Asset And Portfolio Managament

Investing in Japan: Unifying the concepts of risk parity and factor investing

Determining a portfolio’s asset allocation is one of the most challenging decisions that investment professionals face. Overallocating to risky assets can result in large losses; whilst under-allocating to risky assets may prevent a portfolio from earning sufficient returns.

Traditional asset allocation

Traditionally, multi-asset portfolios have allocated 60% of their capital to equities and 40% to bonds. The resulting portfolio is often called a ‘balanced portfolio’. In reality, however, these 60/40 portfolios have had an unbalanced risk exposure with as much as 90% of the portfolio’s risk stemming from the equity component. As such, 60/40 portfolios have been particularly susceptible to stock market crashes, large drawdowns and long recovery periods. These eventualities are particularly significant as they have often occurred when investors have needed their capital most.

An alternative approach: A combination of promising concepts

Concept 1: Risk parity 
The risk parity approach addresses the issue of capital-driven allocations, which result in counter-intuitive risk concentrations, by allocating less capital to riskier assets. By weighting asset classes in inverse proportion to their risk, a risk parity approach creates a truly diversified portfolio whereby each asset class has an equal risk contribution to the portfolio. As a result a risk parity strategy can achieve a consistent performance in diverse market conditions by avoiding risk concentrations and not relying on the forecasting of expected returns, which are inherently hard to estimate.

Concept 2: Factor investing
Factor investing observes that a large part of the investment returns of the main asset classes are attributed to a small number of factors. Factor investing systematically identifies and quantitatively isolates these uncorrelated factors in order to harvest their risk premia.

Combining the concepts: A quantitative strategy for investing in Japan
The dynamic risk parity Japan strategy provides an investor with exposure to the Japanese market by combining the concepts of risk parity and factor investing. The strategy was formulated by exploring and evaluating five alternative investment approaches. Rather than focussing on the return of the strategy, we focused on the return per unit of risk taken.

The simulated strategies explore the following investment underlyings: the Nikkei Total Return Index (Nikkei), 10-year Japanese government bond (JGB) futures, and a dynamic basket of 30 low volatility stocks selected from the TOPIX Index (TOPIX).

The strategies that we considered were as follows:

  1. Buy and hold (Nikkei; 100%)
  2. Invest in a traditional 60/40 ‘balanced portfolio’ (Nikkei and JGB; 60%/40% capital weights)
  3. Invest according to risk parity principles (Nikkei and JGB; 50%/50% risk weights)
  4. Invest according to a combination of risk parity principles and factor investing (TOPIX and JGB; 50%/50% risk weights)
  5. Invest in the dynamic risk parity Japan strategy. This strategy is a combination of risk parity principles, but where the risk weights vary according to the current market cycle, and factor investing (TOPIX and JGB; dynamic risk weights)

Chart 1 and Table 1 show that the benefits of a traditional 60/40 ‘balanced portfolio’ over a pure buy-and-hold, equity-only strategy are evident.

Chart 1: Simulated past performance of a buy-and-hold strategy compared to a traditional 60/40 strategy
Chart 1: Simulated past performance of a buy-and-hold strategy compared to a traditional 60/40 strategy
Source: Bloomberg, Commerzbank as of April 2017
Table 1: Simulated past performance of a buy-and-hold strategy compared to a traditional 60/40 strategy

Buy and hold

Traditional 60/40

Annualised return

8.1%

6.0%

Annualised volatility

24.7%

14.1%

Max. drawdown

60.4%

40.1%

Annualised return/annualised volatility

0.33

0.43

Calmar ratio

0.13

0.15

Source: Bloomberg, Commerzbank as of April 2017

Backtested or hypothetical simulated returns are not indicative of future results. No assurance can be given that any financial instrument or strategy described herein would yield a favourable investment result.

However, the results highlight that a risk parity approach, where risk is distributed across asset classes, further improves risk-adjusted returns (see Chart 2 and Table 2). Moreover, results are enhanced even further by incorporating factor investing; in this case investing in a low volatility factor portfolio rather than the broad market index.

Chart 2: Simulated past performance of a risk parity strategy compared to a risk parity low volatility factor strategy
Chart 2: Simulated past performance of a risk parity strategy compared to a risk parity low volatility factor strategy
Source: Bloomberg, Commerzbank as of April 2017
Table 2: Simulated past performance of a risk parity strategy compared to a risk parity low volatility factor strategy

Risk parity

Risk parity low volatility factor

Annualised return

2.2%

2.7%

Annualised volatility

3.1%

2.9%

Max. drawdown

10.0%

10.4%

Annualised return/annualised volatility

0.7

0.93

Calmar ratio

0.22

0.26

Source: Bloomberg, Commerzbank as of April 2017

Backtested or hypothetical simulated returns are not indicative of future results. No assurance can be given that any financial instrument or strategy described herein would yield a favourable investment result.

In-house research indicated that dynamic asset allocation would further improve the performance of our strategy. This led to the introduction of a timing mechanism, which observes the prevailing market conditions and reacts accordingly. The aim of the timing mechanism is to ensure that an investor’s exposure is tilted towards the right asset class at the appropriate part of the market cycle. Of the strategies explored, the combination of risk parity and factor investing produces the highest risk-adjusted returns (see Chart 3 and Table 3).

Chart 3: Simulated past performance of the dynamic risk parity Japan strategy
Chart 3: Simulated past performance of the dynamic risk parity Japan strategy
Source: Bloomberg, Commerzbank as of April 2017
Table 3: Simulated past performance of the dynamic risk parity Japan strategy

Dynamic risk parity Japan

Annualised return

3.2%

Annualised volatility

3.1%

Max. drawdown

8.0%

Annualised return/annualised volatility

1.02

Calmar ratio

0.40

Source: Bloomberg, Commerzbank as of April 2017

Backtested or hypothetical simulated returns are not indicative of future results. No assurance can be given that any financial instrument or strategy described herein would yield a favourable investment result.

It is of paramount importance that an investor considers returns in risk-adjusted terms rather than in isolation, as the return on any investment must be considered relative to the amount of risk taken. More specifically, the higher return-tovolatility ratio offered by the dynamic risk parity Japan strategy allows an investor to achieve the same return with a lower risk, or a greater return for the same risk, than a traditional 60/40 portfolio.