When Raj started planning his investments for his retirement at the age of 45, he realised that he had only 60 years to build his retirement corpus. He was an extremely risk averse investor and didn’t want any of his investments in risky equity linked assets. However, his financial advisor told him that he would need to balance his risk-return preferences in order to build his corpus in such a short time frame. This is where the concept of ‘Asset Allocation’ comes.

Balancing risk and return is the prime goal of any investment strategy. Asset allocation is one mode of implementing this strategy, wherein the choice of investments varies depending on the investors’ risk aptitude, investment timeframe and goals. As different assets perform differently in different markets, it is important to proactively manage asset allocation and regularly review the same. There are many academic studies revolving around asset allocation which bring out different aspects of the subject. One such study by Roger Ibbotson and Paul Kaplan attempted to evaluate if asset allocation explains 40%, 90% or 100% of the portfolio’s performance. 

In 2000, Roger G. Ibbotson and Paul D. Kaplan undertook the study – Does Asset Allocation Policy Explain 40, 90, or 100 Percent of Performance? The study sought to answer three distinct questions:

1. How much of the variability in the returns across time is answered by the asset allocation policy?
2. How much of the variation in returns among funds is explained by the asset allocation policy?
3. What portion of the return level is explained by the asset allocation policy return?

Ibbotson and Kaplan studied the monthly returns of 94 US balanced mutual funds across 10 years. They also studied the quarterly returns of 58 pension funds. A return based style analysis for the 10 years was used for mutual funds, to assign the policy weights for each fund. However, for pension funds, the actual policy weights and the asset class benchmarks for pension funds was used, instead of estimated weights.

Question 1 – Variability across time: Each fund’s total monthly returns were regressed against the corresponding monthly returns of the fund’s estimated policy benchmark. The value for each fund was recorded and studied. It was found that on an average 90% of the variability of returns of a fund across time was due to the asset allocation policy.

Question 2 – Variability among funds: A cross sectional regression analysis of compound annual total returns was used to compare funds with each other. Each fund’s return was compared with other funds’ return. It was found that 40% of the variation of returns from one fund to another was as a result of policy return differences. The remaining 60% of the variation was as a result of variety of other reasons such as asset class timing, security selection, fees and style within asset classes. So as an example, if one fund’s return is 13% and another fund’s return is 8%, on an average only 2% of the difference is explained by the difference in asset allocation. The remaining 3% is due to the other factors mentioned above.

Question 3 – Return level: The compound annual policy return was divided by the compound annual fund return for the entire period. This was done for each fund. On an average, it was found that a little more than 100% of the return level was explained by the asset allocation policy. Therefore, the average fund’s return to the asset allocation policy was about the same as the return to the benchmark for the respective asset class. Any investor, who had the ability to select a better fund manager before he invested, could earn above average returns.

In conclusion, the study showed that asset allocation explains 90% of the variability of a fund’s returns over time and 40% of the variation of returns among funds. Also, asset allocation policy explains a bit more than 100% of the level of returns. The answer to whether asset allocation policy explains 40%, 90% or 100% of performance depends on the way the question is interpreted.