Portfolio Mix Research Articles

Protecting against Inflation--and Maximizing Yield, by John M. Bragg: Reviewer's Reply

The chief disagreement between Mr. Bragg and me concerns our divergent attitudes about the efficacy of efficient markets theory, and whether or not the costs of an active investment philosophy are more than compensated by the uncertain benefit of abnormal returns. For example, Bragg's of the Economic is dependent on the notion that social psychological moods can be determined, and that a resolute state of social psychology is associated with an optimal portfolio mix which enables investors to maximize yield. My contention with Bragg is not in the accurate gauging of the state of social psychology. In fact, assume that the psychological mood of society can be ascertained. Then the semi-strong version of efficient markets theory would argue that if one investor, or a set of investment experts can sense a mood, and the costs of sensing are low, then other competitive investors must be able to sense a mood as well; therefore, abnormal returns would be non-existent. 1 Certainly the debate between Bragg and me is not new. Since the work on models of equilibrium in capital markets began (Sharpe, 1964; Lintner, 1965; Mossin, 1966), there has been much discussion between advocates of the traditional approach to investment analysis and efficient market theorists. Academic researchers have shown that anomalies such as the small firm effect (Banz, 1981) and the January effect (Keim, 1983) exist, and that markets may not be perfectly efficient. The unresolved question directly related to Bragg's thesis is whether the costs of his active investment strategy are more or less than the benefits of returns in excess of the return on a well-diversified market portfolio; a return that may be obtained with inconsequential transaction costs. If the answer to this question is that the Theory of Economic Series produces superior returns, then one must ask why Bragg would ever let such valuable, proprietary information into the public forum?

A specific decision support system (SDSS) to develop an optimal project portfolio mix under uncertainty

A specific decision support system (SDSS) than can be used as a methodology for choosing an optimal portfolio mix of information systems projects is described. The SDSS is developed by applying a number of well-known management science techniques. A net present value (NPV) of the projects is maximized within limited resources, and the solution of the model provides the optimal project start period, system development language, and staff size. The SDSS conducts a risk analysis on those variables that are deemed critical when determining the optimal solution. The final phase of the SDSS establishes the value of perfect information on these critical variables. To demonstrate the usefulness of the model an example consisting of five information system (IS) projects is presented. It is shown that, given the estimates of the exogenous parameters of the IS project environment, the model can determine when to begin the project development, which systems development language to use, and the number of systems development staff assign to each project. The degree of the variability of the estimates of exogenous parameters is evaluated through empirical probability distributions.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

The Design of an Investment Portfolio Selection Decision Support System Using Two Expert Systems and a Consulting System

Abstract:Two expert systems form the basis for the design of a consulting system that provides advice for selecting an investment portfolio for an individual. Current economic indicators are compared with an investor’s financial goals and constraints by one expert system to augment the other. The first expert system distributes its domain knowledge to the second. Subsequent to this knowledge sharing, the second expert system matches a person’s financial attributes, goals, and risk preferences with investment choices to recommend portfolio mixes. The consulting system suggests portfolio mixes that are preferred by experienced investment counselors within the boundaries of the Decision Support System (dss) output.

Managerial Risk Preferences, Real Pension Costs, and Long-Run Corporate Pension Fund Investment Policy

This paper presents a new methodology for determining long-run corporate pension fund investment policy for a salary-based defined benefit pension plan. A mean-risk dominance model is developed to choose among the ex ante distributions of pension costs. The model identifies the optimum pension portfolio mix that is consistent with the cost objectives of the plan. The optimum mix consists of a minimum investment in stocks, ranging from a low of 40 percent to a high of 100 percent, depending on the risk preferences of the decision agent and corporate pension cost policies. The results are not sensitive to the age structure of the plan population.

Draft of a System for Solvency Control in Non-Life Insurance

AbstractAn outline is given of a proposed system for solvency control in non-life insurance that has recently been discussed within a Working Party appointed by the Norwegian supervisory authorities. According to this system the factual technical reserves must at any time be sufficient to meet, with high probability, all future liabilities stipulated by insurance contracts that have either expired or are currently in force. The system is applied to a provisional, simple model that has been fitted to claims data assembled from Norwegian non-life companies. The numerical examples illustrate, inter alia, how the required reserve depends on the volume of the business, the portfolio mix, and the reinsurance cover.

An asset allocation model for active portfolios

76 s M anagers of active investment portfolios 2 frequently attempt To forecast stock market returns in order to decide how to distribute their funds between stocks and fixed-dollar assets. Our purpose in this study is to develop a model that explains stock market returns and that makes fund allocations more efficient. The project has two phases. The first is to design a multiple regression model that might help to explain the annual rate of return of the stock ma,rket. The second is to determine whether a model that reflects this explanation could provide a portfolio strategy indicator that the manager can apply to making asset allocation decisions in an active portfolio. Before developing a new model and trading rules for determining asset allocations, however, we might ask how successfully managers have performed this task in the past. After all, pension fund administrators as a class have been busier than most institutional investors in varying their portfolio asset mix in equities over the years.’ Table I shows the percent of common stocks held in the portfolios of private non-insured pension funds in each of the past 25 years. Although several different factors can account for shifts in asset mix over time, such as changes in pension plan objectives and legal standards, Table I suggests several policy rules, particularly in conjunction with annual stock market returns as measured by the Standard & Poor’s Stock Index. One rule in evidence is a passive policy, which allows the percent of the portfolios in stocks to fluctuate with the stock market cycle. Observe that the z Fi

A Note on Feldstein's Criticism of Mean-Variance Analysis: A Reply

Joram Mayshar's [3] note is a valuable contribution to the theory of mean-variance portfolio analysis. His two points are correctly and carefully stated. It is important, however, that these results should not be overinterpreted. Mayshar shows first that when the outcome of a portfolio mix can be represented in >u space, these preferences will be quasi-concave for a risk-averse investor. The lognormal distribution that I used [2] as a counter-example to Tobin's [4] conclusion is not quasi-concave but is excluded by Mayshar because such a distribution is not the result of a portfolio mix. Although Mayshar is correct to emphasize the importance of results that are restricted to the distributions generated by portfolio mix problems, this class of distributions is quite limited. Readers may recall that Tobin's original article stated preferences and subjective probabilities in terms of rates of return and therefore, like my comment on it, was not set as a problem of portfolio choice in Mayshar's sense. Mayshar's second point is that a I a representation of preferences always exists when one asset is safe and the other has a given distribution. There can obviously be no quarrel about this conclusion when the form of the given distribution is known. But more generally it remains true that if theform of the two-parameter distribution of the risky asset is unknown, a knowledge of these two parameters is not sufficient to rank the portfolio combinations that result from combining this risky asset with a safe asset. Developments in portfolio theory during the past decade have gone beyond the meanvariance framework to study the portfolio behaviour of risk-averse investors facing general probability distributions (e.g. Arrow [1]). The results of these investigations confirm that I-au analyses may be heuristically useful but should always be regarded as only illustrative of a possible special case.

AN EXAMINATION OF THE YIELDS OF CORPORATE BONDS AND STOCKS: COMMENT

movement. Corporate yields in 1926 implied forward rates for 1969 at the five percent level and the 1969 yield curve was in the eight percent area. This biases the mean of the bond holding period returns downward as an expectation measure and it is only the expected return which is relevant to the kind of conclusion Norgaard develops. The mean of the stock holding period returns will also be effected but the impact here is not nearly so consequential in that dividends are flexible in contrast to coupon payments and face values which are fixed. Since the mean bond return is biased downward as an expectation to a considerably greater extent than is the mean equity return, there is a significant bias in the direction of rejecting the bond or mixed portfolio. Table 1 indicates the extent of the bias for a simple bond case involving unanticipated upward movement in yields. Mean one year holding period returns are developed over 43 years (the length of Norgaard's data) for investment in five percent coupon bonds of varying maturities. It is assumed that yields remain

A Quadratic Programming Model of the Non-Life Insurer

The purpose of this paper is to provide insurers with a flexible construct which can be of assistance in reaching ex ante portfolio mix decisions. Portfolio theory frequently has been applied to risky asset opportunity sets; however, little attention has been focused on risky asset and risky liability opportunity sets in combination. The insurer attempts to allocate premrums and financial assets to generate positive returns. The study's purpose is to specify optimal ex ante (via ex post) portfolios of underwriting and investment alternatives. These areas are specifically related by applying portfolio theory. The sensitivity of the insurer efficient frontier to alternative constraint definitions is evaluated. Underwriting and asset mix are analyzed as relevant institutional constraints are changed. Considerable controversy has developed in the insurance literature as to the measurement of profitability in the property-liability insurance industry. Clearly the profitability of the industry depends on how effectively the underwriting and investment activities are combined. The degree of industry regulation in its pricing of insurance is controlled by state insurance codes. The insurer insists that these regulated prices include a fair for providing insurance. The regulatory apparatus has seen fit to allow reasonable in the development of sale structures. However, investment returns have not been considered relevant in determining the insurance rate structure until recently. Despite the movement toward including investment income in p-l ratemaking, little agreement exists as to the correct approach for its implementation.' When one makes an attempt to calculate the return and risks to the industry both underwriting and investment functions must be explicitly considered simultaneously. John L. Markle, Ph.D., is a member of the sales staff of Salomon Brothers in New York City. Alfred E. Hofflander, Ph.D., CPCU, CLU, is Professor and Director of Graduate Studies in the Graduate School of Management, University of California, Los Angeles. He is current president of ARIA and an Assistant Editor of the Journal of Risk and Insurance. This paper was accepted for publication in June 1975. 1A recent contribution to this field which attempts to derive an objective analysis of the rationale and feasibility of including investment return in the property-liability ratemaking process is In vestnent Return and Property-Liability Insurance Ratemaking by Robert W. Cooper.

Exchange Risk, Political Risk, and Investor Demand for External Currency Deposits

The growth of the external currency market is an institutional response to the costs and risks that derive from national regulation of financial transactions.l The deposits produced within the external market have combinations of exchange risk attributes and political risk attributes different from those associated with domestic deposits. Investors who acquire external deposits necessarily alter the risk attributes of their portfolios; either the currency in which their deposits are denominated or the country in which deposits are produced is changed. Frequently it is asserted that the growth of the external market has led to a change in the currency mix of investor portfolios and that investors hold a larger share of their portfolios in dollardenominated deposits than they would in the absence of the external market.2 Sometimes it is asserted that the links among national money markets have been tightened as a result of the growth of the market or investors are said to be more willing to alter the currency mix of their portfolios

Note on “Devaluation, capital flows and the balance of payments: A respecification”

(2) The argument developed by Todd M. Sandier is the following2: "An increase in the exchange rate (devaluation) makes a given income stream of foreign currency worth more in terms of domestic currency to residents of the devaluing country. Hence, the portfolio mix changes . . . since the relative profitability in terms of domestic currency of assets changes. For instance, starting from portfolio equilibrium if the exchange rate in A increases, there results a situation of disequilibrium in which Α-residents seek to buy foreign assets that now give a higher domestic currency equivalent after the devaluation, ceteris paribus. Also, B-residents try to sell Α-employed capital assets, because the B-currency equivalent of these income stream earnings has fallen. In the process, Qab decreases as B-residents sell Α-employed capital (hence Qaa increases, i. e.,

A Decision Rule for Determining Pension and Endowment Funds' Portfolio Mix

Although no mechanical solution to the problem of determining optimal timing for purchases and sales of common stocks and bonds has been found, this article proposes a potentially useful framework for analysis of the problems faced in the investment of pension funds.

On the Capital Structure Theorem: Reply

IN A RECENT PAPER this journal [1] I suggested a formulation of a cost of capital theorem, stating that in equilibrium, within the range of efficient capital structure the firm's cost of capital is constant. As the firm moves to inefficient capital structure, the firm's cost of capital rises. This theorem was derived from the theory of investors' behavior towards return and risk,' under a constraint, i.e., that the investor can invest his own capital, with any proportion of borrowed capital either one stock or a mixed portfolio of one stock and riskless bonds. The purpose of this paper is first to clarify some issues raised the Comments by Litzenberger and Jones (hereafter L&J) and by Haugen and Pappas (hereafter H&P), and second, to reexamine the theorem when the diversification constraint is removed, i.e., when the investor is allowed to diversify his investment among various assets.

THE CAPITAL STRUCTURE AND THE COST OF CAPITAL: COMMENT

IN A RECENT Journal article [1] Haim Ben-Shahar presented a theory of firm's capital structure from standpoint of return-risk framework of investor behavior. He conducts his analysis under the constraint that investor has opportunity to invest his own capital with any proportion of borrowed capital, either in one stock or in a mixed portfolio of that stock and riskless bonds.' The purpose of this comment is twofold: first, to examine some of implicit assumptions on Ben-Shahar's analysis which severely limit generality of his conclusions; and second, following a suggestion made by Ben-Shahar [1, p. 652], to modify and extend his general analysis to more realistic case where investors hold diversified portfolios of common stocks. I. COMMENTS Our first comment is of a conceptual nature and concerns Ben-Shahar's general approach to problem of firm's capital structure. His capital structure theorem is based on a theory of investor behavior toward return and risk. Implicit in his use of first two moments of probability distribution of firm's operating earnings is assumption that investors have homogeneous probability beliefs about and return. The M&M arbitrage construct is not dependent on these cardinal measures of and return. The term equivalent risk class requires only an ordinal concept of (i.e., that investors feel that same degree of is inherent in operating earnings of all firms in same class). An important strength of M&M analysis is that it is consistent with divergent investor probability beliefs about and return. Our second point concerns semantics. Ben-Shahar seems to imply that expected return and standard deviation of return to investors is synonymous with firm's expected earnings and standard deviation of earnings. He defines expected rate of return, 6., and standard deviation of rate of return, beo, on an all equity firm as:

Risk and Real Estate Ownership: Empirical Evidence of Hotels in Singapore

This paper examined the impact of real property ownership on the systematic risk of hotel companies in Singapore. This issue is interesting because few studies have been carried out on this topic. The hypothesis is that the real estate ownership would impose negative effect on the systematic risk of company in hotel sector because corporate real estate is commonly considered as an instrument for diversification in a mixed portfolio. To examine the effect, two-stage least-square regression was applied. The data was collected from published sources and other data streams. The results indicate that real estate ownership has impact on the systematic risk of companies. The implication is that the different strategies of companies may result in the different directions of impacts.DOI: http://dx.doi.org/10.3329/jbip.v2i0.9579 Journal of Bangladesh Institute of Planners Vol. 2, December 2009, pp. 193-200

SECURITY PRICES, RISK, AND MAXIMAL GAINS FROM DIVERSIFICATION*

SECURITY PRICES, RISK, AND MAXIMAL GAINS FROM DIVERSIFICATION*