The proponent is reluctant to conclude that Coconut Fixed Income Fund's fund manager was able to fully utilize his skill in market-timing and security selection to generate abnormal returns for its clients over the 1- and 3-year horizon due to the fund's heavy reliance on accrual income from significant loan exposures, as highlighted previously. It is difficult to generalize that a fund manager has outperformed his competition on the basis of his market-timing and security selection skills if the accounting revaluation methodology of his portfolio of assets is different from that of its peers and benchmark. These desirable skills can only be ascertained if the portfolio of assets the fund manager invests in is subject to periodic revaluation via mark-to-market. Lastly, it is quite clear from the 5-year Jensen's alpha coefficients of the Philippine bond funds that none of the fund managers were able to create incremental value for their investors beyond what is expected of them at the very minimum.

The final performance metric to be discussed is the information ratio, which, similar to the Sharpe and Treynor ratios, is a risk-adjusted returns measure that looks at the excess returns of a mutual fund (although relative to a benchmark portfolio in this case) after controlling for active risk or the variability in the returns of a portfolio attributable to its fund manager's efforts to actively (and not passively) manage the fund. Since it is a reward-to-volatility ratio, it can likewise be used to rank mutual funds. Table 5 displays the actual information ratios computed for each fund as well as their respective rankings based on this particular composite performance measure. The information ratio of the HSBC benchmark is obviously zero because the numerator of the ratio, which is returns in excess of the benchmark, would yield a zero value.

An analysis of the information ratios of the 10 Philippine bond funds in Table 5 shows that Coconut Fixed Income Fund once again has been able to sustain its dominance in the mutual funds space by being a top performer in terms of active portfolio management. In fact, Coconut Fixed Income Fund's asset manager was able to generate the highest risk-adjusted returns among all bond funds in the study over a 3- and 5-year horizon. Examining the 1-year risk-adjusted performance of the same group of funds meanwhile shows Coconut Fixed Income Fund comfortably in 2nd place. As expected, funds like Coconut Fixed Income Fund, whose portfolio composition and investment strategy differ substantially from the benchmark utilized in this study, would have a higher tracking error relative to its peers. Specifically, the fund's tracking error over the 1, 3-, and 5-year returns evaluation time frames are 3.11%, 0.77%, and 0.47%, respectively, the 2nd highest across all 3 periods. However, notwithstanding the fund's high cost of active portfolio management (high standard deviation of active returns), its consistent industry-best figures for active returns across the 3 returns horizons result in high information ratios for Coconut Fixed Income Fund.

One observation worth pointing out is the failure of a majority of local fixed income funds to generate non-negative average portfolio returns in excess of the HSBC benchmark. To be more specific, for the 1-year returns category, it is only Sunshine Bond Fund, Coconut Fixed Income Fund, and Pacific Ocean Fixed Income Fund, that have positive excess returns amounting to 1.45%, 2.57%, and 0.68% respectively. And this is the main reason why only these 3 funds have positive information ratios. For the 3-year returns, only Coconut Fixed Income Fund had a positive active return of 1.44%; hence, the 1.87 information ratio. None of the fixed income funds was able to yield a positive information ratio in the 5-year returns classification. The implication of this is that active management by Philippine fixed income fund managers over the long-term has been ineffective in outperforming a passive index-tracking investment. This somewhat corroborates the belief of some that beating the market using active portfolio-tilting strategies over the long-run is a difficult task. Moreover, the negative information ratios for the 5-year returns also suggest that investors cannot seek to be compensated for idiosyncratic risks that can be diversified away by passively positioning in a diversified market portfolio of bonds over a long period of time.

CONCLUSIONS AND RECOMMENDATIONS

Performance evaluation is a critical step in the portfolio management process, particularly in determining whether a fund manager has done his job of creating value for his clients more than what is expected. Financial economics has set forth a number of quantitative tools to assess the performance of mutual funds. These analysis techniques look at the historical excess returns of investment portfolios and adjust them for their respective risk qualities to facilitate an apples-to-apples comparison among funds in the investment universe. The four composite measures of portfolio performance that were applied in this study were the Treynor, Sharpe, Jensen, and information ratio risk-adjusted returns metrics. The primary advantage of using these risk-adjusted returns ratios is that they allow analysts and investors to rank mutual funds, and for some of these composite measures, quantify an absolute amount of realized alpha, or abnormal returns. The objective of this research undertaking is to determine whether Philippine fixed income fund managers have been realizing alpha for mutual fund shareholders and to discover which among the local asset management firms have consistently been outperforming or trailing the benchmark or market index.

After having applied all 4 portfolio performance measurement criteria to the returns on Philippine bonds funds over a 1-, 3-, and 5-year evaluation horizon, the proponent was able to identify consistently performing and underperforming mutual funds, both in terms of relative ranking as well as absolute returns performance. Moreover, although the 4 performance appraisal tools provide different approaches to controlling mutual fund returns for uncertainty and for determining relative or absolute outperformance, a number of industry-wide generalizations were arrived at based on the consistent results of the performance metrics. Firstly, after examining the 1-year risk-adjusted returns of fixed income investment companies in the Philippines, it is clear from the numbers that a majority of these funds on average have been able to beat the returns on a passive buy-and-hold T-Bill position. However, after observing the uncertainty-controlled returns of these same mutual funds over the intermediate- and long-term, most local fund managers seem to encounter difficulty on average outperforming the returns on a hold-to-maturity FXTN investment; hence, negative excess returns. These have led to mostly negative Treynor and Sharpe metrics over the 3- and 5-year periods.

Relative to the returns on the market index used by most Philippine fund managers to benchmark fixed income portfolio performance against, the returns on bond funds pale in comparison whether it be in the short-, medium-, or long-run, as seen in the mostly negative active returns and information ratios of these portfolios. Although there are a few fund managers that were able to provide returns in excess of the benchmark over the 1-, and 3-year time periods, these were merely a handful, indicating a general difficulty by asset management firms to generate alpha by focusing on non-market sources of return. This struggle to produce returns via active portfolio management is even more pronounced when viewed in the long-run, where none of the fund managers was successful in beating the benchmark return on a risk-adjusted basis. This finding is coherent with the notion that it is impossible to beat the market in the long-run.

From an absolute excess returns standpoint, the Jensen's alpha measure allowed the proponent to determine whether fixed income fund managers in the Philippines were optimizing their skill in predicting market turns and identifying mispriced bonds to beat the market. With the exception of just 3 funds, none of the other investment companies had positive Jensen's alphas over a 1-year period. And similar to the results of the information ratio analysis, none of the fund managers generated a positive alpha in the long-run, once again validating the perception that it is a daunting task to produce abnormal returns on average over long periods of time.

One of the major shortcomings of the local mutual funds association is its categorization of fixed income funds with different investment styles and portfolio revaluation policies within the same peer group. As observed, funds with significant accrual exposures are compared directly to other funds that have heavy investments in securities designated at fair value. This gives rise to issues in comparability and may render the results of the risk-adjusted performance analysis invalid, especially if these differences are not recognized or highlighted in the appraisal. A workaround to this problem for future studies of similar nature would be to purposively select a sample of funds with similar investment attributes. Although the risk-adjustment process of the composite portfolio performance measures presented in this study are intended to make funds comparable regardless of their differences in risk exposure, performance analysts in the industry have the tendency to compare funds with similar styles.

For future studies, the proponent recommends the use of regression analysis to estimate some of the inputs to the performance metrics, including beta for the Treynor ratio and Jensen's alpha. Alpha estimation using regression is likewise highly recommended for future researchers to ensure that there is enough statistical basis in claiming that a fund manager has indeed shown skill in generating abnormal returns for his clients. Since Jensen's alpha was applied in this study as an absolute measure of abnormal returns, future studies may consider risk-adjusting the alpha measure by way of dividing the computed Jensen's alpha metric with the beta coefficient used, similar to the risk-adjustment procedure under the Treynor performance criterion. Using a multi-factor model of risk and return such as the APT could also be considered in lieu of the CAPM to proxy for the risk premium in the Jensen's alpha methodology.

As for the risk-free rate, the fixing yield on Philippine government securities was used in this research as the representation for the riskless asset return. Future researchers on fund performance assessment may want to consider using the reference rate on done transactions in the secondary government securities market in lieu of the fixing rate. It is noteworthy that both fixing yields and reference yields on government securities are stated gross of applicable withholding taxes in the Philippines. Since the NAVPS of mutual funds from which the rolling returns are derived are reported net of all relevant administrative costs and management fees, future studies could consider restating the fixing or reference yields net of the withholding tax.

Benchmark selection is likewise a major issue in the asset management industry, particularly on the aspect of performance assessment. As discussed previously, drawing conclusions about an asset manager's ability to generate abnormal returns may be misleading if an appropriate benchmark is not used. Purposive sampling or the use of a fund-specific benchmark may address the benchmark selection problem though this comes at the price of not being able to truly determine absolute returns outperformance relative to an index. The proponent suggests that for relative ranking performance appraisal criteria such as the Treynor and information ratios, a fund-specific benchmark be used, whereas for the Jensen's alpha metric, a single index must be utilized.

Sample-wise, future research initiatives focusing on investment fund performance appraisal in the Philippines could be applied to other sub-sectors within the local mutual fund industry. This study concentrated on risk-adjusted returns evaluation for Philippine Peso-denominated bond funds that are members of the local mutual funds association. Money market mutual funds, Dollar-denominated bond funds, and global bond funds can also be assessed using the methodology presented in this research paper. The composite portfolio performance measures could likewise be applied to Philippine fixed income UITFs. An interesting research endeavor would be to implement a comparison of the risk-adjusted returns performance of Philippine fixed income mutual funds versus UITFs as competing asset classes.

REFERENCES

Brown, K.C. & F.K. Reilly (2009). Analysis of Investments and Management of Portfolios (Ninth Edition). Canada: South-Western Cengage Learning.

Bodie, Z., A. Kane, & A.J. Marcus (2009). Investments (Eighth Edition). New York, NY: McGraw-Hill/Irwin.

Chua, E.G., J.G. Geocaniga, N. Pena, A.G. Yambao (2008). Evaluating Data Envelopment Analysis as an Alternative Fund Performance Appraisal Tool to Traditional Composite Measures of Portfolio Performance: A Philippine Mutual Fund Industry Risk-Return Analysis for the Years 2003-2007. Unpublished undergraduate thesis, De La Salle University-Manila.

Cocolife Asset Management Company, Inc. Retrieved January 1, 2011, from http://www.cocolifeasset.i.ph

Elton, E.J., M.J. Gruber, S.J. Brown, & W.N. Goetzmann (2011). Modern Portfolio Theory and Investment Analysis (Eighth Edition). Hoboken, NJ: John Wiley & Sons (Asia) Pte Ltd.

Investment Company Association of the Philippines. Retrieved January 1, 2011, from http://www.icap.com.ph

Unit Investment Trust Fund Online. Retrieved January 1, 2011, from http://www.uitf.com.ph

Valderrama, H.S. (2003). Philippine Mutual Funds and Common Trust Funds: How Different Are They? Retrieved January 1, 2011 from http://www.upd.edu.ph/~cba/docs/hsv_ctf_study.pdf

Valderrama H.S. & C.C. Bautista (2003). The Development and Performance of the Philippine Mutual Funds Industry. Retrieved January 1, 2011 from http://www.upd.edu.ph/~cba/research.htm

Clive Manuel O. Wee Sit, De La Salle University-Manila

Table 1: Philippine Fixed Income Funds--Bond Funds Primarily Invested In Peso Securities December 31, 2010 Bond Fund Asset Manager Net Asset Market Value (PHP Share Millions) (% of Industry NAV) Ayala Peso Bond Number 3 30,164.42 73.93 Fund Bank--Trust Depart. Coconut Fixed Coconut Asset 342.58 0.84 Income Fund Management Church Mutual Number 3 145.98 0.36 Fund Bank--Trust Depart. Education Fixed Number 2 Bank 998.14 2.45 Income Fund Subsidiary Pacific Ocean Pacific Ocean 342.86 0.84 Fixed Income Fund Asset Management Philippine Philippine 3,786.45 9.28 American Bond American Asset Fund Mgt Philippine Stock Philippine Stock 548.83 1.35 Peso Bond Fund Management Prudence Fixed German 287.94 0.71 Income Fund Bank--Trust Depart. Sunshine Bond Sunshine Asset 3,290.29 8.06 Fund Management Sunshine GS Fund Sunshine Asset 895.94 2.20 Management Table 2: Treynor Portfolio Performance Ratios December 31, 2010 Bond Fund 1-year 1-year 3-year Rank Ayala Peso Bond Fund 0.02 6 -0.02 Coconut Fixed Income Fund -0.53 10 0.04 Church Mutual Fund 0.00 7 -0.08 Education Fixed Income Fund 0.03 2 -0.01 Pacific Ocean Fixed Income Fund -0.11 9 -0.02 Philippine American Bond Fund 0.02 5 -0.03 Philippine Stock Peso Bond Fund 0.06 1 -0.01 Prudence Fixed Income Fund -0.04 8 -0.49 Sunshine Bond Fund 0.03 3 -0.02 Sunshine GS Fund 0.02 4 -0.03 HSBC Philippines Local Currency Bond Total Return 0.04 0.01 Bond Fund 3-year 5-year 5-year Rank Rank Ayala Peso Bond Fund 4 -0.09 3 Coconut Fixed Income Fund 1 0.25 1 Church Mutual Fund 9 -0.31 7 Education Fixed Income Fund 3 NA NA Pacific Ocean Fixed Income Fund 6 -0.17 6 Philippine American Bond Fund 8 -0.06 2 Philippine Stock Peso Bond Fund 2 NA NA Prudence Fixed Income Fund 10 -0.13 5 Sunshine Bond Fund 5 -0.10 4 Sunshine GS Fund 7 NA NA HSBC Philippines Local Currency Bond Total Return 0.00 Table 3: Sharpe Portfolio Performance Ratios December 31, 2010 Bond Fund 1-year 1-year 3-year Rank Ayala Peso Bond Fund 0.68 8 -1.30 Coconut Fixed Income Fund 5.25 1 2.44 Church Mutual Fund -0.06 9 -5.22 Education Fixed Income Fund 1.00 5 -1.28 Pacific Ocean Fixed Income Fund 2.01 3 -1.81 Philippine American Bond Fund 0.76 7 -2.87 Philippine Stock Peso Bond Fund 2.12 2 -1.09 Prudence Fixed Income Fund -1.51 10 -10.16 Sunshine Bond Fund 1.02 4 -1.94 Sunshine GS Fund 0.79 6 -2.13 HSBC Philippines Local Currency Bond Total Return 1.59 0.82 Bond Fund 3-year 5-year 5-year Rank Rank Ayala Peso Bond Fund 4 -17.20 4 Coconut Fixed Income Fund 1 -7.45 2 Church Mutual Fund 9 -27.72 7 Education Fixed Income Fund 3 NA NA Pacific Ocean Fixed Income Fund 5 -27.02 6 Philippine American Bond Fund 8 -12.90 3 Philippine Stock Peso Bond Fund 2 NA NA Prudence Fixed Income Fund 10 -4.92 1 Sunshine Bond Fund 6 -20.64 5 Sunshine GS Fund 7 NA NA HSBC Philippines Local Currency Bond Total Return 1.25 Table 4: Jensen Portfolio Performance Ratios December 31, 2010 Bond Fund 1-year 1-year 3-year Rank Ayala Peso Bond Fund -0.0155 8 -0.0106 Coconut Fixed Income Fund 0.0714 1 0.017 Church Mutual Fund -0.0134 7 -0.0276 Education Fixed Income Fund -0.0102 6 -0.0175 Pacific Ocean Fixed Income Fund 0.0646 2 -0.0177 Philippine American Bond Fund -0.0086 5 -0.0157 Philippine Stock Peso Bond Fund 0.0098 3 -0.0101 Prudence Fixed Income Fund -0.0431 10 -0.0365 Sunshine Bond Fund -0.0062 4 -0.0241 Sunshine GS Fund -0.0158 9 -0.0314 HSBC Philippines Local Currency Bond Total Return 0.00 0.00 Bond Fund 3-year 5-year 5-year Rank Rank Ayala Peso Bond Fund 3 -0.0338 4 Coconut Fixed Income Fund 1 -0.0255 1 Church Mutual Fund 8 -0.0389 5 Education Fixed Income Fund 5 NA NA Pacific Ocean Fixed Income Fund 6 -0.0504 7 Philippine American Bond Fund 4 -0.0292 2 Philippine Stock Peso Bond Fund 2 NA NA Prudence Fixed Income Fund 10 -0.0322 3 Sunshine Bond Fund 7 -0.0411 6 Sunshine GS Fund 9 NA NA HSBC Philippines Local Currency Bond Total Return 0.00 Table 5: Information Ratio Portfolio Performance Metrics December 31, 2010 Bond Fund 1-year 1-year 3-year Rank Ayala Peso Bond Fund -4.03 8 -2.80 Coconut Fixed Income Fund 0.83 2 1.87 Church Mutual Fund -2.14 6 -4.06 Education Fixed Income Fund -1.64 5 -8.18 Pacific Ocean Fixed Income Fund 0.16 3 -3.76 Philippine American Bond Fund -2.37 7 -3.37 Philippine Stock Peso Bond Fund -0.91 4 -3.06 Prudence Fixed Income Fund -5.36 10 -4.13 Sunshine Bond Fund 1.26 1 -11.41 Sunshine GS Fund -4.42 9 -6.98 HSBC Philippines Local Currency Bond Total Return 0.00 0.00 Bond Fund 3-year 5-year 5-year Rank Rank Ayala Peso Bond Fund 2 -12.66 5 Coconut Fixed Income Fund 1 -3.34 1 Church Mutual Fund 6 -10.64 3 Education Fixed Income Fund 9 NA NA Pacific Ocean Fixed Income Fund 5 -7.19 6 Philippine American Bond Fund 4 -10.80 4 Philippine Stock Peso Bond Fund 3 NA NA Prudence Fixed Income Fund 7 -3.52 2 Sunshine Bond Fund 10 -21.23 7 Sunshine GS Fund 8 NA NA HSBC Philippines Local Currency Bond Total Return 0.00

Printer friendly Cite/link Email Feedback | |

Author: | Sit, Clive Manuel O. Wee |
---|---|

Publication: | Journal of International Business Research |

Geographic Code: | 9PHIL |

Date: | Dec 1, 2011 |

Words: | 11209 |

Previous Article: | Conflicts and sales information transmission across functional boundaries. |

Next Article: | Business ethics in Japan: taking a closer look at the role of age. |

Topics: |