Following on from the previous post which discussed the unique characteristics of the last 20 years and the impact of this on the returns from balanced investment vehicles, we can now look into the return dynamics more closely and in particular the outsize bond return boost post 2007.
The following chart shows the capital return only of a portfolio with a starting split of 40% (bond)/60% (equity) and no rebalancing, and of a portfolio with rebalancing back to starting allocations with a 5% rebalancing corridor – there are no transaction or management costs or taxation. The source of the data is the Shiller/Yale US market data.
Against these is also shown the capital return on the equity market – this is only meant as a reference point since the yield differential between the two other options would invalidate a correct return comparison.
Now, relative to the non rebalanced portfolio, we can see the rebalanced portfolio underperforming during the bull market of the 1990s, out performing during the bear market, under performing during the mid 1990s stock market recovery and roaring away, post 2007, during the extremely low interest rate conditions of a QE influenced market place.
Indeed, it is also clear from the chart just how supercharged a portfolio, with a significant bond allocation, became during the latter period, especially in QE markets with falling ever lower bond yields. It is this component of the historic return which many commentators are really referring to as evidence of balanced portfolio security that is a concern to me. This is not typical.
The following chart shows the annualised return differential between a rebalanced and a non rebalanced portfolio as a % return relative to the non rebalanced portfolio:
So, at the top of the stock market in 2007, there would have been no benefit from rebalancing within the corridors noted, given the starting point. All the rebalancing benefit has occurred post 2007. So we can see that with no dividends or coupon reinvestment that the benefits of rebalancing appear to be cyclical and dependent on the degree of outperformance of bonds and equities and the starting point. Likewise, the costs of rebalancing need to be taken into consideration. Extremes create both the opportunity for rebalancing returns and the opportunity for rebalancing losses.
Let us look at the reinvestment return differential: that is the differential return between a non rebalanced starting 40/60 portfolio and reinvestment and a rebalanced portfolio with reinvestment:
Reinvesting coupons in a bull market hinders the rebalancing portfolio, but in deep bear markets within a secular bull phase for fixed interest stocks, coupon reinvestment enhances the return.
In this post I have looked specifically at portfolios with no reinvestment of dividends and interest for two reasons:
The first is that many who purchase balanced vehicles or who have balanced portfolios are using the income to meet expenditure needs. They are not reinvesting.
The second is because in truth the actual valuation of assets, including the income streams from them, is the capital value of the assets themselves. Some may find this difficult to comprehend – the capital return is a market representative valuation of an asset class but the total return with dividends and income reinvested is not a representative economic valuation of assets, since it is only a specific return for one segment of the market place.