Six FAQ On Risk Parity

A lot of the criticisms of Risk Parity that I come across share similar themes, some of which are based on misunderstandings of what most Risk Parity providers actually do.

These FAQ are designed to address some of the more common themes that we see and hear in the industry press

1.       Is Risk Parity a bubble?

Let’s think about the generally accepted definition of a bubble : “trade in an asset at prices well above intrinsic value”. Experience shows that often, speculation (buying an asset in the hope of relatively quickly selling it on at a profit) is at the heart of bubbles. We can think of the market for South Sea stocks in the 1700s, Florida Real Estate in the 2000s or Technology stocks in the late 1990s as examples that all fit this description.

Risk Parity does not. An investor does not buy a Risk Parity “asset” with the expectation of selling it on at a profit.

Indeed, Risk Parity is not an asset itself, merely a method of allocating between some of the largest and most liquid asset markets in the world. Could Risk Parity strategies cause a bubble in one of these markets? The sheer size of these markets both in terms of stock and flow compared to the size of Risk Parity strategy holdings (exposures to US Treasuries held in Risk Parity mandates represent less than 1% of the total market for US Treasuries) makes this very unlikely until the assets in Risk Parity strategies are much larger.

2.        Recent past history won’t be repeated, does this make Risk Parity a bad idea?

Global Fixed income markets, one big pillar of a risk parity approach have seen an incredible low-volatility rally over the last 10 years, which has pushed Risk Parity strategies to exceptional risk-adjusted returns, often with Sharpe ratios exceeding 1.

It would be foolish to expect this to be repeated exactly. However several long term Risk Parity simulations (e.g. AQR, Redington) across times when fixed income markets did not perform as well supports a long term Sharpe ratio of 0.4-0.5. This is still substantially better than that achieved by equities, or a traditional fixed weight asset allocation.

On a forward looking basis we would expect Risk Parity strategies to have a Sharpe ratio close to 0.5 over the medium to long-term, making them very attractive for an investor with a similar timeframe for investment (i.e. most investors).

3.       Doesn’t Risk Parity involve leveraging credit and illiquid assets?

Most Risk Parity implementations involve the most liquid asset markets, such as equities, bonds and commodities. Leveraged exposure to illiquid assets should indeed be avoided. The presence of credit, which demonstrates variable levels of liquidity, needs careful thought and attentive risk management. On this front, some of the larger Risk Parity managers, whom have reached their capacity limits in terms of credit, have prudently decided to close those strategies. As a result, the majority of Risk Parity strategies currently open to investors do not contain credit exposure.

4.        Does Risk Parity involve leverage – and doesn’t this make it risky?

The crisis of 2008-9 had excess leverage in the system at its heart, and it was the unwinding of this leverage that contributed to and exacerbated the crisis. Naturally, this should be avoided in the future. Risk Parity, in most implementations, does involve explicit financial leverage (through the use of futures, however, and not through direct borrowing). It is important though to understand the economic equivalence of this leverage, and the flaws in looking at it through only that lens:

  • An allocation of 150% of an investor’s portfolio to 10 year Treasury Futures clearly has more explicit leverage than a 75% allocation to 30 year bonds, but the economic risk to interest rate moves  is roughly the same. Looking solely at the leverage is not a good way to compare the risks of these two positions
  • Though equities are often viewed as “unlevered”, as a company typically takes on debt to finance itself, equities can be seen to be a levered investment in the underlying assets of the company. This means the leverage is “under the hood” but it is nevertheless there.
  • Thus a “traditional” unlevered allocation between stocks and bonds can contain implicit leverage, and indeed it can be shown that on average a Risk Parity portfolio contains less total leverage (implicit plus explicit) than a traditional portfolio. When a Risk Parity strategy does take more leverage, it does so in a dynamic way which responds to market conditions both in terms of increasing and decreasing the amount of leverage

5.       Surely Risk Parity doesn’t work in low interest rate environments?

Experience in Japan shows this not to be the case. The 10 year JBG yield stood at 0.8% at the end of 2012, a very similar level to where it was in 1998, but a long position has delivered a substantial risk adjusted return over this time period by rolling down an upward sloping yield curve. Further, interest rates must rise by more than that implied by the yield curve for the fixed income component to deliver a negative capital return (it earns interest income on top of this).

The times when Risk Parity is most vulnerable to negative returns, are sudden unexpected moves in the underlying asset classes, such as the surprise Fed tightening in 1994. In these cases, there is no chance for a Risk Parity strategy to reduce its exposures.

6.        Isn’t Risk Parity a disaster in the 1970’s environment of sharply rising inflation expectations?

Several studies have sought to quantify the returns that a typical Risk Parity strategy would have experienced in the 1970s. The results do vary according to the exact implementation of Risk Parity that is used, and particularly whether it includes commodities or not.

The 1970s was a time of rising interest rates, and rising expected and realised inflation.

Most quantitative studies agree that there were periods of time when Risk Parity lost money (to be expected in certain scenarios), and also where Risk Parity delivered a negative real return – which was the case for most assets in the face of such high inflation. Studies that include a commodity component in the Risk Parity portfolio generally conclude that the Risk Parity portfolio significantly outperforms a fixed weight portfolio over these periods of time. The commodity component’s correlation with inflation allowed this result to occur (driven largely by the US abandoning the gold standard and the resultant feedback into the commodity complex including oil and gold).

Most quantitative empirical studies that attempt to make a fair representation of real Risk Parity portfolios, agree that over long periods of time, which capture different fixed-income cycles, a Risk Parity strategy would have delivered a better risk-adjusted return than equities, or than a fixed-weight allocation between asset classes.

2 thoughts on “Six FAQ On Risk Parity”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s