Management costs in relation to investment performance

Abstract

Pension fund in the recent years has gained significant attention due to the controversial performance level and controversies involved. With as many as 40% of companies closing down their pension schemes and the pensioners opting out of state pensions, pension fund sector has faced a set back. Despite this fact however, more and more capital inflows have been observed in the managed fund industry with changing investment climate. Observing the institutionalization of the pension fund sector as banks, insurance companies, real estates and the like jumping in to have a bite of the investment share, the researcher is of the opinion that there is great potential in this field. However, the researcher is also cautious of the fact that like all industries and sectors, pension fund investment is most likely subject to market conditions. In particular interest to the researcher is the fact that most of these institutions have taken over the task of the individual investors and offer services for managing their funds at a "small" fee. The researcher is concerned whether the management fee is congruent with the performance of the investment and, whether individual investors should continue to give such high fees for returns on their investments. In the following research the researcher carries out a quantitative and qualitative analysis of these facts based on data derived from Standard Life pension funds, and the results therein are tabulated to reflect the research issues mentioned above.

Table of Content

Chapter 1: Introduction

Background

Statement of the problem

Hypothesis

Scope and advantages of the study

Definition of terms

Plan of research

Chapter 2: Literature Review

- Investment goals and opportunities

- Pension Fund

- Investment Performance Measurement and market efficiency

- Risks

- Costs, price and performance

Chapter 3: Research

Research Design

Methodology

Data Limitation of Analysis

Data collection

Data Analysis

Chapter 4: Results and discussion

Quantitative Results and Discussion

Qualitative Results and Discussion

Chapter 5: Conclusion and recommendation

Bibliography

Appendices

Background

The managed fund industry is one of the most lucrative industries in the UK and Europe. It involves a host of middlemen, closed end and open end investors, and institution investors, and constitutes about 20 percent of the UK corporate equity (Myners Report 2001). The nature of competition as well as the retail based industry environment have deemed the funds industry especially private funds one of the most lucrative arena in asset management. The aim of these managed funds is to provide service to institutional investors such as insurance, asset management companies, and private firms and their employees to contribute significant amount by investing in the capital market. In return the institutional and private investors are charged certain fees for management expertise, advice and investment returns. The provision of management for pension products for instance involves managers of departments of banks, securities houses, insurance companies and large pension funds companies who specialize in investment management by determining the kind of assets to be allocated, categorized into liquid or selling capitals on behalf of the institutional investors. This trend is prevalent in most types of investments. Asset management can be categorized into retail - management of assets such as mutual funds, personal pensions or real estates; and wholesale which involves the management of assets on behalf of institutions such as pension funds of firms. Asset management is further divided into generic, specialized and balanced which stands for quantitative and nondiscretionary exercises; carrying out discretionary security selection; and discretionary asset management based on allocation and selection respectively. In this regard asset managers carry out back office and front office functions which include marketing and development of products; fund management; strategies; transaction processing and settlement; systems support; accounting and administration. The degree of difficulty in these functions differs from the kind of assets and from region to region. Consequently fund managers typically offer a bevy of services such as execution of trade processing; investment advice, custody and marketing to their clients at a bundled price (British Invisibles 1997). According to Davis and Steil (2001) asset management can be analyzed according to the nature of competition and the structure of management performance such as under monopolistic, oliogopolistic and monopolistically competitive or perfectly competitive. Fees from the competitive point of view will take into account of sunk costs, potential competition and strategic interactions of firms as well as markets and the strategic competition prevalent. Typically, Davis and Steil indicate the majority of the retail asset management sector is governed by monopolistic competition with few economies of scale, many firms' involvement and homogenous products. Consequently no supernormal profits exist (Davis and Steil 2001).

On the other hand the wholesale asset management scenario is competitive in the sense that they have few firms, has economies of scale but competitive behaviour is dependent on the potential of the competitors. In this sector too there are no supernormal profits. Considering the low entry barriers and the homogenous characteristics of products, competition therefore is dependent on returns, reputation with customers and reputation of the managing firms. The benefits of size and reputation also offset low operating costs which in turn impact management cost. As a result, the open to all competition has given rise to entry of institutions from various backgrounds such as commercial banks, investment banks and private fund institutions.

Pension fund institutions being part of this market has over the year gained much recognition for its size and lucrative opportunities. In Europe for example pension funds for the year 2000 account for 30% of GDP compared to 50% insurance assets and 40% investment funds (Davis and Steil 2001).

Given the rising scope of pension funds the trend towards real estate assets, pay-as-you-go pension systems, and defined benefit to defined contribution pensions in the private sector has increased the importance of pension funds sector. The dominance of defined benefit occupational pensions for instance has increased the opportunities in portfolio based pension funds to more diversified options such as equity and foreign assets apart from local securities (Davis 2002). Furthermore pension savings is considered to be more attractive to new generation retirees who regard it as a source for high investment returns thereby boosting the industry in general. The scope of pension fund management are however limited by the relative risk of taxation imposed on equities; cost based accounting systems; portfolio regulations; and investor experience in equity risks (Jarion and Goetzmann 1999; Davis 2002). Though the presence of international investment opportunities has perked interests of many investment organizations and investors, nevertheless the high currency risk, liquidity risk and lack of information on investment have directed investors to settle for local based investment opportunities thereby increasing the attractiveness of the local equities.

Asset managers tend to be balanced in this industry offering portfolios rather than specializing in certain type of funds. Pension fund investors' relation with clients is therefore based on the institutional link. In most of the cases the bank of the pension fund sponsor are usually the investment managers regardless of the performance level. For this reason most investors insure their funds and expect at least 4% in returns, a standard approximation. Manager selection process in the private sector therefore differs from investor to investor. In the UK the managers are often consultants who demonstrate their expertise through experience, academic qualification and finance background. The competition therefore is inherent in the performance and their relationship with the fund institution or investors who would select and recommend them to others.

Apart from this fact, fund managers are also gauged based on their performance and profitability. In large firms it is expected that pension managers bring in cumulative 25 percent higher investment returns as compared to smaller firms. This is because of the contribution size as well as the investment cost incur for managing the assets increases with more investments and clients coming their way. The cost involved in management therefore determines the efficiency of the institutions as well as the fund managers'.

The determination of management fees likewise is incorporated in the cost as it has been observed that pension fund asset management cost may be high or low depending on the asset allocation, hidden charges and the type of assets invested, with bonds being the cheapest and international equity the highest. Such allocation also affects competition among fund managers as they are offset by costs of assets under management. The variance of fees and management cost therefore impact the market power and its performance level (McKinsey 2000).

Considering this complex issue of hidden fees, balanced costs and management charges all indicate that there is no clear demarcation or allocation of costs where fund management is concerned. Fund managers therefore set prices of equities, transaction costs, safekeeping costs and management costs according to their reputation, service level, portfolio allocation and a host of other factors. Performance measurement among pension fund managers therefore is difficult to gauge as most fund managers offer services in bulk as retail and wholesale services to institutions, banks and individual investors. Only the investment management firm measure its own performance with the view to estimate its revenue. Low importance is given to performance measurement of asset managers is also due to the importance of asset returns prescribed at certain percentage according to the commonality of asset allocation and returns level. Instead importance is given in risk control and also the reputation of the representative of investment management firms.

Statement of the problem

Given the above background it can be observed that pension fund management and fund performance is largely dependent on a host of issues and factors that may or may not contribute to higher returns. In fact there is no platform upon which firms or investors can measure the performance of managers therefore they depend on these experts to determine the level of returns from investments. Individual investors and institutional alike are not only motivated by high returns but also funds' performance over a long period of time making it prudent that investment managers are efficient and accurate in their performance.

Related to this is the fact that the larger the firm the more experts and consultants hired for the purpose of advising clients and selling the homogenous products. The expertise and wise decisions of these consultants help investors in gaining expected returns. Risk therefore is involved not only in the allocation of assets but also in the choice of managers. The management cost should be reflective of this fact rather than mere reputation and recommendation.

For this reason the researcher is of the opinion that a research should be conducted to establish whether is a correlation between fund managers and fund performance. More importantly the researcher is interested in knowing the level of correlation between the management cost and its relation to investment performance.

Hypothesis

The above rationale gives the researcher the opportunity to pose the follow hypothesis which will be analyzed in the following chapters, and tested against correlation models and theoretical framework:

H0: Investment fund performance does not affect the cost of management.

H1: Management cost affect fund price.

H2. Management cost affects fund efficiency in terms of returns.

Scope and advantages of the study

This research aims to help professionals and students alike in sharing the information on investment pension fund in general and more specifically how management cost can affect fund's performance level, whether there is a measure for the performance of the investment fund managers. The study would entail enumeration of facts relating to pension fund management, its costs, dimensions and relations with investors, private and institutions as well as with the industry and the economy at large. Students may benefit from the findings of this research as it offers a comprehensive overview of the workings of funds management industry, its players and the importance to the economy. Furthermore, for the future researchers may use this platform to research other fund sectors such as technology, imports, telecom, and real estate. The basis of this research can also be used for initiating further research in investment management as a management field. On the other hand professionals are likely to find the research useful in studying the trend and importance of management performance and how it has affected their industry. The realization of the impact of efficiency in management performance and its impact on investment institutions and industrial performance would not only help professionals especially firms' top management to tailor cost structure, allocate expenses and fees accordingly but also enable them to know how they could better the performance of their people.

The research is of the opinion that though other researchers have included studies on managed funds, high risk funds and real estates but few have dedicated studies in pension funds. Consequently, the academia too will benefit from the study as they will have the opportunity to view the potential of pension funds as a separate segment of investment management studies.

Definition of terms

Returns or rate of returns - "The gain or loss of an investment over a specified period of time, expressed as a percentage increase over the initial investment cost." (INvestopedia.com)

Abnormal return - "When return on an asset is in excess of the expected rate of return" (Investopedia.com).

Management fee - A fixed amount investment fund manager charges investors or institutions for services rendered (Investopedia.com)

Transaction cost - Cost incurred when purchasing or selling assets including broker's commissions and spreads (Investopedia.com).

Market Efficiency - Price equal value meaning the degree to which stock prices are reflective of its value and in relevant information (INvestopedia.com)

Plan of research

In the following research, the researcher plans to outline the importance of the pension fund as investment fund. Moreover, the concern is to establish whether there is a correlation between pension fund performance and its cost of investment. To achieve these objectives, the researcher plans to carry out three stage research. In the first stage the researcher shall carry out a theoretical framework through a comprehensive literature review outlining the factors under study namely investment goals, opportunities, pension fund sector, performance measurement and evaluation etc. This is followed by an analysis of data on pension funds - their price, performance and cost correlation. Then, the researcher will combine the theory with the research analysis to come to conclusive results.

- Investment goals and opportunities

The objective of investment is to maximize return on investment capital. The premise is to maximize investor returns and wealth assuming that the individual marginal utility of wealth declines with rising level of consumption. Therefore its performance is also dependent on the utility of the performance of the investment. Measurement of investment is given by return on investment over a time period against the total amount received in return for investment less the amount paid for investment. This equation includes the principal amount, along with the returns and cost of investment.

Return on investments is also sensitive to unsystematic events comprising of company specific performance or local economic boom of a particular industry, or systematic global factors such as worldwide recession and the like. Depending on the events, it may cause a business condition that would make the valuation of the investment uncertain. For this reason, it is important that investment should be measured over a time period to gauge its trend to compute average returns. Considering different assets react different to these events, their impact on the average returns also differ. For some assets returns can be gauged while others cannot as returns on wealth are not always certain. Investors consequently need to analyze the risk on returns from the investments to gauge the kind of loss they may incur. Characteristically investors are risk averse and they tend to avoid high risk on investment wherever possible which is why high risk investments tend to have higher rates of returns to motivate investors to take risks with their money. However, some investors are tolerant of risks and they are willing to pay a high price with the view to have high returns and to avoid tying up their capital for a long duration of time. There are others who are content to have some high risk and some low risk investments, by following a mediocre strategy. Overall their goal is to maximize returns and minimize risk. Balancing the two is the key to successful investment which requires great expertise and experience. For this reason in the recent years investors have largely depended on institutions for the management of their capitals (Murphy 2000).

The goal of the institutions is to act directly or indirectly in the best interests of the investors. The "investor" may be individuals, financial institutions or other organizations that are responsible for making the investment decisions and transactions. These institutions use a process of distribution of investment funds into fixed income bonds; equities or stocks, real assets estate and commodities; and foreign stocks and bonds. The estimated distribution is as follows:

Bonds 20 %

Equities 20 %

Foreign bonds 30%

Foreign equities 30%

The above is an estimation of the distribution of investment which may change according to the market value of the investment locally and globally. The basic premise of having institutions to deal with the investments is because they minimize risks of investing the whole investment amount into a particular category and let the investor enjoy balanced risk at high investment returns. Consequently, institutions play an important role in investment valuation and determining the cost of investment (Murphy 2000).

Another reason why investors tend to rely on institutions to achieve their return targets is because they feel they do not possess the right kind of expertise to find the right price for their investments. Some investors may invest in categories of investment with low prices and long duration due to their limited expertise while others tend to invest too much in certain areas thereby lose out in investment opportunities. To optimize investment, invest managers and advisors are needed who have the expertise, know-how to achieve economies of scale and scope. They actively manage mutual funds, subscribe to investment newsletter and are full service brokers of financial institutions who are able to better provide the necessary advice for the investors to make profitable decisions, avoid taxation and high transaction costs (Coggin, Fabozzi and Rahman 1993). As Dunne and Davies (2005) of association for Institutional Multi-manager Investing and Attica Asset Management respectively opine:

..."consultants use asset models in order to determine the most appropriate combination of asset classes required to meet the expected growth in a scheme's liabilities. The results are hugely reliant on the assumptions made on the return, risk and correlation of these different asset classes."

Moreover, Dunne and Davies also present the fact that fund management is neither a science nor an art. The real power of portfolio management is how it can complement the art of good judgement. It should reflect the manager's ability to process information and take into account of human judgement to decide what to do with the information and resources at hand. Thus they feel that the quantitative managers may represent the scientific approach while the quant managers follow the artistic sense yet it cannot be said that investment decision and processes is purely scientific or art. As investment managers tend to make greater use of quant decision making processes it creates more opportunities to gamble with the market. In the end however, a wise decision is not a play between art and science but rather on the ability of managers to make objective decisions without any biases (Dunne and Davies 2005).

- Pension Fund

The pension industry across the world and especially in the UK has been unexpectedly set back by an aggregate shortfall in assets approximating 200 billion pounds. This has caused a series of events that have de-motivated equity investors. It has been argued that pension funds is responsible for making the economy less efficient and weaken the financial system. This is because the irregular flow of savings into capital markets affecting the prices has made it inefficient in terms of returns on investments. They have cause for concern as Kat et al (2003) note:

"Pension funds are also confronted with the fact that the asset classes traditionally deployed to generate excess return over the risk-free rate are now perceived to contain less premium and more risk than was the case in the last two decades. Recent studies have shed new light on the long accepted equity risk premium level of 4-5%. This historical estimate appears to be biased by measurement and estimation errors and some researchers claim that the actual equity risk premium over the past century might have been closer to 3% with little reason for the expected future risk premium to be any higher than that."

This downward trend has been the result of closures of pension fund schemes by the majority of around 40% of the companies in the UK including household names like Barclays, Marks and Spencer, and Tesco at the beginning of the twenty first century (Kat et al 2003). This catastrophic effect on the industry is further negated by the usual economic trend of labour employment trend, pension fund savings and the appropriate form of savings.

Alternatively, pension funds have gained great support from the finance market due to its flexibility, people's openness to investment and magnitude of contributions to the securities market. Kat et al (2003) have intimated that the potential of pension fund investment improvement based on return-oriented assets and concentration on hiring of active outside fund managers who can introduce different strategies for investment add cost not value or returns. They do not guarantee performance which has resulted in aggregate rise in investment fees rather than performance level. However, there is evidence that indicate otherwise (Kat et al 2003).

The relationship between asset classes through diversification benefits can reduce risks and derive opportunities from alternative equity markets (Kat et al 2003). The contributions in the form of investments are being used to buy real estates, bonds, company shares and international stocks. Unlike its predecessors today pension is more about channelling and inflating securities markets through buy and sell activities, and portfolio investments. As a result the net inflows for pension funds in securities has driven up prices in the securities market creating high demand for managed assets. In the UK the Industrial Ordinary Share Index increased significantly during the 1990s when pension fund schemes have been included in the mainstream of the market. What this does to the economy is drive the prices of security up, yields comes down (Toporowski 2000).

Alternatively, pension fund schemes also have built up more contributions among employees of firms; raised the investment level in the finance market for developing businesses and increased the level of investment funds among institution investors and individual investors. This change in the financing market has given rise to investor's appeal to the pension fund market because it allows employees tax benefits and to opt out of state pension fund (Toporowski 2000).

Similarly, pension fund schemes and the introduction of fund institutions have given rise to immature funded pension schemes. These schemes usually allow the investors to invest in the market which is a better way of gaining returns from funds as it does not involve paying high interest rates on borrowed funds neither do they have to be dependent on the banks for loans or issuance of investment capital. This approach has improved the quality of investment and profitability in operation (Toporowski 2000). Furthermore, the new trend of qualified fund managers with their expertise channel pension funds into the market without exposing the individual investors or institutions they represent making investment safe and protective from the brokers. They are also responsible for making decisions and advising investors by assuring high returns and profitability. With less hassle of payments of interest or time spent in banks, loan institutions, individuals are finding pension funds a better way to inflate their capital through returns in the long run. Although the managers come with a price nevertheless the fund is secured (Toporowski 2000).

- Investment Performance Measurement and market efficiency

Investors depending on their skills, qualification and market knowledge tend to value an asset differently because of the subjective interpretation of information and different expectations of uncertainty. The lack of or poor information may lead investors to interpret data differently which affect trading decisions. Similarly, valuation differences may also occur when investors make decisions pertaining to the performance level of the assets. This is because transaction prices tend to deviate due to the intrinsic value of the assets which may or may not come across as positive to the investors. Opportunities for abnormal profits exist for investors but only for those who are skilled at information processing. Some use the traditional methods while others may depend on information technologies. Nevertheless the basic premise is that investors with quality information, analytical skills and set of expectations may be able to outperform those with lesser experience and quality information.

Having said that, it must be noted that the performance measurement for investors cannot be gauged by their background nor can it be gauged by the market prices. Boudreaux (1973) argues that the divergence of price from net asset value is consistent with the market efficiency. Moreover market efficiency is the result of the price set by the selling and buying behaviour but it does not mean the performance of the asset or its intrinsic value. In this context when market prices approximate the intrinsic value and there are no longer opportunities for abnormal profits, the market is defined as efficient and can be denoted by:

V=P and E (R j )=i

There are two level of market efficiency namely semi strong and strong according to Murphy (2000). These two levels can be defined as:

With semi-strong market efficiency, abnormal profits are not obtainable using all public information.

With strong market efficiency, abnormal profits are not obtainable using any information, including inside information not made public.

Considering the volatile nature of the prices and market efficiency, it is imperative that investors test the market before interpreting investment performance. Market efficiency can be interpreted by testing the performance level through a set of assets, portfolios, strategies or managers by keeping any of these factors constant and testing the others. The performance level can be gauged by the level of risk versus the return and its relation with the price of the assets. Efficiency or inefficiency is evident in a test which shows that investment managers sustain a risk-return relationship and how the asset has achieved the investment goals (Tonks 2001; Murphy 2000).

Experts like Sharpe (1966) have devised measurements of performance by geometric expressions. These methods may not be efficient for measuring performance of particular market for particular kind of asset. However, it is one aspect of measurement. The issue here is that investors are of the belief that if markets are inefficient then abnormal profits opportunities exists whereas profits may take a long time to materialize for asset returns if the interpretation has been incorrect (Murphy 2000).

Alternatively, performance measurement of investment may also be carried out by estimating mispriced stocks. These may result in abnormal profits. Yet only experts and institutions can make decision such as these to generate high rate of returns. Consequently, investors and managers alike do not like to depend on performance measures to make high level risk decisions. Instead, they rely on comparison with peer groups such as using geometric mean of compound rate of return. The average can then be compared with the speciality fund etc.(Murphy 2000).

In this context market efficiency is dependent on a number of factors that would eventually result in the right or wrong decision for investment. For this Hodgson et al (2000) define investment efficiency in a more conservative manner as:

"...a function of the risk, return and total cost of investment management, subject to the constraints within which investors must operate. These constraints include financial elements and non-financial elements such as an investor's time available to manage the investment arrangements, accountability as fiduciary, or legislative requirements. Investment efficiency should therefore be regarded as a combination of financial efficiency and non-financial efficiency."

On the other hand Gruber in 1996 has demonstrated that investors can and do identify superior managers with direct increment when the funds are managed better. Zheng (1999) another researcher has also examined the significance of Gruber's smart money effect and confirms that the existence of fund selection ability greatly influence the resulting performance of the fund.

A third concern is the pricing efficiency. Pricing efficiency depends on the liquidity level of the market while market efficiency relies on quality analysis and active trading. A failure to detect these fundamental factors can lead to loss of profit opportunities as the investors would be misleading value and price to the clients. Mispricing can especially be detrimental in term of limiting the scope of investors and tying up of capital. Often, it can also be observed that the direction and magnitude of returns depend on the recommendation of official institutions like the IPO companies or bank but they do not guarantee its performance. The asset price may be under or over priced that result in negative abnormal returns. For this reason, market efficiency may not be dependent on primary market issuers and its traders alone but it should also refer to secondary markets. The transaction costs likewise should be borne by the issuer rather than the investor. The rationale is that various secondary investment markets may indicate factors that contribute to market efficiency by incorporating fundamental information of the price, the market nature and trading activities that help in advisory information. For instance risks in accounting procedures, stock splits or dividend issuance etc. may increase the level of the asset value and help in determining purchase or selling decisions (Murphy 2000; Fama 1991).

Another aspect is the fact that returns also depend on the past data and efficiency. The trend of time periods can help one to gauge the behaviour of the asset, its value and past prices which may help in determining the picking up trend (Olsen et al 1998). But it should also be noted that past trend cannot be used as the absolute source for future prediction. Another exception is the fact that lower price earning ratio stocks tend to outperform higher P/E ratios stocks because these can adjust risks. Alternatively bonds act in the opposite direction, indicating that no asset type or category perform in the same manner for a particular market; its prices, performance measurement and strategies should therefore be based on some other technique to make effective decisions (Murphy 2000).

- Risks

Since investment involves cash resources, investors tend to evaluate the balance of the portfolio risk tailored to the liquidity level and the interests of the client. To avoid risks, investment managers along with the investors need to analyze the investment in a multi-period scenario to "maximize expected returns for a given constraint on the maximum annual loss an investor could afford." (Murphy 2000). Traditional portfolio managers are conservative with risks and rightly so. However, as a result they are also inaccurate in their estimate allocation of assets and its distribution. The challenge according to Dunne and Davies (2005) is that manager realizes "it is necessary to take risks in order to generate excess return." Risks exist because of the inherent nature of capital market and the system that supports it.

Risks in investment therefore are parallel with the choice of investment manager. The choice of the manager depends on the kind of portfolio they have; sufficient time for investment to increase economies of scale and scope of returns; and trading costs incurred on investment. Managers can no longer afford to follow the traditional benchmark of "think what is right" but should consider the mechanistic and measurement benchmark (Dunne and Davies 2005).

Optimizing investment is thus parallel to optimizing investment management choice. Selection of investment managers and advisors who do not have expertise or do not actively manage funds or are out of circulation may prove compromising as they would not be good financial planners, strategists or advisors. The choice of the advisor must rest on the performance criteria in order to ensure that they can choose investment category and utilize investment strategies to balance the risks of investment (Murphy 2000). As Dunne and Davies write (2005):

"Quantitative tools can bias a manager research and selection process to a significant degree...It is possible to undertake some very sophisticated analysis of past performance in order to decompose returns and to determine if they were delivered as a result of skill or not. However, at best they can tell you very little about the future."

- Costs, price and performance

To maximize returns, investors have to analyze transaction costs, taxes and risks regardless of expertise and superior performance of the investment. It is critical for investment managers to ensure that the transaction costs for conducting trade should not compromise the income the investment aim to achieve. Strategies may be employed and guideline followed for after-tax returns so as to optimize investment, selection of the type of account in which to make investments, and to ensure there is a balance cash inflow and outflows (Murphy 2000). Not only this but gains from investment should be adequate to cover transaction costs and taxes arising out of trading and still leave some leftover for the investor to make the investment worthwhile.

Studies on the performance of investment indicate that professional investment managers tend to make investment outperform the market average on risk-adjusted basis by earning about 1 percent of the managed assets per year. This management cost is a management fee and a transaction cost over a long term period so that it evens out when calculated against the cost of investment. Though the average investment management cost range between 1 percent and 4 percent the total expense incurred by the professional investment managers may be 2 to 5 percent with higher transaction costs for illiquid stock investments. These percentages imply that managers must be able to outperform the market by an average of 3 to 6 percent gross of expenses in order to earn a net of 1 percent (Murphy 2000).

Another aspect of investment management cost is that the average cost is spread over a long time period interval such as 4 years. Studies by Sharpe (1966), and Dunn and Theisen (1983) indicate that mutual funds performance is positively related to the returns over a five year interval or subsequent time interval. Hence, their research negates the fact that high performing average funds are based on investment management efficiency (Hodgson et al 2000).

Evidence also suggests that better trained managers are able to manage funds more efficiently (Shukla and Singh 1994) if they are well educated and qualified in their profession. Alternatively studies also indicate that professional managers may also compromise the performance of investment against market average resulting in high investment expenses. The results imply that efficient and qualified professional investment managers can help identify securities whose values are significantly better than the market prices and they may help investors to make profits from investment opportunities (Simon 1998).

However, studies also indicate performance outcome in relation to management cost depend on a number of factors which may or may not fetch high returns for managed funds. For example opportunities may exist in the market but full utilization for higher returns can only be possible if the managers are actively trading using different market strategies such as buy and hold, discount cash flows and trading against market indices etc. Active traders may or may not incur high cost of management or transaction cost over returns. On the other hand those who trade less actively may fetch a less transaction costs but may also not be able to perform better because they are less active in trading. Thus transaction costs are irrelevant to performance level (Murphy 2000; Hodgson et al 2000).

Transaction cost is perhaps the key to the investment strategy as Williamson (1979) indicates in his study on transaction cost economics. Transaction cost unlike other costs can be minimized and easy to assume that it does not compromise the institutions' choice to reduce costs. Williamson is of the opinion that transaction cost differs in its capacity and effect depending on the environment in which it takes place. In discrete market transactions the traders have the autonomy and flexibility to bargain by predicting the events that lead up to the transaction. However, such opportunities are less likely to happen in the investment market where the identity of the traders are unimportant because the products are homogenous and the scope of bargain is less likely due to the same kind of relationship maintained among investors and managers. (Teece 1999).

Other researchers are of the belief that investment fund performance also depend on the size of the company and efficiency value in the market. This is because stocks of larger companies are widely traded. They tend to be more prominent in the market for traders, have plenty of data on them for analysis and most of the traders are willing to trade in them which make them highly risk-adjustable in returns. Small stocks on the other hand, are difficult to gauge as few traders deal with them and a large transaction cost and management fees cannot be absorbed with low investment funds vested in them. Therefore it is difficult even for the managers to absorb the cost through a spread of the portfolio.

Having discussed the role of costs and its correlation with performance, the researcher now moves to the issue of price. Price of the stocks is the determinant and objective indicator of value. On the outset it is imperative to also point out that pricing of assets also depend on market consensus who determines the worth of the stock. Some may attribute high value for certain securities while others do not. The market price usually represents the consensus value by a large number of investors reflective of the invisible hand. For liquid securities for example the price is dependent on active trading while for illiquid ones the price is determined based on the infrequent trading activities. The consensus value therefore determines the market price and is also the determinant of value of the securities. This implies that investors may employ fundamental analysis in gauging market efficiency but they are not the ones to determine prices, transaction cost or management expense. Pricing efficiency can deviate from value in secondary market and may also be responsible for under pricing investments. With sufficient skills managers may take advantage of this market "inefficiency" and help investors to achieve high returns despite market indicates otherwise.

Research Design

Research design and methodology choice depends on a great number of factors. Depending on the nature of the outcome a researcher may choose to use a qualitative approach or a quantitative approach or a combination of the two. In this case the researcher has chosen the combination method. The rationale is as follow.

Phillip Mayring (2000) is of the opinion that qualitative content analysis refers to bundled techniques for systematic text analysis. Text content analysis usually involves the objective of the content as well as the recorded transcripts such as interviews, discourses, protocols of observations, documents and the like. Content analysis however is not limited to the content of the material only. It takes into account of the differentiated levels of content, themes, main ideas of the texts as well as primary content (Mayring 2000). Furthermore, Mayring also writes that the formal aspects of the material belong to the aims, definition as well as the model of analysis.

Content analysis is also defined by Kripipendorff (1969) as "content analysis as the use of replicable and valid method for making specific inferences from text to other states or properties of its source" (1969, p.103). Thus qualitative content analysis is in itself an empirical method controlled by the texts and its modulation.

Mayring also outlines two processes of qualitative analysis namely - inductive and deductive. Inductive content analysis refers to the classical approach which has few answers to the research questions tailored to the system of categories. Thus depending on the framework, the qualitative approach would be central to the development of aspects of interpretation formulated by content material. The scope of the qualitative analysis developed under the inductive approach is thus reductive in nature and dependent on the psychology of interpreter (Mayring 2000).

On the other hand, the deductive approach refers to the theoretical framework derived from analysis in connection with the text. The text is analyzed through a controlled procedure with defined steps. Each idea is defined, governed by certain code of rules and categorized by the circumstances of the text analytical procedure (Mayring 2000).

Given these two approaches the researcher has chosen the inductive process due to its openness and inclusion of exogenous factors. In the researcher’s opinion for a balanced qualitative analysis, any model should be flexible enough to include other factors.

Alternatively, researchers also have found that qualitative data and analysis alone do not offer exact answers. Consequently, quantitative data is required to support the study results. Qualitative analysis is useful however, sometimes it is critical to include quantitative based information to be able to achieve precision. Quantitative analysis thus has the tendency to clarify conflicting or unclear views. It is able to quantify the research results without being bias thereby offering objective and factual results (McBride and Schostak 2005). According to McBride and Schostak (2005):

"The term 'quantitative paradigm' as to be employed here, refers to those discourses of science which involve throwing a mathematical grid upon the world of experience. Logical deductive reasoning, measurement and reduction to formulaic expressions are its features. The technicist paradigm as employed in this report is a particular kind of version of the more general quantitative paradigm which seeks measurement, observable units of analysis and logical arrangements. The technicist paradigm is reduced in scope in that it takes for granted its frameworks of analysis and its procedures and employs them routinely rather than subjecting them to the judgement of the practitioner. Although this characterisation is an 'ideal type' it has a basis in the data."

Thus in this context the researcher is of the opinion that a combination is best in some cases of research due to the special need, definition of data and the theoretical framework. The data collected may or may not offer a clear view which requires alternatives to support a subjective understanding. Hence as McBride and Schostak (2005) writes "Where quantitative forms of research, employing questionnaires and sampling procedures attempt to eradicate the individual, the particular and the subjective, qualitative research gives special attention to the subjective side of life."

Methodology

The application of econometric methods to time series is widespread. The choice of any of these methods is dependent on the subject of study whether it is investment, consumption, import function or income expenditure. In our case it is the investment problem that requires an econometric application. To begin with the researcher has chosen some econometric models which are discussed below:

In a correlation study the principle task is to determine and evaluate the numerical values for the beta coefficients. This evaluation revolves around the kind of results that the researcher aims to obtain and whether the results are statistically acceptable or not. Even then one cannot guarantee that the initial understanding of the statistical model application is the same as the results at the end. Consequently, the theoretical framework as well as economic reasoning has to be included in order to firmly support the hypothesis (Watson and Teelucksingh 2002).

In this research the researcher has established a general representation model by adopting the linear regression model presented by:

y t = M t, x 1t = 1 for all t, x 2t = Y t, x 3t = (p mt /p dt ) and k = 3.

where the variable Y is endogenous because it is a random variable, and is explained by the X exogenous variable. In this model beta coefficients are linear. To effectively apply the above 20 data sets have been taken to hopefully estimate the true values of the regression analysis (Watson and Teelucksingh 2002).

To support the above another model has been taken into account. This is discussed as follow:

Often in time series analysis, the researcher is faced with the problem of whether the economic variable can help in forecasting another variable. The basic premise is how to establish the causal link between one series of event with another. To resolve Granger in 1969 devised the Granger causality test. This test involves the use of f-tests which test how the lagged information of a variable Y against the lagged variable x. If there is no significance association then Y does not cause X or "Y does not Granger-cause X". Granger's causality test can be used in many tests. One of these tests is the use of correlation efficient.

This can be given by the following unrestricted equation using ordinary least squares:

When an F-test is carried out of the null hypothesis then the equation is given by:

When the F-test is greater than the specified critical value then, the null hypothesis is rejected meaning Y does not Granger-cause X.

Granger-cause X is basically based on a correlation model which has also been used in this research in conjunction with the above. Correlation refers to "When two measures obtained on the same "thing" go together so that it is possible to predict one measure from the other, they are said to be correlated." (Kimble 1978). The value of the correlation is measured by its correlation coefficient denoted by r as given by Pearson Product-moment correlation. The coefficient value ranges from -1.0 to + 1.0. The absolute magnitude value of the coefficient is ignored and it is used to indicate the strengths or closeness of the relationship. A correlation with plus or minus 1.0 represents a perfect relationship while a correlation of 0.0 has no relationship at all (Kimble 1978). One of the clearest ways to demonstrate a correlation is through a graphical representation in the form of a scatter plot. The scatter plot is also used to demonstrate the positive or negative correlation coefficient. A SW to NE direction denotes a positive correlation while the opposite of that in NW to SE direction has a negative relationship (Kimble 1978).

Data Limitation of Analysis

Econometric problems usually involve many variables causing difficulty in analyzing them all together. The two main issues that are faced by the researcher are the causality issue and the marginalization with respect to variables and their lags. Exogeneity problem arises when the researcher seeks to analyze subsets of the variable resulting in problems with validation. Consequently the Granger model (1969) is used to identify the non-causality which is easier to identify and forecast of exogeneity. This model however is limited in its parameters as the researcher finds it difficult to identify determined variables that are outside the system (Koopmans 1950). A variable is exogenous according to Koopmans if it is determined outside the system and is usually independent of the variable under analysis. Therefore any empirical model that employs exogeneity factor should offer consideration for partiality of the variable under study. Furthermore, Koopmans also cautions that it is important to identify the concept of weak exogeneity and "not caused" (Hendry 1995; Charemza and Deadman 1997). For this reason the researcher has included a partial correlation method to avoid neglecting any aspect of data interpretation.

Data collection

For the purpose of this study the data collection process has been divided into two fold procedures.

Firstly, to set a theoretical framework the researcher has followed the qualitative approach by collecting a host of literature on the subject of fund selection, performance, evaluation, management cost and risks etc. These are derived from secondary and primary resources. Primary resources such as books, journals, and information from official websites of the pension fund from Standard Life. These are reviewed but not analyzed in the Literature Review Section. On the other hand, to support the primary data, secondary sources have also been included. These comprise of websites, newspapers, brochures and magazine articles accessible online. The secondary and primary resources combined help to set up the theoretical framework for the qualitative analysis both of which have been outlined in the literature section.

Secondly, a quantitative data collection process has been followed to access live data. Due to the extent and scope of the data sets, the researcher has limited the data on the fund to pension funds offered at Standard Life, one of the most prestigious organizations that offers a host of investment services and renowned for its extraordinary services to individual investors, and institution investors alike. Standard Life has also been chosen as the main source for the data set because of its ease of access to information as well as the reliability in its fund performances. Distribution of investment type, risks, volatility, management fee as well as performance efficiency are all easily accessible through its website. Another website called "My Personal Finance" has also been included as the source for data sets. From this website the researcher has obtained data on the efficiency of performance percentages of pension fund under Standard Life ranging from 1 year to 5 years, and the price (See Appendix). From the Standard Life documents the researcher has found that Standard Life management has set its fee at 1% of the fund. Its annualized growth rate on management fee is also obtained from the website demonstrated in the Appendix.

To avoid collection of redundant data, the researcher has limited the number of data sets to 20 pension funds listed under Standard Life. The data sets are listed in the Appendix.

Data Analysis

Data collected from Standard Life and My Personal Finance is tabulated in MS Excel and MS Word (see appendix). For the calculation of the regression, correlation and F-test for the Granger-cause test have been performed through online software by Wessa.net (2005). The results are given in the appendix. It must be noted due to the lack of space, and time, the calculations of the data sets have been performed on this software to firstly reduce human error and to speed up the process of the research.

For the data analysis, three types of calculations have been carried out to ensure that the researcher has covered the maximum aspects of data acquired. The first test carried out include the partial correlation which uses three variables, namely X, Y, and Z where X is the exogenous variable denoting price; Y the endogenous variable denoting performance for the 1, 3 and 5 years; and Z is the variable denoting management performance for 1, 3 and 5 years. The basic premise for using the partial correlation is to compare how each of the years (1, 3 and 5) performances is comparable with price, performance and management fee. This is critical to this study as it presents the aspects that allows the researcher to establish a strong or weak link between them and help in identifying the relationship between the three variables.

Another data analysis is carried out through a correlation calculation. This includes only the variables X and Y as the researcher wants to endorse the relationship established by the partial correlation. For this purpose the researcher carried out price against performance for 5 years. The results are presented in the appendix. The choice of 5 years aggregate performance against price has been taken into account because some of the data were not available. Also, this 5 year aggregate data also show a realistic view of the data's performance level.

A third test, using correlation is carried out between price and management fee growth rate has been carried out for similar results. The results are to help the researcher to identify whether there is a positive relationship between price and management fee. Here price is indicative of the viability of the fund as relative to the performance level calculated earlier. The level of correlation is dependent on the coefficient of beta and the F-test results. It is estimated that if the beta coefficient is zero then the null hypothesis is rejected whereas a positive coefficient denotes a positive relationship between the variables.

Quantitative Results and Discussion

At the beginning of the research, the researcher has posited the following hypotheses:

H0: Investment fund performance does not affect the cost of management.

H1: Price affect management fee.

H2. Price affects fund performance.

A general view of the observed data indicates that there is a positive relation between performance and annualized growth of management fees. Though the data is not consistent nevertheless the majority of the 5 years performance level of the pension fund at Standard Life affects the 5 year growth in management fee. To quantify this observation, an investigation of the null hypothesis is carried out to analyze through the results of simple regression (given in the Appendix). Using price as the exogenous and y as the endogenous variable, the simple regression are calculated which indicate that the beta value is less than one and nearing to zero. This means that there is little or negative correlation for the null hypothesis. A look at the F-Test result of 0.51 supports this view as according to Granger (1969), an F-test with a greater than specified critical value means Y does not Granger-cause X. The null hypothesis therefore is rejected. What this basically means that performance DOES affect cost of management. How much is this statement true can be analyzed by the partial correlation results discussed next.

A partial correlation is considered to be perfect or correlated if the direction of the coefficient is mapped and shows a direction of SW to NE (mentioned in the Research section). Alternatively, it also has a positive correlation if the coefficient is mapped in the direction of NW to SE. The results from the partial correlation for price and performance indicate a low level of correlation at only 0.035. Alternatively, the correlation between price and management fee demonstrate the same level 0.304. though this result is counteractive of the general observation noted above, nevertheless it explains for the inconsistent trend in investment performance and management fee performance. But what makes this test significant is the fact that the correlation between performance and management fee is positive as it is estimated at 0.95, nearing to 1. As Kimble (1978) indicate there is a positive relationship between the variables if the correlation ranges from 1.0 to -1.0.

A third test, a simple regression is taken first of price against performance which is representative of H2: which states that "Price affect fund performance. Again the beta value is nearing zero at 0.004 indicating that there is little correlation between price and performance, which leaves us with the H1.

For H1, the researcher has taken out a second simple regression correlation of price against management fee. In this case the value of beta is even less than simple regression for H1. By plotting the results on a regression graph and scatter plot it can be observed that there is least correlation between price and management fee.

The results from the simple regression do not show a positive relation between price and performance, and price and management fee. Instead, it leaves room for the researcher to assume that there are chances of a positive relation between performance and management fee. The partial correlation test indicates a positive relationship between performance and management fee performance. This is supported by the initial Granger-cause test which rejects the null hypothesis stating that investment fund performance does not affect cost of management (management fee performance).

Quantitative results as discussed earlier by McBride and Schostak (2005) do not necessarily offer conclusive answers. Mere numbers and statistical calculations do not allow one to deduce that there is an accurate relationship between the variables. The differences in places of decimals do not reflect the true beta values; nor ca