Fixed Index Annuities: The Importance of Diversification
November 15, 2018
Investors rely on annuities to deliver steady returns with low volatility. We believe it is therefore prudent for annuity designers to take advantage of every opportunity to generate those types of returns. Today, many annuities may be limited in their ability to invest broadly. Indeed some may be over-allocated to singlemarket risk, such as U.S. equities.
For Fixed Index Annuities (“FIAs”) a key consideration is the underlying index choice. With a well-designed index, an insurance company can seek to maximize returns for policyholders and add attractive features to the annuity product itself. Diversification can play a key role in index design.
At MSIM we approach designing an index the same way we approach constructing portfolios of large pools of institutional capital. We focus on:
There are a number of different ways to increase diversification: include more asset classes, include different types of investments within asset classes, and include more geographies. In this article we will focus on geographic diversification, both empirically and from a forward-looking perspective. What are the potential benefits of geographic diversification?
In order to achieve portfolio efficiency it is important that individual assets are not highly correlated. When investments behave differently, they can smooth out portfolio volatility. As Display 1 illustrates, correlations across geographies are fairly low, which can help with diversification and portfolio efficiency. While correlations are not always stable, over time this diversification can help
Another benefit of geographic diversification is that it allows for a bigger set of potential investment opportunities. As illustrated in Display 2, more than half of the world’s investable markets are outside the US. At a more granular level, for example, the non-U.S. small cap universe is double the size of the US small cap universe.1 This wider breadth of investments may provide potentially rewarding investment opportunities.
It is a common misconception that non-U.S. equities exhibit higher betas in market sell-offs. As the U.S. equity market is regarded as a higher-quality market, it is expected to sell off less severely in a broad market decline. However, this is not always the case. (Display 3) In fact, non-U.S. equities, as measured by the ACWI ex US Index, can exhibit lower downside beta than U.S. equities. The downside beta of U.S. equities is now at post-crisis highs while that of non-U.S. equities is at post-crisis lows.
It could be beneficial to allow indices to reallocate over time, bearing in mind that U.S. and non-U.S. stocks have historically alternated periods of outperformance, as shown in Display 4. This makes sense because each stock market is influenced by the economic cycle and growth trajectory of its country. For instance, since 2008, the U.S. has performed remarkably well. Yet that hasn’t always been the case.
Indeed, considering the poor performance of U.S. equity markets compared to other asset classes from 2000-2010 highlights the danger of over-emphasizing any one market or asset class. (Display 5)
This is also a reminder not to make assumptions based solely on the recent past. It is widely understood that investors are vulnerable to “recency bias,” meaning that they tend to place outsized importance on what has taken place most recently, underestimating what has happened in the past. In other words, investors tend to “chase returns” or invest where they’ve observed recent historical outperformance. Today, this would lead to being overweight in U.S. equities. While recent back-testing might support such an investment decision, we believe that incorporating forward-looking return and risk assumptions might lead to a more diversified portfolio and improved long-term performance.
Display 6 compares a “base” portfolio of 60% U.S. equities / 40% U.S. bonds to a portfolio that comprises 30% non-US equities, 30% U.S. equities and 40% U.S. bonds. It shows that adding non-U.S. exposure has historically kept portfolio efficiency (as measured by Sharpe ratio) more or less in-line with a U.S.-only portfolio, in some cases improving it. Should an index have the ability to rotate amongst asset classes or geographies, we believe there is greater potential to improve performance results over the long term.
Lastly, there are a number of reasons why having the ability to diversify may be useful, particularly when considered in the context of where U.S. markets are currently priced and where the U.S. is in its economic cycle. As illustrated in Display 7, Morgan Stanley research suggests there is a 70% chance that the U.S. will move from expansion to downturn in the next 12 months. This information should be considered when determining optimal portfolio allocation. Furthermore, forward-looking consensus return expectations, including those of Morgan Stanley research, suggest attractive returns for non-U.S. equities, which further reinforces the potential benefit of being able to allocate to non- U.S. equities.
In our view, any well-designed index should balance a respect for historical data with a sensible forward-looking perspective that facilitates responsiveness to changing market dynamics. We think geographic diversification is a prudent and timely consideration. MSIM’s approach to Fixed Index Annuity design incorporates the flexibility to include both non-U.S. equities and bonds for all of the reasons we’ve discussed here. Over time we believe having the opportunity to diversify allocations geographically may help smooth out volatility and enhance returns.