The Risk Parity Portfolio and the Low-Risk Asset Anomaly

Recently a portfolio management method called risk parity has been attracting attention for its high performance. This method levels out portfolios' risk allocations, but very few explanations have been provided for the high performance. One of the exceptions is Asness et al. [1]. Asness et al. [1] links the discounting of lowrisk assets by leverage aversion to risk parity portfolios. But risks and risk allocations should not be regarded as the same. On the other hand, it is also possible to show that investor overconfidence is a discounting factor for low-risk assets. In this case, as the size of risk allocations in the market is directly linked to either overvaluation or undervaluation, we could get a result supporting risk parity portfolios, which suggests that it is desirable to level out risk allocations by constantly comparing them with the market portfolio. Here we seek the more appropriate explanation for risk parity portfolios between leverage aversion and overconfidence. As a result, we choose the latter as the more plausible explanation. First, we summarize the relationship between the respective implications of the two theories and risk parity portfolios. Next, we conduct research on bond markets where we could detect some difference between leverage aversion and overconfidence. Our empirical study shows that risk parity portfolios demand explanations other than leverage aversion.