Capital for portfolios constructed using the core/satellite principle is invested in two parts: The larger part is invested in a core investment designed to deliver the market return. As a rule, this is accomplished using an equity index. The core investment is complemented by a number of smaller – satellite – investments. These can carry a greater risk and are selected in such a way that the market return from the core investment is enhanced with attractive outperformance opportunities.

The results of this structure are twofold: On the one hand, broadly diversified equity indices are particularly suitable for the core investment. On the other, the costs of implementing the core investment play a decisive role. As this passive approach forgoes the possibility of outperforming the market from the outset, it is important to capture this return in the portfolio at the lowest possible cost. Since broad diversification, high tracking accuracy and low costs are among the traditional strengths of exchange traded funds (ETFs), many investors favor this form of investment for their core investment.

Source: MCSI, UBS Asset Management

The key: optimized risk distribution

Central to this concept, however, is the distribution of risk within the portfolio: Given that passive asset components are used for the greater part of the invested capital, the risk of deviating from the defined benchmark is virtually negligible for the bulk of the portfolio. As a result, the portion of the portfolio designed to outperform the market return can accept a greater risk without significantly increasing the overall portfolio risk.

This brings into sharper relief the unique strength of the core/satellite approach: Its strength lies not in the distribution of capital across various investments but rather in its balanced risk allocation. The risk, i.e. possible deviation from a defined total return, is allocated to those portfolio components designed to ensure an outperformance over the market while at the same time virtually no risk is "consumed" for the core investment, which delivers the market return.

Even assuming that an investor's own portfolio is never – or seldom – actively rebalanced, individual asset class weightings change over time, if for no other reason than the varying performances across individual asset classes owing to financial market fluctuations alone. This carries the risk that the portfolio composition is no longer aligned with the original strategy; as a result, the investor is suddenly exposed to greater risk.

As a large number of major investors such as pension funds and public utilities must satisfy defined performance requirements, however, they cannot afford to deviate from their strategy. The question that not only they but also private investors face is how to remain true to a portfolio strategy without having to constantly rebalance.

Calendar rebalancing – simple and efficient

One concept that is particularly easy to implement is calendar-based portfolio rebalancing. At the end of a specified period, the portfolio is rebalanced back to the originally desired weighting – irrespective of the magnitude of the actual deviation. What makes this concept so appealing is its simplicity.

A comparison between a portfolio consisting of 50% stocks and 50% bonds whose weighting remains unchanged and an identical portfolio whose weighting is rebalanced once a year demonstrates that this simple rebalancing concept can prevent not only a significant shift in portfolio weighting but also an increase in risk as a result.

Equity weighting: Buy and hold vs. Rebalancing

Change in equity weighting in a 50/50 portfolio (stocks/bonds) during the period 1925–2012, buy and hold vs. annual calendar rebalancing

Source: Pictet equity and fixed income index 1925–1997, extended to 2012, our own calculations; constant real annual returns.

However, this concept works only if rebalancing costs are not excessively high. As ETFs make it possible to add or delete entire markets with just a single transaction, they are particularly well suited to implement a simple rule-based rebalancing concept.