Portfolio Management (Finance)

Portfolio management (Strong, 1993) is concerned with distributing investible liquidity across a range of available assets and liabilities with the objective of providing risks and returns that achieve performance objectives. Portfolio management therefore comprises objective setting (establishing the relative importance of delivering capital and income growth and providing stability of principal an d income to actual or prospective investors), asset allocation (where the available funds are distributed across geographic markets and security categories to exploit broad market and currency movements) and security selection (the choice of particular securities in each category that offer the best value in terms of portfolio objectives).

So far as objective setting is concerned this is conducted in either a direct or indirect mode. Direct objectives emerge from detailed customer financial reviews conducted in approved form by financial intermediaries licensed by a regulatory authority. Alternatively pension or insurance fund trustees might set portfolio man agers income and growth objectives relative to a specific benchmark such as the Financial Times/Actuaries All Share Index. Indirect objective setting arises where portfolios in the form of mutual funds (unit trusts and investment trusts in the UK) are offered to the public in which case the basic strategy in terms of exposure to equities or bonds or to UK, European, or Far Eastern markets will be outlined in a prospectus. Arising from this strategy a benchmark in terms of the growth, income, and capital stability characteristics of a particular index (e.g. European equity, North American bond) will be defined and the security reported in that category by the financial press.


Historically the distinction between portfolio management and investment management arises from new ideas about risk diversification introduced in the 1950s by Harry Markowitz (1952) with the observation that the variability of returns for a collection of assets depended on the correlation of asset returns with each other and not just on the weighted average of the individual assets. Diversifying investments across a range of substantially uncorrelated securities, whether within one country or increasingly internationally (Levy and Sarnat, 1970) provides portfolio managers with lower variability for the same return or a higher return for the same variability than any single one of the underlying national or internat1ional securities.

Table 1 Sample Portfolio Recommendations (%)

Equities Bonds Cash
MerrillLynch 50 40 15
LehmanBrothers 65 40 10
NikkoSecurities 45 30 5
DaiwaEurope 55 40 5
CreditAgricole 65 35 0
CreditSuisse 30 48 22

The theory of diversification was developed by Sharpe (1963) and Lintner (1965) to show that where large numbers of securities are used to create a fully diversified portfolio the effect is to eliminate the specific risks relating to each particular asset, leaving only the systematic risk, the common risks to which all securities are exposed. This systematic risk or market risk is effectively equivalent to the riskiness of the market portfolio and provides the reference benchmark for risk pricing used in the capital asset pricing model or CAPM.

Depending on diversification strategy, portfolio management may be active or passive. Passive portfolio management aims to replicate the performance, say, of a particular stock index by neutral weighting whereby asset distribution in the portfolio matches the proportions of each asset or asset class in the index to be proxied. In contrast under active portfolio management elements in the portfolio are either overweight (overrepresented) or underweight (underrepresented) relative to the target index . The intention is to produce outperformance relative to the target index by overrepresentation of assets or asset classes expected to outperform the relevant index. Active management therefore involves frequent rebalancing of both the asset allocation and the individual underlying security holdings to reflect changes in the expected risks and returns.


This rebalancing will aim to exploit timing effects. The relative returns for different countries and for the different types of security such a s equities bonds and money market balances within a country vary with economic conditions of growth, inflation, etc. By overweighting the portfolio with the asset most likely to outperform under the anticipated economic climate the portfolio manager aims to outperform a portfolio that maintains unchanged weightings throughout the economic cycle.

If the choice of assets is simplified to comprise simply high risk (equity) investments that generally outperform under economic recovery and low risk (bonds and cash) that outperform in conditions of economic slowdown and recession, timing effectiveness can be measured relative to a benchmark portfolio with fixed equity and bond/cash proportions. In principle the benchmark portfolio could be fully invest ed in equities with the bond/cash proportion zero, but a fund manager wishing to increase equity exposure relative to the benchmark could borrow cash to invest more than 100 percent of portfolio value in equities. Significant leverage (using debt to purchase equities in excess of the total value of the fund) is encountered both in closed end funds and in the speculative hedge funds, but open-ended mutual funds are prohibited from borrowing and in practice most portfolios contain liquidity either to meet imminent liabilities (pension payouts, insurance claims, fund withdrawals) or from uninvested new contributions. To reflect this, the benchmark portfolio might have 20 percent cash/bonds and 80 percent equity. If the equity index yields 7 percent and money market rates are 5 percent an active fund with a 30 percent/70 percent allocation will earn 0.3

x 5 percent + 0.7 x 7 percent or 6.4 percent, an underperformance relative to the benchmark return (0.2 x 5 percent + 0.8 x 7 percent or 6.6 percent) of 0.2 percent.

The timing stances of a variety of funds in respect of cash, bonds and equities are illustrated in a sample of portfolio recommendations published regularly by The Economist (Table 1).

Strictly performance comparison between portfolios should specifically adjust for the ex ante risks taken by the portfolios, otherwise portfolio managers would simply increase risk levels to improve returns. The CAPM model provides a framework for risk adjustment by using beta or the correlation of returns of a security or portfolio with the returns of the market portfolio as a proxy for riskiness with the market portfolio definitionally having a beta of one. The beta is then multiplied by the risk premium or historical outperformance of equities relative to government bonds to provide a risk adjusted benchmark return. Thus if the risk premium is 7 percent then a portfolio with a beta of 1.5 has to achieve returns 7 percent higher than a portfolio with a beta of 0 .5 before outperformance is demonstrated. Unfortunately in recent years the risk premium has been rather volatile: see Table 2.

Table 2 Real Returns on Investment in US dollar terms, 1984-93 annual average.

Equities Bonds Cash
France 18 14.5 9.5
Holland 17.5 11.5 8
Britain 15 8.5 7.5
Germany 14 9 7.5
Switzerland 13.5 8 6.5
Italy 13 14 9.5
Japan 13 13.5 10
USA 12 11 3
Australia 10.5 11 6
Canada 3.5 11 5.5

As an alternative Merton (1981) argued that as returns of an all-equity portfolio are more variable (risky) than an all-bond portfolio, ri sk differences due to composition should be proxied by using option performance. Perfect timing is equivalent to holding cash plus call options on the entire equity portfolio with benchmark adjustments using reduced options to reflect any equity proportion.

The two best known portfolio performance yardsticks are the Sharpe measure and the Treynor measure. Sharpe (1966) measures return differences from average relative to the standard deviation of returns, while Treyno r measures return differences from average relative to beta, or systematic risk.

Within the overall asset allocation, active portfolio management involves security analysis aimed at picking the best value way of investing allocated funds in asset categories such as bonds, deposits, real estate, equities, and commodities. Portfolios, though, mainly emphasize bonds and equities for the simple reason that they have high liquidity (reasonable quantities can be bought or sold at market price) and l ow transaction costs. In analyzing securities, portfolio managers utilize either fundamental analysis or technical analysis. Fundamental analysis utilizes financial and non-financial data to locate undervalued securities which relative to the market offer growth at a discount, assets at a discount or yield at a discount. Although brokerage houses, among others, invest heavily in such analysis if successful it would contradict the efficient market hypothesis (EMH) which argues that the market prices of securities already incorporate all information in the market and that therefore it is impossible in the long term to outperform the market. Nevertheless, relatively simple transformations such as Gordon’s growth model (Gordon, 1963) relating share prices shares to dividends and dividend growth are widely used in security selection. There is an extensive literature, including Fama (1969), on signaling where factors such as dividend changes or investment announcements are used to explain security price changes.

The arbitrage pricing theory APT developed by Ross (1976) provides a more general formal framework for analyzing return differences based on the basis of multiple factors such as industry, size, market to topic ratio, and other economic and financial variables.

Not surprisingly, the possibility of beating the passive or buy and hold strategies indicated by the EMH has attracted considerable attention, with Banz (1981) among the first to detect an anomaly in the risk-adjusted outperformance of small firms followed by Keim’s (1983) analysis of a January effect. End of month, holiday, and weekend effects together with price/topic anomalies have also been reported, but with an overall effect small relative to transaction costs. Despite this limited success market practitioners continue to offer simple guidelines that they have used to produce exceptional returns. Jim Slater (Slater, 1994) reports favorable results for a stock picking exercise that uses principles developed by the legendary Warren Buffett and more recently by O’Higgins and Downes (1992) who report in Beating the Dow that picking the ten highest yielding shares from the 30 Dow Jones Industrial Index and then investing in the five cheapest (in dollar price) of these shares produced a gain of 2,800 percent against a 560 percent gain on the Dow over eighteen years. It is unclear, though, what these authors have to gain by disclosing such valuable procedures.

Technical analysis or chartism is an alternative and widely used technique in portfolio management. In direct contradiction to the weak-form version of the EMH, which states that all information contained in past securities prices is incorporated in the present market price, technical analysts use past patterns to project trends. These patterns may be simply shapes described for example as ”head and shoulders,” “double tops,” “flags,” and so on or more elaborate short- or long-term trend lines, all of which are used to generate buy or sell signals. Evaluations of technical analysis have generally run into problems because of subjectivity in classifying signals, but recent work in neural networks (Baestans et al., 1994) has provided objective evidence of information in the trend lines used by technical analysts, much of it in non-linear components neglected in some econometric analysis.

The relatively recent development of large, liquid derivative markets - security and index options and futures – has revolutionized the asset allocation process because it allows portfolio managers to proxy the exposure of o ne asset allocation despite holding a portfolio consisting of a completely different set of assets. A bond or money market portfolio together with equity index futures contracts effectively proxies an equity portfolio. An equity portfolio together with the purchase of put options and sale of call options is similarly equivalent to a fixed interest portfolio. Portfolio managers are able to use derivatives to segment risks asymmetrically. An equity portfolio or index future hedged by put options gives the downside stability of a bond portfolio and the upward opportunities of an equity portfolio. This allows the portfolio manager to create funds with partial or full performance guarantees where investors are offered half any upward movement in the equity market plus the return of their original investment.

Index-based derivatives are particularly popular with portfolio managers because they provide market diversification with very low transaction costs and none of the trading and monitoring activity involved in maintaining a portfolio of securities that mimicked the index. A portfolio manager wishing to hold a long-term position in equities but at the same time wanting a flexible asset allocation will typically use an index transaction to adjust exposure. A sale of an index future on 20 percent of the portfolio is equivalent to a 20/80 bond equity portfolio.

The possibility of altering positions in this way without transactions on the spot market has generated a number of new techniques. Program trading, for example, involves buying or selling bundles of shares. A portfolio manager with a bundle of shares that provide an adequate proxy for the market index may use programme trading to arbitrage between the spot market and index futures, with the transaction itself being computer initiated. In other words if index futures rise in value it may be profitable to buy a bundle of shares that proxy the index in the spot market. Alternatively th e index future price may fall and a portfolio manager who has bought in the forward mark et may then program sell in the spot market, depressing the spot market which then transmits a further downward signal to the futures market, arguably increasing the risk of a major price melt down (Roll, 1988).

The second major development is dynamic hedging. Because of the low cost and flexibility of futures markets a portfolio manager can optimize the portfolio on a continuous rather than one off basis. Dynamic hedging incorporates the possibility of new information and the dynamic hedge ratio for a portfolio reflects th e quantity of an option that must be traded to eliminate a unit of risk exposure in a portfolio position. This depends on the delta, which measures the sensitivity of the value of an option to a unit change in the price of the underlying asset, and/or the ratio of the dollar value of the portfolio to the dollar value of the futures index contract multiplied by the beta or systematic risk of the portfolio.

Next post:

Previous post: