اصول سرمایه گذاری برای حساب های بازنشستگی فردی

The phenomenal growth of individual retirement accounts in the US, and globally, challenges both individuals and their advisors to rationally manage these investments. The two essential differences between an individual retirement account and an institutional portfolio are the length of the investment horizon and the regularity of monthly contributions. The purpose of this paper is to contrast principles of institutional investing with the management of individual retirement accounts. Using monthly historical data from 1926 to 2005 we evaluate the suitability for managing individual retirement portfolios of seven principles employed in institutional investing. We discover that some of these guidelines can be beneficially applied to the investment management of individual retirement accounts while others need to be reconsidered.

Extensive academic research on risks and returns of various investment classes has been incorporated in investment primers, technical and statistical books and popular monographs. For example, the investment textbooks of Bodie et al., 2007, Reilly and Brown, 2006 and Viceira and Campbell, 2002, the classic financial econometrics book of Campbell et al. (1997) and the highly successful investment book of Siegel (2002), all collect, document and elucidate numerous stylized facts about asset returns, risks and long-term performance of stocks, bonds, cash and other classes of investments. From these findings certain investment guidelines have been proposed, statistically documented, and debated using very long time series of, mostly annual, returns of various aggregate indexes for capital markets. In this paper we select a few such standard stylized principles or guidelines for long-run institutional investing and explore their relevance in managing individual retirement accounts.

The dramatic recent increase in the number of individual retirement accounts in the US and globally necessitates rational financial management of these accounts. In contrast to the exhaustive research about capital markets summarized in Ibbotson Associates, 2006 and Constantinides, 2002, the academic literature is just beginning to address various aspects of individual retirement accounts. As an increasing number of individuals attempt to manage their own accounts, often with the help of advisors, it is informative to list certain useful guidelines. Furthermore it is helpful to relate these guidelines to ones that have been developed for institutional investing. Our paper proposes several such guidelines. First, calculating individual retirement accumulations is not simple when compared to straightforward lump-sum investing, because both the monthly contributions and returns are unpredictable. The calculations of periodic contributions are not complex but not all investors are skilled to perform them. Using average return and average contributions may simplify the calculations but it also causes such an average accumulation to deviate, often substantially, from the actual one. Using also average returns as an approximation to the actual term structure of returns causes the standard deviations of accumulations to be larger. Second, for investment horizons of 20, 30 and even 40 years of monthly contributions the accumulations, even when funds are invested 100% in stocks are not very large multiples of total contributions because the power of compounding works only partially. Early contributions have long periods to compound while later contributions have shorter time to compound, particularly if they are larger and cause contributions to increase. Third, our calculations support the notion of stability of returns found in the literature. For investment horizons of 40, 30 and even 20 years, geometric means of monthly returns of the S&P 500 Index for numerous generations have similar means but different standard deviations. This stability of returns does not necessarily translate to stability of accumulations for two reasons. First, individuals with longer horizons of 40 years accumulate absolutely much larger sums than those who invest for only 30 or 20 years. The standard deviations of these accumulations are also different because longer horizons imply larger fluctuations. Second, for large numbers of generations the accumulations appear clustered but rapid booms and busts offset such clustering and common measures of risk do not discriminate between periods of stable accumulations and periods of rapid change. In other words, each generation, like vintage wine, is characterized by monthly returns that are path dependent, meaning that each nearby return is correlated with the previous one reflecting the state of the economy that does not act totally at random month to month. This means that individual investors live, invest and consume during their generation and cannot transport their investments at other times. Typical examples are the 1987 October Crash or the bursting of the technology bubble in March 2000. The accumulations of those who retired before such episodes compared to the ones who retired few months later may be significantly different. This fact necessitates careful management of investment allocations for individuals approaching retirement age. In contrast if accumulations were very stable across generations, most generations contributing the same amount would accumulate approximately the same final amounts. Fourth, investing in equities remains the favorite vehicle for highest returns. For horizons of 20–40 years, the overwhelming majority of generations accumulate the largest sums by investing in small company stocks vs large company stocks vs. bonds or bills. Bonds however do not consistently outperform Treasury Bills in our sample. This superior performance of stocks is also associated with relatively higher risk. Fifth, calculations with fixed monthly contributions or contributions increasing annually by a 2% productivity growth and a 3% inflation rate, yield accumulations which are skewed to the right. This pattern of distribution necessitates more careful statistical inference about what percent of generations will achieve certain accumulations. Finally, unlike institutional investing that often has a very long investment horizon, individual retirement accounts have both a terminal goal and horizon. Our analysis suggests that individuals contributing for a 40 year horizon can experience a very high probability of achieving their retirement goal of having sufficient funds to live comfortably for 20 years beyond the age of 65 if they invest in stocks at least 10% of their annual income. Employer contributions and social security payments may reduce the need to invest 10% percent of the investor’s annual income. In our analysis we have ignored transactions costs, taxes and the possibility that the investor lacks the discipline to adhere to his or her 40 year investment plan. These and other real world complications, such as sickness, loss of employment, family problems, all contribute to lowering accumulations and thus lowering probabilities of achieving one’s retirement goals. On the positive side, many individuals after they pay off their mortgages and also pay fully or partially for their children’s education, save for retirement substantial amounts during the last decade of their working lives. As academic research and the practice of managing individual retirement accounts grow the findings discussed here will be further revised and several new guidelines will be proposed to help individuals manage their retirement portfolios more rationally.